Chemical durability of commercial silicate glasses and the impact of surface treatments

Chemical durability of commercial silicate glasses and the impact of surface treatments

Léa Brunswic* 1, Frédéric Angeli 1,Thibault Charpentier 2, Stéphane Gin 1, Laurent Gautron 3, Eric van Hullebusch 4, Mariona Tarragó 4, Daniel Neuville 4, Xavier Capilla 5, Daniel Coillot 6, Ilyes Ben Kacem 7, Justine Fenech 8, Johann Brunie 9

1 CEA, DES, ISEC, DPME, Université de Montpellier, Marcoule, France
2 NIMBE, CEA, CNRS, Université Paris-Saclay CEA Saclay F-91191 Gif-sur-Yvette, France
3 Laboratoire Géomatériaux et Environnement, Université Gustave Eiffel, 5 Bd Descartes 77454 Champs-sur-Marne, Marne-la-Vallée Cedex 02, France
4 Université Paris Cité, Institut de Physique du Globe de Paris, CNRS, F-75005, Paris, France
5 Fédération du verre et du cristal, 114 rue de la Boétie, 75008 Paris
6 Manufacture Baccarat, 54120 Baccarat, France
7 Arc France, 104 avenue du Général de Gaulle, 62510 Arques, France
8 Pochet du Courval, lieu dit Guimerville, 76340 Hodeng au Bosc, France
9 Pyrex, 85 Allée des Maisons Rouges, 36000 Châteauroux, France

Given the present regulatory framework (REACH, Food Contact Materials EU Directives), understanding the mechanisms of glass alteration is increasingly required to guarantee the safe use of daily glassware over their lifetime. A major comprehensive study of a wide range of commercial glass compositions has been pursued on five types of silicate glasses from four major French glass manufacturers. The investigated glasses, lead crystal glass (fine glassware), soda lime glass (containers for the food and cosmetic industries), borosilicate glass (cooking dishes), barium glass (tableware) and opal crystallized glass (tableware) are all bestseller products. A single alteration procedure was used in this work to offer a wide comparison of these glass articles.

Glass alteration experiments were carried out for 3 years in acetic acid (4 % vol.), the reference medium for food contact alteration, imposing a pH of 2.4, at a temperature of 70 °C. It was demonstrated that although the polymerization degree of the glassy network varied between the different glasses, the rate of hydrolysis, determined from the leaching of Si remained similar for all. However, it appears that the structure of the silicate network is strongly correlated with the leaching of Na, considered as a tracing element of alteration. The altered depths of Na ranged between a few tens of nanometres for borosilicate, the most polymerized glass, to a few microns for lead crystal, less polymerized. The retention of lead was effective after 56 days of alteration and lasted until the end of the experiment, explained by the local recondensation of the glassy network. The alteration mechanisms of opal glass (including NaF, BaF₂ and CaF₂ crystals) were different between powder and slab samples highlighting the heterogeneity of the distribution of crystals depth-wise in glass plates, although the hydrolysis rate of the powder was similar to the one measured in fully vitreous glasses.

Another part of this work focused on the impact of surface treatments of industrial interest on glass durability on commercial treated glass slabs. Five surface treatments were selected, three chemical deposits of oxide layers: SnO₂, TiO₂, and SiO₂ as well as two acidic attacks: SO₂ dealkalization and acid polishing, which were applied on each of the five investigated glass compositions. Glass surface characterizations pointed out some changes with different impacts on the alteration behaviour observed for 1.3 years. The unique data collected shows significant reduction of long-term lead leaching, especially in the case of SO₂ dealkalized lead crystal glass. Beneficial effect towards the retention of Ba in Ba-containing glasses was also determined but these surface treatments showed poor interest for borosilicate, as their mechanism relies on the reduction of the interdiffusion rate, while borosilicate’s alteration is controlled by hydrolysis.

Finally, the impact of chromium, added as a colorant in lead crystal, has been investigated because of the high toxicity of hexavalent chromium for human health. Major changes on the leaching of lead crystal glass induced by the presence of Cr were uncovered, even at very low concentrations (Cr < 0.03 mol%). Increasing Cr concentrations demonstrated a dual beneficial effect on lead crystal chemical durability by reducing the leaching of Pb and increasing the polymerization of the silicate network. Both observations were rationalized by the specific insertion of Cr in certain glass domains. Additionally, chromium was never found under its hexavalent form in the pristine nor altered glasses under experimental uncertainties, nor available in solution.

The confrontation of results from local and global scales allowed establishing a referential for the alteration of commercial glasses in acid medium, unfolding the links between structure and alteration for these compositions and highlighting the effects of surface treatments.