Unit Cells (Uc) : Materials Science & Engineering
Unit Cells (Uc) : Materials Science & Engineering
Unit Cells (Uc) : Materials Science & Engineering
MATERIALS SCIENCE
Part of & A Learner’s Guide
ENGINEERING
AN INTRODUCTORY E-BOOK
Anandh Subramaniam & Kantesh Balani
Materials Science and Engineering (MSE)
Indian Institute of Technology, Kanpur- 208016
Email: [email protected], URL: home.iitk.ac.in/~anandh
http://home.iitk.ac.in/~anandh/E-book.htm
Important Note: the term “unit cell” in the current context refers to a crystallographic unit cell.
Other geometrical shapes can be used as a “cell” or a “unit” of the lattice or crystal; but
technically they may not qualify to be called a unit cell.
An unit cell is an unit of the structure [is a line segment in 1D, a parallelogram in 2D and a
parallelepiped in 3D], such that lattice points are at the ends of the line segment (1D) or at
the vertices of the parallelogram (2D) or parallelepiped (3D); which when repeated by the
translational symmetry vector(s) generates the whole structure (i.e. a lattice or a crystal).
A primitive unit cell has only one lattice point per cell, which are at: the ends of the line
segment (1D) and at the vertices of the parallelogram (2D) or parallelepiped (3D).
(A primitive unit cell need not have 1 atom per cell!!).
Funda Check Is the blue hexagon below a unit cell?
By the same token, the triangle below is also not a unit cell. As for the hexagon, the triangle
can be used monohedrally to tile the plane.
1D
1D
Unit cell of a 1D lattice is a line segment of length = the lattice parameter
this is the PRIMITIVE UNIT CELL (i.e. has one lattice point per cell).
Each of these lattice points contributes half a lattice point to the unit cell
Primitive UC
Contributions to the unit cell: Left point = 0.5, Middle point = 1, Right point = 0.5. Total = 2
Doubly Non-
primitive UC
Triply Non-
primitive UC
Unit cell of a 1D crystal will contain Motifs in addition to lattice points.
NOTE
The only kind of motifs possible in 1D are line segments.
Hence in ‘reality’ 1D crystals are not possible as Motifs typically have a finite
dimension (however we shall call them 1D crystals and use them for illustration of
concepts).
We could have 2D or 3D motifs repeated along 1D (hence periodicity and
‘crystallinity’ is only along 1D).
Often unit cells will be drawn like this can be terms as ‘extended unit cell’
Or ‘ball and stick’ model of a unit cell
90
Note: Symmetry of the Lattice or the crystal is not altered by our choice of unit cell!!
IMPORTANT
Symmetry (or the kind) of the Lattice or the
crystal is not altered by our choice of unit cell!!
How does convention come into play in the choice of unit cell for Orthorhombic lattices?
* The lattice may have higher symmetry than the crystal→ but in choosing the unit cell we focus on the
lattice.
E.g. if we decorate a square lattice with a triangle motif, we land up with a rectangle crystal. But we
prefer to chose a square unit cell for the crystal as well.
If the symmetry of the unit cell can be ‘commensurate’ with the lattice/crystal, this is the best.
Centred square lattice = Simple square lattice
& the smaller square UC is preferred (which is primitive).
Continued…
In this case the primitive (square) and the non-primitive square cell both have the same
symmetry
But the primitive square cell is chosen as it has the smaller size
The primitive parallelogram cell is not chosen as it has a lower symmetry
The lattice has 4-fold symmetries as shown
The square cells also have 4-fold symmetry
The parallelogram cell does NOT have 4-fold symmetry
(only 2-fold lower symmetry)
Full symmetry
is 4mm
2mm
The blue and the green unit cells have the same symmetry, which is 2mm. Hence, we now
have to go by size and chose the green UC as the preferred one.
However, by convention we chose the blue unit cell as the ‘preferred one’!!!
(What is the thought process behind this?)
This is the reason (i.e. the choice of a reference rectangle UC) for calling the lattice as a
centred rectangle lattice and not a ‘simple rhombus lattice’. Continued…
Simple rectangular Crystal Now the UC of the crystal will have a motif
(Not a centred crystal)
True
Unit Cell of
Crystal
b UC-1
a b2
UC-2
90
a
Click here
Wigner-Seitz cell
A Wigner-Seitz cell is a particular kind of primitive cell, which has the same symmetry as the
lattice. We will see soon that it is not a crystallographic unit cell.
Asymmetric unit
The asymmetric unit in combination with the space group will tile the entire space.
1D → one.
2D, 3D → Infinite (few examples in 2D given below).
1D
2D
Funda Check + Q &A
Important Note: the term “unit cell” in the current context refers to a crystallographic unit cell (CUC).
Other geometrical shapes can be used as a “cell” or a “unit” of the lattice or crystal; but
technically they may not qualify to be called a unit cell.
We had started with the important note as above. We will now see some examples of cells
which are a “unit” of the structure but do not form an crystallographic unit cell.
Specific cases we will see are (wherein a “cell” is not a crystallographic unit cell):
the vertices of the cell are not lattice points.
the shape of the cell is not a parallelogram (2D) or a parallelepiped (3D).
In such cases the lattice translation vectors cannot fill space without either overlaps or gaps.
We have already noted that the CUC is not unique; this implies that for each type of unit cell
the appropriate lattice translation vectors have to be used to fill entire space (without gaps or
overlaps).
We have also noted before that in some cases we draw a compound/combination of cells,
‘keeping in view’ the symmetry of the lattice/crystal. In that case, we can draw a
combination of three cells for the hexagonal system. Readers may also refer to the chapter
on Miller Indices, wherein we use the Miller-Bravais indices for hexagonal (and sometimes
trigonal) systems, keeping symmetry in view.
Funda Check Should a Unit Cell have Lattice Points only at the Corners?
The conventional unit cell chosen has lattice points at the ends of line segment/corners/vertices …
1D 2D
Conventional UCs
But in principle any unit cell like the ones below (space
filling) should work fine! (all the illustrated cells fill space!)
We had earlier seen that conventional choice of unit cells can ‘cut into’ the lattice points
(and may also cut into entities of motif) (as below).
Choices of some “non-conventional” cells (like the ones drawn before) can alleviate this
problem of ‘cutting into’ lattice points.
The new unit cell may still (or may not as below) cut into parts of the motif.
Corners of
the unit cell
New choice of “non-
conventional” cell
still have to
be lattice
The natural or preferred unit cell points
Problem with this unit cell: By convention
UC has entities of the motif in parts!
The Wigner-Seitz cell is a primitive cell with the symmetry of the lattice.
Technically it is not a unit cell as it does not have lattice points as vertices and may not be a
parallelogram/parallelepiped either.
This is created by Voronoi tessellation of space.
The region enclosed by the Wigner-Seitz cell is closer to a given lattice point than to any
other lattice point. Space is equally apportioned to lattice points.
Wigner-Seitz cell is a primitive cell (in that it contains one lattice point per cell).
The Wigner–Seitz cell in the reciprocal space is the first Brillouin zone.
In 2D the perpendicular bisector lines are drawn to obtain the Voronoi tessellation. In 3D
the perpendicular bisector planes are drawn.
Wigner-Seitz Cell (cotd..)
Square lattice
Wigner-Seitz cells
Not a conventional unit cell as the vertices are not lattice points
mh