180

J. Appl. Cryst. (1993). 26, 180-184

X-ray Powder Diffraction Analysis of Silver Behenate, a Possible Low-Angle Diffraction

Standard
BY T. C. HUANG
IBM Research Division, Almaden Research Center, 650 Harry Road,
San Jose, CA 95120-6099, USA

H. TORAYA
Ceramics Research Laboratory, Nagoya Institute of Technology, Asahigaoka,
Tajimi 507, Japan

T. N. BLANTON
Analytical Technology Division, Eastman Kodak Company, Kodak Park B49,
Rochester, N Y 14652-3712, USA

AND Y. Wu
IBM Research Division, Almaden Research Center, 650 Harry Road,
San Jose, CA 95120-6099, USA

(Received 7 July 1992; accepted 18 September 1992)

Abstract sions and synthetic modulated multilayers with large

Silver behenate, a possible low-angle diffraction
standard, was characterized using the powder
diffraction technique. Diffraction patterns obtained
with 1.54 A synchrotron and Cu K~ radiations showed
thirteen regularly spaced (00/) peaks in the range
1.5-20.0°20. With the National Institute of Standards
and Technology's standard reference material silicon
as an internal standard, the long spacing of silver
behenate was accurately determined from the
profile-fitted synchrotron diffraction peaks, with
doo 1 = 58.380 (3) A. This result was in agreement with
that obtained from the Cu K~ pattern. The profile
widths of the silver behenate peaks were found to be
consistently larger than those of the silicon peaks, indi-
cating significant line broadening for silver behenate.
The average crystallite size along the long-spacing
direction of silver behenate was estimated using the
Scherrer equation, giving Davg = 900 (50),~. Because
silver behenate has a large number of well defined
diffraction peaks distributed evenly in the 1.5-20.0°20
range, it is suitable for use as an angle-calibration
standard for low-angle diffraction. However, care must
be taken if silver behenate is to be used as a peak-profile
calibration standard because of line broadening.
Introduction
In recent years, it has been necessary to analyze
polycrystalline materials with large unit-cell dimen-
0021-8898/93/020180-05506.00
layer periodicities. X-ray powder diffraction is an
indispensable method for the structural characteriza-
tion of these materials. Since the diffraction patterns
of these materials often have strong peaks at low 20
angles, down to a few degrees or less, proper
alignment/calibration of the instrumentation is
essential for a successful diffraction analysis. The
importance of using a low-angle standard was studied
previously by Wong-Ng & Hubbard (1987) and it was
concluded that a standard should have at least one
peak below the lowest-20 peak of the sample so that
an adequate calibration in the low-angle region is
ensured. The National Institute of Standards and
Technology (NIST) has produced and issued several
standard reference materials (SRMs) for powder
diffraction measurements, one of which, SRM 675
(fluorophlogopite, KMg3Si3AIOloF2) , is intended for
use as a low-angle (large d spacing) calibration
standard. SRM 675, however, has only two peaks
below 20 °, at 8.85 and 17.75 °, with d spacings of 9.981
and 4.9905 A, respectively (Hubbard, 1983a; Dragoo,
1986). Because of the recent interest in the
characterization of materials with very large d
spacings/periodicities, it is necessary to develop a new
low-angle diffraction standard which has well defined
diffraction peaks down to very small 20 angles of a
few degrees and less.
A powder diffraction pattern for silver behenate
[CH3(CHz)2oCOOAg] was first reported by Mat-
thews, Warren & Michell (1950). However, the data
(~i~ 1993 International Union of Crystallography
 T. C. H U A N G , H. TORAYA, T. N. B L A N T O N A N D Y. WU
were collected with a powder camera having a
d-spacing cutoff of 24 A for Cr K0t radiation. The use
of silver behenate as a possible low-angle standard
was introduced by one of the authors (Blanton, 1991),
who showed that silver behenate has a set of regularly
spaced Cu K~ diffraction peaks at low 20 angles down
to 1.5°. A repeat axis length of about 58.7 A (without
an internal standard) was determined using a
conventional powder diffractometer. Because it has
well defined low-angle diffraction peaks covering the
entire low-20 range, silver behenate is being considered
by NIST as a candidate for a large-d-spacing powder
diffraction SRM (Cline, 1991).
This paper reports the X-ray diffraction results of
analyzing silver behenate using both synchrotron and
conventional Cu K0~ X-rays. The use of silver behenate
as a possible low-angle calibration standard is also
discussed.
Experimental
Silver behenate powders were supplied by the Eastman
Kodak Company. A scanning-electron-microscope
(SEM) examination of the powders revealed small
platelike crystals with surface dimensions of about
0.2-2.0 ~tm and thicknesses of approximately 1000 A
and less (see Fig. 1). SRM silicon powders (640a for
synchrotron and 640b for Cu K~ measurements) were
used as internal standards (Hubbard, 1983b). The two
Fig. 1. SEM micrograph of silver behenate powders.
181
powders of silver behenate and silicon were thoroughly
mixed (1:1 weight ratio) before being loaded onto
sample holders. The white plate-like silver behenate
powders became oriented very easily with the
application of just a little pressure to the top of the
sample.
Powder diffraction patterns of the silver behenate
and silicon mixture were obtained using both 1.54
synchrotron radiation at the Stanford Synchrotron
Radiation Laboratory and conventional Cu K~
radiation at the Ceramics Research Laboratory of the
Nagoya Institute of Technology, Japan. A high-
intensity vertical-scan diffractometer with a set of very
long parallel horizontal slits with a 0.05 ° aperture in
the diffracted beam was used to collect the synchrotron
diffraction data (Parrish & Hart, 1987) and a
high-resolution vertical-scan diffractometer with a 1°
divergent slit and a 250 mm goniometer radius was
used to collect Cu K~ X-ray diffraction data (Toraya
& Parrish, 1992). Both diffractometers used multiple
parallel slits with a 2 ° aperture to limit possible axial
divergence.
The profile shapes of individual synchrotron and
Cu K~ X-ray diffraction peaks of silver behenate were
analyzed using the PROFIT program (Toraya, 1986).
The precision (or reproducibility) for the peak-position
determination obtained by PROFIT was reported to
be in the range 0.0001-0.0003°20 (Toraya & Parrish,
1992). The 20 positions obtained by PROFIT were
used for the least-squares refinement of the d spacings
of silver behenate. Systematic errors were corrected
using silicon internal standards with asi = 5.43076 A
for SRM 640a (Parrish, Hart, Huang & Bellotto, 1987)
and 5.43081 A for SRM 640b. lit should be noted that
the value of asi for SRM 640b used in this study is
lower than that reported by NIST by a factor of
1.540562/1.5405981. This is because the Cu K~ 1
wavelength of 1.540562 A, obtained by Bearden (1967),
adopted by the International Union of Crystal-
lography and included in International Tablesfor X-ray
Crystallography (1974), was used in this study, while
the value of 1.5405981 A was used in the NIST
calculation.] The displacements in 20 caused by
systematic errors were corrected by a polynomial of
tan 0 as follows:
ASE(20 ) = C O + C~ tan 0 + C2 tan 2 0 + C 3 tan 3 0,
(1)
where Co, C~, C2 and C 3 a r e constants determined
by the least-squares refinement of 20 peak positions
obtained by profile fitting (Toraya & Kitamura, 1990).
The Bragg law shows
2 = 2dsi sin 0Si = 2dsB sin 0sa, (2)
where SB denotes silver behenate. Since the d spacings
of the silicon internal standard are fixed from the
accurately known value of asi, (2) can be used to
182 SILVER B E H E N A T E , A P O S S I B L E L O W - A N G L E D I F F R A C T I O N S T A N D A R D

determine the d spacings of silver behenate. An Table 1. Profile finin9 of synchrotron diffraction peaks
assumption about or a measurement of X-ray of silver behenate (SB) and silicon
wavelength is not required for the analysis (Parrish,
Material hkl 20exp C) FWHM ( ) qt, qn
Hart, Huang & Bellotto, 1987). This method is
especially useful for the analysis of synchrotron SB 001 1.5114 0.132 1.18 0.63
diffraction data where the exact value of the 002 3.0160 0.114 1.17 0.59
003 4.5289 0.108 1.02 0.59
wavelength is not known in advance. It has been shown
004 6.0422 0.102 1.03 0.68
previously that lattice parameters for powder materials 005 7.5575 0.101 0.98 0.68
can be measured to a few parts in 106 using synchrotron 006 9.0733 0.100 0.95 0.74
radiation (Hart, Cernik, Parrish & Toraya, 1990; 007 10.5922 0.102 0.93 0.77
Huang, Parrish, Masciocchi & Wang, 1990) and 008 12.1106 0.102 0.93 0.87
009 13.6320 0.104 0.89 0.81
conventional Cu K~ radiation (Toraya & Parrish, 0,0,10 15.1563 0.104 0.89 0.81
1992). The widths of the fitted profiles were used to 0,0,11 16.6836 0.105 0.89 0.81
examine line broadening and to estimate the average 0,0,12 18.2107 0.114 0.88 0.81
crystallite size of silver behenate. 0,0,13 19.7393 0.139 1.24 0.79
Si 111 28.4288 0,051 0.89 0.79
220 47.2665 0.058 0.95 0.80
Results and discussion 311 56.0732 0.064 0.96 0.82
400 69.0581 0.071 1.07 0.94
The low-angle section of the 1.54/i synchrotron 331 76.2903 0.071 0.94 0.83
diffraction pattern of silver behenate is plotted in Fig.
2. There are 13 well defined (00l) peaks of silver functions and it was found that each peak could be
behenate at 20 _< 20 ~. The silver behenate peaks began fitted by a split pseudo-Voigt function consisting of
to appear at 1.5 ~, continued at regular intervals of equal-width Lorentzian and Gaussian profiles on both
--- 1.5 ~' and ended at 19.T. All peaks except the (0,0,13) high- and low-angle sides (i.e. HWHML = H W H M , ,
peak at 19.7~ were free of overlaps. The first five peaks where H W H M is the half-width at half-maximum),
with 20 below 8 :; were very intense and the (001) had but with a higher Lorentzian content on the low-angle
the strongest intensity. The mixture of silver behenate side of the peak (Toraya & Parrish, 1992). Fig. 3 shows
and silicon has 18 well defined peaks covering the comparisons between the experimental and the
front diffraction region from 1.5 to 76.4~20. fitted profiles of the silver behenate (001), (007) and
(0,0,13) synchrotron diffraction peaks. The split
Profile analysis pseudo-Voigt function was also used successfully in
profile fitting the synchrotron peaks of silicon.
It is well known that high-resolution synchrotron The peak positions (20exp) , full widths at half-
diffraction peaks normally have single symmetric maxima (FWHMs) and Lorentzian contents at the
profiles (Huang, Hart, Parrish & Masciocchi, 1987). low- and the high-angle sides (rh. and I'/,) obtained by
A careful inspection of the peak shapes, however, the PROFIT program are listed in Table 1. Values of
revealed that the profiles for silver behenate peaks at 20 are given to four decimal places because of the high
low 20 angles are slightly asymmetric with relatively precision of the PROFIT analysis. Values of ~/L were
long tails at the low-angle sides. Profile asymmetry is about twice as large as those of qn for the (001) and
quite apparent in the (001) and (002) peaks, which can the (002) peaks because of the asymmetric profiles with
be attributed to severe axial divergence at low
20 (< 3.0°). The synchrotron diffraction profiles were (001) ((X)7 (0013)
analyzed by fitting with different pseudo-Voigt

O01 ) (0O5)

I x 12~

(002) (006) (007)

' I , , I
L~j~((;8) (009) (0011)
IO010) (0012) ((X)]3) A -- ~ .............. ~',, . . . . . -~.,,;i......4,vt'j~/~k,~-~..,,,-~'..-
, , , , , , ,L ,~ , ~ T ~ ~
20 L £---
5 10 15
I 2 10 I1 19 20 20
20( )
Fig. 3. Profile fitting of the (001}, (007) and (0,0,131 synchrotron
Fig. 2. The low-angle section of the 1.54/k synchrotron diffraction diffraction peaks of silver behenate. The squares are the
pattern of silver behenate. experimental data, curves the fitted profiles and A the differences.
 T. C. HUANG, H. TORAYA, T. N. BLANTON AND Y. WU 183

Table 2. Profile fitting o f Cu Kot ~X-ray diffraction peaks Table 3. d-spacing refinement of synchrotron peaks
Material hkl 20,~p(°) FWHM(°) qL qn Material hkl 20expC) 20¢o,(':) 20c~.C) A20 C)
SB 001 1.5222 0.127 1.43 0.68 SB 002 3.0160 3.0221 3.0216 0.0005
002 3.0354 0.114 0.97 0.59 003 4.5289 4.5331 4.5330 0.0001
003 4.5308 0.112 1.06 0.56 004 6.0422 6.0447 6.0453 -0.0006
004 6.0566 0.109 1.09 0.78 005 7.5575 7.5583 7.5586 -0.0003
005 7.5588 0.114 1.07 0.56 006 9.0733 9.0726 9 . 0 7 3 2 -0.0006
006 9.0865 0.110 1.03 0.60 007 10.5922 10.5901 10.5894 0.0007
007 10.5893 0.113 1.03 0.61 008 12.1106 12.1071 12.1075 -0.13003
008 12.1211 0,109 1.01 0.71 009 13.6320 13.6273 13.6277 -0.0003
009 13.6324 0.115 1.08 0.71 0,0,10 15.1563 15.1505 15.1503 0.0003
0,0,10 15.1672 0.111 1.08 0.71 0,0,11 16.6836 16.6770 16.6756 0.0013
0,0,11 16.6827 0.115 1.08 0.71 0,0,12 18.2107 18.2030 18.2039 -0.0008
0,0,12 18.2247 0.082 1.08 0.71 111 28.4288 28.4176 28.4170 0.0006
0,0,13 19.7430 0.166 0.59 0.59 220 47.2665 47.2581 47.2589 -0.0009
Si 111 28.4348 0.071 0.58 0.44 311 56.0732 56.0691 56.0690 0.0001
220 47.3002 0.077 0.76 0.58 400 69.0581 69.0617 69.0611 0.0006
311 56.1221 0.088 0.87 0.62 331 76.2903 76.2972 76.2975 -0.0003
400 69.1122 0.101 0.96 0.76 IA201a~g=0.00051
331 76.3633 0.101 0.85 0.80 dool = 58.380(3)
422 88.0186 0.111 0.94 0.82
Table 4. d-spacing refinement of Cu Ko~l diffraction
relatively long tails on the low-angle sides. The largest
peaks
value of r/L was obtained for the (0,0,13) peak probably Material hkl 20expC) 20¢orC) 20ca,() A20 (~)
because of severe overlap with two other non-(00/)
SB 002 3.0354 3.0335 3.0246 0.0089
peaks at lower angles (see Fig. 3). 003 4.5308 4.5298 4 . 5 2 7 6 -0.0078
D u r i n g the analysis of the Cu Kct X-ray diffraction 004 6.0566 6.0565 6.0514 0.0051
peaks, it was found that the (001) peak was strongly 005 7.5585 7.5592 7.5662 -0.0070
interfered with by the incident beam when using a 1~ 006 9.0865 9.0879 9.0823 0.0056
007 10.5893 10.5913 10.6001 -0.0088
divergence slit. The (001) peak obtained with a small 008 12.1211 12.1237 12.1197 0.0040
divergent slit of (1/6) ° was thus used in the 009 13.6324 13.6355 13.6414 -0.0059
profile-fitting analysis. The results are listed in Table 0,0,10 15.1672 15.1708 15.1656 0.0052
2. It should be noted that the 20 values obtained by 0,0,11 16.6827 16.6867 16.6925 -0.0058
the P R O F I T p r o g r a m are the positions for the Cu 0,0,12 18.2247 18.2291 18.2224 0.0067
Si 111 28.4348 28.4403 28.4424 -0.0021
K ~ peak (Toraya & Parrish, 1992). The 20 positions 220 47.3002 47.3048 47.3028 0.0020
for the diffraction peaks listed in Table 2 are 311 56.1221 56.1268 56.1224 0.0044
consistently larger than those of the synchrotron peaks 400 69.1122 69.1216 69.1301 -0.13085
listed in Table 1. This result indicates that the 331 76.3633 76.3804 76.3763 0.0041
wavelength used in the synchrotron experiment was IA201~vg=0.00573
significantly shorter than that of Cu K71. The dool = 58.373(27) A
exceptionally small value of the F W H M (full width at
half-maximum), 0.082 °, for the (0,0,12) peak was by the refinement) and the peak positions calculated
probably caused by a poor p e a k - t o - b a c k g r o u n d ratio. from the refined d spacings (20cal) is excellent. The
Similarly to the synchrotron analysis, the results for average and m a x i m u m absolute values of A20
the (0,0,13) peak were less reliable because of peak (20¢or -- 20ca~) were 0.00051 and 0.0013 ~', respectively.
The value of dool was determined to be 58.380 (3)A
overlaps.
(the value in parentheses is the e.s.d., a). A least-squares
refinement using all 13 silver behenate (00/) peaks
Lattice spacings including the asymmetric (001) and the overlapped
The (00/) peak positions of silver behenate obtained (0,0,13) was also conducted and virtually the same
by profile fitting were used to determine the values value of dool was obtained. The wavelength of the
ofdoo~, the 'long spacings' (Vand, Aitken & Campbell, synchrotron radiation, determined from the peak
1949; A b r a h a m s s o n & von Sydow, 1954). The results positions of the silicon internal standard, was
of the least-squares refinement of 20~p for 11 (00/) 1.539200,~, which is significantly smaller than the
synchrotron peaks, omitting those of the asymmetric 1.540562 ,~, of the Cu K ~ radiation.
(001) and the overlapped (0,0,13) peaks, are listed in A least-squares refinement using the same 1 1 silver
Table 3. The agreement between the peak positions behenate (00/) peaks obtained with Cu K~ radiation
corrected for systematic errors (20co~ = 20~xp + ASE, was also conducted and the results are listed
where ASE is the total of the systematic errors obtained in Table 4. The agreement between the values of 20cot
184 SILVER BEHENATE, A POSSIBLE LOW-ANGLE DIFFRACTION STANDARD
and 20c, i is good (Id20lavg = 0.00573 and A 2 O m . ~ =
0.0089°). It should be noted that the values of A20 are
an order of magnitude larger than those of the
synchrotron results (see Table 3), probably because of
the Cu Ku~ 2 doublets. The value ofdoo I refined from
the Cu K~I peaks was 58.373 (27)A. The difference
between the values of doo~ obtained from the
synchrotron and Cu K~ radiations is 0.007 A, which
is well within the standard deviation (o" = 0.027 A)
obtained from the Cu K0¢ results.
Since the results obtained from the synchrotron data
had excellent precision, with IA201avg = 0.00051 ° and
o ' = 0 . 0 0 3 A , the value of doo~ = 5 8 . 3 8 0 ( 3 ) A is
recommended for silver behenate.
Line broaden&g
An examination of the values of F W H M s for silver
behenate and silicon listed in Tables 1 and 2 revealed
significant broadening of the silver behenate (001)
profiles. The instrumental broadening at low angles
was extrapolated from the F W H M s of the silicon
peaks. After correcting for instrumental broadening,
the line broadening for silver behenate calculated from
the (007) peak at 10.6°20 was found to be 0.067 and
0.060 ° for the synchrotron and Cu K~ samples,
respectively. It is well known that profile broadening
is inversely proportional to the crystallite size
measurement perpendicular to the surface of a sample.
Using the Scherrer equation (Langford & Wilson,
1978), the crystallite size calculated from the
broadening of the (007) peaks was estimated to be 850
and 950 A for the synchrotron and Cu K~ samples,
respectively. The average crystallite size, Oavg, was
900 (5) A. The results are in good agreement with the
thicknesses of the plate-like silver behenate crystals
observed from the SEM micrograph shown in Fig. 1.
Concluding remarks
It has been shown that silver behenate has a set of
well defined and regularly spaced (001) diffraction peaks
distributed evenly in the low-angle region. The value
of doo I determined from synchrotron diffraction data
was 58.380 (3)A. This material is suitable for use as
an external and an internal low-angle calibration
standard for the diffraction analysis of polycrystalline
materials with big unit-cell dimensions and modulated
multilayers with large layer periodicities. The use of
silver behenate as a calibration standard greatly
extends the coverage of the NIST SRMs down to very
low 20 angles. Care must be taken, however, if silver
behenate is to be used as a diffraction-profile
calibration standard, because of line broadening
arising from small crystallite sizes along the
long-spacing direction.
The authors thank Dr G. Castro of IBM and the staff
of the Stanford Synchrotron Radiation Laboratory for
their aid in carrying out the synchrotron experiment,
J. Vazquez of IBM for the SEM micrograph and J.
Laco of Eastman Kodak for the preparation of silver
behenate powder.
References
ABRAHAMSSON, S. • VON SYDOW, E. (1954). Acta Cryst. 7,
591-593.
BEARDEN, J. A. (1967). Rev. Mod. Phys. 39, 78-124.
BLANTON, T. (1991). Presented at the ICDD-JCPDS
Annual Meeting, Concordville, PA, USA.
CLINE, J. P. (1991). Personal communication.
DRAGOO, A. L. (1986). Powder Diffr. 1,294-298.
HART, M., CERNIK, R. J., PARRISH, W. 8z TORAYA, H. (1990).
J. Appl. Crvst. 23, 286-291.
HUANG, T. C., HART, M., PARRISIt, W. 8~ MAS('IOCCttl, N.
(1987). J. Appl. Phys. 61, 2813-2816.
HUANG, T. C., PARRISH, W., MASCIOCCHI,N. 8¢. WANG, P. W.
(1990). Adv. X-ray Anal. 33, 295-303.
HUBBARD, C. R. (1983a). Adv. X-ray Anal. 26, 45-51.
HUBBARD,C. R. (1983b). J. Appl. Cryst. 16, 285-288.
International Tables for X-ray Crystallography (1974). Vol.
IV. Birmingham: Kynoch Press. (Present distributor
Kluwer Academic Publishers, Dordrecht.)
LANGFORD, J. I. & WILSON, A. J. C. (1978). J. Appl. Crvst.
11, 102-113.
MATTHEWS, F. W., WARREN, G. G. & MICHELL, J. H. (1950).
Anal. Chem. 22, 514-519.
PARRISH,W. 8¢ HART, M. (1987). Z. Kristallogr. 179, 161-173.
PARRISH,W., HART,M., HUANG,T. C. & BEHOTTO, M. (1987).
Adv. X-ray Anal. 30, 373-382.
TORAYA, H. (1986). J. Appl. Crvst. 19, 440-447.
TORAYA, H. 8¢ KITAMURA, M. (1990). J. Appl. Ct:vst. 23,
282-285.
TORAYA, H. 8¢ PARRISlt, W. (1992). Adv. X-ray Anal. 35,
431-438.
VAND, V., AIrKEN, A. & CAMPBELL, R. K. (1949). Acta Ct3'st.
2, 398-403.
WONG-NG, W. 8z HUBBARD,C. R. (1987). Powder D(ffr. 2,
242-248.