Effect of CO on hydrogen storage performance of KF doped 2LiNH2 + MgH2 material

Article abstract of DOI:10.1016/j.jallcom.2014.07.128

(2LiNH₂ + MgH₂) material is one of the most promising hydrogen storage materials. In applications, CO impurity in hydrogen has an effect on the hydrogen storage performance of (2LiNH₂ + MgH ₂) material. In this

Full text of DOI:10.1016/j.jallcom.2014.07.128

Journal of Alloys and Compounds 616 (2014) 47–50
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Effect of CO on hydrogen storage performance of KF doped
2LiNH2 + MgH2 material
⇑
Fei Sun, Min-yan Yan, Jian-hua Ye, Xiao-peng Liu , Li-jun Jiang
Department of Energy Materials and Technology, General Research Institute for Non-Ferrous Metals, Beijing 100088, China
a r t i c l e i n f o
a b s t r a c t
Article history:
(2LiNH2 + MgH2) material is one of the most promising hydrogen storage materials. In applications, CO
impurity in hydrogen has an effect on the hydrogen storage performance of (2LiNH2 + MgH2) material.
In this work, the effect of CO on the hydrogen storage properties of KF doped (2LiNH2 + MgH2) material
was investigated by using hydrogen containing 1 mol% CO as the hydrogenation gas source. The results
indicate that the hydrogen desorption capacity of the hydride decreases from 4.73 wt.% to 3.88 wt.% after
5 hydrogenation cycles. The decrease of the capacity is attributed to the permanent loss of the NH2
caused by the formation of Li2CN2 and KCN. Moreover, the hydrogen desorption kinetics declines
obviously, and the dehydrogenation activation energy increases from 122.1 kJ/mol to 132.0 kJ/mol.
Furthermore, it is found that the hydrogen desorption kinetics cannot be recovered to the initial level
after re-milling. The main reason for these is that the catalytic effect on (Mg(NH2)2 + 2LiH + 0.05KF) sys-
tem disappears due to the transformation from KH to KCN.
Received 4 March 2014
Received in revised form 15 July 2014
Accepted 17 July 2014
Available online 24 July 2014
Keywords:
Hydrogen storage
Lithium–magnesium amide
CO impurity
Hydrogen capacity
Kinetics
Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction
reaction of the (Mg(NH2)2 + 2LiH) system. Furthermore, the KF
doped (Mg(NH2)2 + 2LiH) exhibits
a good hydrogen storage
The (2LiNH2 + MgH2) material has been regarded as a potential
hydrogen storage system due to its suitable operation temperature
and high reversible hydrogen storage capacity of 5.6 wt.% among
numerous of magnesium-based hydrides [1–10]. (2LiNH2 + MgH2)
transforms to Li2Mg(NH)2 in the first hydrogen desorption process,
and then it experiences the reversible process between
(Li2Mg(NH)2 + 2H2) and (Mg(NH2)2 + 2LiH) for the hydrogen
absorption and desorption [3].
The theoretical dehydrogenation temperature of the
(Mg(NH2)2 + 2LiH) is about 90 °C at 1.0 bar equilibrium hydrogen
pressure [4]. However, the high operating temperature and low
kinetics are still the main barriers for the practical application
of the (Mg(NH2)2 + 2LiH) material. In the last decade, efforts
have been focused on lowering the operating temperature and
improving the kinetics of Li–Mg–N–H system [11–25]. Liu et
al. [15] reported that the hydrogen storage performance of
(Mg(NH2)2 + 2LiH) system was enhanced by introducing KF. It
is found that KH, converted from KF, acted as a catalyst to
decrease the activation energy of the first-step dehydrogenation,
and meanwhile a reactant to reduce the desorption enthalpy
change of the second step. This provides a synergetic thermody-
namic and kinetic destabilization in the hydrogen storage
reversibility. In this paper, the experimental material is referred
to as (2LiNH2 + MgH2 + 0.05KF). The hydrogen desorption/
absorption reaction is shown as following:
MgðNH2Þ2 þ 2LiH þ 0:05KH () Li2MgðNHÞ2 þ 2H2
þ 0:05KH
ð1Þ
For practical application, hydrogen source contains reactive and
inert impurities, such as CO, CO2, O2, H2O, N2, and CH4. These impu-
rities may have effects on the performance of hydrogen storage
materials. The influence of the impurities on hydrogen storage
alloys was widely investigated [26–38]. It was reported that the
absorption kinetics of LaNi5 and LaNi4.73Sn0.27 are strongly retarded
by CO contamination [28]. However, thus far there has been little
research on amides [39–42]. Luo and Ruckenstein [41] reported
that water-saturated air could slightly improve the sorption kinet-
ics and final hydrogen capacity. In the present work, hydrogen con-
taining 1 mol% CO is employed as the hydrogenation gas source to
study whether the hydrogen storage performance of the
(2LiNH2 + MgH2 + 0.05KF) system is affected by CO impurity and
how it works in practical application.
2. Experimental
The starting materials LiNH2 (95% purity, Sigma–Aldrich) and KF (99% purity,
Sigma–Aldrich), were used without further purification. The MgH2 powder was pre-
pared by mechanically milling Mg powder under an initial hydrogen pressure of
⇑
Corresponding author. Tel.: +86 10 82241239.
E-mail address: xpgliu@126.com (X.-p. Liu).
http://dx.doi.org/10.1016/j.jallcom.2014.07.128
0925-8388/Ó 2014 Elsevier B.V. All rights reserved.

48
F. Sun et al. / Journal of Alloys and Compounds 616 (2014) 47–50
5. Its reversible hydrogen storage capacity is more than 4.7 wt.%. In
Fig. 1(b), the down triangle represents the first desorption of the
Sample 2 which is hydrogenated with purified H2, and the other
symbols represent the desorption from the 2nd to 5th of Sample
2 which is hydrogenated with H2 containing 1 mol% CO. During
the first cycle, the sample releases 4.73 wt.% H2 within 45 min.
But when H2 containing 1 mol% CO is employed as the hydrogena-
tion source, the capacity declines obviously from the 2nd to 5th
cycle. After 5 cycles, the hydrogen desorption capacity is only
3.88 wt.%. Moreover, the hydrogen desorption kinetics decreases
obviously. It takes 6 min, 9 min, 11 min, 12 min and 14 min to
release 3.5 wt.% H2 from 1st to 5th cycles, respectively. Therefore,
CO in hydrogen source obviously declines the hydrogen desorption
capacity and kinetics of the (Mg(NH2)2 + 2LiH + 0.05KF) system.
3.2. Phase characterization
Figs. 2 and 3 show XRD patterns and FTIR spectra for the
(Mg(NH2)2 + 2LiH + 0.05KF) mixture after hydrogenation (Figs.
2(a) and 3(a)) and dehydrogenation (Figs. 2(b) and 3(b)), respec-
tively. In Figs. 2 and 3, upper lines represent patterns of Sample
1 which is hydrogenated with purified H2, and bottom lines stand
for Sample 2 which has experienced 5 hydrogenation and dehydro-
genation cycles with H2 containing 1 mol% CO. It can be seen from
Fig. 2 that the main components for Sample 1 after hydrogenation
are Mg(NH2)2, LiH and KH, and after dehydrogenation the main
Fig. 1. Hydrogen desorption kinetics of the (Mg(NH2)2 + 2LiH + KF) mixture for (a)
Sample 1 and (b) Sample 2 measured at 220 °C.
3 MPa for 40 h, followed by hydrogenation at 400 °C for 4 h under 8 MPa hydrogen
pressure. A mixture of LiNH2, MgH2 and KF with molar ratio of 2:1:0.05 was
mechanically milled for 10 h under 3 MPa hydrogen pressure. All the milling treat-
ment was done by using a Spex-8000 apparatus, and the ball-to-powder ratio was
around 10:1. The material handing was performed in a glovebox filled with purified
argon to keep the H2O and O2 levels below 1 ppm.
The hydrogen desorption kinetics measurements were carried on a Sieverts-
type apparatus. The as-milled (2LiNH2 + MgH2 + 0.05KF) mixture yielded two
samples (each weighs about 700 mg). Sample 1 was the reference group which
experienced hydrogenation with purified H2 for 5 cycles. Sample 2 was regarded
as the experimental group which was firstly hydrogenated with purified H2, and
then was hydrogenated with H2 containing 1 mol% CO in the following 4 cycles.
The initial pressures of dehydrogenation and hydrogenation were about 0 MPa
and 8 MPa, and the end pressures were about 0.2 MPa and 7.9 MPa respectively.
All the measurements were carried at 220 °C for 4 h.
Differential scanning calorimetry (DSC) was conducted by using Mettler-Toledo
DSC 1 to study the dehydrogenation activation energy of the samples. The samples
(about 18 mg for each) were heated from 30 °C to 350 °C at various heating rate,
e.g., 1, 3, 5, 8 and 10 °C/min, under Ar gas (99.999%) with flow rate of 30 mL/min.
XRD analysis was carried out by using X’pert Pro MPD diffractometer with Cu K
a
radiation at 40 kV and 40 mA over the range of 10–90°. The FTIR absorption
spectrum of N–H and C–N was collected in diffuse reflectance infrared Fourier
transform mode.
3. Results and discussion
3.1. Hydrogen desorption kinetics
Fig. 1 shows the hydrogen desorption kinetics of Sample 1
(Fig. 1(a)) and Sample 2 (Fig. 1(b)). It is seen from Fig. 1(a) that
the hydrogen desorption curves change little from cycle 1 to cycle
Fig. 2. Powder X-ray diffraction patterns of the (Mg(NH2)2 + 2LiH + 0.05KF) samples
after (a) absorption and (b) desorption.

F. Sun et al. / Journal of Alloys and Compounds 616 (2014) 47–50
49
One can see from FTIR spectra that the absorbance around
3200 cmꢀ1 suggests the existence of N–H, and the peaks around
2900 cmꢀ1 are attributed to Nujol oil. Compared with Sample 1,
an absorbance peak at 2131 cmꢀ1 attributed to the C–N vibration
appears in Sample 2.
The XRD and FTIR analysis suggest that a bond break in CO
occurs during hydrogenation when H2 containing 1 mol% CO is
employed as the hydrogen source. Then the carbon reacts with
Li+, K+ and N3ꢀ to form Li2CN2 and KCN. And the oxygen reacts with
Mg2+ to form MgO. The reaction is shown as following:
MgðNH2Þ2 þ 2LiH þ KH þ CO ! Li2MgðNHÞ2 þ Li2CN2 þ KCN
þ MgO þ H2
ð2Þ
The reaction is irreversible because the new products of Li2CN2,
KCN and MgO are found in hydrogenation and dehydrogenation
products. It is well known that nitrogen in NH2ꢀ or (NH2)ꢀ plays
a key role in Li–Mg–N–H system, because the conversion between
NH2ꢀ and (NH2)ꢀ is the main reaction process during hydrogena-
tion and dehydrogenation. So the permanent loss of nitrogen
results in the decline of hydrogen storage capacity.
3.3. Dehydrogenation activation energy
To understand the decline of the dehydrogenation kinetics more
deeply, differential scanning calorimetry (DSC) was conducted to
study the dehydrogenation activation energy of samples. Fig. 4
shows the DSC profiles for Sample 1 and Sample 2 after 5 hydroge-
nation and dehydrogenation cycles. Two Kissinger plots were
drawn to estimate the activation energies. For Sample 2, the overall
peak positions shift to the right direction of 45–50 °C compared to
those of Sample 1. According to Kissinger theory [43], the activa-
tion energies of hydrogen desorption are 122.1 kJ/mol and
132.0 kJ/mol for Samples 1 and 2, respectively. The activation
energy increases 8% for the sample after hydrogenated with hydro-
gen containing 1 mol% CO, resulting in the decline of hydrogen
desorption kinetics of the (Mg(NH2)2 + 2LiH + 0.05KF) sample.
There are two possible reasons for the decrease of the hydrogen
desorption kinetics. One is that the new products of Li2CN2, KCN
and MgO formed on the surface of material particles prevent sub-
stance transmission during the dehydrogenation process. The
Fig. 3. FT-IR spectra of the (Mg(NH2)2 + 2LiH + 0.05KF) samples after (a) absorption
and (b) desorption.
component is Li2Mg(NH)2. After employing H2 containing 1 mol%
CO as the hydrogen source, KH disappears, and peaks of Li2CN2,
KCN and MgO appear.
Fig. 4. Kissinger plots and DSC curves of the (Mg(NH2)2 + 2LiH + 0.05KF) samples.

50
F. Sun et al. / Journal of Alloys and Compounds 616 (2014) 47–50
2010CB631305) under the Ministry of Science and Technology of
China.
References
[1] Z. Xiong, G. Wu, J. Hu, P. Chen, J. Adv. Mater. 16 (2004) 1522–1525.
[2] W. Luo, J. Alloys Comp 381 (2004) 284–287.
[3] W. Luo, E. Rönnebro, J. Alloys Comp. 404–406 (2005) 392–395.
[4] Z. Xiong, J. Hu, G. Wu, P. Chen, W. Luo, K. Gross, J. Wang, J. Alloys Comp. 398
(2005) 235–239.
[5] L. Ouyang, X. Yang, H. Dong, M. Zhu, Scr. Mater. 61 (2009) 339–342.
[6] L. Ouyang, S. Ye, H. Dong, M. Zhu, Appl. Phys. Lett. 90 (2007) 021917.
[7] M. Zhu, H. Wang, L. Ouyang, M. Zeng, Int. J. Hydrogen Energy 31 (2006) 251–
257.
[8] M. Zhu, Y. Gao, X. Che, Y. Yang, C. Chung, J. Alloys Comp. 708 (2002) 330–332.
[9] G. Liu, Y. Wang, L. Jiao, H. Yuan, Int. J. Hydrogen Energy 39 (2014) 3822–3829.
[10] G. Liu, F. Qiu, J. Li, Y. Wang, L. Li, C. Yan, L. Jiao, H. Yuan, Int. J. Hydrogen Energy
37 (2012) 17111–17117.
[11] C. Liang, Y. Liu, M. Gao, H. Pan, J. Mater. Chem. A 1 (2013) 5031–5036.
[12] T. Durojaiye, A. Coudy, Int. J. Hydrogen Energy 37 (2012) 3298–3304.
[13] A. Sudik, J. Yang, D. Halliday, C. Wolverton, J. Phys. Chem. C 111 (2007) 6568–
6573.
[14] J. Hu, Y. Liu, G. Wu, Z. Xiong, Y.S. Chua, P. Chen, J. Chem. Mater. 20 (2008)
4398–4402.
Fig. 5. Hydrogen desorption kinetics of there-milled sample measured at 220 °C.
[15] Y. Liu, C. Li, B. Li, M. Gao, H. Pan, J. Phys. Chem. C 117 (2013) 866–875.
[16] J. Wang, T. Liu, G. Wu, W. Li, Y. Liu, C. Moyses Araujo, R.H. Scheicher, A.
Bomqvist, R. Ahuja, Z. Xiong, P. Yang, M. Gao, H. Pan, P. Chen, Angew. Chem.
Int. Ed. 48 (2009) 5828–5832.
[17] C. Liang, Y. Liu, H. Fu, Y. Ding, M. Gao, H. Pan, J. Alloys Comp. 509 (2011) 7844–
7853.
[18] C. Li, Y. Liu, Y. Pang, M. Gu, M. Gao, H. Pan, Dalton Trans. 43 (2014) 2369–2377.
[19] B. Li, Y. Liu, J. Gu, Y. Gu, M. Gao, H. Pan, Int. J. Hydrogen Energy 38 (2013) 5030–
5038.
[20] Y. Liu, K. Zhong, K. Luo, M. Gao, H. Pan, Q. Wang, J. Am. Chem. Soc. 131 (2009)
1862–1870.
[21] C. Li, Y. Liu, Y. Gu, M. Gao, H. Pan, Chem. Asian J. 8 (2013) 2136–2143.
[22] B. Li, Y. Liu, C. Li, M. Gao, H. Pan, J. Mater. Chem. A 2 (2014) 3155–3162.
[23] H. Pan, S. Shi, Y. Liu, B. Li, Y. Yang, M. Gao, Dalton Trans. 42 (2013) 3802–3811.
[24] B. Li, Y. Liu, J. Gu, M. Gao, H. Pan, Chem. Asian J. 8 (2013) 374–384.
[25] M. Yan, F. Sun, X. Liu, J. Ye, H. Yuan, S. Wang, L. Jiang, J. Alloys Comp. 603
(2014) 19–22.
[26] T. Zhang, X. Yang, J. Li, R. Hu, X. Xue, H. Fu, J. Power Sources 202 (2012) 217–
224.
[27] F.G. Eisemoerg, P.D. Goodell, J. Less-common Met. 89 (1983) 55–62.
[28] E.M. Borzone, M.V. Blanco, G.O. Meyer, A. Baruj, Int. J. Hydrogen Energy 39
(2014) 10517–10524.
other is that the catalytic effect on (Mg(NH2)2 + 2LiH + 0.05KF) sys-
tem disappears due to the transformation from KH to KCN.
Verification experiment was done to find the main reason for
the decrease of the hydrogen desorption kinetics. Sample 2, which
has experienced 5 hydrogenation cycles with H2 containing 1 mol%
CO, was re-milled for 10 h to study whether the hydrogen desorp-
tion kinetics can be recovered. If it is recovered, the main reason is
the slow substance transmission in dehydrogenation. Because the
re-milling treatment can break the coating of the new products
formed on the surface of material particles, providing the effective
channels for the substance transmission. If not, the main reason is
that the catalytic effect of KH disappears. Fig. 5 shows hydrogen
desorption kinetics of the re-milled Sample 2. The data of the 1st
and 5th desorption was collected from Fig. 1(b). It can be seen that
the hydrogen desorption kinetics of the re-milled sample cannot be
restored to the initial level. So it is suggested that the disappear-
ance of the catalytic effect of KH due to the transformation from
KH to KCN is the main reason for the decline of hydrogen desorp-
tion kinetics of the (Mg(NH2)2 + 2LiH + 0.05KF) material.
[29] D. Dunikov, V. Borzenko, S. Malyshenko, Int. J. Hydrogen Energy 37 (2012)
13843–13848.
[30] K.D. Modibane, M. Williams, M. Lototskyy, M.W. Davids, Ye. Klochko, B.G.
Pollet, Int. J. Hydrogen Energy 38 (2013) 9800–9810.
[31] M. Lototskyy, K.D. Modibane, M. Williams, Ye. Klochko, V. Linkov, B.G. Pollet, J.
Alloys Comp. 580 (2013) S382–S385.
4. Conclusions
[32] H. Lin, K. Lin, C. Sung, K. Wu, Int. J. Hydrogen Energy 32 (2007) 2494–2500.
[33] F. Schweppe, M. Martin, E. Fromm, J. Alloys Comp. 253–254 (1997) 511–514.
[34] J. Han, J. Lee, J. Less-common Met. 152 (1989) 319–327.
[35] J. Han, J. Lee, J. Less-common Met. 152 (1989) 329–338.
[36] D.M. Gualtieri, K. Narasimhan, T. Takeshita, J. Appl. Phys. 47 (1976) 3432–
3435.
[37] S. Myhra, E.H. Kisi, E.M. Gray, J. Alloys Comp. 224 (1995) 305–315.
[38] J.I. Han, L.J. Young, J. Less-common Met. 157 (1990) 187–199.
[39] Y. Hu, E. Ruckenstein, Ind. Eng. Chem. Res. 46 (2007) 5940–5942.
[40] Y. Hu, E. Ruckenstein, Ind. Eng. Chem. Res. 47 (2008) 48–50.
[41] S. Luo, T.B. Flanagan, W. Luo, J. Alloys Comp. 440 (2007) L13–L17.
[42] F. Sun, M. Yan, X. Liu, J. Ye, Z. Li, S. Wang, L. Jiang, Int. J. Hydrogen Energy 39
(2014) 9288–9292.
When H2 containing 1 mol% CO is employed as the hydrogen
source for (Mg(NH2)2 + 2LiH + 0.05KF) system, three new stable
phases, Li2CN2, KCN and MgO are formed. The reversible hydrogen
desorption capacity gradually declines due to the permanent loss
of NH2ꢀ in the hydride. Moreover, the hydrogen desorption kinet-
ics decreases and the dehydrogenation activation energy increases
obviously, because of the disappearance of the catalytic effect of
KH due to the transformation from KH to KCN.
[43] E. Homer, Anal. Chem. 29 (1957) 1702–1706.
Acknowledgment
The authors would like to acknowledge the financial supports
from National Basic Research Program of China (Grant No.