See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/26844330

Derivation of Scherrer Relation Using an Approach in Basic Physics Course

Article · January 2008

Source: DOAJ

CITATIONS READS
29 8,698

2 authors, including:

Khairurrijal Khairurrijal
Bandung Institute of Technology
390 PUBLICATIONS   2,313 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Nanoparticle View project

Carbon Nanodots View project

All content following this page was uploaded by Khairurrijal Khairurrijal on 02 June 2014.

The user has requested enhancement of the downloaded file.

Jurnal Nanosains & Nanoteknologi ISSN 1979-0880
Vol. 1 No.1, Februari 2008

Derivation of Scherrer Relation Using an Approach in Basic Physics Course

Mikrajuddin Abdullah(a) and Khairurrijal

Laboratory of Synthesis and Functionalization of Nanomaterial
Physics for Electronic Materials research Division, Faculty of Mathematics and Natural Sciences
Bandung Institute of Technology
Jalan Ganeca 10 Bandung 40132 Indonesia
Tel./Fax.: +62-22-250-0834/250-6452
(a)
E-mail: [email protected]
Diterima Editor : 2 Februari 2008
Diputuskan Publikasi : 11 Februari 2008

Abstract
Size of nanoparticles can be predicted using a simple Scherrer relation, based on the width of X-ray diffraction (XRD)
patterns of materials. In present paper we derive the Scherrer relation using a “multi slits interference” approach studied
in basic physics course. We found the dependence of crystallites size on the broadness of the XRD peak is exactly the same
as the original Scherrer formula. A simple procedure on using the Scherrer formula for predicting the crystallite sizes of
Y2O3 material is also discussed.
Keywords: Scherrer relation, multi slits interference,Y2O3
1. Introduction center to function as common facility for some
universities or research institutions. Much more
Nanotechnology is a broad field of research in
difficulties are faced by institutions in some developing
which scientists and engineers study the behavior of small
countries. It is very often that some developing countries
amounts of matter and try to design sub-microscopic sized
do not have even one such equipment in their universities
objects. When a small number (< 1000) of atoms or
or institutions. At this situation, how can we perform
molecules are brought together, their chemical and
nanotechnology research comprehensively, particularly
physical properties differ from those of individual
when we intend to understand the material sizes we have
particles and their bulk analogues. Thus, the same
produced even we know TEM or AFM is better
substance can show a range of properties depending only
equipment for this characterization?
on its size. Understanding the size of material in
The unavailability of TEM or AFM equipment,
nanoscale is important because many physical and
however, should not restrict our intention to work in
chemical properties are strongly dependent on the
nanotechnology research. It is a challenge to find another
structural size. Manipulation of material properties can be
way to characterize the size of nanostructure. Fortunately,
simply done by manipulation of the structural size.
there are some approximate methods for this purpose.
We can take advantage of nanoscale phenomena to
This method is helpful if the accuracy is not strictly
create miniaturized circuits for the next generation of
important, instead the approximate size is enough just for
computers, more environmentally friendly industrial
proving our successful in producing nanostructure
processes, more sensitive medical tests, better
materials in our laboratory. One well known method for
pharmaceuticals and many applications that we have not
this purpose is a Scherrer method [1-10]. This method
even considered yet. These new materials and capabilities
uses X-ray diffraction patterns to predict the size of
will transform all aspects of society, not just isolated
crystallites based on the width of the diffraction peak.
fields like computer science or cancer research. Consider
Indeed, the Scherrer method predicts the size of
how the industrial revolution has changed the world. The
crystallites, instead of the size of particle. Large particles
anticipated “nanotech revolution” will have a similar
might contain several crystallites. However, it is common
impact on our lives.
that nanometer-sized particle contains only one crystallite.
Research in nanotechnology, however, needs
Therefore, the size of crystallites in nanometer-sized
some sophisticated equipments for characterization of
particle as predicted by the Scherrer relation also
material properties. Transmission electron microscope
indirectly figures out the size of particle itself.
(TEM), scanning electron microscope (SEM), atomic
Experiments proved that the smaller the crystallite
force microscope (AFM), scanning tunneling microscope
size the broaden the diffraction peak. Very large crystal
(STM), and some spectroscopic facilities are examples of
with a single orientation produces diffraction peaks which
equipments required for nanomaterials research. However
are nearly vertical lines in shape. On the other hand, small
those equipments are very expensive for some
crystal produces very wide peak. Therefore, the width of
institutions, either in price, operational, and maintenance.
the diffraction peaks gives information on the crystal
Even in the developed countries, not all universities and
sizes.
research institutes are equipped with those facilities.
Why do small crystallites produce wide diffraction
Often, those facilities are placed at a certain research
peaks? The answer is small crystallites have restricted

28
J. Nano Saintek. Vol. 1 No. 1, Feb 2008 29

number of reflection planes. The diffraction peaks are a By inspecting Fig. 1B the difference in the
result of constructive interference of X-ray reflected by pathways of rays passing through two consecutive slits is
crystal planes.
In basic physics course, the students have studied δ = d sin θ B + d sin θ (1)
wave interference phenomenon from double slits (Young
experiment) or multi slits, and found that as the number of The displacements of wave oscillation originate from
slits increases the interference patterns on the screen different slits when reaching the detector are
become sharper and brighter. Interference from From the first slit: y1 = Aeikx
approximately infinity number of slits is very sharp, From the second slit: y2 = Aeik ( x +δ )
shaping like a very tiny line. Large slit number
From the third slit: y3 = Aeik ( x + 2δ )
corresponds to large crystallite sizes. Because X-ray
diffraction is basically an interference phenomenon of a .
large number of wave sources, it is acceptable to find the .
relation between the crystallite sizes and the width of .
diffraction peak using a basic physics approach. From the n-th slit: yn = Aeik [ x + ( n −1)δ ] (2)
The object of this work is to derive Scherrer The total deviation of waves at the detector is a
formula using a wave interference approach discussed in superposition of all waves originate from all slits, i.e.,
basic physics course. We found that this simple derivation
can produce the exactly the original form of Scherrer y = y1 + y2 + y3 + ... + yn
formula. = Aeikx + Aeik ( x +δ ) + Aeik ( x + 2δ ) + ... + Aeik [ x + ( n −1)δ ]
2. Derivation (
= Aeikx 1 + eikδ + ei 2 kδ + ... + ei ( n −1) kδ )
Figure 1 shows multi slits spaced by d. The slit 1− e inkδ

space corresponds to distance between reflection planes in = Aeikx

a crystal. If the thickness of the crystal is t then the 1 − eikδ
number of reflection planes (the number of slits) is einkδ / 2 e − inkδ / 2 − einkδ / 2
= Aeikx ikδ / 2 − ikδ / 2 ikδ / 2
n = t / d . Part of light originates from the first slit, other e e −e
parts originate from the second slit, the third slit, and so ⎛ [t / d ]k[ d sin θ B + d sin θ ] ⎞
on. Beams originate from different slits have different inkδ / 2
sin ⎜ ⎟
ikx e ⎝ 2 ⎠
phase when reaching the detector because they have = Ae
passed different distances. e ikδ / 2
⎛ k[d sin θ B + d sin θ ] ⎞
sin ⎜ ⎟
If the angle of incidence, θNB, is different from ⎝ 2 ⎠
einkδ / 2 sin ( kt [sin θ B + sin θ ] / 2 )
Bragg reflection angle, as shown in Fig. 1A, there is
approximately zero intensity detected by detector placed = Aeikx (3)
at the center of the screen. On the other hand, at Fig. 1B, eikδ / 2 ⎛ kt [sin θ B + sin θ ] / 2 ⎞
sin ⎜ ⎟⎟
if the incident angle is the same as the Bragg angle, θB, ⎜ n
⎝ ⎠
the constructive interference is detected by detector
placed at the center of the screen.
A B
Detector position
Detector position

d
θ θ
d sin θB
d sin θ

θNB θB

Figure 1 Diffraction by many slits. (A) If incident angle is different from Bragg angle, no intensity peak is detected by
detector placed at the center of the screen. (B) Detector at the center of the screen detects interference peak at the center of
the screen when the incident angle is equal to the Bragg angle.
J. Nano Saintek. Vol. 1 No. 1, Feb 2008 30

Although the crystallite size is small, however, n is B 1

usually large enough (suppose larger than 10). This kt cos θ B ≈ 2sin c −1 (10)
2 2
crystallite has a size of larger than one nanometer. One
nanometer consists of about ten atoms. Most of
nanoparticles synthesized by every method usually have a
size of larger than one nanometer. Based on this Io
assumption, we can approximate the sine function as
following

⎛ kt [sin θ B + sin θ ] / 2 ⎞ kt [sin θ B + sin θ ]

sin ⎜
⎜ ⎟⎟ ≈ (4) Io/2
⎝ n ⎠ 2n

We then find the approximated form for the total wave

deviation at detector as

einkδ / 2 sin ( kt sin θ )

y ≈ Aeikx
eikδ / 2 ⎛ kt sin θ ⎞
⎜ n ⎟ θB-Δθ θB θB+Δθ
⎝ ⎠ (5)
Figure 2 Profile of diffraction peak produced by
einkδ / 2 ⎛ [sin θ B + sin θ ] ⎞ interference from many slits.
= nAeikx sin c ⎜ kt ⎟⎟
e ik δ / 2 ⎜ 2
⎝ ⎠

The intensity of the detected wave is Although it is not perfect, for simple purpose, the
condition of sinc2 x = 1/2 can be approximated by x which
2 ⎛ [sin θ B + sin θ ] ⎞ is located at the middle of the locations of main maximum
I ∝ y = I o sin c 2 ⎜ kt ⎟⎟ (6) and the first minimum of the sinc2 x function, i.e., at
⎜ 2
⎝ ⎠ x =π /2 . Therefore we can approximate
We define the width of diffraction pattern as the sin c (1/ 2 ) ≈ π / 2 and Eq. 10 approximates to
−1

width at half of maximum of intensity, FWHM (full width

at half maximum). This is the width of the peak when the B π
kt cos θ B ≈2
intensity satisfies I = Io/2, or 2 2
⎛ [sin θ B + sin θ ] ⎞ 1 Using the relation k = 2π / λ with λ as the wavelength
sin c ⎜ kt
⎜ ⎟⎟ = (7) we finally find
⎝ 2 ⎠ 2
λ
If the angles satisfying the above conditions are t≈ (11)
θ B − Δθ and θ B + Δθ , we can write B cos θ B

kt
[sin θ B + sin(θ B + Δθ )] = sin c −1 ⎛ 1 ⎞
(8a)
The general form of Scherrer equation is
t ≈ K λ / B cos θ B , with K is a constant, closely to unity.
2 ⎜ ⎟
⎝ 2⎠ Different values of K have been used by authors, but all
are not far from unity. For examples some authors use K
kt
[sin θ B + sin(θ B − Δθ )] = − sin c −1 ⎛ 1 ⎞
(8b)
= 1 [2,5], K = 0.9 [3,6-10], and K = 0.94 [4]. Our result as
2 ⎜ ⎟ shown in Eq. 9 exactly produces the original Schrerrer
⎝ 2⎠
relation.
Subtracting Eq. 8a with 8b we obtain 3. Example of Application
sin(θ B + Δθ ) − sin(θ B − Δθ ) ⎛ 1 ⎞ Now we try to apply Scherrer equation to predict
kt = 2sin c −1 ⎜ ⎟ the crystallite size based on the XRD patters. Initially we
2 ⎝ 2⎠ record the X-ray diffraction patterns of sample at a wide
angle span, for example from 20o to 70o. Figure 3 is
⎛ 1 ⎞ example of diffraction patterns of yttrium oxide (Y2O3)
kt cos θ B sin Δθ = 2sin c −1 ⎜ ⎟ (9) synthesized by method of simple heating of nitrous
⎝ 2⎠ precursor in a polymer solution [11,12]. Those who
interested to understand more about this method can refer
Generally, the width of the diffraction peak is not to those references. Based on SEM picture in Fig. 4, the
very wide to allow the use of the following approximation particle (grain) sizes are several hundreds of nanometers.
sin Δθ ≈ Δθ = B / 2 , with B is the FWHM. Therefore we We are intending to determine the crystalline size in other
find the approximate form for Eq. 9 as to understand whether it is a single crystal or a
 J. Nano Saintek. Vol. 1 No. 1, Feb 2008 31

polycrystal. We see several XRD peaks appear in the function. The fitting result is shown as curve in Fig. 5.
measured angle range. By comparing the measured The parameters for the curve are following: curve area =
patterns with XRD database we believe that the material 29.097, center = 29.1o, width = 0.30187o, offset = 1178.1,
is a Y2O3. and height = 614.32. The important parameters for us are
the location of the peak and the width of the curve since
1800 they are used to determine the crystalline size using
Scherrer formula. Because the horizontal axis is in 2θ, we
1700 used half of it for B, i.e., B = 0.30187o/2 = 0.1509o =
Counts [arbitrary units]

0,002632 rad. The length of X-ray used in diffractometer

1600 is 0.1540598 nm. Therefore, the predicted crystallites size
is
1500
0.1540598
1400 t≈ = 67 nm
0.002632 × cos ( 29.1)
1300
This size is smaller that the particle (grain) size
1200 obtained from SEM, suggesting that the particles are
20 30 40 50 60 70
polycrystallites.
2θ/o
Figure 3 XRD patterns of Y2O3 1800

Counts [arbitrary units] 1700

1600

1500

1400

1300

1200
28.6 28.8 29.0 29.2 29.4 29.6
o
2θ/
Figure 5 Lorentzian fitting for peak that is located at
around 29.1o
600nm 4. Conclusion
Using multi slits interference studied in basic
Figure 4 SEM picture of Y2O3 prepared by heating of physics course we succeeded to derive Scherrer formula
metallic nitrous in a polymer solution. The corresponding for predicting the crystallites size based on the broadness
XRD patterns is displayed in Fig. 3 of XRD peak. We found a formula which is exactly the
same as the original Scherer formula.
How do we determine the crystalline size? First we
determine the FWHM of a selected diffraction peak which is
Acknowledgement
thought as the best one. Let us select a peak located at
International Linkage Grant 2007 from Department
around 30o because it is rather smoother and intense. We
of Physics, Bandung Institute of Technology is gratefully
redraw the XRD data in a very restricted angle span so that
acknowledged.
only the selected peak presents in our figure. Symbols in
Fig. 5 are the redrawing results in the form of symbols in the
angle span of 28.6o to 29.6o. References
It is commonly assumed that the diffraction profile
[1] K.-Y. Chan and B.-S. Teo, Microelectron. J. 38, 60
can be well fitted with a Lorentzian function
(2007).
1 Γ/2 [2] S.C. Chanda, A. Manna, V. Vijayan, Pranaba K.
L( x) = A + (12) Nayak, M. Ashok and H.N. Acharya, Mater. Lett. 61,
π ( x − xo ) 2 + (Γ / 2) 2
5059 (2007).
[3] B. Pejova, and I. Grozdanov, Thin Solid Films 515,
where A the offset, xo the center, and Γ a parameter 5203 (2007).
specifying the width of the curve. We used Origin [4] C. Gümüş, C. Ulutaş and Y. Ufuktepe, Opt. Mater.
software (Originlab.co) to fit the data using a Lorentzian 29, 1183 (2007).
 J. Nano Saintek. Vol. 1 No. 1, Feb 2008 32

[5] E.A. Souza, J.G.S. Duque, L. Kubota and C.T.

Meneses, J. Phys. Chem. Solids 68, 594 (2007).
[6] Sin-iti Kitazawa, Y. Choi, S. Yamamoto and T.
Yamaki, Thin Solid Films 515, 1901 (2006).
[7] P. Manisankar, C. Vedhi, G. Selvanathan and H.
Gurumallesh Prabu, Electrochim. Acta 52, 831
(2006).
[8] Y. Xie and C. Yuan, Appl. Catal. B: Environmenta,
46, 251 (2003).
[9] Mikrajuddin, F. Iskandar, K. Okuyama, and F.G. Shi,
J. Appl. Phys. 89, 6431 ( 2001).
[10] M. Abdullah, C. Panatarani, T.-O. Kim and K.
Okuyama, J. Alloys Comp. 377, 298 ( 2004).
[11] M. Abdullah, I. W. Lenggoro, B. Xia, and K.
Okuyama, J. Ceram. Soc. Jpn.113, 97 (2005).
[12] M. Abdullah and K. Okuyama, Proc. ITB Eng. Sci.
35B, 81 (2003).

View publication stats