JOURNAL OF SOLID STATE CHEMISTRY 35, 167-175 (1980)

Neutron Powder Diffraction Study of the Structures of CeTaO,,

CeNbO,, and NdTaO,

A. SANTORO, M. MAREZIO,* R. S. ROTH, AND D. MINOR

National Measurement Laboratory, National Bureau of Standards,
Washington, D.C. 20234; and “Laboratoire de Cristallographie, CNRS,
38042 Grenoble Cedex, France

Received October 24, 1979; in revised form January 7, 1980

The crystal structures of CeTaO,, CeNbO,, and NdTaO, have been refined with the Rietveld method
using neutron powder diffraction data collected at room temperature. The results of these refinements
show that the coordination of the Ce and Nd cations is 8 with an average Ce-0 distance of 2.520 A in
CeTaO, and 2.480 A in CeNbO,, and an average Nd-0 distance of 2.451 A in NdTaO,. The Nb and Ta
cations in CeNbO, and NdTaO, are surrounded by six oxygen ions with Nb-0 distances varying
between 1.851 and 2.482 A, and Ta-0 distances varying between 1.861 and 2.353 A. The structural
arrangement of CeTaO, can be regarded as that of an oxidized ABO, perovskite. The difference in
structure of CeTaO, and NdTaO, is most likely due to the difference in size of the rare-earth cations.

Introduction features of the title compounds is desirable

for the understanding of the oxidation
Of the tantalates and niobates of rare- mechanisms leading to the formation of the
earth metals only those of Ce3+ oxidize in oxygen-rich phases and for conjecturing
air (Z-4), and the change of composition is possible phase relationships between stoi-
accompanied by changes of the structural chiometric and oxidized materials. Such
and physical properties of the compounds knowledge can be gained by using neutron
involved. Although the starting oxides have diffraction techniques, because the values
been characterized chemically, little is of the neutron scattering amplitudes of the
known about their detailed crystal struc- atoms involved do not differ greatly from
tures. X-Ray powder and precession pat- one another.
terns have shown (2) that CeTaO, is mono- A neutron diffraction study of CeTaO,,
clinic with a LaTaO,-type structure (5), CeNb04, and NdTaO, was therefore under-
while CeNbO, and NdTaO, have a fergu- taken and the results of the analysis are
sonite-like structure which is a monoclinic reported in the following sections.
distortion of scheelite (24). No
refinements of the structures were at- Experimental
tempted in these experiments, however,
because of the large differences in the X-ray The compounds were prepared by mixing
scattering factors of the oxygen and metal the appropriate amounts of tantalum,
atoms. niobium, cerium, and neodymium oxides
A detailed knowledge of the structural and by heating the mixtures overnight at
167 0022-45%/80/140167-09$02.00/O
Copyright @ 1960 by Academic Press. Inc.
All rights of reproduction in any form reserved.
168 SANTORO ET AL.

1000°C in platinum crucibles. After they TABLE I

were cooled to room temperature, the prod- EXPERIMENTALCONDITIONS USEDTO COLLECTTHE
ucts of this heat treatment were ground, POWDERINTENSITY DATA FORC~T~O,,C~N~O~,
heated again at 1400°Cfor CeNbO,, 1500°C AND NdTaO,
for CeTaO,, and 1600°Cfor NdTaO,, over a Monochromatic beam: reflection 220 of a Cu mono-
period of 16-20 hr; quenched in water, chromator in transmission geometry.
filtered, and dried. The only single crystals Mean neutron wavelength: 1.5416(3)A
obtained for these materials were too small Horizontal divergences”
for neutron diffraction and many of them (i) In-pile collimator: IO-min arc;
were twinned, especially in the case of (ii) Monochromatic beam collimator: 2@min arc;
(iii) Diffracted beam collimator: IO-min arc.
CeTa04. It was therefore decided to ana- Monochromator mosaic spread:” 1%min arc.
lyze the materials with the powder tech- Sample container: V can of l-cm diameter.
nique and to refine the structures with the
Rietvelt method (6). ” For a definition of these quantities see Caghoti er
Neutron diffraction measurements were al. (9).
made at room temperature on a powder
diffractometer at the National Bureau of initial values of the structural parameters
Standards Reactor, using the experimental were those obtained in the X-ray study of
conditions shown in Table I. As a general LaTaOI (5) for CeTaO, and those obtained
procedure, the contribution to the profile in the X-ray study of YNb04 (8) for
from the background is determined at por- CeNb04 and NdTaO,. In all cases individ-
tions of the powder pattern free from dif- ual isotropic temperature factors were used
fraction effects, and estimated by extrapo- in the calculations and were allowed to
lation from these regions to those where vary. The starting lattice constants were
overlapping is more severe. The reliability determined with the X-ray powder method
of this procedure depends on the materials and the initial values of the profile parame-
under study and on the adopted experimen- ters U, V, and W were calculated theoreti-
tal conditions. In our case the distribution cally (9). The total number of refined struc-
of the reflections and the resolution of the tural, lattice, and profile parameters was 33
diffractometer used for the intensity mea- for CeTaO, and 21 for CeNbO, and
surements leave many regions of the pow- NdTaO,. Refinement was continued until
der patterns free from diffracted intensity all shifts were less than 0.3o.
over the entire 28 angular intervals consid- From the early stages of the analysis it
ered. This is a favorable circumstance, as was apparent that the isotropic temperature
the reliability with which thermal parame- factor of the Nd atom in the structure of
ters are determined is strongly dependent NdTaO, was converging to an impossible
on the evaluation of background. The struc- value [b(Nd) = -0.19 A2]. Such abnormal
tural refinements were based on space behavior of the thermal parameter may be
group P2,/c for CeTaO, and 12/a’ for caused by the use of an improper neutron
CeNbO, and NdTaO,. The neutron scatter- scattering amplitude for the Nd atom, and,
ing amplitudes used were b(Ta) = 0.70, in fact, a check of the literature revealed
b(Ce) = 0.48, b(Nb) = 0.71, b(Nd) = that the value b(Nd) = 0.72 x lo-l2 cm used
0.72, and b(0) = 0.58 x lo-l2 cm (7). The in the calculations was too low. More rea-
sonable results and lower R factors were
obtained by using an updated amplitude of
1The standard setting for space group 12/a is C2/c. 0.77 x lo-** cm (IO), but b(Nd) still con-
The transformation matrix from our axes to the con- verged to an implausibly low value (0.13
ventional ones is (ooi/oio/lOt). A2). At this point b(Nd) was allowed to
 NEUTRON STUDY OF CeTaO,, CeNbO,, NdTaO, 169

TABLE II Recently, Sakata and Cooper (II) con-

COHERENTNEUTRONSCATTERINGAMPLITUDEOF cluded that in the Rietveld method the
Nd observations are incorrectly weighted and
the estimated standard deviations are
NW 0.72” 0.77* 0.80'
smaller by a factor of 2 or more than those
BW) -0.19(4) AZ 0.13(4) 0.31(4)
B(Ta) 0.70(5) 0.57(5) 0.50(S) obtained in a refinement based on inte-
B(01) 0.67(4) 0.52(4) 0.43(4) grated intensities. These conclusions, how-
NO3 0.71(4) 0.58(4) 0.51(4) ever, do not seem to be justified. As
R” 5.41 5.23 5.24 pointed out by Prince (12), the high preci-
4’ 9.76 9.41 9.42
11.07 10.73 10.70
sion of the profile method is a consequence
4
4, 7.08 7.06 7.06 of the fact that in this method each peak is
X 1.57 1.52 1.52 assumed to be Gaussian, so that each data
point becomes an independent measure-
(1Reference (7). ment of the same intensity. This results in a
b Reference (IO).
c From refinement of NdTaO,. more precise estimate of the intensity. In
c, X]l(obs) - 1(cal)l what follows, therefore, we quote the esti-
R=lOOx
i Xl(obs) I’ mated standard deviations obtained in the
I$(obs) - y(cal)j profile refinement.
R, = 100 x
1 Zy(obs) I’
Zwb(obs) - y(cal)]’ lR Results
R, = 100 x
1 Zw Iy(obs)]* I ’
(a) Profile Parameters
R, = 100 x (;;k-&g”.
The profile parameters obtained in the
In the above formulas N is the number of statistically refinements are shown in Table III, where
independent observations, P the number of parame- they are compared with those calculated
ters refined, C the number of constraints, I the inte- theoretically. The experimental values
grated intensities, and y the profile intensities of
weight W. agree closely with one another and with
those of other compounds studied with the
same experimental conditions, but they are
significantly higher than the theoretical
vary. Although correlation between the
values. Discrepancies of this kind have been
scattering amplitude and the temperature
factor was large (78%), the refinement pro-
ceeded to a stable minimum. A final value TABLE III
b(Nd) = 0.80 (1) x lo-l2 cm was obtained
PROFILEPARAMETERS
with reasonable thermal parameters for all
atoms in the structure. The results of these W
calculations are summarized in Table II. (It u V (deg’)
is worth noting that the atomic positional
in the three Calculated 0.212 -0.231 0.107
parameters obtained
CeTaOl 0.43(2) -0.47(2) 0.221(6)
refinements of the structure did not vary by CeNbO, 0.418(8) -0.47(l) 0.219(3)
more than two times the standard devia- NdTaO., 0.39(l) -0.45(2) 0.221(5)
tion.)
Refinements of the structures of CeNb04 Note. The parameters II, V, and W appear in the
and NdTa04 in the centric space group 12/a equation HZ = U tan28 + V tan 8 + W, where H is the
fulI width at half maximum of a reflection occurring at
were satisfactory; therefore, the centric the Bragg angle 0. The theoretical values of U, V, and
space group was assumed for the two com- W have been calculated with the expressions derived
pounds. by Caglioti et al. (9).
170 SANTORO ET AL.

TABLE IV tron study agree well with those determined

LATTICE PARAMETERS with the X-ray powder method. The
former, however, are systematically
CeTaOl NdTa04 CeNbO, smaller than the latter, as shown in Table
a
IV. This effect is probably the result of
Neutrons 7.6161(3) 5.5115(2) 55350(l) underestimating the neutron wavelength of
X rays 7.618(l) 5.520(l) 5.542(l) about 1 part in 1000. Clearly, the standard
b
deviation quoted in Table I has to be con-
Neutrons 5.5254(2) 11.2320(3) 11.3991(3) sidered a measure of precision, rather than
x rays 5.531(l) 11.244(2) 11.414(l) one of accuracy. The X-ray parameters,
c therefore, should be preferred over those
Neutrons 7.7588(3) 5.1112(l) 5.1590(l) obtained in the neutron refinement, and, in
X rays 7.767( 1) 5.117(l) 5.165(l) fact, these parameters were used in the
P calculation of the interatomic distances so
Neutrons 100.87(2) 95.71(l) 94.6ql) that none of the conclusions discussed in
X rays 100.94(2) 95.72(1) 94.6ql) the paper is affected by the error on the
wavelength.
Note. Theerrorsonthecelldimensionsdeterminedby
neutron diffractiondo not include the error in the neutron
wavelength.
(c) Structure of CeTa04
noted previously (23, 14) and are probably The results of the structure refinement of
caused by approximations made in deriving CeTa04 are shown in Table V. A projection
the original theory rather than by crystallite of the unit cell on the (010) plane is shown
size effects or structural distortions. in Fig. 1 and cation-oxygen distances are
given in Table VI. In the asymmetric unit
(b) Lattice Parameters there are one type of tantalum and one type
The lattice constants obtained in the neu- of cerium cation, whereas there are four
TABLE V
RESULTSOFTHETOTALPROFILEREFINEMENTOFTHE STRUCTUREOFCeTaO,a,b

Atom Position X Y Z B

Ce 4el 0.3441(6) 0.7728(7) 0.0999(5) 0.73(8)

Ta 4el 0.1668(4) 0.2670(6) 0.3045(4) 0.65(6)
O(l) 4el 0.1680(5) 0.1604(6) 0.0544(5) 0.69(8)
O(2) 4el 0.0568(5) 0.5869(5) 0.2067(5) 0.9W3)
O(3) 4el 0.3823(4) O&42(6) 0.3312(5) 0.52(7)
O(4) 4el 0.3338(4) 0.0084(6) 0.3671(4) 0.78(6)
R = 3.82, R, = 8.29, R, = 10.04; R, = 8.08.
Number of points above background: 1277.
2@initial) = ll.W, 28(final) = 94”, step = 0.05”.
Number of independent Bragg reflections: 290.
Parameters refined
(i) Structural (including scale factor) = 25,
(ii) Lattice (including zero point for 28) = 5,
(iii) Profile = 3.

n Figures in parentheses are standard deviations in the last decimal figure.

b Space group P&/c.
 NEUTRON STUDY OF CeTaO,, CeNbO,,NdTaO, 171

atoms of type O(l), two of type O(2), one of

type O(3), and one of type O(4). The aver-
age Ta-0 distance is 190 A. The octahe-
dra share only vertices among themselves;
they form slabs which are infinite in two
dimensions (along the b and c axes) and are
two octahedra thick in the third (the a*
axis). These are actually perovskite-like
slabs and the cerium atoms are located in
sites similar to the A sites of the AB03
perovskite structure. The cerium cations
are surrounded by eleven oxygen anions,
belonging to the same slab [three O(l), four
0 0 0 O(2), two O(3), and two O(4)]. The twelfth
FIG. 1. Projection on (010) of the structure of is replaced by two oxygen anions of the
CeTaO+ The Ta cations are located at the centers of adjacent slab. These two come closer to the
the distorted octahedra outlined in the figure. The y cerium cation because each slab is dis-
coordinate (X 100) is given for each atom.
placed with respect to the adjacent one by
half the height of an octahedron. The slabs
types of oxygen anion. The tantalum cat- are linked together by these interslab Ce-0
ions are located at the centers of the oxygen bonds. In order to form these bonds, the
octahedra which are outlined in Fig. 1. cerium cations move out of the centers of
Each octahedron is formed by two oxygen the 11-oxygen polyhedra. As a result, the
polyhedra become so distorted that only six
TABLE VI oxygen anions are nearest neighbors to the
CATION-OXYGEN cerium cation. The Ce-0 distances corre-
DISTANCES (A) IN
sponding to the interslab bonds (2.361 and
CeTaO.,
2.826 A) are of the same order of magnitude
Tantalum-octahedron as those of the intraslab Ce-0 bonds. Each
Ta-O( 1) 2.032(5) cerium cation is bonded to one 0( 1)) one
O(1) 1.980(5) O(2), three O(3), and three O(4) oxygen
O(2) 2.041(4)
atoms. The coordination number of the
O(2) I .960(5)
O(3) 2.013(4) cerium cations is thus 8, and the average
O(4) 1.914(S) Ce-0 distance is 2.520 A. The coordination
Average = I.990 number of the oxygen anions is 3 for 0( 1)
Cerium-polyhedron and O(2) and 4 for O(3) and O(4). The first
Ce-O( 1) 2.518(5) two are bonded to two Ta and one Ce
O(2) 2.688(6) cations, whereas the other two are bonded
O(3) 2.380(5) to one Ta and three Ce cations. If one
O(3) 2.361(5)"
2.545(6)
calculates the charge of these four oxygen
O(3)
O(4) 2.464(5) anions without taking into account the dif-
O(4) 2.826(6)" ferences in the anion-cation distances (an-
O(4) 2.375(5) ionic charge = z$Si, where Si, the bond
Average = 2.520 strength, is the ratio Z/n, Z is the cationic
* These are in- charge, and n the cationic coordination
terblocks Ce-0 dis- number), one finds for 0( 1) and O(2), CiSi =
tances. 2.04 and for O(3) and O(4), ZiSi = I.%.
172 SANTORO ET AL.

These values indicate that only a small part (d) Structures of CeNbO, and NdTaO,
of the distortion is needed to balance the The results of the profile refinements of
electrostatic charge. CeNbO, and NdTaO, are given in Tables
It was pointed out recently (I, 2) that VII and VIII and relevant cation-oxygen
CeTaO, can absorb oxygen according to distances in Tables IX and X. The two
the oxidation-reduction reaction compounds are isostructural and, there-
fore, many of the considerations made for
CeTaO, + 1/202 + CefL.CeWa04+,. CeNbO, also apply to NdTaOl. Figure 2
shows the arrangement of cerium and
The x parameter is variable and tempera- niobium polyhedra in CeNbO,, as viewed
ture dependent. Three distinct ranges in x down the monoclinic b axis. This structure
(x = 0.49, x = 0.37, x = 0.12) were found to is a monoclinic distortion of the tetragonal
yield three different phases. structure of scheelite, CaW04.
The structural arrangment of CeTaOl can There are on cerium, one niobium, and
also be regarded as that of an oxidized two oxygen ions in the asymmetric unit.
ABO, perovskite. An extra oxygen layer is The cations have a distorted, face-centered
inserted in the structure every three anion cubic arrangement. The cerium cations are
layers, and this causes the separation of the surrounded by eight oxygen ions, four O( 1)
perovskite blocks into finite two-octahe- and four O(2), forming two interpenetrated
dron-thick slabs. For instance, it can be tetrahedra. The average Ce-0 distance is
seen from Fig. I that if the oxygen anions 2.480 A. The niobium cations are sur-
O(3) at y = 1.02 were coincident with O(4) rounded by four close oxygen ions, two
at y = 0.99, and if O(4) at y = 0.49 were O(1) and two O(2), arranged as a dis-
coincident with O(3) at y = 0.52, the struc- torted tetrahedron. The average Nb-0
tural arrangement would be truly an ABO3 distance over these four neighbors is
perovskite-like arrangement. When these 1.886 A. There are four other oxygen ions
oxygen anions are not coincident, the ABO, around the niobium cations, each one cap-
compound is obtained instead. ping a face of the closest tetrahedron. Two

TABLE VII
RESULTSOFTHETOTALPROFILEREFINEMENTOFTHE STRUCTUREOFCeNbO,”

Atom Position x Y Z B

Ce 4e2 a 0.1202(2) 0 0.41(S)

Nb 4e2 f O&470(1) 0 0.40(4)
O(l) W 0.0129(3) 0.7171(1) 0.2031(3) 0.68(3)
O(2) 8fl 0.8999(3) 0.4553(1) 0.2399(3) O&O(3)
R = 3.32, R, = 7.33, R, = 8.94, R, = 6.34.
Number of points above background: 1520
2B(initial) = 14.5”, 28(final) = 111.7”, step = 0.05”.
Number of independent Bragg reflections: 212.
Parameters refined
(i) Structural (including scale factor) = 13,
(ii) Lattice (including zero point for 28) = 5,
(iii) Profile = 3.

I1Space Group 12/a.

 NEUTRON STUDY OF CeTaO.,, CeNbO,, NdTaO, 173

TABLE VIII
RESULTS OF THE TOTAL PROFILE REFINEMENT OF THE STRUCTURE OF NdTaO,”

Atom Position X Y Z B

Nd 4e7 a 0.1178(2) 0 0.31(4)

Ta 4e2 a 0.6515(2) 0 OSO(5)
O(l) 8fl 0.0152(4) 0.7188(2) 0.2165(5) 0.43(4)
00) 8fl 0.9014(4) 0.4563(2) 0.2396(4) 0.51(4)
R = 5.24, R, = 9.42, R, = 10.70, R, = 7.06.
Number of points above background: 1516.
20(initial) = 15.0”, 28(final) = 114.5”, step = 0.05”.
Number of independent Bragg reflections: 220.
Parameters refined:
(i) Structural (including scale factor) = 13,
(ii) Lattice (including zero point for 28) = 5,
(iii) Profile = 3.

‘1Space group 12/0.

of these oxygen ions are close enough (the sten cations because the two sets of cation-
corresponding Nb-0 distance being 2.482 oxygen distances in the calcium case are
A) that they may be considered bonded to very close, 2.44 and 2.48 A.
the niobium atoms, whereas the other two One possible way of looking at the
are removed much farther and cannot be scheelite structure is to consider it as a
considered as involved in the bonds. In the distorted superstructure of fluorite, CaF,.
scheelite CaW04 structure, the eight oxy- The fluorite structure contains a simple
gen ions surrounding the tungsten cation cubic array of anions so that all cations
are arranged as two interpenetrated tetra- which form a perfect face-centered cubic
hedra of very different sizes. The two sets lattice have an eightfold coordination. In
of W-O distances are 1.78 and 2.91 A (1.5). the scheelite structure, the cations are con-
The calcium cations are also surrounded by siderably distorted from the fee lattice and
eight oxygen ions but differ from the tung- the anions move away from the simple

TABLE IX
CATION-OXYGEN DISTANCES
(A) IN CeNbO,

Cerium polyhedron
Ce-O( 1) x2 2.461(2)
O(1) x2 2.535(2)
O(2) x2 2.428(2)
O(2) x2 2.494(2)
Average = 2.480
Niobium polyhedron
FIG. 2. Projection of (010) of the structure of Nb-O(l) x2 1.920(2)
CeNbO,. The solid lines identify the Nb polyhedron, O(2) x2 1.851(2)
the broken lines the Ce polyhedron. They coordinate O(1) x2 2.482(2)
(x 100)is indicated for each atom. The relevant Ce-0 Average = 2.084
and Nb-0 distances are also given.
174 SANTORO ET AL.

TABLE X into account the individual distances, the

CATION-OXYGEN DISTANCES former anion is overbonded (&Si = 2.42),
(A) IN NdTaO, whereas the latter one is underbonded (&Si
= 1.58). Therefore, in this structure the
Neodymium polyhedron
2&l(3)
differences in these distances are mainly
Nd-O(1) x2
O(1) x2 2.534(3) due to balancing the electrostatic charge.
O(2) x2 2.380(2) As we mentioned earlier, NdTaO, and
O(2) x2 2.449(3) CeNbO, are isostructural. A di@erence
Average = 2.451 which is worthwhile to point out lies in the
Tantalum polyhedron coordination of the pentavalent cations. As
Ta-O(1) x2 1.941(2) can be seen from Tables IX and X the first
O(2) x2 1.861(3) two sets of M5+-0 distances increase (from
O(1) x2 2.353(3)
Average = 2.052
1.851 and 1.920 to 1.861 and 1.941 A),
whereas the third set decreases (from 2.482
to 2.353 A) on going from the niobate to the
cubic positions in the following manner: tantalate. These values indicate that the
four of the eight anions around the posi- oxygen atoms in this third set behave, at
tions occupied by the small cations move least in the niobate, as second-nearest
closer to these cations and in this way a neighbors. Therefore the coordination
tetrahedral coordination is achieved for the number of these cations should be taken as
small cations. The remaining four anions 4 + 2, although the Ta5+ ion appears to be
move farther away from these cations in approaching sixfold coordination.
such a way that the large cations still have The change in structure which takes
an eightfold coordination. The resulting place in the tantalum series is probably due
polyhedron around the large cations has to the size of the rare-earth cation. When
mm symmetry, i.e., lower than the m3m this size becomes too large for the A site of
symmetry of the cubic structure. From this the fergusonite-like structure, the tantalates
point of view the fergusonite-like structures crystallize with the LaTaOl-type structure.
could be considered as intermediate be- The same structural change does not take
tween the fluorite and the scheelite struc- place in the niobates because of the prefer-
tures, even though the fluorite is cubic, the ence of the Nb5+ cations for a coordination
scheelite is tetragonal, and the fergusonite- smaller than 6.
like structures have monoclinic symmetry.
In the latter type of structure, two anions of
the four forming the larger tetrahedron References
around the small pentavalent cations, come
closer to Nb5+ cations in CeNbO, and to the T. NEGAS, R. S. ROTH, C. L. MCDANIEL, H. S.
Ta5+ cations in NdTaOl. It should also be PARKER, AND C. D. OLSON, Mater. Res. Bull. 12,
1161 (1977).
pointed out that in the scheelite structure R. S. ROTH, T. NEGAS, H. S. PARKER, D. B.
the cubic arrangement of the anions is so MINOR, AND C. JONES, Mater. Res. Bull. 12, 1173
distorted that it is better described as a (1977).
puckered hexagonal network. R. S. ROTH, T. NEGAS, H. S. PARKER, D. B.
In the structure of CeNbOl, the oxygen MINOR, C. D. OLSON, AND C. SKARDA, in “The
Rare Earths in Modem Science and Technology”
atom O(1) is surrounded by two niobium (G. J. McCarthy and J. J. Rhyne, Eds.), p. 163,
and two cerium atoms and the O(2) atom is Plenum, New York/London (1978).
surrounded by one niobium and two cerium R. J. CAVA,T. NEGAS, R. S. ROTH, H. S. PARKER,
atoms. Consequently, if one does not take D. B. MINOR AND C. D. OLSON, in “The Rare
 NEUTRON STUDY OF CeTaOl, CeNbO,, NdTaOl 175

Earths in Modem Scienceand Technology” (G. J. 10. W. B. YELON, Neutron Diffraction Newsletter
McCarthy and J. J. Rhyne, Eds.), p. 181,Plenum, (1977).
New York/London (1978). II. M. SAKATAAND M. J. COOPER,MPD/NBS/95
5. T. A. KIROVA AND V. B. ALEXANDROV, Dokl. AERE, Harwell, July 1978.
Akad. Nauk USSR 201, 1095(1971). 12. E. PRINCE,submittedfor publication.
6. H. M. RIETVELD,J. Appl. Ctystallogr. 2, 65 13. G. CAGLIOTIAND F. P. RICCI, Nucl. Instrum.
( 1969). Methods 15, 155(1962).
7. G. E. BACON,Acta Crystallogr. A 28, 357 (1972). 14. A. SANTORO, R. S. ROTH, AND D. B. MINOR,
8. C. T. PREWITT,private communication. Acta Crystallogr. B 33, 3945 (1977).
9. G. CAGLIOTI,A. PAOLETTI,AND F. P. Rrccr, 15. A. ZALKIN AND D. H. TEMPLETON, J. Chem.
Nucl. Instrum. 3, 223 (19.58). Phys. 40, 501(1964).