Lunn, Geoffrey David
(2018)
Wire flame sprayed aluminium coatings onto polymer and CFRP substrates.
PhD thesis, University of Nottingham.
Abstract
Fibre reinforced polymer matrix composites (FRPs) are used in an increasing number of applications because of their attractive mechanical properties such as high specific strength and stiffness. However, for a number of applications, enhanced surface properties such as thermal and electrical conductivity and resistance to wear are required. Thermal spray coatings have the potential to be a technically suitable and cost effective means of achieving these surface properties. This thesis is a systematic study of the deposition of wire flame sprayed (WFS) pure aluminium onto two different carbon fibre reinforced polymer (CFRP) substrates; one epoxy and one PEEK based as well as unreinforced epoxy, PEEK and steel substrates.
The work investigates the following aspects: (i) characterisation of the flame spray plume; (ii) formation of splats on polymer and CFRP; (iii) how surface preparation (grit blasting), affects the surface features of PEEK and epoxy polymers and their respective CFRPs; (iv) the development of thick wire flame sprayed aluminium coatings deposited onto FRPs and the resultant adhesion between coating and substrate.
Particle diagnostic techniques as well as physical collection of in-flight particles were used to characterise the WFS plume. In order for the diagnostic system to detect the particles, external lighting was required, indicating that the particles were at temperatures of <1000°C. The particles detected by the system were shown to be travelling at high velocities ≈ 250 – 300 m/s, 50 – 200 m/s faster than most previous literature would suggest. Collection of the particles from the plume indicated that the particle diameters were log-normally distributed with a mean diameter of ≈ 25 µm. However, the particle diagnostic system could not detect particles ⪅45 µm, and thus only the largest particles in the plume could be characterised in terms of size and velocity. Overall, wire flame spraying generates a surprisingly stable particle stream.
Wipe tests, where isolated particles are deposited onto a substrate, were performed. The different splat characteristics on polymer, CFRP and metallic substrates were compared in both a qualitative and semi-quantitative manner. The splat morphology showed that the particles deposited onto the substrates were molten on impact, and thus > 660°C. Splats deposited onto the polymer and CFRP substrates had large bulbous regions at their centre, due to gas evolution at the substrate interface and this limited the effective bonding area. Is postulated that bubbles could explode and formed ring shaped splats. This latter splat type was more prevalent on the epoxy based substrates, indicating a greater tendency for gas evolution on these substrates.
The different substrates were subjected to either a light or heavy grit-blast (LGB or HGB), and the surfaces characterised. The epoxy substrate was only slightly roughened and the damage induced was indicative of a brittle substrate, with material loss due to crack propagation. In contrast, grit-blasting of the PEEK substrate significantly roughened the surface via extensive plastic deformation. Finally, the CFRP substrates were extensively roughened by the grit-blasting process, exposing fibres and creating a distinctive bi-modal height distribution on the surface with deep valleys created due to the extensive loss of carbon fibres and high resin ridges left due to the resilience of the matrix.
WFS aluminium coatings were sprayed onto the different substrates with smooth, LGB and HGB surfaces. These experiments enabled pull-off adhesion tests to be conducted which showed clear differences between different sample types and surface preparations with results ranging from 1 MPa for as received CF-epoxy to 17.5 MPa for lightly grit-blasted CF-PEEK. The thermoplastic PEEK substrates typically had greater adhesion than the epoxy based ones and well adhered coatings could be deposited even onto smooth PEEK based substrates. The fracture face of substrates obtained from pull off tests showed structures indicative of local melting and flow in the case of thermoplastic PEEK which may account for the greater adhesion. Sectioned samples showed that there was extensive local interfacial cracking when fibres came into contact with the coating, indicating that the polymer phase has better coating adhesion than the fibre phase. Despite this, the FRP substrates typically had better coating adhesion than the polymer ones. It is believed that this is due to the significantly rougher surface of the FRP substrates.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)