Toward Anti-icing and De-icing Surfaces : Effects of Surface Topography and Temperature

Abstract: Icing severely affects society, especially in the Nordic countries. Iceaccumulation can result in critical performance problems and safetyconcerns for instance in road, air and sea transportation, transmissionlines, marine and offshore structures, wind turbines and heat exchangers.Present active ice-combating approaches possess environmental,efficiency and cost drawbacks. Thus, fabricating icephobic surfaces orcoatings impeding ice formation (anti-icing), but facilitating ice removal(de-icing) is desired. However, different conditions in the environmentduring ice formation and growth add to the complexity of the problem.An icephobic surface that works for a certain application might not be agood candidate for another. These surfaces and the challenges are infocus in this thesis.Wetting properties are important for ice formation on surfaces fromthe liquid phase (often supercooled water), where the water repellency ofthe surfaces could enhance their anti-icing effect. Considering this,different hydrophobic and superhydrophobic surfaces with differentchemistry, morphology and roughness scale were prepared. Since anyinduced wetting state hysteresis on hydrophobic surfaces could influencetheir performance, the wetting stability was investigated. In particulardynamic wetting studies of the hydrophobic surfaces revealed whatsurface characteristics benefit a stable wetting performance. Further, theeffect of temperature, particularly sub-zero temperatures, on the wettingstate of flat and nanostructured hydrophobic surfaces was investigated.This was complemented with studies of the wetting stability of sessilewater droplets on flat to micro- and multi-scale (micro-nano) roughhydrophobic samples in a freeze-thaw cycle. To be consistent with mostapplications, all temperature-controlled experiments were performed inan environmental condition facilitating frost formation. Further, antiicingproperties of hydrophobic surfaces with different topography butsimilar chemistry were studied by freezing delay measurements.A dynamic wetting study using hydrophobic samples with similarchemistry but different topography revealed that multi-scale roughnesscould benefit the wetting stability. However, when these surfaces areutilized at low temperatures the wetting hysteresis observed during acooling/heating cycle is significant. Such a temperature-inducedhysteresis is also significant on superhydrophobic surfaces. I attributethis to condensation followed by frost formation facilitating spreading of the supercooled water droplet. The freezing delay measurementsdemonstrate no significant effect of surface topography on anti-icingproperties of hydrophobic surfaces, however the flat surfaces showed thelongest delay. These findings are in agreement with heterogeneous icenucleation theory, suggesting preferential ice nucleation in concave sites,provided they are wetted.In the second part of this thesis, I consider the findings from theprevious part illustrating the limitations of (super)hydrophobic surfaces.The de-icing properties of hydrophilic surfaces with a hydration waterlayer, hypothesized to lubricate the interface with ice, were studied. Heretemperature-controlled shear ice adhesion measurements, down to -25oC, were performed on an adsorbed layer of a polymer, either bottle-brushstructured poly(ethylene oxide) or linear poly(ethylene oxide). The iceadhesion strength was reduced significantly on the bottle-brushstructured polymer layer, specifically at temperatures above -15 oC,whereas less adhesion reduction was observed on the layer formed by thelinear polymer. These findings are consistent with differential scanningcalorimetry (DSC) data, demonstrating that the hydration water, boundto the bottle-brush structured polymer, is in the liquid state at thetemperatures where de-icing benefit is observed. Further, continuingwith the hypothesis of the advantage of surfaces with a natural lubricantlayer for de-icing targets, I studied shear ice adhesion on the molecularlyflat basal plane of hydrophilic mica down to -35 oC. Interestingly, ultralowice adhesion strength was measured on this surface. I relate this to theproposed distinct structure of the first ice-like but fluid water layer onmica, with no free OH groups, followed by more bulk liquid-like layers.This combined with the molecularly smooth nature of mica results in aperfect plane for ice sliding.