Ru-based bimetallic alloys for hydrogen generation by hydrolysis of sodium tetrahydroborate

Article abstract of DOI:10.1016/j.jallcom.2007.08.077

The present study contemplates the application of Ru-based bimetallic alloys for hydrogen generation by hydrolysis of sodium tetrahydroborate (NaBH₄). Ru and Pt, RuCu, RuPd, RuAg and RuPt (atomic ratio 1:1), PtAg, and Ru_xPt_y (atomic ratios x:y of 2:1 or 1:2), all supported over titanium oxide, were prepared. Their activity decreased in the order Ru～Ru₂Pt₁ > RuPt～Ru₁Pt₂ > RuPd > RuAg～Pt > RuCu > PtAg. Alloying Ru with an inactive metal like Cu, Pd or Ag did not improve the performances of Ru. The catalytic ability of Ru₂Pt₁-TiO₂ is in the range of the highest values reported so far in the literature with a hydrogen generation rate of 15.2 L(H₂) min^-1 g^-1(RuPt). After separation from the reaction medium and rinsing with deionised water, the used Ru₂Pt₁-TiO₂ catalyst was re-evaluated and almost the same catalytic activities as fresh catalyst were obtained during several cycles.

Full text of DOI:10.1016/j.jallcom.2007.08.077

Journal of Alloys and Compounds 463 (2008) 107–111
Ru-based bimetallic alloys for hydrogen generation by
hydrolysis of sodium tetrahydroborate
U.B. Demirci∗, F. Garin
LMSPC, ECPM, ULP, CNRS, 25 rue Becquerel, 67000 Strasbourg, France
Received 11 June 2007; accepted 29 August 2007
Available online 4 September 2007
Abstract
The present study contemplates the application of Ru-based bimetallic alloys for hydrogen generation by hydrolysis of sodium tetrahydroborate
(NaBH4). Ru and Pt, RuCu, RuPd, RuAg and RuPt (atomic ratio 1:1), PtAg, and RuxPty (atomic ratios x:y of 2:1 or 1:2), all supported over titanium
oxide, were prepared. Their activity decreased in the order Ru∼Ru2Pt1 > RuPt∼Ru1Pt2 > RuPd > RuAg∼Pt > RuCu > PtAg. Alloying Ru with an
inactive metal like Cu, Pd or Ag did not improve the performances of Ru. The catalytic ability of Ru2Pt1-TiO2 is in the range of the highest values
reported so far in the literature with a hydrogen generation rate of 15.2 L(H2) min−1 g−1(RuPt). After separation from the reaction medium and
rinsing with deionised water, the used Ru2Pt1-TiO2 catalyst was re-evaluated and almost the same catalytic activities as fresh catalyst were obtained
during several cycles.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Alloys; Metals; Liquid–solid reactions; Catalysis
1. Introduction
a function of temperature and pH [4]. Controlled production
of hydrogen can be obtained by buffering the aqueous solution
of NaBH4 at a high pH and then using a catalyst [2]. Cata-
lysts are essential for controlled rate of hydrogen generation by
NaBH4 hydrolysis [5]. The catalysts used are transition metals
and these ones have been used in various forms: metal salts [6];
metal borides [7]; La- and Ni-based intermetallic compounds
(LaNi4.5M0.5, M: metal) [8]; fluorinated metals [9]; bulk met-
als [10]; Raney metals [11]; and, metal-promoted metal oxides
[12]. Today, the main objective of the researchers is to find highly
capable catalysts with superior activity and durability.
The best metals for hydrogen generation by hydrolysis of
NaBH4 are Ru, Rh and Pt. The order of these metals, made
out from their catalytic activity, is somewhat different from
one paper to the other, e.g., Ru∼Rh > Pt [6]; Pt > Rh > Ru [12];
Rh > Pt∼Ru [13]. An alternative way for searching for new cat-
alysts could be the bimetallic alloys AB (where A would be Ru,
Pt or Rh, and B another transition metal). For example, Krishnan
et al. [14] noticed that the efficiency of PtRu was almost double
of that of either Ru or Pt (all were supported over LiCoO2). Liu
et al. [11] observed that alloys Raney NiCo led to faster hydro-
gen generation rates than the pure metals and concluded “this
kind of phenomenon is very interesting from a practical as well
as academic point of view”.
Sodium tetrahydroborate NaBH4 is considered as a prospec-
tive safe chemical hydride for hydrogen storage [1]. Its
hydrogen content is of 10.66 wt.%, what makes itself one of the
highest hydrogen containing compounds. Molecular hydrogen
can be recovered by thermal decomposition of NaBH4 [1]
but, to use its high hydrogen capacity, its hydrolysis reaction
is the most attractive approach [2]. The hydrolysis reaction
is generally written with the quantity of water just required
to evolve hydrogen: NaBH4 + 2H2O → NaBO2 + 4H2. Hence,
10.92 wt.% of hydrogen can be recovered. However, the
by-product NaBO2 is a compound that is usually hydrated and
hydrolysis must involve four molecules of water [3], namely,
NaBH4 + 4H2O → NaB(OH)4 + 4H2. Hence, 7.34 wt.% of
hydrogen can be recovered, what is smaller than the former
value.
NaBH4 is water-sensitive. On its own, it reacts exothermally
with water: its self-hydrolyses [2]. The self-hydrolysis rate is
∗
Corresponding author. Tel.: +33 3 90 2427 58; fax: +33 3 90 2427 61.
E-mail addresses: umitbilgedemirci@yahoo.fr,
umit.demirci@ecpm.u-strasbg.fr (U.B. Demirci).
0925-8388/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2007.08.077

108
U.B. Demirci, F. Garin / Journal of Alloys and Compounds 463 (2008) 107–111
The present paper reports the catalytic behaviour of Ru- and
Pt-based bimetallic alloys supported over TiO2 for hydrogen
generation by hydrolysis of NaBH4.
2. Experimental
Titanium oxide (Aeroxide® TiO2 P25, Degussa, 55 m2 g−1) was chosen
as support. Its X-ray diffraction (XRD) measurement was taken on a Bruker
axs D8 advance powder diffractometer (Cu K␣1 radiation, λ = 0.15406 nm).
TiO2 had a mixed crystalline structure anatase-rutile in the proportions 80–20%
(XRD pattern not given; the phase identification used the JCPDS index
89-4921 to reference the anatase peaks and the JCPDS index 89-4920 to ref-
erence the rutile peaks). The crystallite sizes were determined (from the XRD
pattern) as being 24 and 37 nm for the anatase and rutile structures, respec-
tively.
The wet impregnation of the metals was as follows. The desired amount of
TiO2 was wetted and the slurry pH was adjusted to 10 (0.5M NaOH solu-
tion). The desired amount(s) of the metal(s) (1 wt.%, atomic ratios of 1:1,
2:1 or 1:2) was(were) simultaneously added from aqueous solution(s) of the
metallic salt(s) (CuCl2, RuCl3, Pd(NO3)2, AgNO3, H2PtCl6 Strem). The mix-
ture was stirred for 1 h and was dried in an oven for 12 h. The catalysts
are denoted as Cu-TiO2, Ru-TiO2, Pd-TiO2, Ag-TiO2, Pt-TiO2, RuCu-TiO2,
RuPd-TiO2, RuAg-TiO2, RuPt-TiO2, Ru2Pt1-TiO2, Ru1Pt2-TiO2 and PtAg-
TiO2.
In typical hydrogen generation experiments (apparatus adapted from ref.
[5]), 20 mL of an aqueous alkaline (NaOH, 1 M, Aldrich) solution of NaBH4
(0.394 g, Aldrich) filled a sealed flask. The sealed flask was fitted with an outlet
tubeforcollectingevolvedhydrogen. Anamountof0.250 gofcatalystwasadded
to the solution under stirring (not to have external diffusion effects, 400 rpm).
The temperature was maintained within the range of (20.0 0.5) ◦C thanks to
a water bath. The outlet tube exhaust was connected to a water-filled graduated
U-type glass cylinder and generated hydrogen was measured by monitoring
water displacement from the cylinder as the reaction proceeded. The hydrogen
generationrate(HGR)wasthencalculatedinthetimerange0–5 min. Therelative
error on the measured volume was of about 2%.
Fig. 1. Mechanism for the catalytic production of hydrogen from an aqueous
solution of NaBH4; adapted from ref. [15].
−
surface via the putative chemiadsorbed species Pd–BH3 and
Pd–H.
Nørskov and co-workers [17] have developed a systematic
and affordable computational framework for the rational screen-
ing of over 700 bimetallic alloys for the hydrogen evolution
reaction: 2H+ + 2e− → H2. This reaction had been chosen by the
authors for its chemical simplicity and it looks like the reaction
(M–H + e−) + HOδ−–Hδ+ → M• + H2 + HO− given above. For
this reason, the present study uses the Nørskov and co-workers’
work [17] as a ‘tool’ in order to attempt to predict the reactivity
of the bimetallic alloys that will be selected.
Before preparing bimetallic alloys AB, it is essential to make
a selection for A. Today, the best transition metals for hydrogen
generation by NaBH4 hydrolysis are known (see Section 1):
they are Ru, Rh and Pt. On the basis of the Krishnan et al.’s work
[14], which reports the catalytic abilities of Ru- and Pt-promoted
LiCoO2, it was decided to evaluate Ru and Pt (both supported
over TiO2). Fig. 2 shows the relations of generated hydrogen
volume with time for both supported metals. The HGRs were of
38.8 and 9.1 mL min−1 for Ru-TiO2 and Pt-TiO2, respectively.
Consequently Ru was selected.
3. Results and discussion
On the basis of the computational high-throughput screen-
ing on 256 pure metals and alloys established by Nørskov and
co-workers [17], four Ru-based bimetallic alloys were selected:
RuCu, RuPd, RuAg and RuPt. These four bimetallic combi-
nations should yield materials predicted by the calculations to
have activities comparable to, or even better than, pure Ru for
the hydrogen evolution reaction [17].
3.1. Preliminaries
The self-hydrolysis rate of the aqueous solution of NaBH4
was measured and was found as being <0.01 mL(H2) h−1
(<0.2 ␮L(H2) min−1), what is negligible.
HGR with metal-free TiO2 was assessed. In our experimental
conditions, itwasof0.07 mL(H2) h−1 (1.2 ␮L(H2) min−1). TiO2
was inactive towards NaBH4 hydrolysis.
3.2. Reactivity of the bimetallic alloys
˜
Pena-Alonso et al. [15] have proposed a mechanism for
the catalytic production of hydrogen from an aqueous solu-
tion of NaBH4. This mechanism is partly given in Fig. 1. It
involves two essential kinetic steps, namely, the tetrahydrobo-
rate dissociative chemisorption to the surface metal atoms and
−
the transfer of the negative charge of M–BH3 to H of M–H.
The hydrogen evolution reaction may be summarised as fol-
lows: (M–H + e−) + HOδ−–Hδ+ → M• + H2 + HO− (and HO−
interacts then with adsorbed BH3). A similar reaction mech-
anism has also been suggested by Guella et al. [16] for a Pd
catalyst; the authors have claimed that the activation of the sub-
strate towards hydrolysis is caused by electron-withdrawing and
electron-releasing effects of the palladium atoms on the catalyst
Fig. 2. Volume of hydrogen generated as a function of time: Ru-TiO2 vs. Pt-
TiO2.

U.B. Demirci, F. Garin / Journal of Alloys and Compounds 463 (2008) 107–111
109
proportion of Ru in the alloy is, the more active RuxPty-TiO2
is. Because of the occurrence of segregation for each of the
Ru-based bimetallic alloys [18], RuCu, RuPd and RuAg are less
active than pure Ru because the surface of the alloy is very likely
enriched with the inactive metal (i.e., Cu, Pd and Ag); hence, the
observed activity is that of Ru, which is present at the surface
of the alloy in lower concentrations than for pure Ru. Regard-
ing RuPt, its activity is better than those of the three other alloys
because Pt is an active metal that likely participate to the reaction
even if the global activity is lower than that of pure Ru. Simi-
larly, Liu et al. [11] have remarked that Raney Ni25Co75 > Raney
Ni50Co50 > Raney Ni75Co75 while Raney Co > Raney Ni; the
reactivity of the Raney alloy seems to depend more on the con-
tent of the most active metal, i.e., Co. A last observation from
Fig. 4 is that Ru-TiO2 and Ru2Pt1-TiO2 showed similar activi-
ties; it was also the case for Ru1Pt1-TiO2 and Ru1Pt2-TiO2. This
observation may suggest that the content of active sites at the
surface of these pairs of catalysts would be alike.
Fig. 3. Volume of hydrogen generated as a function of time for the Ru-based
bimetallic alloys.
To summary, the activity of the different catalysts
decreases in the order Ru∼Ru2Pt1 > RuPt∼Ru1Pt2 > RuPd >
RuAg∼Pt > RuCu > PtAg. Alloying Ru with an inactive metal
like Cu, Pd and Ag is uninteresting since Ru-TiO2 is more active
thanthe bimetallic catalysts. New tracks ofinvestigationsremain
to explore. The reactivity of Ru might improve if impurities (a
few percent of metal) are added [19]. Such study, in which Ru
is “doped”, is in progress.
For bimetallic alloys, segregation may occur. Segregation is
the phenomenon by which one metal segregates to the surface
of another metal and this controls then the surface composition
of alloys. Nørskov and co-workers [18] have compiled the cal-
culated surface segregation energies in transition-metal alloys.
According to these data, Cu, Pd, Ag and Pt should segregate
when they alloy to Ru; in other words, the surface of the alloys
RuCu, RuPd, RuAg and RuPt should be enriched with Cu, Pd,
Ag and Pt, respectively; the Ru aggregates at the alloy surface
should be smaller and better dispersed than for a surface of pure
Ru but, likely, their number should be lower.
Fig. 3 plots generated volume of hydrogen as a function of
time for the Ru-based bimetallic alloys. Their activity decreases
in the order Ru > RuPt > RuPd > RuAg > RuCu, with HGRs of
38.8, 25.0, 14.9, 9.6 and 7.8 mL(H2) min−1, respectively. Note
that Ag-TiO2, Cu-TiO2 and Pd-TiO2 were rather inactive with
HGRs of 0.02, 0.04 and 0.3 mL(H2) min−1, respectively. The
addition of a metal to Ru does not improve the reactivity of
Ru. PtAg-TiO2 was also prepared and compared to Pt-TiO2,
the former being less competitive than the latter: HGRs of 6.4
and 9.1 mL(H2) min−1, respectively. Krishnan et al. [14] have
reported PtRu > Ru > Pt (all supported over LiCoO2) but no
explanation about the synergetic effect of Pt and Ru is given.
An electronic effect (variation in the d-band centre of the met-
als and so in the metals electronic structure) and/or a geometric
effect (segregation occurrence and heterogeneity in the com-
position of the alloy surface) might justify the best activity of
this bimetallic alloy. In the present study, such synergetic effect
was not observed (Fig. 3), and therefore, alloying Ru is inef-
fective. The predictions from ref. [17] are unfortunately not
confirmed by our experiments. Hydrogen generation by NaBH4
hydrolysis may be more complicated than the hydrogen evolu-
tion reaction because of the presence of adsorbed boron-based
compounds (Fig. 1). Aside from that it seems that the reactiv-
ity of the Ru-based bimetallic alloys is closely linked to that
of Ru (Fig. 3). From this observation, RuxPty-TiO2 catalysts,
for which the atomic ratios x:y were varied (1:0, 2:1, 1:1, 1:2,
0:1), were prepared. Fig. 4 shows their reactivity. The larger the
3.3. Durability of Ru2Pt1-TiO2
Catalyst durability is crucial for NaBH4-based hydrogen
application. InaccordancewiththeprocedurereportedbyYeand
co-workers [20,21], used Ru2Pt1-TiO2 was separated from the
NaBH4 solution, washed seven times with deionised water, dried
(at 120 ◦C for 12 h) and re-used. Fig. 5 shows the cycle behaviour
on hydrogen generation activity for Ru2Pt1-TiO2. There was no
noteworthy decrease in the catalytic activity. The HGRs were
similar: 37.9, 36.7 and 37.1 mL(H2) min−1 for cycles 1, 3 and
5, respectively. As it was the case for Ye and co-workers [21],
Fig. 4. Volume of hydrogen generated as a function of time for the RuxPty-TiO2
catalysts (atomic ratio x:y = 1:0, 2:1, 1:1, 1:2, 0:1).

110
U.B. Demirci, F. Garin / Journal of Alloys and Compounds 463 (2008) 107–111
3.4. Achievable power levels
Amendola et al. [5] have estimated achievable power levels
for their Ru-catalysed hydrogen generator. According to their
assumptions and calculations, the same estimation is done in
the present section.
Before that, it is necessary to report some HGRs
(Table 1) that have been published in the open literature
[5,10–12,14,15,20–24]. The 1.5 wt.% Pt-LiCoO2 catalyst is
outstanding with a HGR of 3.1 L(H2) min−1 g−1(catalyst) or
206.5 L(H2) min−1 g−1(Pt) [12]. Except 10 wt.% PtRu-LiCoO2
[14] and Ru [23], the other catalysts have HGRs below
1 L(H2) min−1 g−1(catalyst). Our Ru2Pt1-TiO2 catalyst is
among this latter group of catalysts but, when the HGRs in
L(H2) min−1 g−1(metal) are considered, Ru2Pt1-TiO2 is the
second best catalyst.
Fig. 5. Multi-cycle operations of 1 wt.% Ru2Pt1-SZ.
From the HGRs in Table 1, the achievable power levels (P)
for each catalyst can be calculated [5]. It is assumed that 100%
stoichiometric yield of hydrogen is achieved and that a polymer
electrolyte membrane fuel cell operates at 0.7 V. Table 2 shows
these data. Since P is proportional to HGR (despite small vari-
the precipitation of NaB(OH)4, a by-product of the hydrolysis
reaction, was not observed. The durability is a major criterion
for a catalyst and the results show that Ru2Pt1-TiO2 had good
stability to hydrolysis of stabilised NaBH4 solution.
Table 1
Hydrogen generation rates (HGRs) reported in the open literature
Catalyst
NaBH4 (wt.%);
NaOH (wt.%)
Temperature
(◦C)
HGRa in L(H2) min−1 g−1
(catalyst)
HGRa in
L(H2) min−1 g−1 (metal)
Reference
5 wt.% Ru dispersed on anionic resin
Filamentary Ni mixed Co
Raney Ni50-Co50
1.5 wt.% Pt-LiCoO2
10 wt.% PtRu-LiCoO2
PtPd-carbon nanotubes
9 wt.% Co-␥Al2O3
2 wt.% Pt-C
NixB
Co-B
Ru
5 wt.% Ru-C
1 wt.% Ru2Pt1-TiO2
20; 10
12.5; (0.01 M)
1; 10
20; 10
5; 5
(0.03 M); (pH 13)
1; 5
5; 5
1.5; 20
20; 5
20; 5
25
0.2
0.1
0.7
3.1
1.2
0.2
0.1
0.2
0.3
0.9
1.6
0.7
0.15
4.1
0.1
0.7
206.5
12.4
9.0
1.0
8.5
0.3
0.9
[5]
[10]
[11]
[12]
[14]
[15]
[20]
[21]
[22]
[23]
[23]
[24]
Room temperature
20
22
25
29
30
30
40
20
20
25
20
1.6
12.9
15.2
0.993; 3.75
2; 4
Present study
a
Approximate values.
Table 2
Achievable power levels (P) for the catalysts given in Table 1
Catalyst
Pa,b in W g−1 (catalyst)
Pa,b in W g−1 (metal)
Reference
5 wt.% Ru dispersed on anionic resin
Filamentary Ni mixed Co
Raney Ni50-Co50
1.5 wt.% Pt-LiCoO2
10 wt.% PtRu-LiCoO2
PtPd-carbon nanotubes
9 wt.% Co-␥Al2O3
2 wt.% Pt-C
NixB
Co-B
Ru
5 wt.% Ru-C
1 wt.% Ru2Pt1-TiO2
18
9
370
9
66
19130
1126
793
87
743
24
[5]
[10]
[11]
[12]
[14]
[15]
[20]
[21]
[22]
[23]
[23]
[24]
66
290
109
18
9
18
24
84
150
64
14
84
150
1170
1426
Present study
a
P = [I (26.8 A) × V (0.7 V) × MwH (2 g mol−1) × HGR (L(H2) min−1 g−1(catalyst) or L(H2) min−1 g−1(metal))]/[Vm (L mol−1)]; adapted from reference [5].
2
b
Approximate values.

U.B. Demirci, F. Garin / Journal of Alloys and Compounds 463 (2008) 107–111
111
ations of Vm(T) with temperature (20–40 ◦C)), the comments in
the previous paragraph are also valid here.
To summary, the catalytic abilities of Ru2Pt1-TiO2 in hydro-
gen generation are in the range of the highest values reported so
far in the literature. Improvement ways exist, e.g., higher metal
loading, doping of Ru with ‘impurities’ [19] and optimisation
of the catalyst preparation. Such studies are in progress.
[3] V.C.Y. Kong, F.R. Foulkes, D.W. Kirk, J.T. Hinatsu, Int. J. HydrogenEnergy
24 (1999) 665–675.
[4] M.M. Kreevoy, R.W. Jacobson, Ventron Alembic 15 (1979) 2–3.
[5] S.C. Amendola, S.L. Sharp-Goldman, M.S. Janjua, M.T. Kelly,
P.J. Petillo, M. Binder, J. Power Sources 85 (2000) 186–
189.
[6] H.C. Brown, C.A. Brown, J. Am. Chem. Soc. 84 (1962) 1493–1494.
[7] C.M. Kaufman, B. Sen, J. Chem. Soc. Dalton Trans. (1985) 307–313.
[8] I.I. Korobov, N.G. Mozgina, L.N. Blinova, Kinet. Catal. 48 (1995)
380–384.
[9] S. Suda, Y.M. Sun, B.H. Liu, Y. Zhou, S. Morimitsu, K. Arai, N. Tsukamoto,
M. Uchida, Y. Candra, Z.P. Li, Appl. Phys. A 72 (2001) 209–212.
[10] J.H. Kim, K.T. Kim, Y.M. Kang, H.S. Kim, M.S. Song, Y.J. Lee, P.S. Lee,
J.Y. Lee, J. Alloys Compd. 379 (2004) 222–227.
[11] B.H. Liu, Z.P. Li, S. Suda, J. Alloys Compd. 415 (2006) 288–293.
[12] Y. Kojima, K.I. Suzuki, K. Fukumoto, M. Sasaki, T. Yamamoto, Y. Kawai,
H. Hayashi, Int. J. Hydrogen Energy 27 (2002) 1029–1034.
[13] V.I. Simagina, P.A. Storozhenko, O.V. Netskina, O.V. Komova, G.V.
Odegova, T.Yu. Samoilenko, A.G. Gentsler, Kinet. Catal. 48 (2007)
168–175.
4. Conclusion
Hydrogen generation by hydrolysis of NaBH4 was stud-
ied using Ru- and Pt-based bimetallic alloys, which were
supported over titanium oxide TiO2. The activity
of the investigated catalysts decreased in the order
Ru∼Ru2Pt1 > RuPt∼Ru1Pt2 > RuPd > RuAg∼Pt > RuCu > PtAg.
Alloying Ru with an inactive metal like Cu, Pd and Ag was
ineffective because Ru-TiO2 was more active than the bimetallic
catalysts.
[14] P. Krishnan, T.H. Yang, W.Y. Lee, C.S. Kim, J. Power Sources 143 (2005)
17–23.
The catalytic abilities of Ru2Pt1-TiO2 in hydrogen gen-
eration are in the range of the highest values reported so
far in the literature with a hydrogen generation rate of
15.2 L(H2) min−1 g−1(RuPt). Durability test was performed
for this catalyst. Used Ru2Pt1-TiO2 was separated from the
NaBH4 solution, washed with deionised water, dried and re-
evaluated: almost the same catalytic activities as fresh catalyst
were obtained during several cycles. The Ru2Pt1-TiO2 catalyst
showed quick response and good durability. These results are
rather promising and improvement ways exist (e.g., optimisation
of the metal(s) loading or doping of the active metal).
[15] R. Pen˜a-Alonso, A. Sicurelli, E. Callone, G. Carturan, R. Raj, J. Power
Sources 165 (2007) 315–323.
[16] G. Guella, C. Zanchetta, B. Patton, A. Miotello, J. Phys. Chem. B 110
(2006) 17024–17033.
[17] J. Greeley, T.F. Jaramillo, J. Bonde, I. Chorkendorff, J.K. Nørskov, Nature
5 (2006) 909–913.
[18] A.V. Ruban, H.L. Skriver, J.K. Nørskov, Phys. Rev. B 59 (1999)
15990–16000.
[19] A. Ruban, B. Hammer, P. Stoltze, H.L. Skriver, J.K. Nørskov, J. Mol. Catal.
A 115 (1997) 421–429.
[20] W. Ye, H. Zhang, D. Xu, L. Ma, B. Yi, J. Power Sources 164 (2007)
544–548.
[21] D. Xu, H. Zhang, W. Ye, Catal. Commun. 8 (2007) 1767–1771.
[22] D. Hua, Y. Hanxi, A. Xinping, C. Chuansin, Int. J. Hydrogen Energy 28
(2003) 1095–1100.
[23] S.U. Jeong, R.K. Kim, E.A. Cho, H.J. Kim, S.W. Nam, I.H. Oh, S.A. Hong,
S.H. Kim, J. Power Sources 144 (2005) 129–134.
References
[1] A. Zu¨ttel, A. Borgschulte, S.I. Orimo, Scripta Mater. 56 (2007) 823–828.
[2] H.I. Schlesinger, H.C. Brown, A.E. Finholt, J.R. Gilbreath, H.R. Hoekstra,
E.K. Hyde, J. Am. Chem. Soc. 75 (1953) 215–219.
[24] J.S. Zhang, W.N. Delgass, T.S. Fisher, J.P. Gore, J. Power Sources 164
(2007) 772–781.