Al13Fe4 selectively catalyzes the hydrogenation of butadiene at room temperature

Article abstract of DOI:10.1039/c3cc44987d

The hydrogenation of butadiene has been investigated for the first time on Al₁₃Fe₄. The model (010) surface of this non-noble metal combination appears to be both active and selective under mild reaction conditions. The performances of Al₁₃Fe₄ for CC bond hydrogenation are compared with those of the reference noble metal, palladium.

Full text of DOI:10.1039/c3cc44987d

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Al13Fe4 selectively catalyzes the hydrogenation of
butadiene at room temperature
Cite this: Chem. Commun., 2013,
49, 9149
Laurent Piccolo
Received 3rd July 2013,
Accepted 2nd August 2013
DOI: 10.1039/c3cc44987d
www.rsc.org/chemcomm
The hydrogenation of butadiene has been investigated for the first
In this Communication, we report on the catalytic properties
time on Al13Fe4. The model (010) surface of this non-noble metal of Al13Fe4(010) in butadiene hydrogenation. The Pd-catalyzed
combination appears to be both active and selective under mild hydrogenation of 1,3-butadiene to butene is an industrially
Q
reaction conditions. The performances of Al13Fe4 for C C bond important reaction for butene purification in polymer synthesis
hydrogenation are compared with those of the reference noble and an interesting model-reaction for sequential CQC bond
metal, palladium.
hydrogenation.12–16
The Al13Fe4(010) sample was purchased from Mateck GmbH
Recent work by Armbru¨ster et al. has shown the unexpected ability (Germany). An ultrapure crystal was grown following the
of Al13Co4 and Al13Fe4 alloys to catalyze the semi-hydrogenation Czochralski method17 and cut perpendicular to its [010] direc-
of acetylene, a reaction industrially applied in polyethylene tion into a small disk (10 mm  1 mm). The final roughness
production.1,2 This finding is of great importance due to the and disorientation of the as-furnished Al13Fe4(010) surface were
low cost of the metallic constituents as compared to conven- smaller than 0.03 mm and 0.11, respectively. The surface was
tionally used Pd and Pd–Ag. The same research group had further cleaned by repeated cycles of Ar+ sputtering and annealing
previously evidenced that ordered compounds of the Ga–Pd at 600–800 1C under ultrahigh vacuum. The cleaned surface
system were efficient acetylene hydrogenation catalysts.3–6 The was contaminant-free, as attested using Auger electron spectro-
results have been conceptualized in terms of ‘‘site isolation’’, scopy (Fig. 1, middle spectrum), and exhibited a sharp low energy
meaning that Pd atoms embedded in a Ga matrix form electron diffraction (LEED) pattern (Fig. 1, inset). The LEED
active, selective, and stable reaction sites.4 Such discoveries pattern corresponds to a (1 Â 1) structure with an oblique unit
have also been driven by the recent advances in computa- mesh, in agreement with the previous findings.11 The increase of
tional screening of catalysts, which has led to the identifica-
tion of new non-noble bimetallic alloys, such as Ni–Zn, as
potential alternatives to platinum-group metals for acetylene
hydrogenation.7,8
Al13Fe4 is considered, like other Al13M4 complex intermetallic
compounds (M = Fe, Co, Ni. . .), to be a periodic approximant to
decagonal quasicrystals.9,10 Recently, Ledieu et al. have revealed
the complex surface structure of the (010) surface of Al13Fe4
using scanning tunnelling microscopy and density functional
theory calculations.11 The authors have proposed isolated Fe
atoms protruding above pentagonal Al motifs as the catalytically
active sites for acetylene conversion. On a similar single-crystal
surface, in situ X-ray photoelectron spectroscopy (XPS) has
revealed that the surface phase is essentially unaltered under
acetylene hydrogenation conditions.2
Institut de recherches sur la catalyse et l’environnement de Lyon (IRCELYON),
´
UMR 5256 CNRS & Universite Lyon 1, 2 avenue Albert Einstein,
Fig. 1 Auger electron spectra for Al13Fe4(010) after sputtering (bottom spectrum),
annealing (middle), and reaction (top). Inset: LEED pattern of the clean annealed
surface.
F-69626 Villeurbanne, France. E-mail: laurent.piccolo@ircelyon.univ-lyon1.fr;
Fax: +33 472445399; Tel: +33 472445324
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 9149--9151 9149

View Article Online
Communication
ChemComm
Table 1 Effect of temperature on reaction rates for clean Al13Fe4(010)
the Al/Fe low-energy Auger peak ratio from 1.9 to 2.8 upon
annealing of the sputtered surface can be explained either
by preferential sputtering of Al during bombardment, or
surface enrichment in Al during annealing as for AlFe3
surfaces.18 However, surface segregation has been ruled out
by Ledieu et al.11
a
a
Temperature r1
(1C)
r2
b
(10À8 mol minÀ1
)
(10À8 mol minÀ1
)
r2/r1 S1 (%)
24
110
200
9.3
79
51
38
98
15
4.1
1.2
0.3
89
96
100
a
The butadiene hydrogenation reaction was carried out in a
r1 and r2 are the butene and butane formation rates averaged on the first
dedicated static catalytic reactor (volume ca. 120 cm3) coupled and second hydrogenation periods, respectively (initial rates were hardly
b
to the surface preparation/analysis chamber.19,20 Similar experi-
ments were performed on Al13Fe4(010) and Pd(100), which
serves here as a reference. The Al–Fe sample could be heated
from the back by an infrared laser beam and its surface
measurable due to their high values). S1 is the selectivity to butene
averaged on the first hydrogenation period (the initial selectivity is always
100%). S1 = r1/r1tot, where r1tot is the averaged butadiene consumption rate.
temperature was measured by an infrared pyrometer (surface Conversely, some butane is produced during the first hydro-
emissivity set to 0.3). A mixture of ultrapure gases (5 Torr genation period on Al13Fe4(010).
hydrogen, 0.5 Torr butadiene, and 0.5 Torr Ar for internal
The reaction was also carried out at 110 1C and 200 1C. As
calibration)† was prepared in a separate chamber before injec- reported in Table 1, both rates increase from RT to 110 1C, and
tion into the reactor. During the reaction, the gases were decrease from 110 to 200 1C. This suggests that above 110 1C, the
continuously sampled through a leak valve and analyzed by reaction rate is limited by hydrocarbon desorption. However,
mass spectrometry (MS).‡ At the end of the reaction run, the the extent of these variations is different between butadiene
products were evacuated, and the reaction–evacuation cycle and butene hydrogenations. As a matter of fact, the butene
was again performed two times to assess the catalyst stability. formation rate becomes higher than the butane one at 200 1C
The hydrogen and hydrocarbon partial pressures were obtained (r1 E r2 at 110 1C).
from the MS signal by taking into account product-dependent
Only in the case of palladium, a gradual increase of the
sensitivity, ion fragmentation, and gaseous matter conservation.21 butadiene conversion rate is observed (Fig. 2). This is due to
Fig. 2 shows the results of a catalytic test of the clean the dissolution of hydrogen in the subsurface, as previously
annealed Al13Fe4(010) surface at room temperature (straight demonstrated.21 The more hydrogen dissolved, the faster
lines). Strikingly, butadiene is fully converted into butane the surface hydrogenation rate.§ Conversely, Al13Fe4 does not
within 50 min on this material. This is exactly the time needed absorb hydrogen, which is a further advantage of this material.2
for the same conversion on clean Pd(100) (see Fig. 2, dotted
Unlike Pd, Al13Fe4 deactivates in the course of the reaction
lines). Moreover, the hydrogenation of butadiene is highly cycles (not shown). For example, at 200 1C, r1 is decreased by
selective to butene. The selectivity is 100% at initial time, and more than 40% from the first to the second reaction run, then
89% when averaged over the whole semi-hydrogenation period by less than 30% from the second to the third run. However, the
(Table 1). Only after butadiene has been fully converted, the selectivity to butenes is preserved. In addition, butene hydro-
butane formation rate increases steeply. Interestingly, the butene genation appears less sensitive to deactivation. As suggested
hydrogenation rate (r2) is greater than the butene formation rate using post-reaction Auger electron spectroscopy (Fig. 1, top
(r1), whereas it is the opposite on palladium (r1 > r2). This is spectrum), the deactivation is caused by chemisorbed oxygen
consistent with an even greater selectivity of Pd(100), which is (and not carbon), which is most probably due to adsorption of
100% whatever the extent of butadiene conversion (Fig. 2). water impurities inherently present in UHV chambers and
research-grade gases.¶ On-run deactivation may explain the
gradual decrease of the butene formation rate in Fig. 2, and
the corresponding decrease of butene selectivity. Nevertheless,
this behaviour disappears at higher temperatures. Together
with previously mentioned observations, this excludes that
the decrease of r1 and r2 averaged rates from 110 to 200 1C
(Table 1) could be explained by thermally-enhanced deactiva-
tion rates. Note that a flash desorption above 600 1C allowed an
at least partial regeneration of the surface. The deactivation
issue was also mentioned by Armbru¨ster et al. during long-term
acetylene hydrogenation on Al13Fe4 powder.2 Besides, the authors
detected oxygen (in the form of Al oxides) and carbon on their
single-crystal surface using XPS.
It should be noted that the propensity of Al13Fe4 to catalyze
the hydrogenation of CQC bonds is not limited to butadiene
and butene. The same experiment as that reported in Fig. 2 was
carried out with ethylene instead of butadiene. Ethylene was
fully converted into ethane in less than one minute on the clean
Al13Fe4(010) surface.
Fig. 2 Hydrocarbon partial pressures during butadiene hydrogenation on
Al13Fe4(010) (straight lines) and Pd(100) (dashed lines) at 24 1C.
c
9150 Chem. Commun., 2013, 49, 9149--9151
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
For butadiene at RT under our pressure conditions, the Notes and references
turnover frequency (TOF) is of the order of 10 molecules
† 1 Torr = 133 Pa. This unit is the most widely employed in surface and
vacuum sciences.
converted per second per surface palladium atom. If we assume
that protruding Fe atoms constitute the centres of the cata-
lytically active sites in the case of Al13Fe4(010),11 the TOF per
active site is of the order of 100 sÀ1, which is extremely high.8
Importantly, the freshly sputtered surface (see the corresponding
Auger spectrum in Fig. 1) was found ca. 10 times less active than
the freshly annealed one (not shown), in spite of the increases in
surface area and roughness due to Ar+ sputtering. This implies that
the high activity of Al13Fe4(010) is related to the specific structure
of the well-crystallized and chemically well-ordered surface.
Although the physical origins of the high activity and
selectivity of Al13Fe4 in butadiene hydrogenation still have to
be determined, some insights can be gained from the theoretical
‡ MS intensities for m/z = 2, 40, 54, 56, and 58 were recorded for
hydrogen, argon, butadiene (C4H6), butene (C4H8), and butane (C4H10),
respectively. Butene isomers are not distinguished here.
§ Consequently, the Pd surface is more active for butene formation in
the subsequent reaction runs, and r1 c r2 (not shown).
¶ Nothing can be rigorously said from AES on a possible Al–Fe composi-
tional change at the surface during the reaction, due to the large amount
of oxygen significantly masking Al and Fe after the reaction.
8 In the authors’ model, the surface unit mesh contains 4 Fe atoms
including 2 protruding ones, and 22 Al atoms. The surface atom density
is 14.0 nmÀ2, vs. 13.2 nmÀ2 for Pd(100).
¨
1 M. Armbru¨ster, K. Kovnir, J. Grin, R. Schlogl, P. Gille, M. Heggen
and M. Feuerbacher, US patent, 2012, 0029254.
2 M. Armbru¨ster, K. Kovnir, M. Friedrich, D. Teschner, G. Wowsnick,
´
M. Hahne, P. Gille, L. Szentmiklosi, M. Feuerbacher, M. Heggen,
¨
F. Girgsdies, D. Rosenthal, R. Schlogl and Y. Grin, Nat. Mater., 2012,
11, 690–693.
ˇ´
work of Krajcı and Hafner in the case of acetylene hydrogenation
over Al13Co4(100).22,23 The authors have shown that the cata-
lytically active sites consist of pentagonal CoAl5 clusters with
strong internal Co–Al bonding and low atom coordination at
the cluster edges, the latter being induced by the complex
surface topology. The resulting surface electronic structure
provides optimal surface bonding and reaction energies to
the adsorbates. The high selectivity to ethylene is explained
by the lower desorption barrier for this molecule with respect to
its hydrogenation barrier. Similar arguments can be formulated
for selective butadiene hydrogenation on Al13Fe4(010), as was
previously done in the case of highly selective Au–Pd surfaces,
in which gold favours butene desorption.19
3 K. Kovnir, J. Osswald, M. Armbru¨ster, R. Giedigkeit, T. Ressler,
Y. Grin and R. Schlogl, Stud. Surf. Sci. Catal., 2006, 162, 481–488.
4 K. Kovnir, M. Armbru¨ster, D. Teschner, T. V. Venkov, F. C. Jentoft,
A. Knop-Gericke, Y. Grin and R. Schlogl, Sci. Technol. Adv. Mater.,
¨
¨
2007, 8, 420–427.
5 J. Osswald, K. Kovnir, M. Armbru¨ster, R. Giedigkeit, R. E. Jentoft,
U. Wild, Y. Grin and R. Schlogl, J. Catal., 2008, 258, 219–227.
6 M. Armbru¨ster, K. Kovnir, M. Behrens, D. Teschner, Y. Grin and
R. Schlogl, J. Am. Chem. Soc., 2010, 132, 14745–14747.
7 F. Studt, F. Abild-Pedersen, T. Bligaard, R. Z. Sorensen,
C. H. Christensen and J. K. Nørskov, Science, 2008, 320, 1320–1322.
8 F. Studt, F. Abild-Pedersen, T. Bligaard, R. Z. Sørensen, C. H.
Christensen and J. K. Nørskov, Angew. Chem., Int. Ed., 2008, 47,
9299–9302.
9 K. Saito, K. Sugiyama and K. Hiraga, Mater. Sci. Eng., 2000, 294–296,
279–282.
¨
¨
In conclusion, the first work on alkene hydrogenation over
the Al13Fe4 system has been reported. The hydrogenation of
1,3-butadiene was performed under mild conditions (20–200 1C,
800 Pa) on an in situ cleaned Al13Fe4(010) surface, as well as on
Pd(100) for comparison. The non-precious metallic compound
is highly active, both for butadiene and butene hydrogenation,
and highly selective to butene during butadiene hydrogenation,
even at room temperature. Based on previous work on Al13M4
systems, these unexpected catalytic performances are ascribed
to complex ensemble effects on FeAl5 active sites. Future studies
will aim at addressing the deactivation issue and understanding
the reaction mechanism through further investigation of the
kinetics.
I greatly thank my colleague Dr F. Morfin for inestimable
technical help and regular scientific discussions. The French
government is acknowledged for financial support via the ANR
DINAMIC programme. French GDR CNRS 3182 ‘‘Nanoalliages’’
and European COST Action MP0903 ‘‘Nanoalloy’’ are acknowl-
edged for stimulating networking activities.
ˇ
´
ˇ
10 P. Popcevic, A. Smontara, J. Ivkov, M. Wencka, M. Komelj, P. Jeglic,
ˇ ´
S. Vrtnik, M. Bobnar, Z. Jaglicic, B. Bauer, P. Gille, H. Borrmann,
U. Burkhardt, Y. Grin and J. Dolinsek, Phys. Rev. B, 2010, 81, 184203.
11 J. Ledieu, E. Gaudry, L. N. S. Loli, S. A. Villaseca, M.-C. de Weerd,
M. Hahne, P. Gille, Y. Grin, J.-M. Dubois and V. Fournee, Phys. Rev.
Lett., 2013, 110, 076102.
12 J. P. Boitiaux, J. Cosyns and S. Vasudevan, Appl. Catal., 1983, 6,
41–51.
13 J.-C. Bertolini, Appl. Catal., A, 2000, 191, 15–21.
14 J. Silvestre-Albero, G. Rupprechter and H.-J. Freund, Chem. Commun.,
2006, 80–82.
ˇ
´
15 C. Breinlich, J. Haubrich, C. Becker, A. Valcarcel, F. Delbecq and
K. Wandelt, J. Catal., 2007, 251, 123–130.
16 L. Piccolo, A. Valcarcel, M. Bausach, C. Thomazeau, D. Uzio and
G. Berhault, Phys. Chem. Chem. Phys., 2008, 10, 5504–5506.
17 P. Gille and B. Bauer, Cryst. Res. Technol., 2008, 43, 1161–1167.
18 L. Piccolo and L. Barbier, Surf. Sci., 2002, 505, 271–284.
19 L. Piccolo, A. Piednoir and J. C. Bertolini, Surf. Sci., 2005, 592,
169–181.
20 F. Morfin and L. Piccolo, Rev. Sci. Instrum., 2013, DOI: 10.1063/
1.4818669.
21 A. Valcarcel, F. Morfin and L. Piccolo, J. Catal., 2009, 263, 315–320.
ˇ´
22 M. Krajcı and J. Hafner, J. Catal., 2011, 278, 200–207.
ˇ´
23 M. Krajcı and J. Hafner, Phys. Rev. B, 2011, 84, 115410.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 9149--9151 9151