Ice crystal structure – University of Copenhagen

Forward this page to a friend Resize Print Bookmark and Share

Centre for Ice and Climate > Research > Flow of ice > Ice crystal structure

Ice crystal structure

Ice crystal structure

In an ice crystal the water molecules are arranged in layers of hexagonal rings. These layers are called the basal planes of the crystal, and the normal to the basal plane is called the c-axis or the optical axis of the crystal. 

Hexagonal ring structure

The hexagonal ring structure of an ice crystal (the blue and black spheres represent the oxygen atoms from the H2O).

Basal planes

The basal planes of the crystal are perpendicular to the c-axis. The crystal deforms by gliding along the basal planes.

The bonds between molecules situated in the same basal plane are much stronger than the bonds between molecules located in different basal planes. This causes the ice crystal to deform by gliding on its basal planes. The planes glide past each other like the cards in a deck that is pushed from the side.

 

If the crystal is oriented unfavorably for basal gliding - if the stress is applied perpendicular to the basal planes - it can still deform, but the stress needed is 100 times higher than that required for basal gliding.  

Poly crystal


Glacier ice is built up from many individual ice crystals that are packed closely together. In the top of an ice sheet the ice crystals are randomly oriented because the snow flakes have settled randomly. Some crystals are oriented favorably for basal gliding and others are not. This means that the deformation proceeds much more slowly than for a single ice crystal.

 

Poly crystal deformed

As the ice deforms, the individual crystals in the ice slowly change shape as the basal planes glide past each other, just like a deck of cards changes shape when it is pushed from one side. This causes the individual crystals to rotate. Generally, the c-axes of the crystals rotate towards an axis of compression and away from an axis of extension. The effect of this is that deep down in the ice sheet the crystals are no longer randomly oriented but have a preferred direction, which depends on the flow history. Thus the flow history of the ice can be found from investigation of the crystal orientation at different depths.

The crystal orientation is determined by studying thin sections of the ice. A thin section is a slab of ice approx. 0.5 mm thick. When this slab of ice is placed between two crossed polarization filters, the individual ice crystals can be seen. The colour of the crystal depends on its orientation.  

Crystals

Thin sections showing the crystal structure at a depth of a few hundred metres (left) and from the middle of the ice sheet (right). In the top of the ice sheet the crystals have random orientation. This is seen as the crystals of the thin section to the left having many different colours. Deeper down, the deformation of the ice has lead to the crystals having a preferred direction. Therefore, most of the crystals in the thin section to right have similar colours – blue.

The size of the individual crystals also changes with depth. In the Greenland ice sheet these crystals are between 1 mm and 10 cm in diameter. In the top layers the crystals are generally small, but with time the smallest crystals are 'eaten up' by larger neighbouring crystals causing the size of the crystals to increase with depth. Close to bedrock the crystals can grow very big because the geothermal heat released from the bedrock increases the growth rate of the crystals.

The crystal size also depends on the impurity content of the ice. When there is a high impurity content the crystals tend to be smaller because the impurities inhibit the growth of the crystals.

Read more about impurities and crystal sizes here.

Read more about how crystal size distribution and orientation are determined from analysis of thin sections.

Relevant reading:

  • G. Durand, A. Persson, D. Samyn, A. Svensson
    Relation between neighbouring grains in the upper part of the NorthGRIP ice core - Implications for rotation recrystallization
    Earth and Planetary Science Letters 265 (2008) 666-671
    Abstract | doi 

  • J. Mathiesen, J. Ferkinghoff-Borg, M. H. Jensen, M. Levinsen, P. Olesen, D. Dahl-Jensen and A. Svensson
    Dynamics of crystal formation in the Greenland NorthGRIP ice core
    Journal of Glaciology, Vol. 50, No. 170, 325-328, 2004
    Abstract

  • A. Svensson, P. Baadsager, A. Persson, C. S. Hvidberg and M.-L. Siggaard-Andersen
    Seasonal variability in ice crystal properties at NorthGRIP: a case study around 301 m depth
    Annals of Glaciology, Vol. 37, p. 119-122, 2003
    Abstract

  • A. Svensson, K. G. Schmidt, D. Dahl-Jensen, S. J. Johnsen, Y. Wang, S. Kipfstuhl and T. Thorsteinsson
    Properties of ice crystals in NorthGRIP late - to middle-Holocene ice
    Annals of Glaciology, Vol. 37, p. 113-118, 2003
    Abstract

  • K. M. Hansen, A. Svensson, Y. Wang and J. P. Steffensen
    Properties of GRIP ice crystals from around Greenland interstadial 3
    Annals of Glaciology, 35, p. 531-537, 2002
    Abstract