Ba3AlO4H: Synthesis and Structure of a New Hydrogen-Stabilized Phase

Article abstract of DOI:10.1006/jssc.1998.8022

The title compound and its deuteride were synthesized in high yield by sintering appropriate mixtures of BaH₂(BaD₂), BaO and Al₂O₃in welded Ta containers at 1100°C under a hydrogen (deuterium) pressure of 600-700 torr. The structure was determined by a combination of single crystal X-ray and time-of-flight powder neutron diffraction means (orthorhombicPnma,Z=4,a=10.4911(9),b=8.1518(7),c=7.2399(6) A for the deuteride). The compound contains nominal Ba²⁺cations, AlO^5-₄complex anions and H^-(D^-) anions. The hydride is bound in a strongly distorted barium octahedron while the tetrahedral AlO^5-₄anion is surrounded by eleven barium atoms in a peculiar polyhedral configuration. The HBa₆octahedra are linked together by corner-sharing to form a three dimension framework with large voids defined by eight of the HBa₆octahedra that are occupied by the tetrahedral AlO^5-₄anions. The structure is closely related in a hierarchical way to that of orthorhombic inverse perovskitic Ba₃GeO.

Full text of DOI:10.1006/jssc.1998.8022

JOURNAL OF SOLID STATE CHEMISTRY 141, 570—575 (1998)
ARTICLE NO. SC988022
Ba3AlO4H: Synthesis and Structure of
a New Hydrogen-Stabilized Phase
Baoquan Huang and John D. Corbettꢀ
Ames Laboratory-DOEꢁ and Department of Chemistry, Iowa State University, Ames, Iowa 50011
Received May 21, 1998; in revised form August 25, 1998; accepted September 7, 1998
Ba Ga (9), and ternary suboxides such as Ba ¹ O
ꢀꢆ
ꢁꢀ ꢁ ꢂ
(¹"Ge, Si, Ga, In Tl) (10, 11) because hydrogen as an
The title compound and its deuteride were synthesized in high
yield by sintering appropriate mixtures of BaH2(BaD2), BaO
interstitial or anionic component had not been recognized
and Al2O3 in welded Ta containers at 1100 °C under a hydrogen or considered.
(deuterium) pressure of 600–700 torr. The structure was deter-
mined by a combination of single crystal X-ray and time-of-flight
powder neutron diffraction means (orthorhombic Pnma, Z ؍ 4,
a ؍ 10.4911(9), b ؍ 8.1518(7), c ؍ 7.2399(6) A for the deuter-
ide). The compound contains nominal Ba2؉ cations, AlO45؊ com-
plex anions and H؊(D؊) anions. The hydride is bound in
a strongly distorted barium octahedron while the tetrahedral
AlO54؊ anion is surrounded by eleven barium atoms in a peculiar
polyhedral configuration. The HBa6 octahedra are linked to-
gether by corner-sharing to form a three dimension framework
with large voids defined by eight of the HBa6 octahedra that are
occupied by the tetrahedral AlO54؊ anions. The structure is close-
ly related in a hierarchical way to that of orthorhombic inverse
perovskitic Ba3GeO. ( 1998 Academic Press
During attempts to synthesize a possible hydrogen-stabil-
ized version of the suboxide Ba Al O (12), we accidentally
ꢀꢇ ꢃ ꢇ
ꢅ
discovered the new phase Ba AlO H, which shows a close
structural relationship to the orthorhombic inverse perov-
ꢃ
skite Ba GeO (13). Here we present its synthesis and struc-
ꢃ
tural definition by X-ray single-crystal and neutron powder
diffraction means.
EXPERIMENTAL
Synthesis
Sublimed barium (Aldrich-APL, 99.99%), its binary hy-
dride, and the product of the present studies are very sensi-
tive to air and moisture, and they were therefore handled
only in He- or N -filled glove boxes (H O(1 ppm vol).
INTRODUCTION
ꢁ
ꢁ
Binary barium hydride was prepared from barium metal by
reaction with hydrogen (600 torr, Matheson, 99.999%) at
800°C for 12 h. Barium deuteride was prepared in a parallel
manner with D (Matheson, 99.5%) instead of H . In this
Recently, we have systematically reexamined some com-
pounds formed by the alkaline-earth metals (Ae) and the
later main-group (p-block) elements because their forma-
tion, stoichiometry, and structure are controversial and
incomplete. Many are complicated by hydrogen impurities
in the Ae reagents (1). We have quantitatively synthesized
a series of hydrogen-stabilized Zintl phases, such as
Ae Pn H (Pn"As, Sb, Bi), (formerly b-Yb Sb type) (1, 2),
ꢁ
ꢁ
case, impurity hydrogen in the stock barium metal was first
removed from Ba sealed in Ta by heating in high vacuum at
1100°C. Guinier X-ray powder patterns showed only the
binary hydride (or deuteride) was formed. Barium oxide was
prepared by decomposition of barium carbonate (Baker,
99.8%) in an open Ta container for 24 h at 1000°C under
a dynamic vacuum (410\ꢂ torr). The Guinier X-ray pat-
tern showed only the presence of BaO. The other materials
used were Al (United Mineral and Chemical, 99.999%) and
ꢂ
ꢂ
ꢃ
ꢃ
ꢂ
ꢂ
ꢃ
Ae ¹t H (¹t"Si, Ge, Sn, Pb), Sr Tl H (Cr B type)
ꢂ ꢃ
ꢂ ꢃ
(3), Ba Ga H (4), Ba ¹t O H (¹t"Ge, Si), and
ꢁꢀ ꢁ ꢂ ꢁꢅ
ꢄ ꢁ
Ba ¹r O H (¹r"Ga, In, Tl) (5). These materials had
ꢁꢀ ꢁ ꢂ ꢁꢁ
initially been described as binary intermetallic compounds,
Al O (Baker, 99.0%).
such as eight examples of Ae Pn (6), Sr Tl (7), Ba Ga (8),
ꢁ ꢃ
ꢂ
ꢃ
ꢂ
ꢃ
ꢂ
ꢄ
Red transparent single crystals of Ba AlO H were ini-
ꢃ
ꢅ
ꢀTo whom correspondence should be addressed.
tially isolated from the product of a reaction loaded as
ꢁThe Ames Laboratory is operated for the U.S. Department of Energy
by Iowa State University under Contract W-7405-Eng-82. The research
was supported by the Office of Basic Energy Sciences, Materials Sciences
Division, DOE. The U.S. Government’s right to retain a nonexclusive
royalty-free license in and to the copyright covering this paper, for govern-
mental purposes, is acknowledged.
Ba Al O H that was heated at 1100°C and cooled at
20°C/h. A single crystal was also obtained during a syn-
thesis in the ternary Ba—Ge—Al system that was con-
taminated by very small amounts of oxygen and hydrogen.
It was later found that single-phase samples could be
ꢀꢇ ꢃ ꢇ V
570
0022-4596/98 $25.00
Copyright ( 1998 by Academic Press
All rights of reproduction in any form reserved.

STRUCTURE OF Ba AlO H
571
ꢃ
ꢅ
synthesized from stoichiometric mixtures of BaH or BaD , refined with the TEXSAN package (16) on a VAX station.
ꢁ
ꢁ
BaO and Al O with an overall composition Ba AlO H(D) Absorption corrections were found to be crucial for this
ꢁ ꢃ
ꢃ
ꢅ
(total amount: 500—2000 mg) that were thoroughly ground compound because Ba is a strong absorber of MoKa radi-
and welded in a Ta container under Ar. This was in turn ation (k"189.4 cm\ꢀ). The corrections were carried out
placed in fused-silica tubing and connected to a high- according to t-scan curves for three strong reflections at
vacuum source (410\ꢂ torr). The assembly was heated at different h, and this gave isotropic convergence and a good
300°C for 4 h to remove some H O from the jacket, then heavy-atom model with R(F)"0.033 and R "0.043.
ꢁ
ꢀ
charged with hydrogen or deuterium to 600—700 torr, A subsequent absorption correction by DIFABS (17) was
heated to 1100°C, and held there for 12 h (under H or D ) applied in an attempt to locate the hydrogen atom. The
ꢁ
ꢁ
followed by cooling at 20°C/h. The final quaternary product orthorhombic structure according to the X-ray data and
was a white powder that was extremely sensitive to air and including one hydrogen atom with a fixed position from
gave a Guinier powder pattern that appeared to be single neutron diffraction studies was later refined anisotropically
phase according to the powder pattern calculated after the on F by full-matrix least-squares methods. A total of
ꢁ
structure had been solved. A 4.0-g deuteride sample was 543 reflections with F '3p and 32 variables gave
ꢁ
'
sealed under He into a gasketed vanadium cylinder with an R(F)"0.021 and R "0.023. The ellipsoids were then no-
ꢀ
inner diameter of 8 mm for the neutron diffraction experi- tably smaller and more nicely spherical than before the
ment.
second absorption correction. Some X-ray single-crystal
A Ta container was also loaded with &500 mg of this data collection and refinement parameters are listed in
Ba AlO D product, welded under Ar, and then heated to Table 1. The structure factor data are available from J.D.C.
ꢃ
ꢅ
1100 °C under high vacuum for 4 h followed by cooling to
room temperature. The final product had a gray color, and
the X-ray pattern indicated a mixture of Ba Al O and
Neutron Powder Diffraction
ꢅ
ꢁ ꢇ
Ba Al O (14). A microscopic examination showed the
The atomic parameters and isotropic displacement am-
ꢈ
ꢁ ꢀꢀ
product contained many small black particles, which plitudes for a deuteride sample were established with the aid
were presumed to be Ba metal. (8Ba AlO DP4Ba# of neutron powder diffraction data collected at room tem-
ꢃ
ꢅ
3Ba Al O #Ba Al O #4D ).
perature on the Special Environment Powder Diffrac-
tometer (SEPD) at the Intense Pulsed Neutron Source
(IPNS) facility at Argonne National Laboratory. The SEPD
is a time-of-flight powder diffractometer with multidetector
ꢅ
ꢁ ꢇ ꢈ ꢁ ꢀꢀ
ꢁ
X-Ray Powder Diffraction
Guinier patterns were obtained from finely powdered arrays at fixed scattering angles, and the high-resolution
samples mixed with NIST silicon (a"5.43088 A) with the data from the 2h"144.8° backscattering detector bank
aid of monochromatic CuKa radiation (j"1.54056 A). were used for this refinement. The powder data were refined
ꢀ
These were mounted between pieces of cellophane tape by the Rietveld method (18) with the PC version of the
and held under forepump vacuum within an Enraf-Nonius General Structure Analysis System (GSAS) by Larson and
Guinier camera. Cell parameters were refined from indexed Von Dreele (19), starting with the positional data from the
lines and their 2h values determined with the aid of a non- X-ray study. The nuclear scattering lengths were taken from
linear fit to the positions of the standard Si lines.
Ref. (20). The final refinement included data in the range
0.50(d(4.0 A (5252 profile points, 2865 reflections).
The heavy-atom structure from X-ray studies readily re-
fined to reveal a single deuterium site in a distorted octahed-
X-Ray Single-Crystal Diffraction
Several crystals were sealed in thin-walled glass capillar- ral barium interstice. (The same position was deduced from
ies and checked for singularity by Laue photographs. Room X-ray data, but a second ghost peak of the same size was
temperature data were collected from one on a Rigaku also present near Ba.) Preliminary refinements gave a D
AFC6R diffractometer (MoKa radiation from a rotating atom occupancy of 99.2(5)% (º fixed), which was then
*ꢂꢁ
anode and a graphite monochromator) with an u—2h scan fixed at unity. One impurity phase, BaO, was also observed,
mode for 2h up to 50°. Three standard reflections measured and its proportion and lattice parameter were included in
every 150 reflections during the data collection did not the refinement; the relative amount was found to be 2.2(1)%
change in intensity significantly. Statistical analysis of the (by weight). The background function was a cosine Fourier
intensity distribution showed primitive orthorhombic cen- series with 12 coefficients. The other variables were as fol-
trosymmetry, and the systematic absences k#l"2n#1 in lows: two phase fractions, four lattice parameters, sixteen
0kl, h"2n#1 in hk0 indicated the unique space group atomic positional parameters, and seven isotropic displace-
ꢀꢄ
D -Pnma.
ment parameters. The final conventional agreement indices
ꢁF
The structure was solved by direct methods via SHELXS were R "3.47% and R "5.22% with a reduced sꢁ
(15). The atomic positions and thermal parameters were of 1.748. Some neutron data collection and refinement
ꢃ
ꢀꢃ

572
HUANG AND CORBETT
TABLE 1
barium aluminates whose structures have not yet been de-
termined. The Ba metal particles led to the gray sample
color product.
Crystallography and X Ray Refinement data for Ba3AlO4H
Crystal data
The structure of Ba AlO H was refined from a combina-
Crystal size (mm), color
Crytal system
Space group, Z
Lattice constants (A)?
&0.15;0.15;0.15; red, transparent
Orthorhombic
Pnma (No. 62), 4
a"10.505(4)
b"8.155(3)
c"7.234(4)
619.6(8)
5.402
189.35
ꢃ
ꢅ
tion of X-ray single-crystal and time-of-flight neutron power
data. The final atomic coordinates, the isotropic-equivalent
temperature factors, and their estimated standard devi-
ations from both studies are listed in Table 3. The final
neutron results are clearly more precise, and it is from these
that the distance and angular data in Table 4 were obtained.
The compound crystallizes in a new structure type with
orthorhombic symmetry, space group Pnma. The structure
»(Aꢃ)
d
(g/cmꢃ)
k(MoKa) (cm\ꢀ)
#!ꢄ#
Data collection
ꢂ\
isolated D\(H\) anions as shown in two views in Figs. 2
is made up of Baꢁ> cations, AlO complex anions, and
ꢅ
Diffractometer
Rigaku AFC6R
296
MoKa, 0.709580
u—2h
Temperature (K)
Radiation, j (A)
Scan type
and 3. The result is an orthorhombic distortion of a quasi-
inverse perovskite type X AB brought about by the order-
ꢃ
Reflections measured, 2h
#h, #k, $l, 50°
ꢂ\
B (D\) anions, but not from a simple distortion because the
ing of the three X (Baꢁ>) cations, A (AlO ) anions, and
ꢅ!ꢆ
ꢅ
Refinement
ꢂ\
AlO
anion is not a sphere. The Ba positions about
Number of measured reflections
Number of unique observed
1248
ꢅ
deuterium are strongly distorted from octahedral geometry,
as shown in Fig. 4, with Ba—D bond distances from 2.843(3)
to 2.995(3) A and cis Ba—D—Ba bond angles between
reflections (I'3p )
543
0.020
'
R(int)
Number of variables
Absorption correction
Relative transmission coefficient
range
32
79.69(7) and 106.00(7)°. The deformed DBa octahedra
ꢄ
3 t-scans, DIFABS
share corners to form a three-dimensional framework
ꢂ\
with the AlO
corner-sharing octahedral DBa units. The oxygen
tetrahedron surrounding aluminum is nearly regular
units filling large voids defined by eight
0.922—1.00
ꢅ
R(F ), R @ for I'3p
ꢅ!ꢆ
0.021, 0.023
0.97, !1.33
ꢁ
ꢀ
'
ꢄ
*F peak, hole (e/Aꢃ)
(Al—O1"1.749(2) A (2;), Al—O2"1.773(4) A, Al—O3"
?Rigaku data.
1.784(4) A; angles 107.2(1), 108.1(1), 108.6(2), and 117.5(2)°).
@R"+ " "F "!"F " "/+"F ", R "[+( "F "!"F "ꢁ/+wFꢁ]ꢀꢉꢁ, w"p\ꢁ.
ꢁ
#
ꢁ
ꢀ
ꢁ
#
ꢁ
$
ꢂ\
The AlO
tetrahedron is surrounded by eleven barium
ꢅ
atoms in a peculiar polyhedron (Fig. 5) in which seven
barium atoms define a polyhedron that allows the four
parameters are listed in Table 2. The neutron diffraction oxygen vertices of the tetrahedron to lie near the bases of
patterns in Fig. 1 show the raw intensity data and, from top four distorted square pyramids of Ba. This agreement leads
to bottom in each range, the background, the individual to optimized Ba—O and Al—O bond distances and O—Al—O
contributions from the main phase Ba AlO D and the im- bond angles and reflects the ‘‘rigid’’ or ‘‘hard’’ chemical
ꢃ
ꢅ
purity BaO, each with tick marks for the location of the character of the aluminate anions. On the other hand, the
reflections included, and the difference between the cal- low-field, polarizable deuteride lies in a strongly distorted
culated and observed intensity profiles.
barium octahedron because it is relatively ‘‘anomalous’’ (21)
or ‘‘soft’’ (22). This also explains why the isotropic thermal
parameter of the D atom is relatively large compared with
those of the other atoms.
RESULTS AND DISCUSSION
Reactions among BaH or BaD , BaO, and Al O at the
In the Ba AlO D structure, all three cystallographicaly
ꢁ
ꢁ
ꢁ ꢃ
ꢃ
ꢅ
proper stoichiometry under a pressure of 600—700 torr H
independent oxygen atoms are octahedrally coordinated to
one A1 and five Ba atoms. The average Ba—O bond lengths
(2.90 (O1), 2.80 (O2), and 2.83 A (O3)) are significantly
longer than either the sum of crystal radii of Baꢁ> and Oꢁ\
(2.70 A, CN 6) (23) or the mean Ba—O distances in
Ba Ge O H and Ba GeO (2.73 A for both). However,
ꢁ
or D at 1100°C followed by slow cooling gave quantitative
ꢁ
yields of Ba AlO H(D) according to Guinier patterns. This
ꢃ
ꢅ
must be a gas—solid—solid reaction process since the product
remained in powder form. Experiments showed that a thor-
ough grinding of the starting materials is the key step in
a high-yield synthesis. Single crystals were obtained only in
ꢁꢀ ꢁ ꢂ ꢁꢅ
ꢃ
bond valence calculations performed with bond length
parameters derived by Brese and O’Keeffe (24) show bond
orders sums of 1.82 for O1, 2.00 for O2, and 1.95 for O3,
close to the expected value of 2.0. The increase of Ba—O
distances probably correlates with the presence of the sixth
the presence of excess BaH or Ba metal as a flux. If the
ꢁ
Ba AlO D product is heated again (in Ta) to 1100°C but
ꢃ
ꢅ
under high vacuum, the D is lost, and the decomposition
ꢁ
gives Ba metal and, according to the powder pattern, two

STRUCTURE OF Ba AlO H
573
ꢃ
ꢅ
TABLE 2
Powder Neutron Data Collection and Refinement Parameters
for Ba3AlO4D
TABLE 3
Positional Parameters for Ba3AlO4D and Ba3AlO4H?
Atom
Ba1
Ba2
A1
x
y
z
10ꢁº (Aꢁ)
*ꢂꢁ
Temperature (°C)
23
0.3346(1)
0.33497(3)
0.4625(2)
0.46330(5)
0.1029(3)
0.1037(3)
0.0725(1)
0.0719(4)
0.0119(2)
0.0116(6)
0.2680(2)
0.2673(6)
0.2310(2)
0.0259(2)
0.0251(5)
1/4
0.1010(2)
0.10172(7)
0.5990(3)
0.5988(1)
0.3110(5)
0.3108(4)
0.1938(2)
0.1949(7)
0.5172(3)
0.519(1)
0.14(3)
0.62(2)
0.07(4)
0.48(2)
0.07(6)
0.4(1)
0.53(3)
1.1(1)
0.37(4)
0.6(1)
Lattice constants (A)?
a"10.4911(9)
b"8.1518(7)
c"7.2399(6)
619.2(1)
5252
1/4
1/4
1/4
»(Aꢃ)
Number of data
Number of reflections
Variables
2865
42
O1
0.0665(2)
0.0657(6)
1/4
Residuals; R , R @ (%)
3.47, 5.22
1.748
ꢃ
ꢀꢃ
O2
sꢁ
1/4
1/4
1/4
1/4
Scattering lengths (10\ꢀꢁcm)A
Ba, 0.507
Al, 0.346
O, 0.580
D, 0.667
O3
0.3706(3)
0.372(1)
0.8393(3)
0.46(4)
1.1(1)
1.81(5)
D
?Guinier data.
? Values in the first line are from refinement of neutron powder data from
the deuteride, and those on the second line, from X-ray single-crystal data
for the hydride.
ꢁ
@R "(+("I !I ")/+I ; R "[+(I !I )ꢁ/+wI ]ꢀꢉꢁ.
ꢃ
ꢁ
#
ꢁ
ꢀꢃ
ꢁ
#
ꢁ
AReference (23).
FIG. 1. Neutron powder pattern of Ba AlO D divided into two regions, d"0.5—1.6 and 1.6—4.0 A. Shown in each are, from top to bottom, the
ꢃ
ꢅ
experimental pattern with fitted background; the calculated patterns for the major and impurity BaO phases and the locations of the reflections used (tick
marks); and the difference spectrum on the same scale.

574
HUANG AND CORBETT
TABLE 4
Important Bond Distances (A) and angles (Deg) in Ba3AlO4D
Ba1—D
Ba1—D
2.851(2)
2.915(2)
2.744(2)
2.764(2)
2.832(2)
2.851(2)
2.9223(2)
3.000(2)
3.195(2)
2.843(3)
2.995(5)
2.626(3)
2.695(2)
2.826(3)
2.839(3)
1.749(2)
1.773(4)
1.784(4)
O1—Al
O1—Ba2
O1—Ba2
O1—Ba12
O1—Ba1
O1—Ba1
O2—Al
O2—Ba1 (2;)
O2—Ba1 (2;)
O2—Ba2
1.749(2)
2.695(2)
2.839(3)
2.851(2)
2.923(2)
3.195(2)
1.773(4)
2.744(2)
2.832(2)
2.826(3)
1.784(4)
2.626(3)
2.764(2)
3.000(2)
2.843(3)
2.851(2)
2.915(2)
2.995(3)
Ba1—O2
Ba1—O3
Ba1—O2
Ba1—O1
Ba1—O1
Ba1—O3
Ba1—O1
Ba2—D
Ba2—D
O3—Al
O3—Ba2
Ba2—O3
Ba2—O1 (2;)
Ba2—O2
Ba2—O1 (2;)
Al—O1 (2;)
Al—O2
O3—Ba1 (2;)
O3—Bal (2;)
D—Ba2
D—Ba1 (2;)
D—Ba1 (2;)
D—Ba2
FIG. 3. [011] view of Ba AlO D showing the network of corner-shar-
ꢃ
ꢅ
Al—O3
ing DBa octahedra.
ꢄ
O1—Al—O1
O1—Al—O2
O1—Al—O3
O2—Al—O3
117.5(2)
Ba1—D—Ba1
Ba1—D—Ba2
Ba1—D—Ba2
Ba1—D—Ba1
Ba1—D—Ba2
Ba1—D—Ba1
Ba1—D—Ba2
79.69(7)
81.12(6)
82.13(6)
89.34(2)
94.17(7)
100.99(8)
106.00(7)
108.1(1)
107.2(1)
108.6(2)
In comparison, Ba GeO crystallizes in the orthorhombic
ꢃ
inverse perovskite GdFeO type (24) with the same space
ꢃ
group, Pnma (13). The slightly deformed corner-sharing
OBa octahedra here make up a three-dimensional frame-
ꢄ
work, and Ge atoms are bound in barium voids that are
defined by eight OBa units. This structure is similar to that
ꢄ
of Ba AlO D except for the stacking of X A since the
ꢃ
ꢅ
ꢃ
ꢂ\
difference in formal charge between the Geꢅ\ and AlO
ꢅ
high-field Alꢃ> neighbor. Calculations on the deuterium site
gave bond orders of 0.18 (2;), 0.15 (2;) (Ba1—D), and 0.19
and 0.12 (Ba2—D) and a bond order sum about deuterium of
0.97, close to the ideal value of 1.0. Finally, an oxidation
state count indicates Ba AlO H is a valence compound, as
anions is just compensated by the charge difference between
divalent oxygen and monovalent deuterium. In the struc-
ꢂ\
ture of Ba AlO D, the larger AlO complex anion with
ꢃ
ꢅ
ꢅ
only near-tetrahedral, not spherical, symmetry causes a
strong deformation of the cubic-close-packed arrangement.
It is interesting that reduced charges on both the cations
ꢁ\
ꢃ
ꢅ
supported by its transparency.
(A"K, Rb, Cs) and MnX
tetrahedra (X"Cl, Br, H)
ꢅ
make the related A (MnX )X phases tetragonal (I4/mcm)
ꢃ
ꢅ
with linear chains of elongated A H octahedra (26).
ꢄ
There is also as series of basic orthosilicates (A'') O(SiO )
for A"Ca, Sr, Ba, Cd, etc. that contain the same type of
ꢃ
ꢅ
FIG. 2. View down the orthorhombic [100] axis of Ba AlO D show-
ꢃ
ꢅ
ing the DBa octahedra sharing corners to form a network.
FIG. 4. Environment around the deuterium atom in Ba AlO D.
ꢄ
ꢃ
ꢅ

STRUCTURE OF Ba AlO H
575
ꢃ
ꢅ
by the U.S. Department of Energy, BES-Materials Sciences, under Con-
tract W-31-109-ENG-38.
REFERENCES
1. E. A. Leon-Escamilla and J. D. Corbett, J. Alloys Compd. 206, L15
(1994).
2. E. A. Leon-Escamilla and J. D. Corbett, J. Alloys Compd. 265, 105
(1998).
3. E. A. Leon-Escamilla, Ph.D. Dissertation, Iowa State University, 1996.
4. R. W. Henning, E. A. Leon-Escamilla, J.-T. Zhao, and J. D. Corbett,
Inorg. Chem. 36, 1282 (1997).
5. B. Huang and J. D. Corbett, Inorg. Chem. 37, 1892 (1998).
6. G. D. Brunton and H. Steinfink, Inorg. Chem. 10, 2301 (1971).
7. A. Bruzzone, E. Franceschi, and F. Merlo, J. ¸ess-Common Met. 60, 59
(1978).
FIG. 5. The tetrahedral AlOꢂ\ anion and its barium neighbors.
8. M. L. Fornasini and M. Pani, J. Alloys Compd. 205, 179 (1994).
9. M. L. Fornasini and F. Merlo, Rev. Chim. Miner. 16, 458 (1979).
10. C. Rohr, Z. Anorg. Allg. Chem. 621, 1496 (1995).
11. C. Rohr, Z. Kristallogr., Suppl. Issue No. 9, 24 (1995).
12. C. Rohr and R. George, Angew. Chem., Int. Ed. Engl. 34, 2115 (1995).
13. B. Huang and J. D. Corbett, Z. Anorg. Allg. Chem. 324, in press (1998).
ꢅ
components as does Ba H(AlO ). Several are polymorphic.
ꢃ
ꢅ
Those closest in cation size, the strontium (27) and barium
(28) salts, have primitive and centered tetragonal struc- 14. P. Appendino, Ann. Chim, (Rome) 61, 822 (1971).
15. G. M. Sheldrick, SHELXS-86, Universta
¨ t Gottingen, Gottingen, Ger-
tures, respectively, but no orthorhombic examples seem
to be known. This general contrast with the structure
of Ba AlO H seems to be a logical consequence of the
many, 1986.
16. TEXSAN, Version 6.0, Molecular Structure Copr., The Woodlands,
TX, 1990.
17. N. Walker and D. Stuart, Acta Crystallogr., Sect. A 39, 158 (1983).
ꢃ
ꢅ
presence of the hard oxide instead of the lower charged,
polarizable hydride. For example, the barium coordination 18. H. M. Rietveld, Acta Crystallogr. 22, 151 (1967).
19. A. C. Larson and R. B. Von Dreele, Report LAUR 86-748; Los Almos
National Laboratory, Los Almos, NM, 1995.
20. V. F. Sears, Neutron News 3, 26 (1992).
21. W. Bronger, Angew. Chem., Int. Ed. Engl. 30, 759 (1991).
22. K. Yvon, in ‘‘Encyclopedia of Inorganic Chemistry’’ (R. B. King, Ed.),
Vol. 3, p. 1401. Wiley, New York, 1994.
by oxygen changes from 10 and 8 in the oxide, two each
involving the lone oxygen atom (4/m), to six oxygen (on the
ꢂ\
AlO ) and two hydrogen in the present case. The lower
ꢅ
symmetries about both the lone anion and the MO group
ꢅ
are clear (Figs. 4 and 5) and the polyoxoanion is also more
23. R. D. Shannon, Acta Crystallogr., Sect. A 32, 751 (1976).
24. N. E. Brese and M. O’Keeffe, Acta Crystallogr., Sect. B 47, 192 (1991).
25. S. Geller, J. Chem. Phys. 24, 1236 (1956).
26. W. Bronger, S. Hasenberg, and G. Auffermann, J. Aloys. Compd. 257,
75 (1997).
distorted.
ACKNOWLEDGMENTS
We are indebted to S. Short and J. Jorgensen for provision of the neutron 27. W. Eysel, Neues Jahrb. Mineral. Monatsh. 534 (1970).
diffraction data. This research has benefited from the use of the Intense 28. E. Tillmanns and H.-P. Grosse, Acta Crystallogr., Sect. B 34, 649
Pulsed Neutron Source at Argonne National Laboratory, which is funded
(1978).