Marzook, Mariam; (2025) Icephobic and multifunctional composite coatings for helicopter rotor blades. Doctoral thesis (Ph.D), UCL (University College London).

Text Marzook_Thesis.pdf - Accepted Version Access restricted to UCL open access staff until 1 July 2026. Download (95MB)

Abstract

Ice formation on helicopter blades can compromise mobility and lead to catastrophic failures. To address this, significant research has been dedicated to developing passive methods for mitigating ice accumulation as energy-efficient alternatives to current techniques. A prominent strategy involves designing superhydrophobic textured surfaces, which repel water through high water contact angles and low surface energy. These surfaces reduce ice adhesion by creating micro- and nanoscale rough textures that minimise liquid/solid contact area, thereby delaying ice formation. However, superhydrophobic surfaces often exhibit enhanced ice adhesion under high humidity and dynamic icing conditions. Additionally, maintaining icephobic properties after mechanical erosive damage poses a significant challenge. To overcome these limitations, smoother elastic materials have emerged as promising alternatives, demonstrating low ice adhesion due to stress concentrators that facilitate ice removal while offering superior mechanical durability. This thesis explores the development of superhydrophobic nanocomposites with micro- and nano-scale features. The surface properties and icephobic performance of these materials were analysed and compared. Durability was assessed through harsh particle impact tests, revealing a loss of superhydrophobic properties following erosion damage. Based on these findings, smoother, tougher materials were designed by modifying one of the initial nanocomposites. These new materials underwent extensive mechanical testing, demonstrating low ice adhesion strengths and superior resistance to erosion by sand impact tests compared to harder base materials and initial nanocomposites. Lastly, a machine learning model was implemented to aid in the discovery of new icephobic materials and surfaces. Using data from published studies and controlled laboratory experiments, the model predicted ice adhesion strength based on wetting properties, surface roughness, and temperature. Leveraging concepts of Patch Learning and Deep Neural Fuzzy Networks, the model accurately predicted ice adhesion strength even with imprecise data, providing a valuable tool for material and surface design.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item