In situ synchrotron radiation powder X-ray diffraction study of the 2LiNH2 + LiH + KBH4 system

Article abstract of DOI:10.1016/j.jallcom.2013.02.175

In the present work we focus on the 2LiNH₂ + KBH₄ + LiH system: the phase-structural transformations occurring during the desorption process on the powder mixture are described by in situ synchrotron radiation powder X-ray diffraction (SR-PXD), high-pressure differential scanning calorimetry and manometric measurements. It is observed that LiNH₂ transforms into Li₂NH during heating, at about 160 C, while the reflections related to KBH₄ disappeared at 380 C. At higher temperature, the formation of Li₃BN₂ is detected, together with an evident increase of the background, ascribable to the presence of a further phase in the molten state. Patterns at room temperature, after cooling down the sample, confirm the presence of Li₃BN₂ and KH as reported in the theoretical study. For the as prepared mixture it is possible to achieve the theoretical hydrogen gravimetric capacity of 7.4 wt%.

Full text of DOI:10.1016/j.jallcom.2013.02.175

Journal of Alloys and Compounds 580 (2013) S278–S281
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
In situ synchrotron radiation powder X-ray diffraction study
of the 2LiNH + LiH + KBH system
2
4
a
b
a
a
c
Maddalena Sale , Claudio Pistidda , Alessandro Taras , Emilio Napolitano , Chiara Milanese ,
b
b
a,
⇑
a
a
Fahim Karimi , Martin Dornheim , Sebastiano Garroni , Stefano Enzo , Gabriele Mulas
a
Department of Chemistry and Pharmacy, University di Sassari and INSTM, Via Vienna 2, Sassari I-07100, Italy
Institute of Materials Research, Materials Technology, Helmholtz-Zentrum Geesthacht, Max-Planck, Str. 1, Geesthacht D-21502, Germany
Pavia H Lab, C.S.G.I. & Dipartimento di Chimica, Sezione di Chimica Fisica, Università di Pavia, Viale Taramelli 16, Pavia I-27100, Italy
2
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 14 March 2013
2 4
In the present work we focus on the 2LiNH + KBH + LiH system: the phase-structural transformations
occurring during the desorption process on the powder mixture are described by in situ synchrotron radi-
ation powder X-ray diffraction (SR-PXD), high-pressure differential scanning calorimetry and manomet-
ric measurements. It is observed that LiNH transforms into Li NH during heating, at about 160 °C, while
Keywords:
Hydrogen storage in solid state
Complex hydrides
In situ synchrotron radiation powder x-ray
diffraction (SR-PXD)
Lithium amide
2
2
4 3 2
the reflections related to KBH disappeared at 380 °C. At higher temperature, the formation of Li BN is
detected, together with an evident increase of the background, ascribable to the presence of a further
phase in the molten state. Patterns at room temperature, after cooling down the sample, confirm the
presence of Li
to achieve the theoretical hydrogen gravimetric capacity of 7.4 wt%.
Ó 2013 Elsevier B.V. All rights reserved.
3 2
BN and KH as reported in the theoretical study. For the as prepared mixture it is possible
Lithium hydride
1
. Introduction
the recharging process evolves important quantities of heat, hardly
to be dispersed in a short time.
ꢀ
1
Hydrogen storage in complex hydrides has been extensively
Optimal windows are estimated to lie in the 15–25 kJ mol and
ꢀ1
ꢀ1
investigated as a safest and promising solution compared with
the other technologies such as gas compression and liquefaction
90–110 J K mol ranges for dehydrogenation enthalpy and en-
tropy, respectively [7]. Thus, the present focus of the hydrogen
storage research is to produce a material, in pure or composite
form, which can satisfy such values.
[
1,2].
The wide interest towards the classes of light and complex
hydrides, as borohydrides M(BH ) and alanates M(AlH ) is justified
by their low molecular mass and high gravimetric storage capaci-
ties, up to 18 wt% H in the case of LiBH [3]. Unfortunately, due
to the high thermodynamic stability, dissociation enthalpies of
4
4
Thermodynamic calculations based on first-principles have
represented a useful tool in order to find promising mixture of
hydrides [8,9]. In this contest, thanks to the implementation of
crystal database (>200 solid compounds), Alapati and coauthors
have explored a wide range of metal hydrides at different temper-
atures, detecting 43 very promising single-step reactions [10].
Most recently, Sholl et al. presented the largest set of thermody-
namic calculations of reversible hydrogen storage systems using
a database of 359 crystalline materials [11]. Among several sys-
2
4
ꢀ1
these compounds are near 200 kJ mol
2
H , and working tempera-
tures are too high for application purposes. To overcome it, multi-
component hydride systems based on the mixture of complex
hydrides and light-metal hydrides represents an attractive class
2
of materials [4,5]. Similarly, systems with N–H bonds, as LiNH –
LiH, have high storage density (i.e. 9.6 wt.%) but desorption occurs
above 400 °C [6]. The challenge is to develop synthetic strategies
that allow modulating thermodynamic and kinetic properties in
order to achieve an effective hydrides destabilization. The thermo-
dynamic performance of each hydride is controlled by the enthalpy
and entropy of hydrogenation/dehydrogenation. The two quanti-
ties are correlated, and in the case of high enthalpy values, a con-
siderable amount of energy is required to extract hydrogen, while
tems identified, the LiNH
2:1:1 represents a very attractive composite: it shows a hydrogen
gravimetric capacity of 7.48 wt% and a theoretical 300 value
equal to 28.1 kJ/mol H [11]. The van’t Hoff plot based on
the DFT calculations shows that for the one-step reaction
2LiNH + KBH + LiH ? Li BN + KH + 4H the sorption tempera-
tures are expected between 27 °C and 147 °C in a range of H pres-
sures from 1 to 100 bar, i.e. appropriate for reversible hydrogen
storage applications [11]. These encouraging theoretical results
suggest that this system is a good mixture for experimental study.
In this work, we investigated the thermal hydrogen
2 4
–KBH –LiH mixture in the molar ratio
D
H
2
2
4
3
2
2
2
⇑
Corresponding author. Tel.: +39 079 229524; fax: +39 079 229559.
E-mail address: sgarroni@uniss.it (S. Garroni).
2 4
desorption process of 2LiNH + KBH + LiH system, by using
0
925-8388/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jallcom.2013.02.175

M. Sale et al. / Journal of Alloys and Compounds 580 (2013) S278–S281
S279
in situ synchrotron radiation powder X-ray diffraction (SR-PXD),
high-pressure differential scanning calorimetry (HP-DSC), and
manometric measurements. The combination of these techniques
gives a first tangible result of the reaction mechanism of the ter-
nary system.
The measurement, shown in Fig. 2, reveals three main peaks
during the heating step: two exothermic peaks with onset at
160 °C (A) and 330 °C (B), respectively, and one strong endothermic
peak at 360 °C (C). Concerning the cooling step, three exothermic
events at 260 °C (D), 210 °C (E) and 160 °C (F), are detected.
According to the manometric measurement reported in Fig. 1 a
significant release of gas is only observed for the reactions that
correspond to A and C peaks, confirming a multi-step desorption
reaction.
2
. Experimental details
Commercial powders of KBH
4
(99% purity), LiNH
2
(95% purity) and LiH (95%
–LiH–
in a molar ratio 2:1:1 was prepared by manual mixing and subsequently grin-
ded using a mortar inside a glove box device (MBraun-20-G), with O , N and H
purity) were purchased from Sigma–Aldrich. The ternary mixture of LiNH
KBH
2
To clarify exhaustively the desorption reaction mechanism,
in situ SR-PXD were performed on the 2LiNH –KBH –LiH mixture,
2 4
4
2
2
2
O
applying the same thermal treatment conditions used in the HP-
DSC analysis.
Fig. 3 shows the series of patterns collected from room tempera-
ture to 400 °C with a scanning rate of 5 °C/min. At room temperature
the starting material reflections corresponding to the crystalline
levels below 1 ppm. The phases involved during the desorption process were char-
acterized by means of in situ synchrotron powder X-ray diffraction (SR-PXD), per-
formed at the beamline D3 in the research laboratory HASYLAB, DESY, Hamburg.
For the experiment, a new high pressure sample cell designed for in situ monitoring
of solid/gas reactions was utilized [12,13]. The sapphire capillary was loaded with
the as-prepared powders inside a glove box machine. The in situ measurement
was performed at 1 bar of Ar pressure. Each XRPD pattern was collected with an
exposure time of 60 s in the 2h range of 2–40°, using a wavelength, k, of
phases of KBH
4
2
and LiNH are detected. LiH peaks are instead over-
lapped to those of KBH
4
. In addition, small traces of LiOH are also de-
0
.49902 Å. The material was first heated from 30 °C to 400 °C with a heating rate
tected. The presence of LiOH is most likely due to a partial oxidation
of as-received LiH (see also Fig. 4A). During heating, all peaks are
shifted to lower 2h angle because of the continuous increase in the
lattice cell parameters of the phases associated with the thermal
expansion. As also indicated by DSC (event A in Fig. 2), around
of 5 °C/min and then kept for 15 min at 400 °C. XRPD patterns were also collected
during cooling step (5 °C/min). All the raw SR diffraction data were elaborated
and converted to powder patterns by the use of the FIT2D program [14]. The differ-
ent phases and their microstructural parameters were evaluated by fitting the XRD
patterns using the MAUD (Materials Analysis Using Diffraction) Rietveld refinement
software [15]. Manometric and calorimetric measurements were performed by a
manometric apparatus (PCTPro-2000, Setaram & Hy-Energy) and a high-pressure
calorimeter (Sensys DSC, Setaram, pressure measurement accuracy: 1% of reading),
respectively. For the manometric apparatus, the experiments were carried out by
heating about 300 mg of powder from room temperature to 400 °C at 5 °C/min un-
der static vacuum, followed by 150 min of isothermal step at 400 °C. Regarding the
calorimetric analysis (HP-DSC), the high-pressure cell of the calorimeter was loaded
with 50 mg of the as-prepared sample in the glove-box under continuously purified
atmosphere. The temperature programmed desorption (TPD) curves were obtained
by heating the samples from room temperature up to 400 °C with a scanning rate of
1
60 °C a first transformation occurs (blue dash in Fig. 3), correspond-
ing to the formation of Li NH from the reaction of LiNH and LiH. To
reinforce this, single XRPD scan (#45) recorded at 166 °C is shown in
Fig. 4 B. As reported in literature [16] the decomposition of LiNH
2
2
2
starts at 200 °C; however it was demonstrated that the addition of
LiH in different molar ratio contributed to decrease the decomposi-
tion temperature [17]. Furthermore, this reaction is accompanied by
the evolution of ammonia and hydrogen as shown in the following:
5
°C/min in an atmosphere of 1 bar of pure helium.
2
LiNH
2
! Li
2
NH þ NH
3
ðgÞ
ð1Þ
ð2Þ
3
. Results and discussion
NH
3
ðgÞ þ LiH ! LiNH
2
þ H
2
The desorption analysis of the reacting 2LiNH
ture reveals a multi-step path, displayed in Fig. 1 as H
2
–KBH
4
–LiH mix-
wt% vs.
time. The first desorption step starts at around 150 °C with a re-
lease of about 1.0 wt% of H , while the second step takes places
with much faster kinetics at 380 °C resulting in a larger release
of H . The full dehydrogenation is reached after 20 min of thermal
treatment at 400 °C with a total weight loss of 7.40 wt% of H
The global reaction is assumed to be [18]:
2
2LiNH
2
þ LiH ! LiNH
2
þ Li
2
NH þ H
2
ð3Þ
2
As displayed in Fig. 3, lithium immide and lithium amide are
still present at temperatures over 350 °C (scan #110), and in par-
ticular the intensity associated with the LiNH reflections signifi-
2
cantly increase during this step, as also confirmed in Fig. 4C. This
aspect is probably ascribed to a coherent scattering domain growth
2
2
,
which is rather consistent with the theoretical gravimetric capacity
of the system (7.48 wt%).
In order to define the sequences of reactions taking place during
the desorption process, HP-DSC analysis was performed on the
as-prepared hydride mixture.
of the LiNH
80 °C, corresponding to the scan number 115 in Fig. 3, some yet
unassigned peaks appear with a significant decrease of the LiNH
2
crystals before of the full decomposition. Around
3
2
,
Fig. 2. Calorimetric signal corresponding to the dehydrogenation process of the
2LiNH –KBH –LiH mixture). Calorimetric profile: solid line. Temperature profile:
dashed line.
Fig. 1. Thermal programmed desorption profile acquired on the 2LiNH
LiH mixture. Manometric signal: solid line. Temperature profile: dash-dotted line.
2
+ KBH
4
+ -
2
4

S280
M. Sale et al. / Journal of Alloys and Compounds 580 (2013) S278–S281
intensities are vanished. As also resulted from the scan 120, the
KBH disappearance is followed by the background increasing at
low 2h angle that can be associated to a molten phase. At the
end of the annealing process, only Li BN and molten phases are
detected in the patterns as evinced by Fig. 4D. The formation of
the Li BN phase was also reported in a previous work by Orimo
et al. [19]. They proved that the LiNH /LiBH system in the molar
ratio 2:1 desorbs hydrogen forming Li BN , Li N–BN and Li BN
To the contrary, in this work the formation of Li BN is not ob-
served in the in situ XRD measurements. A possible explanation
for the not observed phase is that, in the ternary system, KBH
starts to decompose only after 380 °C, whereas Li BN melts al-
ready at about 200 °C. Further experimental investigations are re-
quired to clarify the Li BN formation from the reaction of Li NH,
LiH and KBH
4
3
2
3
2
2
4
3
H
2 8
3
3
2
.
3
2 8
H
4
3
2 8
H
3
2
2
4
.
During the cooling step (not shown here), at roughly 260 °C the
formation of KH and the crystallization of potassium, the latter in a
form stabilized at high pressure, with the vanishing of the diffuse
halo are achieved. Both processes are linked to the exothermic
peaks observed in the HP-DSC analysis during the cooling. As
shown in Fig. 4E, no further changes occur during cooling until
the measurement was stopped at 25 °C. It is interesting to note
that, in accordance with the theoretical study of Sholl et al. [11],
2 4
Fig. 3. In situ SR-PXD measurements of the 2LiNH –KBH –LiH mixture (2D plot).
The analysis was carried out under 1 bar of Ar, heating the material from RT to
4
4
00 °C (5 °C/min) and then keeping it for 15 min under isothermal conditions at
00 °C. The thermal profile (red line) is reported as a function of the time and
pattern scans. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
3 2
the formation of KH, as well as that of the ternary phase Li BN ,
is experimentally demonstrated. However, one step desorption
reaction is not observed and partial decomposition of KH is
achieved at 400 °C. On this way, more efforts have to be devoted
to decrease the hydrogen desorption temperature and understand
the chemical nature of the unknown phase.
4
. Conclusions
In this work we reported an experimental study on the
decomposition of the 2LiNH
that the desorption of the 2:1:1 mixture follows a multi-step reac-
tion with a total weight loss of 7.40 wt% of H , consistent with the
2 4
–KBH –LiH system. It was observed
2
theoretical gravimetric capacity of the system (7.48 wt%). In situ
SR-PXD and HP-DSC experiments were performed to study the
dehydrogenation mechanism and characterize the decomposition
products. During heating, the partial decomposition of LiNH
accompanied by the formation of Li NH at 160 °C with a small re-
lease of gas. At higher temperature, around 380 °C, the reflections
of the LiNH , Li NH and KBH phases completely disappeared to
2
was
2
2
2
4
give a diffuse halo corresponding to a liquid-like phase plus an
unstable unassigned phase. This step was characterized by a strong
release of hydrogen. In addition, the ternary phase Li
detected. Upon cooling, the halo was vanished and, in addition to
the Li BN phase, there was weak appearance of KH and K cubic
phases. Interesting, the formation of Li BN and KH was also ex-
3 2
BN was also
3
2
3
2
pected by theoretical calculations, but the experimental tempera-
ture at which the reaction occurred was considerably higher
(Texp = 400 °C vs. Tth = 30 °C). However, considering that no experi-
mental evidences have been achieved up to now for this attractive
system, the present results bring new contributes to the study of
2 4
the reaction mechanism of the 2LiNH + LiH + KBH mixture.
Fig. 4. In situ SR-PXD patterns of the mixture collected at the temperature of 40 °C
(
(
A), 166 °C (B), 355 °C (C), 400 °C (D) during heating, followed by cooling at 300 °C
E) and 40 °C (F). Dots are experimental data, full red lines are from the Rietveld fit.
Acknowledgments
The bars at the bottom indicate the line positions expected for each phase
appearing in the various patterns examined. (For interpretation of the references to
color in this figure legend, the reader is referred to the web version of this article.)
This work was funded by COST Action MP1103: ‘‘Nanostructured
Materials for Solid State Hydrogen Storage’’, by MIUR (Italian Minis-
try for University and Research), in the frame of the PRIN Project
‘‘Synthesis, characterization and functional evaluation of light
hydrides-based nanostructured materials and nanoparticles for
solid state hydrogen storage’’, and by University of Sassari. S.G.
acknowledges his postdoctoral fellowship, supported by the
2 4
Li NH and KBH diffracted intensities. The unassigned phase is
found to be stable just below 400 °C, and then its diffraction peaks
disappear during the isothermal treatment.
The formation of the tetragonal Li
3
BN
2
phase is observed at
diffracted
4
00 °C (scan #120), while the LiNH , Li
2
2
NH and KBH
4

M. Sale et al. / Journal of Alloys and Compounds 580 (2013) S278–S281
S281
Fondazione Banco di Sardegna. Authors acknowledge beamline D3
in the research laboratory HASYLAB, DESY.
[10] S.V. Alapati, J.K. Johnson, D.S. Sholl, J. Phys. Chem. C. 112 (2008) 5258–5262.
[
[
11] K.C. Kim, A.D. Kulkarni, J.K. Johnson, D.S. Sholl, Phys. Chem. Chem. Phys. 13
2011) 7218–7229.
12] U. Bösenberg, S. Doppiu, L. Mosegaard, G. Barkhordarian, N. Eigen, A.
Borgschulte, T.R. Jensen, Y. Cerenius, O. Gutfleisch, T. Klassen, M. Dornheim,
R. Bormann, Acta Mater. 55 (2007) 3951–3958.
(
References
[
13] C. Pistidda, S. Garroni, C.B. Minella, F. Dolci, T.R. Jensen, P. Nolis, U. Bosenberg,
Y Cerenius, W. Lohstroh, M. Fichtner, M.D. Baró, R. Bormann, R. Dornheim, J.
Phys. Chem. C. 114 (2010) 21816–21823.
14] http://www.esrf.eu/computing/scientific/FIT2D/.
15] P. Scardi, L. Lutterotti, P. Maistrelli, Powder Diffr. 9 (1994) 180–186.
16] P. Cheng, Z. Xiong, J. Luo, J. Lin, Hydride J. Chem. B107 (2003) 10967–
[
[
1] L. Schlapbach, A. Zuttel, Nature 414 (2001) 353–358.
2] S. Orimo, Y. Nakamori, J.R. Eliseo, A. Zuttel, C.M. Jensen, Chem. Rev. 107 (2007)
4
111–4132.
[
[
[
[
[
3] J. Graetz, J. Chem. Soc. Rev. 38 (2009) 73–82.
4] S. Garroni, C. Milanese, A. Girella, A. Marini, G. Mulas, E. Menéndez, C. Pistidda,
M. Dornheim, S. Suriñach, M.D. Baró, Int. J. Hydrogen Energy 35 (2010) 5434–
1
0970.
17] G. Miceli, C.S. Cucinotta, M. Bernasconi, M. Parrinello, J. Phys. Chem. B114
2010) 15174–15183.
5
441.
[
[
[
[
5] C. Milanese, S. Garroni, A. Girella, G. Mulas, V. Berbenni, G. Bruni, S. Suriñach,
M.D. Baró, A. Marini, J. Phys. Chem. C. 115 (2011) 3151–3162.
6] P. Chen, Z.T. Xiong, J.Z. Lou, J.Y. Lin, K.L. Tan, Nature 420 (2002) 302–304.
7] W.I.F. David, Faraday Discuss. 151 (2011) 399–414.
8] A.R. Akbarzadeh, V. Ozolins, C. Wolverton, Adv. Mater. 19 (2007) 3233–3239.
9] V. Ozolins, E.H. Majzoub, C. Wolverton, J. Am. Chem. Soc. 131 (2009) 230–237.
(
18] W.I.F. David, M.O. Jones, D.H. Gregory, C.M. Jewell, S.R. Johnson, A. Walton, P.P.
Edwards, J. Am. Chem. Soc. 129 (2007) 1594–1601.
19] Y. Nakamori, A. Ninomiya, G. Kitahara, M. Aoki, T. Noritake, K. Miwa, Y. Kojima,
S. Orimo, J. Power Sources 155 (2006) 447–455.
[
[
[
[