: Le rouge à lèvres est un produit incontournable du marché des cosmétiques. C’est un produit de formulation complexe et aux fonctionnalités multiples, parmi lesquelles la sensorialité est centrale et dépend principalement des corps gras mis en œuvre. Face à la multiplicité des ingrédients disponibles et aux pratiques de formulation s’appuyant largement sur l’expertise empirique des formulateurs, cette thèse répond à une problématique industrielle de mise en place d’une méthodologie de formulation raisonnée pilotée par la fonctionnalité sensorielle. Mettre en place une telle démarche d’ingénierie sensorielle nécessite d’intégrer la connaissance des liens entre la formulation, la structure et les fonctionnalités du produit. Pour ce faire, nous avons proposé une démarche en sept étapes, s’appuyant sur la réalisation d’une formule simplifiée et réaliste. Ce système a permis de structurer une approche hypothético-déductive concernant le rôle de différentes catégories d’ingrédients sur les propriétés sensorielles du produit. Nous avons ainsi pu montrer que les huiles et les cires ont un rôle majeur sur ces propriétés. Les huiles affectent principalement le glissant, le fondant, l’huileux, la douceur, le collant et la présence sur les lèvres. Les cires, quant à elles, ont une influence sur l’opacité et la quantité de rouge à lèvres déposé. Ce travail s’appuie sur une compréhension approfondie des mécanismes physicochimiques à l’origine de la structure du rouge à lèvres, observée à différentes échelles, et des propriétés résultantes. Enfin, sur deux cas concrets d’ingénierie inverse, nous avons validé cette démarche ainsi que les relations entre les propriétés des ingrédients, la structure et les propriétés mécaniques et sensorielles du rouge à lèvres. / Lipstick is a must-have product of the cosmetics market. It is made from a complex formulation and it has multiple functionalities, among which the sensoriality is central and depends mainly on fats and oils used. Given the multiplicity of available ingredients, formulation practices are largely based on the formulators’ empirical expertise. In this industrial context, this thesis aims at setting up a reasoned formulation methodology driven by the sensory functionality. Implementing such a sensory engineering approach requires integrating knowledge of the relationships between formulation, structure and product functionalities. In order to do this, we proposed a seven-step approach, based on the realization of a simplified and realistic formula. This system allowed to structure a deductive reasoning approach concerning the role of different categories of ingredients in the sensory properties of the product. Thanks to this approach, we have shown that oils and waxes have a major role in these properties. The oils mainly affect the slipperiness, the melting quality, the oiliness, the softness, the stickiness and the presence on the lips. Waxes, on the other hand, have an influence on the opacity and the amount of lipstick deposited. This work is based on a thorough understanding of the physicochemical mechanisms at the origin of the structure of lipsticks, observed at different scales, and of the resulting properties. Finally, we validated this approach, as well as the relationships between the properties of the ingredients, the structure and the mechanical and sensory properties of lipsticks, on two concrete cases of reverse engineering.

Red Lustrous Wheelmade Ware (RLWm ware) transport and storage vessels have been excavated from Late Bronze Age (LBA) sites across the eastern Mediterranean. These distinctive vessels were traded for the valuable commodity they contained so far unidentified. Seventy-three sherds (61 RLWm ware, 12 in local fabrics) and two visible residues were analysed for organic residues using standard lipid extraction techniques. Seven residues from a previous study were re-examined. Gas chromatography-mass spectrometry identified four materials - beeswax, bitumen, fat/oil and resin. Beeswax, found only in vessels from Hittite sites in Turkey, was probably used as a post-firing treatment. Fat/oil, present in some sherds from every site, represents the contents of the vessels and showed many of the characteristics of degraded plant oil. Two examples contained a plant sterol and three yielded ricinoleic acid, a biomarker for castor oil. Gas-chromatography compound-specific isotope ratio mass spectrometry of selected residues excluded dairy products, ruminant animal fats and fish oils as source materials for the fats/oils, while comparison with a small database of modern oils created during this study does not exclude plant oils. Selected samples analysed by high performance liquid chromatography-tandem mass spectrometry did not reveal wine residues. Data on the elemental composition of the fabric collected during another study was re-analysed and compared with data from a further published study, confirming the remarkable consistency of RLWm ware fabric. Volume calculations were also attempted to give an estimate of the capacity of the main vessel forms.

Red Lustrous Wheelmade Ware (RLWm ware) transport and storage vessels have been excavated from Late Bronze Age (LBA) sites across the eastern Mediterranean. These distinctive vessels were traded for the valuable commodity they contained so far unidentified. Seventy-three sherds (61 RLWm ware, 12 in local fabrics) and two visible residues were analysed for organic residues using standard lipid extraction techniques. Seven residues from a previous study were re-examined. Gas chromatography-mass spectrometry identified four materials ¿ beeswax, bitumen, fat/oil and resin. Beeswax, found only in vessels from Hittite sites in Turkey, was probably used as a post-firing treatment. Fat/oil, present in some sherds from every site, represents the contents of the vessels and showed many of the characteristics of degraded plant oil. Two examples contained a plant sterol and three yielded ricinoleic acid, a biomarker for castor oil. Gas-chromatography compound-specific isotope ratio mass spectrometry of selected residues excluded dairy products, ruminant animal fats and fish oils as source materials for the fats/oils, while comparison with a small database of modern oils created during this study does not exclude plant oils. Selected samples analysed by high performance liquid chromatography-tandem mass spectrometry did not reveal wine residues. Data on the elemental composition of the fabric collected during another study was re-analysed and compared with data from a further published study, confirming the remarkable consistency of RLWm ware fabric. Volume calculations were also attempted to give an estimate of the capacity of the main vessel forms. / Arts and Humanities Research Council / Extensive folders of data and appendices which accompany this thesis are not presently available online.