Very basic xtallography questions

John Osborn j.c.osborn at
Tue Jan 11 05:21:23 EST 2000

Rob W. W. Hooft wrote:
> >>>>> "BV" == Bev Vincent <MaxDevore at> writes:
>  BV> To build a crystal in theory, you need the unit cell dimensions,
>  BV> space group and the contents of the asymmetric unit of the unit
>  BV> cell (i.e. tell where all the atoms are). However, there is no
>  BV> way to determine the crystal faces from this information. A
>  BV> crystal's habit (description of the crystal shape) is as much a
>  BV> characteristic of the substance as its color or unit cell, though
>  BV> the habit may be altered by changing the crystallization method
>  BV> (different solvent, for example). The only way to determine the
>  BV> faces is through experimental methods.
> True, but I'd like to add that using energy considerations it is
> possible to "guess" what the crystal habit might become knowing only
> the unit cell and the space group extinctions. These methods
> assume that the crystal grows one layer at a time, and that the
> attachment energy is isotropic.

If you try to guess the habit knowing only the cell dimensions and
space group (i.e. no atomic coordinates) then you cannot make use of
energy considerations. The morphology model which ignores atomic
coordinates and therefore ignores energy is called the
Bravais-Friedel-Donnay-Harker model. Better models which do use
energy considerations require the atomic coordinates. At least some
of them assume growth one layer at a time, but I think some of the
more recent models go beyond this.

I am not sure what you mean by saying that the attachment energy is
assumed to be isotropic. The attachment energy is certainly not
assumed to be the same for each face and there is no requirement for
the individual interatomic interactions which make up the attachment
energy to be assumed to be isotropic. (Some software packages may
require these interatomic potentials to be isotropic, but not all
packages have that limitation.) However the "growth morphology" model
which uses attachment energies does assume that the attachment energy
is the same for two opposite faces and hence predicts opposite faces
to be equal in area, which is not actually the case except when the
faces are required to be equivalent due to symmetry.

> IIRC, these theories are referred to as "Hartman-Perdok"

The name Hartman-Perdok is sometimes restricted to a certain class
of morphology models developed by Hartman, Perdok and their
collaborators and successors.

    John Osborn
    University of Bradford, UK.
    To reply by email, replace "br" by "bradford" in my email address.

More information about the Xtal-log mailing list