Fracture energy of columnar freshwater ice: Influence of loading type, loading rate and size

Abstract

This work investigates the influence of loading type, loading rate, and test size on the fracture energy of columnar freshwater S2 ice. The ice sheet in the Ice tank at Aalto University was very warm (above -0.5 ∘C) and thick (0.34<h<0.41 m). A program of nineteen mode I fracture tests using deeply cracked edge-cracked rectangular plates of various sizes (size range 1:39), loading types, and loading rates were carried out. Fourteen displacement-controlled tests (DC) were loaded monotonically to fracture, and five load-controlled tests (LC) were conducted with creep/cyclic-recovery and monotonic loading to fracture. Different methods for computing the fracture energy were applied and compared. The apparent fracture energy at crack growth initiation was obtained via Rice's J-integral expression (JQ) modified to be applicable to the special case of a deeply cracked edge-cracked plate as well as via a viscoelastic fictitious crack analysis (GVFCM). The work-of-fracture (Wf) was also evaluated. Both JQ and Wf were measured from the load-displacement record at the crack mouth. GVFCM was obtained from the back-calculated stress-separation (σ−δ) relation within the fracture process zone. A rather good agreement was obtained between GVFCM and JQ, especially for the large specimens. JQ and Wf exhibited interrelated size and rate effects. The Wf/JQ values for the DC tests were affected by rate and confined to a narrow range of 1 to 3. The creep-recovery loading did not affect the JQ values but led to an increase in the Wf and Wf/JQ values for most of the LC tests.