Ice-Ice and Ice-Steel Friction in Field and in Laboratory

Abstract

Friction is one of the key phenomena during the interaction of ice with offshore structures and ships, ice rafting and ridging processes, and the brittle compressive failure of ice. Therefore, knowledge of the ice friction laws and a better understanding of this phenomenon are needed for the design of safe and reliable offshore structures and successful operations in ice-infested waters. This thesis presents experimental studies on ice friction in the field and in the laboratory and can be divided into three parts: field study of ice friction and the effects of different parameters on the kinetic and static friction coefficients;investigation of geometrical stick-slip interaction between ice and stainless steel in small-scale laboratory tests;field investigation of the vertical ice forces exerted on isolated vertical piles frozen in a level sea due to changes in the water level. The first version of the experimental setup for the field tests to study ice friction was relatively simple and was used only in 2010. Significant modifications and improvements were made to the setup in 2011 that allowed us to better control the test conditions (i.e., velocity, normal load). The available range of sliding velocities (between 6 mm s-1 and 43 mm s-1) did not cover the whole range of our interest. Therefore, a new pulling mechanism was introduced in the tests in 2012 that allowed us to extend the sliding range up to 110 mm s-1.

The field experiments were conducted with first year sea ice in the Barents Sea and in fjords at Spitsbergen during three springs (2010-2012). In September of 2012, the ice friction tests were performed with multiyear sea ice northeast of Greenland. Most of the tests were performed to investigate the friction between sea ice and sea ice, while the rest were to study the friction between sea ice and corroded steel. The aim of the study was to survey the most important factors that affect ice friction in field conditions and to determine whether the existing friction models correctly predict the dependences observed in the field tests. The effects of the sliding velocity (6 mm s-1to 105 mm s-1), air temperatures (-2°C to -20°C), normal load (300 N to 2000 N), presence of sea water in the interface, and ice grain orientation with respect to the sliding direction on the friction coefficient were investigated. The effect of the hold time on the static friction coefficient was also studied. The test campaigns showed that ice surface roughness is likely to be the most important parameter in determining the friction coefficient. Repeated sliding over the same track led to surface polishing and decreased the kinetic friction coefficient from 0.48 to 0.05. When sliding occurs between unsmoothed surfaces, the friction coefficient was found to be independent of the sliding velocity. As the contacting surfaces become smoother, the kinetic friction coefficient begins to depend on the velocity, as predicted by existing ice friction models. Some attempts were made to characterise ice surface irregularities and real contact area using two techniques: 1) the production of an ice surface cast and its further analysis using an optical microscope and 2) measurements of the pressure distribution and real contact area using tactile sensors. The static friction coefficient increases logarithmically with the hold time and changes from approximately 0.6 at 5 s to 1.26 at 960 s.

The small-scale laboratory studies on the stick-slip interaction between ice and stainless steel were performed at the University Centre in Svalbard (UNIS). A specially designed device allowed us to study the effect of well controlled steel surface roughness on the interaction between ice and steel under various test conditions. The effects of the relative sliding rate (1.67 × 10-6 m s-1 to 0.83 × 10-3 m s-1), temperature (-5°C to -25°C) and applied constant pushing force were also investigated. Stick-slip interaction was always observed in the tests on samples with an arithmetic average roughness (Ra) between 2 μm and 15 μm. Both steady state sliding and stick-slip were observed in the tests on samples with the lowest (Ra = 2 μm) and the highest (Ra = 25.2 μm) roughness. The slip distance was found to be equal to the mean pitch profile (mean profile wavelength) for all velocities studied. The elasticity of the ice was found to be an important factor in the stick-slip interaction.

The field studies of the vertical ice forces exerted on isolated vertical piles frozen in the ice were performed in Svea, Spitsbergen, in March of 2010. Four different piles made of steel and aluminium were used in the experiments. The piles were pushed through the ice using a hydraulic jack. The first peak load measured in the tests is associated with the onset of the ice-pile relative movement. Because we always observed failure at or very near the ice-pile interface, the first peak load can be treated as the strength of adhesion or the cohesive failure of ice in the vicinity of the interface. Elastic plate theory was used to estimate theoretically the vertical forces exerted on a pile.