Search

The tests described in this data report have been designed primarily to test the repeatability of results from Phase I of testing. In Phase I, indentation tests were performed as a joint collaboration between Memorial University of Newfoundland and the Institute for Ocean Technology (NRC-IOT). The test specimens were made of polycrystalline ice with a grain size of ~4 mm. Two cm cubed monocrystals were placed at specific locations within the specimen with the c-axis of the monocrystal placed either parallel or perpendicular to the axis of indentation. Four groups of tests were completed as part of Phase II. Group 1 tested specimens that did not include an embedded monocrystal. This will allow for a more thorough investigation of the influence of the presence of an embedded crystal. Group 2 consisted of a number of repeats of experiments from Phase I. This group will be used to investigate the influence of the orientation of the c-axis of the embedded monocrystal. Group 3 used ice specimens with no embedded crystal and were performed at a velocity expected to produce cyclic loading. This will allow the link between cyclic loading and ice crushing to be studied in greater detail. Finally, Group 4 consisted of tests performed on specimens with the same dimensions as in phase I but with different compositions (compressed snow, ~4mm polycrystalline ice with air bubbles, and bubble-free ice with a highly irregular grain structure). This will allow the influence of ice composition to be studied in greater detail.

At very slow loading rates, ice will creep; as the rate is increased, microstructural changes occur and the rate of creep is enhanced as a result. These microstructural changes are referred to as 'damage'. As the rate of loading is increased further, the damage becomes localized into a layer adjacent to the indentor. This layer is associated with 'high-pressure zones' (hpz's). A bulb of pressure develops over these zones with values up to 100 MPa at the centre. Processes within the layer vary with distance from the centre; microfracturing and recrystallization occur near the outside with recrystallization and pressure melting (it is supposed) along grain boundaries in the central part. Cyclic loading can result since the cycle of pressure softening and subsequent hardening upon release of the pressure results in a repetitive cycle. An analysis has been performed using the ABAQUS computer program that encapsulates the principal features of the process. Some results of this analysis are given. It is also shown that the mechanics can be scaled geometrically without any basic change. This explains why the hpz's are found over widely differing scales. Laboratory and field data involving cyclic loading, including Molikpaq and medium scale data, are reviewed and discussed in the context of feedback mechanisms and ice-induced vibration. Finally, data and results from impact tests on other materials, as well as dynamic recrystallization at high speeds are reviewed and discussed in the context of ice compressive failure., 19th International Symposium on Ice, July 6-11, 2008, Vancouver, BC

Results are presented for a series of indentation tests that were designed to study ice failure processes under varying loading conditions. A high-speed video camera was used to link load trace events (i.e. load spikes and drops, etc) with failure processes. It was shown that large-scale fractures and spalls are directly correlated with significant drops in load. Thin-sectioning of the samples showed the existence of a damaged layer of ice during testing. Micro-cracking, crushing, and recrystallization were all present in a zone beneath the indenter. This is consistent with the current description of a high-pressure zone, providing confidence in the current ice-structure interaction models understanding the role played by such mechanical processes as large-scale fracture, spalling, and damage on global load reduction. During an ice-structure interaction the majority of the force will be transferred to the structure through a series of relatively small zones of high pressure (HPZs). These HPZs vary in both time and location throughout the interaction (Fig. 1 (a)). They are influential in the evolution of the interaction since their presence is linked to localized spalling, large-scale fracture, damage and softening of ice in the interaction area. When the HPZ occurs near the edge of an ice-sheet large-scale fractures can be precipitated. These fracture events significantly reduce the load that is felt by the structure (Jordaan, 2001). In the vicinity of the HPZs, damaged ice forms in a layer as shown in Fig. 1 (b). In the center of the HPZ, where the confining pressures are highest, the damaged layer likely consists of a region of recrystallized and pressure., 19th International Conference on Port and Ocean Engineering under Arctic Conditions, 27-30 June 2007, Dalian, China

A series of indentation tests were performed as a joint collaboration between Memorial University of Newfoundland and The Institute for Ocean Technology (NRC-IOT). The tests were performed on ice specimens with dimensions of 20 x 20 x 10 cm, which were made of polycrystalline ice with a grain size of ~ 4 mm. In order to explore the role of grain boundaries, the test specimens included 2 cm cubed monocrystals placed at specific locations within the specimen. It is hypothesized that the large grain boundaries provided by the monocrystal will help to precipitate a fracture in the vicinity of the crystal. This should affect both the maximum load attainable during the test as well as the maximum duration of the tests as compared to a specimen that does not contain an embedded crystal. The tests were performed using an MTS located at the NRC-IOT. The total force was recorded during the tests using a 25 kN load cell with a sampling frequency of 20 kHz. A high-speed video camera was used in these tests to identify the location where fracture begins. A vertical sheet of laser light was used to aid in this process. The high-speed video was also used to help find a correlation between fluctuations in the load traces and visual events such as crushing, spalling and fracture. Additionally, the tests included the use of a tactile pressure sensor in the indentation area. The sensor was included in order to explore the pressure distribution in the area of the indenter. This area was found to consist of isolated areas of very high pressure compared to the average pressure found during the test.

Under high hydrostatic pressure combined with shear, the structure of polycrystalline ice breaks down into fine-grained material by a process of recrystallization. Accompanying this process is a substantial softening of the material. Measurements of activation energy at low strains prior to the breakdown of structure, at temperatures of -10 °C, show an increase of activation energy from about 80 kJ mol-1 to about 120 kJ mol -1 as the pressure increases, with values at about 70 MPa similar to those found by other researchers for temperatures between -10 and 0 °C. These results suggest that grain boundary effects might result in the increased strain rates. The various energies that might drive the subsequent recrystallization process have been investigated. The results suggest strongly that elastic strain energy is the driving force in the process., 18th International Conference on Port and Ocean Engineering under Arctic Conditions, POAC 2005, June 26-30, 2005, Potsdam, NY; USA

High speed uniaxial compression tests were conducted on columnar, granular and monocrystalline ice. Results were consistent with previous work and showed an increasing strength with strain rate in this range. The monocrystals, oriented with the c-axis parallel to the loading axis, were slightly stronger than the other types of ice.

Using indentation experiments on freshwater ice as a starting point, numerical models for ice damage, clearing and flexure have been developed together with a mechanical model of ice-induced vibration. Indentation speeds were such as to cover a range of ice behaviour, from essentially pure crushing to situations where both flexural failure and crushing took place. Modelling of the damage process was used to estimate the thickness and equivalent viscosity of the crushed layer under various circumstances, and also to demonstrate the change in pressure distributions at different stages in the loading cycle. Crushing is complicated by the presence of spalling as well as distributed damage. Analysis of these effects and associated local high pressure zones is important for determining when crushing failure occurs. With regard to flexural failure, important issues that are dealt with are the influence of local crushing, the effect of speed and the hydrodynamic effect on the distance at which circumferential cracks are formed. The distance to primary flexural cracks was predicted successfully using analytical plate bending solutions and an equivalent beam model was calibrated. Hydrodynamic effects did not account for observed secondary cracks at shorter distances. A simple mechanical model was developed to integrate the above behaviour with regard to crushing and flexure. The model includes the structural and far field ice responses, and was successful in simulating peak forces and the frequency of dynamic loading cycles in a variety of ice and loading conditions., 11th International Conference on Port and Ocean Engineering Under Arctic Conditions, September 24-28, 1991, St. John's, Nfld.

A fundamental approach has been taken to address problems related to the resistance to ship transit in level ice and the forces exerted by moving ice on fixed structures. Using indentation experiments on freshwater ice as a starting point, numerical models for ice damage, clearing and flexure have been developed together with a mechanical model of ice-induced vibrations.