Towards a generic protocol for measuring the constituent particle size distribution of E171 in food by electron microscopy

Abstract

Food additive E171 (titanium dioxide, TiO 2 ) is authorized to be used as food colorant in the European Union. E171 is a particulate material, containing a fraction of nanoparticles. A major challenge for food control laboratories is the lack of harmonized, validated analytical methods for measuring the physicochemical properties of nanoparticles in food. This study proposes a generic method for the determination of the number-based size distribution of the constituent particles of E171 in food products based on transmission electron microscopy (TEM). As proof of concept, the protocol was evaluated on 20 food products containing E171 and one personal care product containing titanium dioxide. The total titanium concentration in the products measured by inductively coupled plasma - mass spectrometry (ICP-MS) ranged from 0.1 to 4.0 g/kg, corresponding to 0.2–6.7 g TiO 2 /kg. The required initial mass of starting product needed for TEM analysis was estimated from the ICP-MS concentration. Sample preparation included matrix digestion, a first centrifugation step to remove the remaining food matrix, sonication to reduce agglomeration, and a second centrifugation step to concentrate the sample. Descriptive TEM analysis showed that five food products, labelled to contain E171, contained smaller rutile TiO 2 particles (20–40 nm) coated with mica, which are potassium aluminium silicate-based pearlescent pigments. For 14 other food products and the personal care product, the protocol resulted in a homogenous distribution of the TiO 2 particles over the grid surface. In general, variation between products in the number of particles on the grids was limited; on average, 672 ± 173 particles were measured in 10 images. The median of the minimum Feret diameter distribution varied from 72 ± 7 nm to 135 ± 14 nm. The percentage of nanoparticles (minimum Feret diameter smaller than 100 nm) ranged from 23% to 82%. The corresponding mass percentages of these fractions ranged from 3% to 41%. The percentages of E171 particles smaller than 30 nm was very low (3% or less) while corresponding mass fractions were negligible. For only one food product, the TiO 2 concentration was too low to apply the protocol without adaptations. • A generic protocol for particle size measurement of E 171 in food is proposed. • The protocol is designed for transmission electron microscopy. • The protocol was evaluated on 20 food products and one personal care product. • The median Fmin varied from 72 ± 7 nm to 135 ± 14 nm. • The percentage of nanoparticles (<100 nm) ranged from 23% to 82%.