MSE 2021: Single- and Multi-Phase Identification via XRD Lab Report

Student Name: _________________________________________________________

Section _______________________________________________________________

Lab Session Date: ______________________________________________________

Lab Report Due Date: ___________________________________________________

TA:

Lab Participation Lab Report

Possible Points Possible Points

Points Awarded Points Awarded
Attendance 25 Questions 20
Participation 5 Procedure 10
Attentiveness 5 Results 30
Teamwork 5 Data analysis 40
Safety 10 Discussion 32
SUBTOTAL 50 Conclusion 18
SUBTOTAL 150

School of Materials Science and Engineering

Georgia Institute of Technology

Prepared

21 June 2021

1
 Lab 6&7: Utilizing X-ray Diffraction to Identify Unknown Powder

Abstract
An x-ray diffractometer and a Hanawalt Search Manual were used in conjunction to first identify an
unknown powder and secondly a mixture of unknown powders. For the single phase powder peak intensity
values for a range of 2 values are analyzed, and considerations of experimental variation are discussed.
With the aid of spreadsheets, the plot/replot method will be used to separate peak intensities, allowing for
the identification of the phases present in the powder mixture.

Introduction

When incident light waves have a wavelength on the order of repeating scattering centers,
diffraction can occur. One example is electromagnetic radiation, such as a beam of x-rays, incident on a
crystal lattice. Crystal lattices are comprised of many parallel planes of atoms (each atom is equivalent to a
scattering center). Each particular crystallographic plane (ie. 110, 100, 220) has unique spacing distances,
dhkl. These interplanar spacings depend on both crystal structure and composition. At certain angles of
incidence, the diffracted parallel waves constructively interfere and create detectable peaks in intensity.
W.H. Bragg identified the relationship illustrated in Figure 1 and derived a corresponding equation, eqn. 1.

Figure 1: An illustration of Bragg’s law showing constructive interference occurs at particular angles with
respect to a lattice, constructive interference only occurs integer multiples of the incident wavelength, λ.

n = 2d’ sin  or ( = 2dhkl sin , in which ‘order’ of diffraction has been taken into account) [ eqn. 1]
When a peak in intensity is observed, eqn. 1 is necessarily satisfied. Consequently, one can calculate d-
spacings based on the angles at which peaks are observed. By calculating the d-spacings of the three
strongest peaks, a single-phase material can readily be identified using a Hanawalt Search Manual, which
lists d-spacings of thousands of materials in order of observed intensity. Often more than one material may
seem to fit the experimental data. One way to determine which indexed material best matches an
experimental pattern is to calculate a figure of merit. One popular figure of merit is given by eqn. 2, where
Nposs is the number of independent diffraction lines listed in the powder diffraction file (pdf) for the 2θ
range scanned, Δ2θ is the average absolute discrepancy between indexed and observed 2θ values, and N is
the number of peaks in the experimental pattern. A higher figure of merit, FN, means a better match.
Obviously the figure of merit described does not include analysis of the relative peak heights in the pattern;
so, for complete analysis, some manual interpretation must be utilized.

2
 [eqn. 2]

Because the possible 2 reflections depend on crystal structure and satisfying Bragg’s law,
predicting the diffraction angle for any set of planes in a particular structure is possible through a general
relationship produced by combining Bragg’s law and a particular structure’s plane spacing equation. Eqn.
3 is the plane spacing equation for cubic lattices where dhkl is the interplanar spacing, a is the lattice
parameter and h, k, and l are the Miller indices for a particular plane.
1 (h2  k 2  l 2 )

d hkl 2 a2 [eqn. 3]

In this lab, we will be using the powder diffraction method, which is easily the most popular
diffraction technique. Powder diffraction requires a polycrystalline sample and monochromatic x-rays
(fixed ). These requirements simplify how diffraction peaks are measured. If the sample were a single
crystal, it would need to be rotated until the Bragg condition is satisfied to see a spike in diffracted
intensity. Because the powder is assumed to be made up of randomly oriented grains, by chance, certain
crystals will be oriented for diffraction of one plane, while another group will be oriented for the diffraction
of another plane. As a result, we observe every possible lattice plane capable of diffraction.

Identifying multiple phases in a particular X-ray pattern, especially when containing more than
two phases, can be quite cumbersome. There are now many software programs that provide automated
searches of the ICDD database for matches to a particular pattern. However, these programs generally only
rank possible identification matches by some figure of merit. For complete analysis some manual
interpretation must be utilized, such as the fit of relative peak intensities. The purpose of this lab is to
identify two separate single phase materials in a powder mixture by using an x-ray diffraction pattern,
possible matches identified in a Hanawalt search manual, and spreadsheet calculations. By performing
manual phase identification, a better understanding of identification software and its limitations will be
garnered.

It is assumed that the examined powder mixture is crystalline and randomly oriented. Often in the
case of a multiple constituent material, peaks from individual phases will add, possibly producing double
peak structures and/or single peaks with increased intensity due to overlap at particular 2θ values. To
determine the identities of the separate phases, one can use a technique known as the plot/replot method.
An example of the final result of a plot/replot method is shown in Figure 2.

3
 Unknown with Identified Overlays

120

100

80
Relative Intensity (%)

Unknown
60 Copper
Cuprous Oxide

40

20

0
0.5 1 1.5 2 2.5 3 3.5
d (Angstroms)

Figure 2: Stick plot generated from unknown powder diffraction pattern with overlays of identified
compounds. Generated from data in Elements of X-ray Diffraction by Cullity and Stock, 3rd edition, 2001.
See example spreadsheet on t-square.

Notice the two materials identified (overlays) in Figure 2 account for all of the peaks present in the original
diffraction pattern. Using a phase identification procedure similar to Laboratory #4, one can calculate d-
spacings of the strongest peaks and use a Hanawalt Search Manual to determine the first phase present in
the sample. Once the first phase is identified, the normalized intensities will be subtracted off and the
result renormalized. The second phase will then be identified from the remaining peaks, and a residual
error will be calculated.

Questions to be answered for Background/Introduction

Often what are actually diffracted beams are referred to as “reflected beams.” Diffraction of x-rays by
crystals and the reflection of light by mirrors are sometimes confused because both phenomena result in the
angle of “reflection” being equal to the angle of incidence.

1) (2 Points) Name one way in which diffraction and reflection differ. (Hint: consider the qualities of
the incident beam, any equations governing either, what is required for each to occur, etc.)

A: Reflection involves the direction of the light ray changing when it bounces of a barrier while diffraction
occurs when the direction of a ray is changed after passing through a certain slit.

2) (2 Points) Because interatomic distances for most crystalline solids are on the order of 10-9, x-ray
wavelength beams can be used to measure lattice spacing. Knowing this, could visible light be used
for crystal diffraction? Please circle one of the following: YES NO

3) (2 Points) In a typical X-ray pattern, intensity is plotted versus INCREASING 2 which

corresponds to Decreasing d-spacing.

4
4) (2 Points) If the atomic radius of aluminum (FCC crystal structure) is 0.143 nm, what is the d
spacing between (100) planes? (show your work)

5) (3 Points) If the spacing between the (3 1 1) planes of Al is 0.122 nm, calculate the lattice
parameter (a) for Al.

6) (3 Points) Assuming the crystal structure is cubic, rearrange the list of Miller’s Indices to put them
in order of decreasing (largest -> smallest) interplanar spacing (3 1 1), (2 2 2), (2 1 1), (3 0 0).

7) (2 Points) What are some of the difficulties associated with the identification of multiple phases in
a mixture?

8) (2 Points) Repacking a powder sample for an additional scan (repacked because the scan varied
from expected relative peak heights) resulted in drastically different relative intensities, which
eliminated the primary source of _____________as the primary source of relative intensity error.
(Hint: how might repacking affect the orientation of the crystals?)

9) (2 Points) You produced an extremely thin coating (phase X) on a small amount of powder
material (phase Y). Because phase X is black and phase Y is white you know you were successful in
forming a coating because the resulting material was black and SEM/EDS analysis revealed a
continuous coating containing the coating elements was present; however, only phase Y was detected
in a diffraction pattern of this material. How would you set up a subsequent scan to detect phase X?
(Hint: how would you reduce the influence of the bulk in your collected signal?)

Procedure

It is assumed that the unknown powder sample is homogeneous in composition, crystalline in structure, and
randomly oriented. Different sample holders may be required for different specimen volumes (e.g. if you
only have a small amount of sample). Two different specimen volumes were extracted from the unknown
sample, prepared in different sample holders, and scanned in an X'Pert PRO Alpha-1 diffractometer. Each
student will be given a unique set of parameters to test the unknown sample with.

1) (5 Points) In your own words, briefly describe the sample preparation procedure. Name the two
types of sample holders used in this lab. What is the primary reason for choosing one sample holder
over the other? What affect does sample volume have on the diffracted beam intensity?

A: For this lab the sample was put into a large powered sample holder and then compressed lightly as
compressing it too much would lead to alignment while the sample would have too much air in it if not
compressed. The larger sample holder is used as

5
2) (5 Points) Provide the instrument parameters used for data collection on the sample you prepared
below

Monochromatic x-ray source: Copper with wavelength 1.54 A (include units).

X-ray tube operating settings: 45kV, 40 mA

Start 2 θ:15o

End 2 θ:100o

Incident Soller Slit: 0.04 Rad

Mask Size: 200

Incident Divergence Slit: 0.5o

Incident Anti-scatter Slit: 0.5o

Diffracted Soller Slit: 0.04Rad

Diffracted Anti-scatter Slit: 5mm

Detector: X’celerator

Results
Insert a plot showing the diffraction pattern generated with your instrument settings. Normalize the
intensity in your plot such that the most intense peak has an intensity of “100.0” and all other intensities are
relative to this value. Use the provided label below with your sample name inserted in the blank space.

1) (10 Points) Figure 3: Diffraction Profile for Single Phase Sample (Insert Figure Below):

6
A:

2) (10 Points) List each observed peak for the single phase sample in the Table 1. In the space below,
show your work for the conversion of the observed 2 value to the corresponding value of d-spacing.
State any assumptions needed for this calculation.

Table 1: PEAK DATA FOR SAMPLE _________

2 I d I/Imax

3) (10 Points) Figure 4: Diffraction Profile for Multi-Phase Sample (Insert Figure Below):

7
A:

Data Analysis
Use the Hanawalt Search Manual to select at least 2 possible matches for your single phase sample. Report
the values of d-spacing, relative intensity and corresponding hkl values for each entry in the tables below.
Then, calculate the figure of merit for each possible match to determine which possible match is the best
candidate for the unknown sample. Show an example calculation for figure of merit in the space below.

1) (5 Points) Table 2 PDF for

d I hkl 2θpdf 2θexp |∆ 2θ|

FN =

8
2) (5 Points) Table 3 PDF for
d I hkl 2θpdf 2θexp |∆ 2θ|

FN =

3) (5 Points) The peak matching software “X’Pert Highscore Plus” was also used to identify the
single phase material. List the top three matches below.

4) (10 Points) For the multi-phase sample, attach the spreadsheet results (list as Appendix A)
determined from the diffraction data provided after the lab practicum. Be sure to identify 3 primary
peaks for each phase, and include the manual plot/replot analysis.

5) (10 Points) Briefly describe the procedure you used to identify the phases in the two-phase
mixtures you were given. Draw some simple plots showing the mixture XRD pattern before and
after separating out the different phases for one of the samples you analyzed. Be sure to list the PDF
#’s corresponding to your identified phases below.

6) (5 Points) Does the analysis confirm that only two powders are present? What phases would you
conclude to be present in the powder sample? What crystal structures are associated with the
different phases?

Discussion
1) (8 Points) Bragg’s Law gives us discrete values for diffraction peaks. However in the lab, we find
that diffraction peaks have width. Why?

9
A: We find that the peaks have some width as Bragg’s law assumes that the crystal is perfect, i.e. it does
not have any defects of impurities. However, real samples have both thus the broadening is caused by both
these factors.

2) (8 Points) What are some of the causes of phase misidentification when using the automated
search/match program for multi-phase analysis?
A: Some causes could be:
 Crystal Size not being accounted for.
 Defects in the crystals.
 Impurities in the sample.

3) (8 Points) In this lab we assumed that the sample grains were randomly oriented. How would the
measured diffraction pattern change if they were preferentially oriented? (For example, in a textured
polycrystalline thin film).

A: There would not be any resemblance of peaks in the sample and the graph would just show a curve of
many non-defined peaks.

4) (8 Points) Draw what would your data look like if instead of a crystalline material, we looked at an
amorphous material? Why does it look this way? (Remember proper axis labels!)

Conclusions

(18 Points) Summarize your results, and explain what the analysis of your data shows for both single-
and multi-phase analysis. (Include numbers)

References
B. D. Cullity and S. R. Stock. Elements of X-Ray Diffraction, 3rd edition. Prentice Hall, 2001. Chapters 3, 4,
and 9.

10