Dr.

Vineet Kumar Singh

Assistant Professor
Department of Physics
DDU Gorakhpur University

1
Lecture 10

 Direct lattice versus reciprocal lattice

 210 in reciprocal lattice
 Geometrical construction of reciprocal lattice
 Properties of reciprocal lattices
 Diffraction conditions
 Ewald construction
 Ewald construction (XRD versus TEM)

2
 Illustration of direct lattices versus reciprocal lattices for a cubic system

 Each vector of the reciprocal lattice is normal to a set of lattice planes of the direct lattice
From previous lecture-
eiG.T = 1, i.e.
Exp[iG.T] = Exp[i {2u1v1 + 2 u2v2 + 2 u3v3 }] = Exp[2i {u1v1 + u2v2 + u3v3 }]
=Exp [2i  N] = 1 = when N is an integer only

(here, b1.a1 = 2[a1 .a2  a3 / a1 .a2  a3 ] = 2)

i.e. G.T = 2N (N is an integer)

Which tells us that the projection of the vector T on the direction of the vector G has the
length-
i.e. |G| |T| cos  = 2N
|G| dN = 2N (|T| cos  = dN )
dN = 2N / |G|

This length (dN) represents the distance from the origin to this plane. This further implies
that the reciprocal lattice vector Ghkl is perpendicular to the set of (hkl) lattice planes. It is
further noted that the value of d = dN /N corresponds to the distance between the adjacent
planes and is given by d = 2 / |G|. In summary, a single reciprocal lattice vector Ghkl is
uniquely assigned to a set of (hkl) lattice planes and is perpendicular to it.
3
 G = (110)

Figure: Cross-section cut across the x-y plane for a simple cubic crystal with lattice
constant a. The z-axis is perpendicular to the page. The cross-section of the family of (110)
lattice planes and the corresponding (110) reciprocal lattice vector are shown. They are
perpendicular to each other.

Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press
4
 Fig. (100) and (310) lattice planes with the corresponding reciprocal lattice vectors.
From figure, atoms are more densely distributed on the (100) planes than the (310)
planes.

Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press
5
 Illustration of direct lattices versus reciprocal lattices for a cubic system

Direct lattice space Reciprocal lattice space

a2 b2
(010)
4Å 040

(100)

2Å (with parenthesis) 020 b1

110
a1 210

a3 2Å 4Å b3 200 400

(without parenthesis)
(210) plane in direct lattice
Axial length 4 4 0
Intercept length 2 4 
Fractional length 2/4 4/4 
Miller Indices 1/(1/2) 1/1 1/
2 1 0

Reference: Elements of X-Ray diffraction, B.D. Cullity, S.R. Stock, third edition, Prentice Hall 6
 Illustration of direct lattices versus reciprocal lattices for a cubic system

210 in reciprocal lattice

 (210) plane in direct lattice will be represented by a dot in reciprocal lattice. The coordinate
of the dot in reciprocal lattice is 210 i.e. parenthesis is omitted.

 A vector connecting the 210 point through origin in reciprocal lattice will be perpendicular
to the (210) plane in direct lattice

 The length of this vector is equal to the 1/dhkl where dhkl is the interplaner separation
corresponding to (hkl) plane

 Different dots in reciprocal lattice space represent the different sets of parallel planes

Different position of dots in reciprocal lattice space represent the inverse of respective
interplaner separations from origin

7
 Illustration of direct lattices versus reciprocal lattices for a cubic system

 Let us discuss it a very simple way. Suppose a crystal has 10000 planes parallel to (110)
plane in direct lattice. Now, in reciprocal lattice we will represent these all 10000 planes by
a single dot corresponding to 110 coordinate. The separation of this dot from the origin in
reciprocal lattice is equal to the length 1/d110 (see next slide figure of reciprocal lattice
space) where d110 is the interplanar spacing (see next slide figure of direct lattice space).
Again, there are several possible position corresponding to the length of 1/d110 from the
origin. Finally, the position of the dot will be fixed at the end of the G110 reciprocal lattice
vector which is perpendicular to the (110) plane.

8
 Illustration of direct lattices versus reciprocal lattices for a cubic system

Direct lattice space Reciprocal lattice space

a2 (010) b2
4Å 040

90 (100)
2Å 020
90
110
a1 210 b1
a3 2Å 4Å
b3 200 400
This length is dN corresponding to
(110), it’s not d110 1/d110 1/d210

d210 = dN /N
(dN = interplanar separation for N planes)

Note: Reciprocal lattice vectors G210 (1/d210) and G110 are perpendicular to the
corresponding planes (210) and (110) in direct lattice space in cubical system as shown in
figure (above) [fig is not properly scaled].
9
 Geometrical construction of reciprocal lattice

1. Fix up some point in the direct lattice as a common origin

2. From this common origin draw normal to each and every set of parallel planes in
the direct lattice.
3. Fix the length of each normal equal to the reciprocal of the inter planar spacing
(1/dhkl) of the set of parallel planes (hkl) it represents.
4. Put a point at the end of each normal

Direct lattice space Reciprocal lattice space

a2 (010) b2
4Å 040

90 (100)
2Å 020
90
a1 110
210 b1
a3 2Å 4Å
d110 b3 200 400
d210
1/d110 1/d210
10
 Properties of reciprocal lattice

T= u1a1 + u2a2 + u3a3 (for direct lattice)

G= v1b1 + v2b2 + v3b3 (for reciprocal lattice)
a1, a2 and a3 = primitive vectors in for the direct lattice
b1, b2 and b3 = primitive vectors in for the reciprocal lattice
u1, u2 and u3 = integers (0, 1, 2, 3…)
v1, v2 and v3 = integers (0, 1, 2, 3…)

 A reciprocal lattice bi can be defined for every direct space lattice ai by

b1 = 2 a2 *a3 /[a1 .a2*a3 ] {b1 is perpendicular to a2 and a3 }
b2 = 2 a3 *a1 /[a1 .a2*a3 ]
b3 = 2 a1 *a2 /[a1 .a2*a3 ]
Unit of reciprocal lattice: Å-1
 bi . aj = 2ij
ij = 1 if i = j Orthogonal
ij = 0 if i  j
 Dimension of vector in the direct lattice = length
Dimension of vector in the reciprocal lattice = 1/length
 Reciprocal of the reciprocal lattice = 1
 b1 .b2*b3 = (2)3/[a1 .a2*a3 ]

Question: How is reciprocal basis vector related to the direct basis vectors ?
11
 Diffraction condition

k (scattering vector)

Wave vector of incident wave = k

Wave vector of outgoing wave= k’
Change in the wave vector or scattering vector = k
Scattering amplitude F = dV n(r) Exp [i(k-k’).r]
= dV n(r) Exp [i(-k.r)]

where k + k = k’ and …….(1)

n(r) = G nG Exp(iG.r) =electron number density

=G  dV nG Exp[i(G- k).r]

= V nG for k = G …….(2)
For, elastic scattering energy (h/2) is conserved that means energy of incident wave
h/2 = h’/2 (energy of outgoing wave). This implies
 = ’
ck = ck’ (from c=)
k = k’
k2 = k’2 12
From equation (1) and (2)
k + G = k’
Or (k + G)2 = k’2
k2 + G2 + 2k.G = k2 [for elastic scattering]
G2 + 2k.G = 0 ………(3)
If G is a reciprocal lattice vector then equation (3) becomes
G2 + 2k.(-G) = 0
G2 - 2k.G = 0
2 k. G = G2
2 k. =
k. =
i.e. the component of the incident wave vector k along the reciprocal lattice vector G must be
half the length G.
Thus an incident wave vector k will satisfy the Laue condition if and only if the tip of the
vector lies in a plane that is the perpendicular bisector of a line joining the origin of k-space to
a reciprocal lattice point G. such k-space planes are called Bragg planes.

k. (G/2) = (G/2)2 ………(4) [Another statement of Bragg condition]

Ref: Introduction to solid state physics, C. Kittel, seventh edition, Wiley 13

Solid State Physics, Ashcroft Mermin
 Diffraction condition

k
000
G
G/2

Constructing the plane normal to G at the midpoint, G/2, then any vector k drawn from
the origin 000, to any position on this plane satisfies the Bragg Diffraction condition i.e.

k. = or k. (G/2) = (G/2)2 .

Prob: X-rays with a vector kx are scattered from a simple cubic lattice with lattice spacing
a=2, the scattered X-rays have wave vector kx‘. The possible values of kx = kx - kx‘ for
which there are peaks in the scattered intensity are

(a) 0  kx /4 (b) kx = integer

(c) kx = integer multiple of 2 (d) kx = integer/2

Solution: k =G
G = 2N/dhkl
= 2N/2
= N = Integer 14
 Ewald Construction

Fig. Bragg’s reflection in reciprocal space

Suppose that an incident X-ray with wavelength  falls at a glancing angle  on the crystal
plane. We consider this situation in reciprocal space. As shown schematically in the above
figure. The center of the circle with the radius 2/ is positioned at a point O, where the
incident X-ray intersects with the lattice plane DE. The line AO refers to the direction of
the incident X-ray beam. The point B, at which the extrapolated line AO intersects with
the circle, is chosen as the origin in reciprocal space. The point C represents the
intersection of the reflected X-Ray beam with the circle. A simple geometric consideration
immediately leads to the relation AC//DE and, hence, the plane AC would possibly belong
to the same set as the lattice plane DE.
Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press 15
 In addition, we see that the vector BC is always perpendicular to the set of the planes
(AC & DE) and thus, it coincides with the direction of the reciprocal lattice vector. If the
point C coincide with the reciprocal lattice point, then the vector BC obviously
represents the reciprocal lattice vector Gn associated with the set of the lattice planes
(DE and AC).
dN =2N/|G|

The above equation confirms that the magnitude of the vector BC must be equal to
multiples of 2/d or 2N/d, where N is a positive integer and the set of the lattice planes
(DE and AC) is characterized by the interplanar distance d.

Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press 16
 2d sin  = n
Vector AO = wave vector k of the incident X-ray beam
= 2/
= radius of the Ewald Sphere
Vector OC = wave vector k’ of the reflected X-ray beam
= 2/

|k| = |k’| and

k’ = k + Gn

This is known as the Laue condition.

Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press 17
 Bragg’s law

Direct lattice Reciprocal lattice

2dhkl sin =  Ghkl = 2/dhkl

2dnh nk nl sin = n Gnh nk nl = 2/dnh nk nl = 2n/ dhkl

i.e. first order X-ray reflection from the set of (nh nk nl) planes with a spacing (d/n) or the
nth order reflection from the set of (hkl) lattice planes with the spacing d is similar thing.
For, example in previous lecture we have already derived the above concepts for cubic
lattice
sin (3rd order 111) =3/2d111
=/2(d111 /3)
= 1/2d333 where d333 = d111 /3
= sin (1st order 333)
The 3rd order reflection of d111 occur at same angle  as 1st order reflection of d333 but d333
is just one third of the d111.

 Finally, d111 = 2d222 = 3d333

Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press 18
 Bragg’s law

Let us consider another example for the set of (220) planes in the simple cubic lattice. The
length of the reciprocal lattice vector (220) is twice as large as that of the reciprocal lattice
vector (110), as discussed in previous slide.

Note that the (220) planes in the simple cubic lattice do not contain atoms but the
reciprocal lattice vector (220) is meaningful and refers to second order reflections from the
(110) planes.

Ref: Introduction to the electron theory of metals, Uichiro Mizutani, Cambridge University Press
19
 Ewald Construction

Incident beam
(a) (b)
1/

Reciprocal
lattice of crystal

(c) (d)
1/
1/

Diffracted
beams

Fig: (a) the reciprocal lattice, (b) a vector of length L=1/, drawn parallel to the beam
direction, (c) a sphere passing through the 000 reflection, drawn using 1/ radius, (d) the
positions of the diffracted beams
Ref: Crystals and crystal structures: R.J.D. Tilley, John Wiley & Sons LTD 20
Application: X-ray diffraction versus electron diffraction pattern

X-ray wavelength corresponding CuK  0.154 nm

Ewald sphere radius  1/ = 6.49 nm-1

Electron beam wavelength  0.0037 nm (at 100 KV acceleration)

Ewald sphere radius 270.27 nm-1
For Cu, lattice constant a = 0.360 nm
Reciprocal lattice spacing a* = 2.78 nm-1

For simple explanation, let us write all the numbers together:

Reciprocal lattice spacing for copper a* = 2.78 nm-1
X-ray beam Ewald sphere radius = 6.49 nm-1
electron beam Ewald sphere radius = 270.27 nm-1

Due to larger radius of Ewald sphere for electron beam (nearly 97 times higher than copper
reciprocal lattice Spacing) the surface of the Ewald sphere is nearly flat and interact with
numerous reciprocal lattice planes than X-ray Ewald sphere. That is what when the Ewald
sphere is much larger than the spacing between reciprocal lattice points, it will produce larger
number of diffraction pattern for the same sample.
21
 X-ray diffraction pattern (XRD)

Selective area electron

diffraction pattern (SAED)

Note the scale dimension

(1/nm) i.e. length-1 TEM Image

Ref: Nanomaterials 2018, 8(9), 717; https://doi.org/10.3390/nano8090717 (only for teaching

purpose, not for commercial use) 22
Note the scale
dimension (1/A)
i.e. length-1

Note down the value of Ghkl along the red line (shown in figure) and justify that the G800 has
just double length than G400 . Similarly, the length of G800 is four times of the length G200 (as
shown in figure).

Ref: http://crystalmaker.com/singlecrystal/ (only for teaching purpose, not for commercial use)
23
References

1. Introduction to Solid State Physics, Charles Kittel, Wiley

2. Introduction to the electron theory of metals, U. Mizutani, Cambridge University
Press, 2001
3. Elements of X-Ray diffraction, B.D. Cullity, S.R. Stock, third edition, Prentice Hall
4. Crystals and crystal structures: R.J.D. Tilley, John Wiley & Sons LTD
5. R. Farzana, R. Rajarao, P. Ranjan Behra, K. Hassan and V. Sahajwalla,
Nanomaterials 2018, 8(9), 717; https://doi.org/10.3390/nano8090717
6. Solid State Physics, S.O. Pillai, New Age International Publishers
7. http://crystalmaker.com/singlecrystal/

24