Further Reading:

An Outline of Structural Geology, 1976. Hobbs, Means & Williams p 73-104

Crystalline Plasticity and Solid State Flow in Metamorphic Rocks, 1976 Nicholas & Poirier p 52-121

Creep of Crystals, 1985 Poirier p 38-63

Microtectonics, 1996, C. W. Passchier & R.A.J. Trouw, Springer-Verlag, Berlin.

The first part of the lab consists of a series of computer animations of various small scale deformation mechanisms.

To view the images in this course as a movie, click on the right arrow at the bottom left of the picture (or fast forward and reverse with the arrows on the lower right hand side).

These are numerical simulations where the interactions between atoms in a crystal are calculated for systems typically with 100x100x100 atoms. These allow the investigation of VERY small scale deformation processes.

In the movie below, the dislocations generated around a crack tip have been modelled numerically by a technique known as non-equilibrium molecular dynamics. In this technique the interactions between atoms are modelled, and the visualisation shows only those atoms with high energy states (ie near dislocations). The big loops that develop near the crack tip are typical of real dislocations. Further information on this study can be found in (S. J. Zhou, et al., Physical Review Letters,  78, 1997).


• If these simulations have 10 million atoms in them, how many times larger would they have to be to simulate a single 1 mm cube of quartz? (You will have to find lattice cell parameters for quartz from a mineralogy text book). Similarly these simulations run at strain rates of 10⁺⁸s^-1, how many orders of magnitude is this from a "geological" strain rate?

If you have access to the web and a PC or a PC emulator for the Mac, get a copy of atomdemo from http://www.ims.uconn.edu/centers/simul/progdoc/atomdemo/atomdemo.html and you will be able to run your own 2D molecular dynamics simulations.

The following links are to dislocation dynamics simulations calculated by Benoit Devincre and Ladislas Kubin at the Laboratoire d'Etude des Microstructures (CNRS/ONERA) and show a variety of dislocation motions and interactions.


• Dislocation Frank-Read source.
• Dynamics of two interacting Frank-Read sources, repelling and attracting.
• Dislocation dipole.
• Dislocation annihilation by cross-slip.
• Dislocation double cross-slip.
• Dislocation-dislocation reaction.
• Dislocation "forest" interactions.
• Dislocation 3D dynamics.



• Give a brief description of each simulation.

• Explain how each process below leads to the straining and formation of microstructures in the crystals.

• What might cause a non-random motion of vacancies in a natural crystal?

D) Edge Dislocations Motion. In these movies a single horizontal glide plane is activated by elastic strain build up in a crystal. Watch as the lattice bonds in the glide plane get stretched, and then switch one by one to a new orientation, and then once the dislocation has passed through the crystal, how this allows the further slip of the crystal on the glide plane.

• Notice how the distortion of the lattice is restricted to a small area around the tip of the extra half plane. This zone of elastic distortion (known as the dislocation core) is a fast diffusional pathway for fluid migration, and also provides the driving force for various other deformation processes

• Notice also that in this cubic symmetry crystal another perpendicular slip system may well be active at the same time, to say nothing of possible slip systems at 45 degrees to the existing one.

F) Dislocation loops and Frank-Reed Sources of Dislocations. If you have a dislocation which forms a complete loop, the loop diameter can grow as a shear stress is applied, causing strain in the crystal. How do new dislocations form? Well one really good way of generating an endless supply of dislocations is to have a Frank-Read source, which you have already seen above, which is really just a glide plane with two sticking points on it at which a glissile dislocation sticks. The stress on the initially straight dislocation bends it out until it bends so far around it joins up with itself and forms a closed loop, which then just grows in size. The remaining pinned dislocation then starts again and generates an endless supply of free loops.

This type of source will necessarily result in a concentration of slip on just a few planes, rather than even glide throughout the body of the crystal, and this is what is seen in experimental deformation of single crystals.

G) Tilt-wall formation. This movie of ice deformation made by Chris Wilson has us looking side on on a bending crystal, and shows the initial bending of a crystal (eg the light green grain in the top-left), with the regeneration of perfect crystal by the addition of dislocations of the same sign to a developing sub-grain (tilt wall boundary). This process will convert a grain showing an undulatory extinction microstructure to one with a sub-grain microstructure.

• In the same grain a twist wall boundary forms on the left hand side. How do we know this is a twist boundary (assuming that only basal glide is active)?
• What would happen if two tilt-wall sub-grain boundaries with the same sign coalesce? What would happen if two tilt-wall sub-grain boundaries with the opposite sign coalesce?

• How could you distinguish a kink band from a twin in a natural crystal.

Copyright Mark Jessell & Paul Bons 2000