Synthesis of Li2PtH6 using high pressure: Completion of the homologous series A2PtH6 (A=alkali metal)

Article abstract of DOI:10.1016/j.jssc.2010.05.033

Li₂PtH₆, the missing member of the complex transition metal hydride series A₂PtH₆ (A=alkali metal), was prepared by reacting mixtures of LiH and Pt in the presence of BH ₃NH₃ as hydrogen source at pressures above 8 GPa. According to powder X-ray diffraction analysis, Li₂PtH₆ is isostructural to its heavier homologues and crystallizes in the cubic K ₂PtCl₆ structure (space group Fm3m, a=6.7681(3), Z=4). However, PtH₆^2- octahedral complexes in Li ₂PtH₆ approach each other closely and its structure may likewise be regarded as a defective perovskite structure where half of the octahedrally coordinated atoms (cations) are missing. The IR spectrum of Li ₂PtH₆ reveals the PtH T_1u stretch and bend at 1840 and 889 cm^-1, respectively.

Full text of DOI:10.1016/j.jssc.2010.05.033

Journal of Solid State Chemistry 183 (2010) 1785–1789
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Synthesis of Li2PtH6 using high pressure: Completion of the homologous
series A2PtH6 (A¼alkali metal)
Kati Puhakainen, Emil Stoyanov, Michael J. Evans, Kurt Leinenweber, Ulrich Ha¨ussermann n
Department of Chemistry and Biochemistry, Arizona State University, Tempe, AZ 85287-1604, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Li2PtH6, the missing member of the complex transition metal hydride series A2PtH6 (A¼alkali metal),
was prepared by reacting mixtures of LiH and Pt in the presence of BH3NH3 as hydrogen source at
pressures above 8 GPa. According to powder X-ray diffraction analysis, Li2PtH6 is isostructural to its
Received 28 March 2010
Received in revised form
24 May 2010
¯
heavier homologues and crystallizes in the cubic K2PtCl6 structure (space group Fm3m, a¼6.7681(3),
Accepted 30 May 2010
Available online 8 June 2010
Z¼4). However, PtH26ꢀ octahedral complexes in Li2PtH6 approach each other closely and its structure
may likewise be regarded as a defective perovskite structure where half of the octahedrally coordinated
atoms (cations) are missing. The IR spectrum of Li2PtH6 reveals the Pt–H T1u stretch and bend at 1840
and 889 cmꢀ1, respectively.
Keywords:
High pressure materials synthesis
Multi-anvil techniques
& 2010 Elsevier Inc. All rights reserved.
Complex transition metal hydrides
1. Introduction
of 150 MPa afford Na2PtH6 with octahedral Pt(IV)H26ꢀ complex
ions [10].
While the synthesis conditions applied for Na2PtH6 also yield
the heavier homologues A2PtH6 (A¼K–Cs) [11,12], for A¼Li
Li5Pt2H9 is obtained [13]. This compound contains the complex
anion Pt2H59ꢀ in which two square planar entities PtH24ꢀ are
bridged by Hꢀ. Thus, for A¼Li platinum remains in the oxidation
state II. Recently Parker et al. [14] provided a comparative analysis
of the series A2PtH6 while performing a detailed study of the
vibrational properties of Rb2PtH6 and Rb2PtD6. These authors
suggested that the missing start member, Li2PtH6, should be
accessible when applying even higher hydrogen pressures.
High hydrogen pressures offer exciting prospects for metal
hydride synthesis because the activity of hydrogen fluid increases
sharply above 1 GPa [15]. The pressure limit of autoclave
techniques may be around 500 MPa (0.5 GPa), especially when
simultaneous high temperatures (up to 700 1C) are applied [6].
Pressures well above 10 GPa are accessible with diamond anvil
cells (DACs) [16]. However, the DAC is not a practical tool for
synthesis because of the small sample size and, in the particular
case of hydrogenation reactions, the difficulties associated with
loading elemental H2. A more feasible high pressure strategy for
metal hydride synthesis is the utilization of multi-anvil (MA) [17]
or toroid devices [18], which provide larger sample volumes.
Hydrogen is introduced via an internal source that releases
irreversibly H2 when heating the pressurized sample. Pressures
are usually restricted to around 10 GPa. When employing an
internal source, hydrogen activity does not relate directly to
applied pressure as it is the case for H2 loaded autoclaves or
DACs [19]; the hydrogen concentration provided by a source is
rather low.
Complex transition metal hydrides are a peculiar class of solid
state compounds, which consist of homoleptic hydrido complexes
[THn]mꢀ and active metal (alkali, alkaline earth, rare earth)
cations [1,2]. Typically the transition metal T is from groups 7 to
10. The complex anions [THn]mꢀ, in which hydridic H is covalently
bonded to T, exhibit a perplexing range of coordination numbers
and geometries [3]. An interesting feature is the occurrence of low
formal oxidation states of T with a ligand (Hꢀ) that does not
afford the conventional ‘‘
p-back-donation’’ mechanism. This is
especially seen with group 10 metals (e.g. tetrahedral NiH44ꢀ
,
trigonal planar PdH33ꢀ, or linear PdH22ꢀ with zero-valent T) where
tetrahedral complexes with zero-valent Ni and Pd show extra-
ordinarily weak T–H bonds [4]. Thus actual bonding situations can
be intricate and sometimes the cation environment plays a
decisive role in stabilizing such complexes [5].
Complex transition metal hydrides are conveniently prepared
by hot sintering active metal hydrides with the corresponding
transition metal in a hydrogen atmosphere employing autoclave
techniques [6]. The hydrogen pressure applied will often
determine the oxidation state of T and its coordination number
in the hydrido complex. For example, Na2PdH2 with linear
complex ions Pd(0)H22ꢀ forms at hydrogen pressures below
100 bar (10 MPa) [7], while Na2PdH4 with square planar Pd(II)H42ꢀ
requires kbar pressures (above 180 MPa) [8]. Similarly, Na2PtH4 is
obtained at pressures below 1 MPa [9], while pressures in excess
n Corresponding author.
E-mail address: Ulrich.Haussermann@asu.edu (U. Ha¨ussermann).
0022-4596/$ - see front matter & 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2010.05.033

1786
K. Puhakainen et al. / Journal of Solid State Chemistry 183 (2010) 1785–1789
Here we show that Li2PtH6 can be readily obtained by
Highly crystalline Li2PtH6 was obtained with a purity of480%
from reaction mixtures LiH:Pt:BH3NH3¼3:1:3.333 at pressures
above 8 GPa and temperatures between 450 and 500 1C. The
powder X-ray pattern is shown in Fig. 1. The amount of hydrogen
source corresponded to a theoretical molar ratio Pt:H2¼1:10. As a
side product the intermetallic compound LiPt7, which crystallizes
in a 2 ꢂ 2 ꢂ 2 superstructure of the simple fcc structure [28], was
identified. Compared to Li2PtH6 LiPt7 appears poorly crystalline
which could indicate the incorporation of some hydrogen. Li2PtH6
forms also when using a stoichiometric ratio LiH:Pt¼2:1 and/or
lower pressures (down to 5 GPa). However, at these conditions
the amount of side product is substantially increased. Further
increasing the excess of LiH did not improve the yield of Li2PtH6. It
appears that the formation of LiPt7 competes with the formation
of Li2PtH6 and the latter is favored by applying an excess of LiH
and higher pressures. To support this hypothesis we attempted
the synthesis of Na2PtH6 by MA hydrogenation using the same
reaction temperature and Pt:BH3NH3 ratio as in the Li2PtH6
experiments. NaPt7 does not exist and a reaction performed at
hydrogenating mixtures of LiH and Pt in a MA device. The
structural properties of Li2PtH6 fit perfectly the trend of A2PtH6
compounds, which, however, results in a peculiar H substructure
for A¼Li. For synthesizing Li2PtH6 we facilitated MA gigapascal
hydrogenations by introducing ammonia borane, BH3NH3, as
superior internal hydrogen source, and by employing recently
developed large-volume pressure cells [20].
2. Results and discussion
High pressure hydrogenations for metal hydride synthesis
were pioneered by Russian researchers using toroid type devices
[21] and later followed up by Japanese research groups using
more common MA techniques [22]. In particular, it was shown
that middle and late transition metals T – for which hydrides had
not been known – form compounds TH
(e.g FeH, MoH, NiH)
ꢁ1
when pressures between 1 and 9 GPa are applied. These hydrides
are recoverable and metastable at liquid nitrogen temperature.
Further examples include the synthesis of a vacancy-ordered Pd
hydride (Pd3VacH4) at a pressure of 5 GPa [23] and the formation
of AlH3 from the elements (at 9 GPa) [24]. Attempts to prepare
ternary hydrides by high pressure hydrogenations, however,
resulted only in few conclusively characterized products [25].
An important ingredient to high pressure hydrogenations is
the hydrogen source being part of the sample. Ideally this source
releases H2 rapidly and irreversibly at slightly elevated tempera-
tures (preferably below 300 1C) and the decomposition residual is
inert toward the metal or intermetallic precursor to be hydro-
genated as well as the resulting hydride. Hitherto employed
sources such as NaBH4+2Ca(OH)2-NaBO2+2CaO+4H2, LiAlH4-
LiH+Al+3/2H2, or AlH3-Al+3/2H2 all have drawbacks in this
respect. Also, the decomposition behavior of a source under high
pressure conditions is usually not known.
5 GPa with
quantitatively Na2PtH6.
According to its powder X-ray diffraction pattern Li2PtH6 is
isostructural to the heavier homologues A2PtH6 (A¼Na–Cs),
which crystallize in the cubic K2PtCl6 structure type (space group
a
stoichiometric ratio NaH:Pt¼2:1 yielded
¯
Fm3m, cf. Table 1). This structure corresponds to a CaF2-type
arrangement of PtH26ꢀ octahedral units and alkali metal (A)
cations, which are coordinated by 12 H-ions (i.e. by four
tetrahedrally arranged faces from four different octahedra).
Accordingly, the Pt and A-type atoms (ions) occupy the special
positions 4a (0,0,0) and 8c (1/4,1/4,1/4), respectively, while the H
We perform high pressure hydrogenations in a MA device and
employ ammonia borane (BH3NH3) as internal hydrogen source.
BH3NH3 is isoelectronic to ethane but due to its intermolecular
dihydrogen bonding it is a crystalline solid. The effects of high
pressure (up to 9 GPa) on the thermal decomposition of BH3NH3
have been recently studied. In contrast with the three-step
decomposition at ambient pressure, thermolysis under pressure
releases irreversibly almost the entire hydrogen content of the
molecule in two distinct steps [26]. The residual after the second
decomposition, where two equivalents of hydrogen are released,
corresponds to a derivate of hexagonal BN. The temperature of the
second decomposition step increases from 180 1C at 1 GPa to
about 320 1C at 9 GPa. With its high hydrogen capacity and well-
defined decomposition behavior leading to inert residuals,
BH3NH3 appears as superior internal hydrogen source for high
pressure hydrogenations. Also, the isotopologue BD3ND3 can be
easily prepared [27], which allows access to deuterized samples
for e.g. neutron diffraction studies.
LiPt
7
20
30
40
50
60
70
80
90
100 110
2Θ (deg)
Fig. 1. The Rietveld fit to the X-ray powder diffraction pattern of Li2PtH6 (CuKa1).
The grey line represents the measured pattern, the dotted black line the calculated
one. Bragg peak positions are marked by horizontal bars (black: Li2PtH6; red: side
product LiPt7). (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)
For MA hydrogenations the precursor material is pressed into a
pellet which in turn is sandwiched between two pellets of
Table 1
Crystal data and structure refinement of Li2PtH6.
BH3NH3. The three-pellet arrangement is sealed in
a NaCl
Compound
Li2PtH6
container. In contrast with sources yielding reactive intermedi-
ates/residuals no protective separator between source and sample
is needed. NaCl is a popular capsule material because it provides
tight seals in the case of air/moisture sensitive precursors and
resists hydrogen diffusion. In our lab we apply two differently
sized NaCl capsules. Their dimensions (inner diameter ꢂ length)
are either 4 ꢂ 8 mm2 (V¼82 mm3) or 6.15 ꢂ 8 mm2 (V¼238 mm3)
for attainable pressures slightly above 10 GPa and up to 7 GPa,
respectively. This represents an excellent pressure–volume
capability for MA techniques [20].
¯
Space group
Z
Fm3m
4
6.7681(3)
˚
a, A
3
310.0(1)
˚
V, A
Temp, K
w2
Rp, %
Rwp, %
RB, %
295
1.73
2.84
3.78
1.22

K. Puhakainen et al. / Journal of Solid State Chemistry 183 (2010) 1785–1789
1787
position 24e (x,0,0) represents the only flexible structural
parameter. Although deuterized samples (from LiD/Pt/BD3ND3)
can be prepared using the same conditions as for Li2PtH6, the high
pressures necessary for obtaining high yield samples (48 GPa)
prohibited the preparation of a sample volume large enough for
neutron diffraction measurements. Therefore we were not able to
determine the H(D) atom positional parameter experimentally.
Instead, we performed an optimization of the K2PtCl6 structure for
Li2PtH6 by first principles calculations using density functional
theory. In this study we also included the heavier homologues.
Fig. 2 summarizes the result.
The structural trend in the series A2PtH6 (Fig. 2a–c) correlates
well with the Pauling ionic radii of A. This has already been
suggested in Ref. [14]. The agreement between experimental
parameters (where available) and computed ones is very good
with perhaps the exception of the H atom positional parameter
for Na2PtH6 (Fig. 2b). In particular, the lattice parameter decreases
linearly with decreasing size of the ionic radius of A. The H atom x
parameter increases linearly and approaches ¼ (calculated value:
0.246) for A¼Li. This leads to a peculiar situation when analyzing
interatomic distances (Fig. 2c). The Pt–H distance in the
9.0
8.5
8.0
7.5
7.0
˚
octahedral complex ion is around 1.65 A and virtually not affected
by A. As a consequence distances between H atoms within the
octahedral units (H–H(1)) and in between (H–H(2)) – which are
considerably different for A¼Na–Cs – become almost identical for
˚
Li2PtH6 (2.35 and 2.43 A, respectively). This is shown in Fig. 3
6.5
where the structures of the two end members, Li2PtH6 and
Cs2PtH6, are compared. As a matter of fact, the substructure
of H atoms in Li2PtH6 approaches closely that of O atoms in the
cubic perovskite structure and Li attains an almost regular
cuboctahedral coordination by H atoms.
0.6 0.8 1.0 1.2 1.4 1.6 1.8
0.25
0.24
0.23
0.22
0.21
0.20
0.19
0.18
The ‘‘strained’’ situation of the Li2PtH6 structure is also
apparent when comparing computed formation energies (refer-
ring to 0 K) for the reaction 2AH+Pt+2H2¼A2PtH6 (Fig. 2d). For
A¼Cs, Rb, K the value is very similar, around ꢀ3 eV. It decreases
to ꢀ2.5 eV for Na2PtH6 and becomes just ꢀ1.7 eV for Li2PtH6. The
temperature-dependent Gibbs free energy (DG(T)H2) for a H2 gas
molecule with respect to 0 K is around ꢀ0.32 eV at 300 K [29].
Therefore Li2PtH6 should represent a thermodynamically stable
compound at room temperature and ambient pressure [30].
Li2PtH6 decomposes at temperatures between 150 and 200 1C
when heating in a 1 bar Ar atmosphere for 20 h. The X-ray powder
pattern of the decomposition product obtained at 200 1C shows
very broad reflections which can be related to fcc-Pt. These
reflections constitute – somewhat sharpened – also the pattern of
the 250 1C decomposition product. When decomposing Li2PtH6 at
300 1C trigonal LiyPtHx [31], LiH and a small amount of LiPt7 can
be indentified in the powder pattern of the product. Thus, the
thermal behavior of Li2PtH6 is different from Li5Pt2H9, which
decomposes quantitatively to Li2PtH2 (with zero-valent Pt) and
LiH at 220 1C when applying the same conditions [32]. Li2PtH2
decomposes at 280 1C to LiyPtHx [32]. Finally, somewhat surpris-
ing is the observation that Li2PtH6 degrades only slowly outside a
protective atmosphere while the heavier homologues are
extremely air/moisture sensitive.
0.6 0.8 1.0 1.2 1.4 1.6 1.8
4.0
H-H(2)
3.5
A-H
3.0
H-H(1)
2.5
2.0
Pt-H
1.5
0.6 0.8 1.0 1.2 1.4 1.6 1.8
-1.6
-1.8
-2.0
–2.2
-2.4
-2.6
-2.8
-3.0
In Fig. 4 we compare the IR spectra of Na2PtH6, Li2PtH6 and
Li2PtD6. The vibrational properties of K2PtH6 and Rb2PtH6 have
been extensively studied previously both by optical (IR and
Li2PtH6
Cs2PtH6
-3.2
0.6 0.8 1.0 1.2 1.4 1.6 1.8
Li
Na
K Rb Cs
Pauling ionic radius(Å)
Fig. 2. Lattice parameter (a), H atom x parameter (b), and interatomic distances (c)
for the series A2PtH6 (A¼alkali metal) as a function of the Pauling ionic radius of A.
Theoretical and experimental values (this work, Refs. [10–12]) are denoted as
circles and squares, respectively. The lines are least square fits to the theoretical
values (correlation coefficients 40.995). H–H(1) and H–H(2) denote H–H
distances within and between octahedral units, respectively. In (d) the computed
zero-temperature formation energy for the reaction 2AH+Pt+2H2-A2PtH6 is
shown.
Fig. 3. Structure of Li2PtH6 (left) and Cs2PtH6 (right). The H atom framework in
Li2PtH6 is emphasized by red lines. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of this article.)

1788
K. Puhakainen et al. / Journal of Solid State Chemistry 183 (2010) 1785–1789
proton-exchanged. Earlier Bublitz et al. [34] prepared purposely
(a)
a series of mixed complexes K2[PtHnD6ꢀn] for spectroscopic
studies. Based on their results the bands at 1992 and 1384 cmꢀ1
can be assigned to Pt–H and Pt–D stretches, respectively, in
heteroleptic Li2[PtHnD6ꢀn]. The bends of such complexes appear
in the region 700–900 cmꢀ1. The Pt–D stretch for homoloeptic
Li2PtD6 is at 1315 cmꢀ1 (corresponding to an isotopic shift of
1.399), while the bend should coincide with the sharp edge of the
broad intensity feature at low frequencies. This edge is at
638 cmꢀ1 (corresponding to an isotopic shift of 1.393); the
bending mode is camouflaged by – what we think – a broad
*
895
889
1681
1735
1793
(b)
(c)
571
impurity band. The intensity ratio of stretches stemming from
2ꢀ
and homoleptic PtD26ꢀ complexes in
1840
heteroleptic [PtHnD6ꢀn
]
the IR spectrum of Fig. 4c matches very well that for K2[PtHnD6ꢀn
]
858
806
754
638
n¼1.5 reported by Bublitz et al. [34]. Thus we assume that our
isotopomer has a similar composition.
1992
1384
1315
3. Conclusions
500
1000
1500
Wavenumber (cm-1)
2000
2500
Li2PtH6, the missing member of the complex transition metal
hydride series A2PtH6 (A¼alkali metal), has been prepared by high
pressure hydrogenations employing multi-anvil (MA) techniques.
According to powder X-ray diffraction analysis Li2PtH6 is
isostructural to its heavier homologues and crystallizes in the
cubic K2PtCl6 structure. However, whereas PtH26ꢀ octahedral
complexes are well separated for the heavier homologues, H–H
distances within and between octahedra become almost equal for
Li2PtH6. Therefore the Li2PtH6 structure may likewise be regarded
as a defective perovskite structure where half of the octahedrally
coordinated atoms (cations) are missing. The IR observable T1 u
Pt–H stretching mode is highest for A¼Li (1840 cmꢀ1).
Fig. 4. IR spectra of Na2PtH6 (a), Li2PtH6 (b), and ‘‘Li2PtD6‘‘ (c). ‘‘Li2PtD6’’ has an
estimated composition Li2PtH1.5D4.5 The asterisk in spectrum (a) denotes an
impurity associated with the decomposition of extremely air/moisture sensitive
Na2PtH6. Italicized wave numbers in spectrum (c) refer to modes for heteroleptic
[PtHnD6ꢀn]
2ꢀ. The unlabeled mode at 1840 cmꢀ1 in spectrum (c) belongs to
[PtH6]2ꢀ (cf. spectrum (b)).
Raman) and by inelastic neutron scattering (INS) spectroscopy
[14,33,34]. The findings from these studies are excellent points of
reference.
High pressure hydrogenations offer exciting prospects for
metal hydrides synthesis because of the sharply increased activity
of hydrogen above 1 GPa. We facilitated this kind of synthesis for
MA techniques by introducing ammonia borane as superior
hydrogen source and by employing large-volume pressure cells.
The advantage of BH3NH3 is a well-defined decomposition to inert
residuals under high pressure and the easy access of BD3ND3 for
the preparation of deuterized samples. Employing large-volume
pressure cells ensures appreciable quantities of reaction product
despite the fact that a considerable fraction of sample volume is
occupied by the hydrogen source.
The IR spectrum of Na2PtH6 (Fig. 4a) shows the expected T1u
(asymmetric) Pt–H stretching and bending modes at 1793 and
895 cmꢀ1, respectively. The stretching mode appears peculiarly
split which – in exactly the same way – is also observed for the K
and Rb compounds [34]. This feature has been lately attributed to
extra bands observable because of Fermi resonance [33]. The Pt–H
stretching/bending mode frequencies increase slightly when
going from the Rb (1743/877 cmꢀ1) to the K compound (1748/
881 cmꢀ1). This trend continues with the Na compound, however,
the amount of increase is substantially larger.
The Pt–H stretching mode for Li2PtH6 appears at 1840 cmꢀ1
,
which represents a further substantial increase compared to
Na2PtH6, while the frequency of the bending mode (889 cmꢀ1) is
lower than that for the Na compound (Fig. 4b). The feature of
additional bands accompanying the stretch, which is character-
istic for the heavier homologues, is only vaguely recognizable for
Li2PtH6. The most obvious difference compared to the heavier
homologues is the occurrence of a broad and intense band at low
frequencies (with a maximum around 571 cmꢀ1). The origin of
this band is not clear. It is unlikely that it relates to libration or
translation. Librations (or torsions) of PtH26ꢀ octahedra are
inactive in optical spectroscopy for the heavier homologues
[14]. In the INS spectrum of Rb2PtH6 the librational mode is at
366 cmꢀ1 and translational modes are observed at around
100 cmꢀ1 [14]. Although translation has an IR active T1 u
component and the mass difference between Li and Rb is huge,
its frequency should not rise above 400 cmꢀ1 (OmRb/OmLiE3.5).
Therefore we attribute this band to an impurity, perhaps
stemming from oxidation/hydrolysis of the side product LiPt7
(LiPt7Hx).
4. Methods
High pressure hydrogenations were carried out in a 6–8
Walker-type multi-anvil (MA) high pressure module [35]. Reac-
tants were LiH (Aldrich, powder, 95%), LiD (from reacting Li
(Aldrich, rod, 99.99%) with D2 at a pressure of 5 bar at 750 1C for
12 h), NaH (Aldrich, powder, 95%), Pt (Alfa Aesar, powder, 99.9%),
BH3NH3, and BD3ND3. BH3NH3 was synthesized according to the
procedure described in Ref. [36]. BD3ND3 was obtained by
reacting NaBD4 with ammonium formate and subsequent deu-
teration of the amine protons with D2O (according to Ref. [37].
Here an additional deuteration cycle should have been performed
to achieve complete proton exchange.
The preparation of the sample capsule was preformed in an
Ar-filled glove box (O2 concentrationo0.1 ppm). Mixtures of
metal hydride(deuteride) and Pt with a total amount of about
100 mg were pressed into a pellet, which was placed in a NaCl
capsule between two pellets of BH3NH3(BD3ND3). Subsequently
the NaCl capsule was sealed. The sealed capsule was taken out
from the glove box and positioned inside a graphite furnace. This
arrangement was then placed together with a zirconia thermal
The spectrum of Li2PtD6 (Fig. 4c) reveals the presence of
heteroleptic D/H complexes. Apparently, the deuterium source
BD3ND3 used for its synthesis had not been completely

K. Puhakainen et al. / Journal of Solid State Chemistry 183 (2010) 1785–1789
1789
insulating sleeve in a magnesia octahedron with 18 mm edge
References
length. Samples were pressurized with 25 mm tungsten carbide
cubes truncated to 12 mm edge length. After reaching the target
pressure the samples were heated slowly to temperatures
between 450 and 500 1C and quenched after 60 min by turning
off the power to the furnace (quench rate ꢁ200 1C/min and at
approximately constant pressure). When BH3NH3 decomposes,
the sample volume shrinks considerably which may cause a
blowout. We found that a slow heating rate (1 1C/min) reduces
substantially the risk for such a blowout. The temperature was
measured close to the sample using a thermocouple type C
(W5%Re–W26%Re wire) in an Al2O3 sleeve. Afterward, the
pressure was released over a period of several hours and the
sample recovered in the glove box. Typically the three-pellet
arrangement of the sample preparation is still recognizable after
the run. All hydrides/deuterides had a grey color (from contam-
ination of small amounts of Pt or LiPt7) and could be easily
separated from the surrounding white residual of decomposed
BH3NH3(BD3ND3). Preparations for subsequent sample analysis
were performed in the glove box.
The products Li2PtH6, Li2PtD6 and Na2PtH6 were analyzed by
powder X-ray diffraction and IR spectroscopy. Samples were
ground, loaded into 0.3 mm capillaries, and measured on a Bruker
D8 Advance diffractometer fitted with an incident beam Ge
monochromator (transmission geometry; CuKa1 radiation). Lat-
tice parameters were obtained from least-squares refinement of
the measured and indexed lines of the corresponding powder
diffractograms [38]. The reflection positions were corrected by
use of an internal Si standard. The Rietveld refinements were
performed with the TOPAS software [39]. For the Rietveld ana-
lysis, the following parameters were refined: background, unit
cell, sample displacement, profile, isotropic atomic displacement
parameters, and absorption correction [40]. Fourier-transform
infrared spectroscopy (FTIR) investigations were carried out on a
Bruker IFS 66v/s instrument. KBr pellets were prepared in the
glove box and transferred in a closed container to the spectrometer.
For thermal decomposition experiments 10–20 mg amounts of
Li2PtH6 sample were pressed into a pellet, which was loaded in a
sealable stainless steel container in the glove box. The steel
container was heated for 20 h at different temperatures between
150 and 300 1C.
[1] K. Yvon, Chimia 52 (1998) 613.
[2] W. Bronger, J. Alloys Compd. 229 (1995) 1.
[3] T.K. Firman, C.R. Landis, J. Am. Chem. Soc. 120 (1998) 12650.
[4] M. Olofsson-Ma˚ rtensson, U. Ha¨ussermann, J. Tomkinson, D. Noreus, J. Am.
Chem. Soc. 122 (2000) 6960.
[5] U. Ha¨ussermann, H. Blomqvist, D. Noreus, Inorg. Chem. 41 (2002) 3684.
[6] W. Bronger, G. Auffermann, Chem. Mater. 10 (1998) 2723.
[7] D. Noreus, K.W. To¨rnroos, A. Bo¨rje, T. Szabo, W. Bronger, H. Spittank, G.
Auffermann, P. Mu¨ller, J. Less-Common Met. 139 (1988) 233.
[8] W. Bronger, G. Auffermann, J. Alloys Compd. 228 (1995) 119.
[9] W. Bronger, P. Mu¨ller, D. Schmitz, H. Spittank, Z. Anorg. Allg. Chem. 516
(1984) 35.
[10] W. Bronger, G. Auffermann, J. Alloys Compd. 219 (1995) 45.
[11] W. Bronger, G. Auffermann, Angew. Chem. Int. Ed. 33 (1994) 1112.
[12] W. Bronger, G. Auffermann, Z. Anorg. Allg. Chem. 621 (1995) 1318.
[13] W. Bronger, L.A. Brassard, Angew. Chem. Int. Ed. 34 (1995) 898.
[14] S.F. Parker, S.M. Bennington, A.J. Ramirez-Cuesta, G. Auffermann, W. Bronger,
H. Herman, K.P.J. Williams, T. Smith, J. Am. Chem. Soc. 125 (2003) 11656.
[15 (a) H. Hemmes, A. Driessen, R.J. Griessen, C Phys., Solid State Phys. 19 (1986)
3571;
(b) A. Driessen, P. Sa¨nger, H. Hemmes, R. Griessen, J. Phys., Condens. Matter 2
(1990) 9797.
[16] (a) H.-K. Mao, P.M. Bell, Science 203 (1979) 1004;
(b) J.V. Badding, R.J. Hemley, H.-K. Mao, Science 253 (1991) 421;
(c) R. Burtovyy, M. Tkacz, Solid State Commun. 131 (2004) 169.
[17] (a) N. Kawai, S. Endo, Rev. Sci. Instrum. 41 (1970) 1178;
(b) H. Huppertz, Z. Kristallogr. 219 (2004) 330.
[18] L.G. Khvostantsev, V.N. Slesarev, V.V. Brazhkin, High Pressure Res. 24 (2004)
371;
(b) L.G. Khvostantsev, V.N. Slesarev, Phys. Usp. 51 (2008) 1059.
[19] M. Tkacz, Defect Diffusion Forum 208-209 (2002) 107.
[20] E. Stoyanov, U. Ha¨ussermann, K. Leinenweber, High Pressure Res. 30 (2010)
175.
[21] (a) V.E. Antonov, J. Alloys Compd. 330-332 (2002) 110;
(b) V.E. Antonov, M. Baier, B. Dorner, V.K. Fedotov, G. Grosse, A.I. Kolesnikov,
E.G. Ponyatovsky, G. Schneider, F.E. Wagner, J. Phys.: Condens. Matter 14
(2002) 6427.
[22] Y. Fukai, J. Alloys Compd. 404-406 (2005) 7.
[23] Y. Fukai, N. Okuma, Phys. Rev. Lett. 73 (1994) 1640.
[24] H. Saitoh, A. Machida, Y. Katayama, K. Aoki, Appl. Phys. Lett. 93 (2008)
151918.
[25] (a) E. Ro¨nnebro, N. Kitamura, T. Sakai, J. Alloys Compd. 358 (2003)
216;
(b) J. Chen, T. Sakai, N. Kitamura, H.T. Takeshita, N. Kuriyama, J. Am. Chem.
Soc. 123 (2001) 6193;
(c) A. Kamegawa, Y. Goto, R. Kataoka, H. Takamura, M. Okada, Renew. Energy
33 (2008) 221;
(d) H. Takamura, H. Kakuta, Y. Goto, A. Kamegawa, M. Okada, Mater. Trans.
42 (2001) 1301;
(e) D. Kyoi, T. Sato, E. Ro¨nnebro, N. Kitamura, A. Ueda, M. Ito, S. Katsuyama, S.
´
Hara, D. Noreus, T. Sakai, J. Alloys Compd. 372 (2004) 213.
Theoretical calculations were performed in the framework of
the frozen core all-electron projected augmented wave (PAW)
method [41], as implemented in the program VASP [42]. The
energy cut-off was set to 500 eV. Exchange and correlation effects
were treated by the generalized gradient approximation (GGA),
usually referred to as PW91 [43]. The integration over the
Brillouin zone was done over a 15 ꢂ 15 ꢂ 15 Monkhorst-Pack
grid [44]. Total energies were converged to at least 1 meV/atom.
[26] J. Nyle´n, T. Sato, E. Soignard, J.L. Yarger, E. Stoyanov, U. Ha¨ussermann, J. Chem.
Phys. 131 (2009) 104506.
[27] M.G. Hu, J.M. van Paasschen, R.A. Geanangel, J. Inorg., Nucl. Chem. 39 (1977)
2147.
[28] W. Bronger, G. Klessen, P. Mu¨ller, J. Less-Common Met. 109 (1985) L1.
[29] X. Ke, I. Tanaka, Phys. Rev. B 71 (2005) 024117.
[30] This estimate neglects zero point energies, the effect of pressure and the
temperature dependence of the internal energy.
[31] B. Nacken, W. Bronger, J. Less-Common Met. 52 (1977) 323.
[32] W. Bronger, L.A. Brassard, Z. Anorg. Allg. Chem 622 (1996) 462.
[33] S.F. Parker, Coord. Chem. Rev. 254 (2010) 215.
[34] D. Bublitz, G. Peters, W. Preetz, G. Auffermann, W. Bronger, Z. Anorg. Allg.
Chem. 623 (1997) 184.
[35] D. Walker, M.A. Carpenter, C.M. Hitch, Am. Mineral. 75 (1990)
1020.
[36] Y. Lin, W.L. Mao, V. Drozd, J.H. Chen, L.L. Daemen, J. Chem. Phys. 129 (2008)
234509.
For all A2PtH6 systems the
H atom positional parameter
was optimized for a fixed volume of the unit cell. By repeating
this process for different volumes the equilibrium cell and
positional parameter are determined from the global minimum
energy.
[37] V.M. Parvanov, G.K. Schenter, N.J. Hess, L.L. Daemen, M. Hartl, A.C. Stowe,
D.M. Camaioni, T. Autrey, Dalton Trans. (2008) 4514.
[38] P.-E. Werner, Ark. Kemi 31 (1969) 513.
Acknowledgments
[39] Bruker AXS, TOPAS V4: General profile and structure analysis software for
powder diffraction data, User’s Manual, Karlsruhe, Germany, 2008.
[40] T.M. Sabine, B.A. Hunter, W.R. Sabine, C.J. Ball, J. Appl. Crystallogr. 31 (1998) 47.
[41] (a) P.E. Blo¨chl, Phys. Rev. B 50 (1994) 17953;
(b) G. Kresse, J. Joubert, Phys. Rev. B 59 (1999) 1758.
[42] (a) G. Kresse, J. Hafner, Phys. Rev. B 47 (1993) 558;
(b) G. Kresse, J. Furthmu¨ller, Phys. Rev. B 54 (1996) 11169.
[43] J.P. Perdew, J. Wang, Phys. Rev. B 45 (1992) 13244.
[44] H.J. Monkhorst, J.D. Pack, Phys. Rev. B 13 (1976) 5188.
This research has been supported by the ACS Petroleum
Research Fund under the Grant no. 45796-AC10 and the National
Science Foundation through Grants DMR-0638826 and
CHE-0742006. We are grateful to Dr. Stewart F. Parker (ISIS
facility, STFC Rutherford Appleton Laboratory (UK)) for valuable
advice in the analysis of IR spectra.