ICE ABRASION ON MARINE CONCRETE STRUCTURES

Abstract

A significant amount of the world’s undiscovered oil- and gas resources are located North of the Arctic Circle. In the light of this, there is currently an increasing interest in developing necessary infrastructure to support sustainable development of these resources. Also increased availability to the Arctic due to reduced amounts of ice for longer periods of the year will ease the access for ship transport and increase the seasonal window for construction activities in these areas. Towards the end of the last millennium, the exploration of oil and gas in the North Sea was the main driving force behind development in concrete technology for use in large offshore concrete structures. Knowledge and expertise derived from this period has benefited in other sectors of the concrete industry as well, such as costal bridges and quay structures where durability is important. Going forward, the ability of concrete structures to withstand exposure from sea ice and low temperatures is decisive and calls for increased knowledge and new technology. Ice abrasion is a severe degradation mechanism responsible for reduced service life in marine concrete structures exposed to sea ice. The phenomenon is caused by ice-concrete friction forces and results in gradual loss of concrete cover. In extreme cases, ice abrasion has brought about complete deterioration of the reinforcement cover of marine concrete structures, which represents a potential threat regarding the safety against load-carrying failure. Current regulations with regard to material selection and design of concrete structures exposed to ice abrasion are defined in ISO 19903:2006 and ISO 19906:2010, where most of the regulations are given as functional descriptions. The interpretation of the regulations is therefore mainly based on considerations without sufficient documentation, which calls for reliable and research-based estimation methods. As a first approach, an investigation was performed in order to develop equations for rough estimation of expected abrasion rates based on the experimental results in the current study. The equations were derived on the basis of a regression analyses where the measured ice abrasion rates from the experiments were taken as the response parameter and the three explanatory variables in the experiments; compressive strength, ice pressure and ice temperature, as predictors. Based on simplified assumptions, a direct application of the equations provided fairly good approximations of maximum abrasion depths when benchmarked against a marine concrete structure with known abrasion depths. The best prediction was achieved for the parts of the structure with orientation parallel to the dominating ice drift directions. For the faces with lower ice exposure, the equations underpredicted the associated abrasion depths significantly. The ice-concrete coefficient of friction is reported to increase with decreasing drift speed and cause higher abrasion rates. As our equations did not include the speed of ice as predictor, they were not able to account for this effect. The distribution of ice abrasion on structures is highly dependent on the direction of the ice drift. In previous field studies dealing with concrete abrasion due to ice drift, reliable information about site specific ice conditions is sparse. On this basis, results from in-situ measurements of three severely damaged concrete lighthouses due to long term sea ice exposure are presented and discussed. As one of the inspected lighthouses showed particular severe damage, this structure formed the main basis of the investigation. The focus was on local ice conditions with main emphasise on the amount of ice drift per direction and its influence on distribution of abrasion depths along the perimeter of the lighthouse. The investigation revealed that the largest abrasion depths were observed on the faces oriented parallel with the dominating ice drift direction(s). Further, the rate of abrasion increased significantly with decreasing annual ice drift. This trend was most pronounced for the faces with the lowest ice exposure. The investigation revealed that abrasion depths on faces oriented diametrically opposite to each other increased linearly with the amount of ice drift along these faces. In previous experimental- and field investigations, ice abrasion rates were traditionally reported as average values, often without quantification of the uncertainty in the results. In order to allow a probabilistic approach in the results analysis, an appropriate probability distribution function to represent abrasion rates is needed. Based on our experimental results we performed goodness of fit tests and found that the abrasion rates were well represented by a 3-parameter Weibull distribution. The distribution was statistically significant based on a confidence level of 95%. An ice abrasion test rig was developed where a vertical oriented fresh-water ice cylinder was sliding in a repetitive back and forth motion on the surface of the concrete specimens. Ice pressure and concrete compressive strength were the most important parameters governing ice abrasion rates. Abrasion rates increased with increasing ice pressure. For some experiments the abrasion rates showed a quadratic increase with the ice pressure. As a result, it is important to seek to reduce the ice loads for concrete structures exposed to abrasion. This can be achieved by giving the structure a sloped cross section in the ice exposed zone, which causes the ice to fail in bending which is associated with lower loads as compared to failure in crushing. The results were ambiguous with regard to the effect of ice temperature on abrasion and no clear conclusion was reached based on the conducted experiments. We suspect that this was related to a specific feature of the experimental set-up and recommend that further studies investigate this closer. Ice abrasion on concrete is a complex process which may be investigated on several scales, pending on the aim of the research. In order to seek a fundamental understanding of the governing tribology processes involved, it would also be of interest to study concrete ice abrasion at nano scale.