2 2 6
New Synthetic Route to Mg Na NiH
that the complex needs support from the small and strongly
2 4
fine powders of Mg NiH and NaH were mixed in a 1:1 molar
2
+
5,6
ratio by grinding them together in a mortar. This mixture (∼2 g)
was compacted in a tablet press at 3 kbar and sintered in a tube
furnace at 300-315 °C in hydrogen of 50 bar for about 20 h. After
cooling, the tablets were crushed and a small amount was removed
for X-ray analysis. To obtain a fairly “single-phased” sample, the
tablet had to be repressed again for two or three subsequent sintering
periods. Additional starting material was added until we reached a
polarizing Mg counterions for stability. Similarly, all of
the transition metal hydrido complexes of the first row
2
+
transition metals have mainly been synthesized with Mg
counterions either alone or in combination with other
counterions. There are now a number of these as shown by
the examples of Mg
3
CrH∼6, Mg
3
MnH∼6, Mg
3 7 2
MnH , Mg -
FeH
LaMg
6
, Mg
2
CoH
5
-12
, CaMgNiH , SrMgNiH
Of all of these complexes, only [FeH ]
6
4
4
, YbMgNiH
4
, and
2 4
molar ratio of 1:2 for Mg NiH to NaH, whereby an olive-green
7
4-
2
NiH
7
.
powder was obtained. Several attempts were made in this iterative
search before the optimum preparative conditions could be found.
A typical successful synthesis started with a 1:2 molar mixture of
2+
has been synthesized without Mg counterions, but then only
with low yield in incomplete reactions (i.e., in Ca FeH and
with a not too electron-dense d configuration).
2
6
6
13
Sr
2
FeH
6
2 4
Mg NiH to NaH, which was compacted and sintered as described
above. Regrinding, recompacting, and resintering were more
essential than the length of the sintering period. During this period,
we were able to add the balancing amounts of NaH and Ni powders.
With each such cycle the amount of impurity phases was reduced.
As this consumed a lot of deuterium, we stopped after three sintering
periods for the larger sample used in the diffraction experiment.
X-ray and Neutron Powder Diffraction Data Collection and
Structure Refinements. X-ray diffraction photographs were ob-
tained from a subtraction-geometry Guinier-H a¨ gg focusing camera,
We also tried to make the corresponding hydrides with alkali
metal counterions by reacting powders of all the first row
transition metals with the alkali metals of Li, Na, and K or
corresponding binary hydrides at temperatures and hydrogen
pressures of up to 600 °C and 150 bar. In no case could we
obtain a reaction. On the other hand, transition metal hydrido
complexes of the second and third row transition metals,
which are held together by stronger bonds, have been
synthesized in large numbers with counterions of most of
the alkali earth as well as alkali metals group. This shows
that magnesium is especially important for stabilizing first
row transition metal hydrido complexes, but in this paper
we report the first quaternary hydride of this type, where an
alkali metal is directly involved in the stabilization of a
1
using strictly monochromatized Cu KR radiation. A single-coated
X-ray film (CEA Reflex 15) was used in all the work. The films
were evaluated by means of a computer-controlled single-beam
microdensitometer designed for the scanning analysis of X-ray
powder photographs.14 The θ scale was calibrated by means of the
internal (silicon) standard technique, using a parabolic correction
curve. To localize the hydrogen atoms in the new hydrides we
[
4
NiH ] complex as its closest neighbor. It is also supported
2 2 6
prepared a Mg Na NiD sample for neutron diffraction. The
2+
by Mg ions but at a significantly longer distance from the
complex. This hydride could not be synthesized directly from
diffraction patterns were recorded at room temperature at the R2
reactor in Studsvik, Sweden, using a cylindrical vanadium sample
holder (diam ) 5 mm) and a neutron wavelength of 1.47 Å.
2 4
the elements but was made by first creating Mg NiH , which
was subsequently reacted with NaH.
15
With the use of the TREOR program and the least-squares
refinements program PIRUM,16 the final X-ray powder pattern of
Experimental Section
2 2 6
Mg Na NiH could be indexed on the basis of an orthorhombic
The starting sodium metal ingot, with 99.95% purity, was
obtained from Aldrich. The Mg Ni alloy ingot was obtained from
Ergenics, USA (as their HYSTOR-301 Alloy). This alloy contains
about 10 wt % extra magnesium to prevent the formation of the
structure with unit cell dimensions of a ) 11.428(2), b ) 8.442(2),
and c ) 5.4165(9) Å and unit cell volume of 523 Å . The deuteride,
3
2
Mg Na NiD , has unit cell dimensions of a ) 11.403(2), b )
2
2
6
3
8.411(3), and c ) 5.405(2) Å and a unit cell volume of 518 Å .
Metal deuterides often have smaller unit cell dimensions than those
of the corresponding hydrides due to unharmonicity in the hydrogen
vibrations. Several syntheses of both the hydride and the deuteride
were made to verify that the difference in unit cell volume was not
due to differences in hydrogen content. All syntheses, how-
ever, gave the same unit cell dimensions as given above, indicating
that the compound had a preference for formation with full
stoiciometry.
non-hydrogen absorbing MgNi
Mg NiH /D were synthesized by reaction with H
forming fine powders. The extra Mg in the Mg Ni alloy formed
MgH /D . The X-ray powder diffraction patterns for the MgH
, NaH/D, and Mg NiH /D produced were consistent with data
2
phase during casting. NaH/D and
2
4
4
2
and D gases
2
2
2
2
2
/
D
2
2
4
4
in the JCPDS tables. To protect the material from air or moisture,
all storing and handling was done in a glovebox under continuously
purified argon gas.
As nickel powder could not be made to react with Na or NaH,
the quaternary hydride was made in a two-step reaction. First, the
Systematic absences were consistent with space group Pnma (No.
2).17 The positions of the heavy atoms were found by direct
6
18
methods (SHELXS ), and subsequent Rietveld refinements RI-
(
5) H a¨ ussermann, U.; Blomqvist, H.; Nor e´ us, D. Inorg. Chem. 2002, 41,
19,20
21
ET94
and FULLPROF based on X-ray and neutron powder
3684.
(
(
6) Miller, G. J.; Deng, H.; Hoffmann, R. Inorg. Chem. 1994, 33, 1330.
7) R o¨ nnebro, E.; Kyoi, D.; Blomqvist, H.; Nor e´ us, D.; Sakai, T. J. Alloys
Compd. 2004, 279 (1-2), 368.
(14) Johansson, K. E.; Palm, T.; Werner, P. E. J. Phys. 1980, E13, 1289.
(15) Werner, P. E.; Eriksson, L.; Westdahl, M. J. Appl. Cryst. 1985, 18,
367.
(16) Werner, P. E.; Eriksson L. World Directory of Powder Diffraction
Programs, Release 2.12; International Union of Crystallography:
Chester, 1993.
(17) International Tables for Crystallography: Space Group Symmetry;
Hahn, T., Ed.; Kluwer Academic Publishers: Dordrecht, Germany,
1989; Vol A.
(
8) Blomqvist, H.; R o¨ nnebro, E.; Kyoi, D.; Sakai, T.; Nor e´ us, D. J. Alloys
Compd. 2003, 358 (1-2), 82-86.
(
9) Bortz, M.; Bertheville, B.; Yvon, K.; Movlaev, E. A.; Verbetsky, V.
N.; Verbetsky, V. N.; Fauth, F. J. Alloys Compd. 1998, 279 (2), L8-
L10.
(
10) Selvam, P.; Yvon, K. Int. J. Hydrogen Energy 1991, 16 (9), 615-
617.
(11) Renandin, G.; Guenee, L.; Yvon, K. J. Alloys Compd. 2003, 350, 145.
(12) Huang, B.; Yvon, K. J. Alloys Compd. 1994, 204, L5.
(13) Huang, B.; Bonhomme, F.; Selvam, P.; Yvon, K. J. Less-Common
Met. 1991, 179, 301.
(18) (a) Sheldrick, G. M. SHELXS-97 Acta Crystallogr. 1990, A46, 467.
(b) Sheldrick, G. M. SHELXL-97, Computer Program for the Refine-
ment of Crystal Structures; University of G o¨ ttingen: G o¨ ttingen,
Germany, 1997.
Inorganic Chemistry, Vol. 46, No. 6, 2007 2221