Cr(VI)-free pre-treatments for adhesive bonding of aerospace aluminium alloys

For more than six decades, chromic acid anodizing (CAA) has been the central process in the surface pre-treatment of aluminium for adhesively bonded aircraft structures in Europe. Unfortunately, this electrolyte contains hexavalent chromium (Cr(VI)), a compound known for its toxicity and carcinogenic properties. The approaching ban on the use of hexavalent chromium (Cr(VI)) makes its elimination a high-priority R&D topic within the aerospace industry and the Cr(VI)-era will soon have to come to an end. Anodizing aluminium in acid electrolytes produces a self-ordered porous oxide layer with a thin barrier layer underneath. This special type of oxide readily adheres to the organic resin and provides protection against corrosion. Although Cr(VI)-free candidates such as sulphuric acid- (SAA), phosphoric acid- (PAA) and mixtures of phosphoric-sulphuric acid anodizing (PSA) can be used to create this type of structure, the excellent adhesion and corrosion resistance that is currently achieved by the Cr(VI)-based process is not easily matched. To gain a better understanding of the underlying physical and chemical mechanisms that contribute to the adhesion and durability in these structures, this study investigates the correlation between the oxide’s chemical and morphological characteristics, as influenced by the anodizing electrolyte, and bond performance. The major challenge in the mechanistic understanding of the adhesion in bonded components is to differentiate between the different forces acting at the oxide/resin interface. In the first part of this PhD thesis, studies focus on the role of surface chemistry. To exclude the contribution of mechanical interlocking between the oxide and the resin, featureless oxides were prepared by stopping the anodizing during the formation of the barrier layer. Surface characterization of the different anodic oxides by means of Fourier transform infrared (FTIR) and X-ray Photoelectron Spectroscopy (XPS) revealed no significant net change in the acid-base properties of the different anodic oxides. It was found that local chemical changes were introduced due to the incorporation of electrolyte-driven anions. Therefore, a model was developed to quantify the relative amounts of O2-, OH−, PO4 3−, and SO4 2−, showing significant changes in the type and amount of surface species. Consequently, measurements showed that the pretreatments and the molecule type affected oxide/molecule interfacial interactions. To evaluate the contribution of adsorptive interaction in practice, peel tests were performed on featureless oxides bonded with commercial aerospace adhesives. Results showed that significant initial dry adhesion is achieved with FM 73 epoxy without mechanical interlocking, and independent of the type of pretreatment. However, the formed bonding was not water resistant, with the amount of applied stress needed for peeling linearly increasing with the amount of surface hydroxyls. Moreover, the application of a thin γ-APS silane layer before bonding with epoxy has confirmed that the stability of the interface is also determined by the nature of the bond, showing much more stable interfaces in the presence of covalent interactions. When peel tests were performed with a phenolic-based adhesive (Redux 775), no correlation to the surface chemistry was found. Nevertheless, the bonded joints on the basis of the weakly acidic character of the phenolic adhesive showed better resistance to corrosion in salt spray tests, compared to those on the basis of the epoxy adhesive. Therefore, we conclude that both oxide surface- and adhesive chemistries play a role in the formation and long-term stability of the oxide/resin interface. In the second part of this thesis industrial porous oxides were applied. Fundamental investigations show that changing the voltage during anodizing can produce morphological variations across the oxide thickness. The effect of the initial voltage sweeps, however, was limited by the oxide dissolution action of phosphoric acid in PSA, since prolonged anodizing in this electrolyte not only leads to an increase of the pore diameter, but also completely dissolves the upper most part of the oxide. Morphological changes were distinguished between geometrical modifications that affect the pore size and changes in the surface roughness that was caused by extended chemical dissolution at higher anodizing temperatures and/or phosphoric acid concentration. Measured carbon concentration profiles within the pores using high-resolution transmission electron microscopy (TEM) coupled with energy-dispersive X-ray spectroscopy (EDS) indicated that resin penetration is affected by both aspects. Moreover, mechanical performance in peel tests indicates that these parameters, rather than the oxide layer thickness are critical for moisture-resistant adhesion. Both adhesion mechanisms: adsorption and mechanical interlocking seem to contribute to the adhesion in these structural bonds. A higher degree of dissolution during anodizing is beneficial for the adhesion, facilitating a composite-like interphase. Too much dissolution, however, reduces the resistance to bondline corrosion. Overall, the presented results illustrate the need to consider both chemical and morphological changes in the selection of Cr(VI)-free alternatives for structural adhesive bonding.