Gold(III) - metal organic framework bridges the gap between homogeneous and heterogeneous gold catalysts

Article abstract of DOI:10.1016/j.jcat.2009.04.021

A MOF containing an Au(III) Schiff base complex lining the pore walls has been prepared by a post-synthesis method. The Au(III)-containing MOF is highly active and selective for domino coupling and cyclization reactions in liquid phase, the Au(III) species remain after the reaction, and the catalyst is fully recyclable. This gives higher activity than homogeneous and gold-supported catalysts reported up to now. The well-defined Au(III) sites are active for dissociating H₂ and proved to be active for the gas-phase selective hydrogenation of 1,3-butadiene into the butenes.

Full text of DOI:10.1016/j.jcat.2009.04.021

Journal of Catalysis 265 (2009) 155–160
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Gold(III) – metal organic framework bridges the gap between homogeneous
and heterogeneous gold catalysts
*
X. Zhang, F.X. Llabrés i Xamena, A. Corma
´
´
Instituto de Tecnologıa Quımica UPV-CSIC, Avda. de los Naranjos, s/n, 46022 Valencia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A MOF containing an Au(III) Schiff base complex lining the pore walls has been prepared by a post-syn-
thesis method. The Au(III)-containing MOF is highly active and selective for domino coupling and cycli-
zation reactions in liquid phase, the Au(III) species remain after the reaction, and the catalyst is fully
recyclable. This gives higher activity than homogeneous and gold-supported catalysts reported up to
now.
Received 5 February 2009
Revised 24 April 2009
Accepted 25 April 2009
Available online 22 May 2009
The well-defined Au(III) sites are active for dissociating H2 and proved to be active for the gas-phase
selective hydrogenation of 1,3-butadiene into the butenes.
Keywords:
Au(III)–MOFs
Au(III)–MOF catalysts
3D coupling reaction on Au(III)–MOFs
Hydrogenations with gold–MOFs
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
strategies can be adopted in order to prepare MOFs with open me-
tal sites [5,6]. One strategy [5a,5b] consists in using labile ligands
The interest on the catalytic chemistry of gold has undergone a
marked increase in recent years [1]. Supported gold nanoparticles
and gold salts/complexes are the basis of the present gold rush in
heterogeneous and homogeneous catalysis, respectively [1,2].
While catalyst recovery and process work-up are much easier in
the case of solid gold catalysts in comparison with gold salts/com-
plexes [3], it becomes more difficult to know the exact nature of
the active sites associated with gold, and there is still an intensive
debate on the nature of the gold species involved in the reaction
and the potential role played by the support [1a,2a,2b]. It is not
then surprising that research efforts are being made today to
bridge the gap between homogeneous and heterogeneous gold
catalysis [3a].
In the last years, a class of porous metal organic frameworks
(MOFs) have aroused great interest for their potential application
in gas separation and storage, sensors, and catalysis [4]. These
materials are formed by metal ions or clusters acting as nodes
and polyfunctional organic ligands as linkers. Unlike other materi-
als (such as metal oxides), the lack of non-accessible bulk volume
and the fully exposed metal sites in metal organic framework
compounds provide an ultimately high degree of metal dispersion.
However, if the coordination sphere of the metal ions is completely
blocked by the organic linkers, then there are not free positions
available to interact/activate with the reactants and this could lim-
it the possibilities of MOFs for catalysis. Nevertheless, different
in the synthesis of the MOF, that can be removed during the acti-
vation stage prior to the catalytic reaction. In most cases, the labile
ligands are solvent molecules that expose a free coordination posi-
tion when removed. Kitagawa and co-workers [5c] proposed an
alternative approach to prepare MOFs with open metal sites, by
using a pre-assembled metalloligand that can be further reacted
with a second metal acting as a node in the framework.
Very recently, a new preparative strategy [6] consisting of the
covalent post-synthesis modification of an existing MOF with suit-
able functional groups has been proposed, as opposed to the ‘‘pre-
synthesis” strategy used so far. Wang et al. [6a,6b] described the
modification of the amino groups of IRMOF-3 with alkyl anhy-
drides. This idea was later used by Costa et al. [6c] to prepare a
MOF containing Gd(III) and 2-aminoterephthalate ligands. The
amino group of the organic ligand was modified post-synthesis
with two different functionalities, by reacting either with an isocy-
anate or with a carboxylic acid. Post-synthesis modifications of
MOFs have also been applied very recently in different scenarios.
Ingleson et al. [6d] described the preparation of a material contain-
ing Brønsted acid sites by post-synthesis protonation of the car-
boxylate ligands of a MOF with anhydrous HCl. Hwang et al. [6e]
described the grafting of an amine pendant group to the chromium
sites of the chromium terephthalate MIL-101, using a 1,
x-diam-
inoalkane. Morris et al. [6f] modified zeolite imidazolate frame-
works (ZIFs).
During the preparation of this manuscript, a new work by Ingle-
son et al. [6g] has appeared describing the preparation of a mate-
rial obtained by reacting the –NH2 groups of IRMOF-3 with
* Corresponding author. Fax: +34 963877809.
E-mail address: acorma@itq.upv.es (A. Corma).
0021-9517/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.04.021

156
X. Zhang et al. / Journal of Catalysis 265 (2009) 155–160
salicylaldehyde to form the corresponding imine. The authors used
this ligand to complex metal ions, as demonstrated by covalently
anchoring vanadyl acetylacetonate. This material was found to be
active for the oxidation of cyclohexene with tBuOOH, but the turn-
over frequencies obtained were low and the material was found to
loose its framework integrity [6g].
amine (0.24 mmol), and aldehyde (0.40 mmol) in 1.0 mL of 1,4-
dioxane with n-octane as an internal standard. The pure product
was obtained by flash chromatograph and identified by GC-MS
and 1H NMR, 13C NMR (see Supporting Information).
Selective hydrogenation of 1,3-butadiene: Hydrogenation of 1,3-
butadiene was performed in a fixed-bed quartz reactor (i.d.=
6 mm) at atmospheric pressure and 130 °C. The reactants of
In a parallel approximation, we have also used the covalent
post-synthesis methodology to prepare MOFs with open metal
sites with potential use in catalysis. We have prepared a MOF con-
taining a Au(III) Schiff base complex lining the pore walls. We have
anticipated that if coordinative unsaturated gold ions (i.e., open
cationic gold sites) can be created in MOF, the resulting heteroge-
neous gold catalyst would emulate the catalytic properties of
homogeneous counterparts, i.e. well-defined active sites with dis-
persed metal ions. It is presented here that the resulting MOF con-
taining gold is a highly active, selective, and reusable catalyst for
domino coupling and cyclization reactions in liquid phase, and
hydrogenation of 1,3-butadiene in gas phase. To the best of our
knowledge, this material represents the first example of a catalyt-
ically active MOF containing isolated cationic gold species, and has
allowed to set the discussion on the nature of the gold active sites
for selective diene hydrogenation. It represents a novel catalyst
that helps to bridge the gap between homogeneous and heteroge-
neous catalysis [7].
2.0 vol% 1,3-butadiene in H2 (13.0 mL minÀ1) and Ar (28 mL minÀ1
)
were introduced to the catalyst (containing 0.0011 mmol gold, di-
luted with 450 mg SiC), Specifically, the Au/TiO2 catalyst was
in situ pretreated in flowing Ar at 130 °C or H2 at 250 °C, respec-
tively, before reaction. The reactor effluent was in situ analyzed
using gas chromatograph (FID, a HP-PLOT/Al2O3 capillary column).
3. Results and discussion
3.1. State of gold and stability in the gold MOF IRMOF-3-SI-Au
The starting material chosen for this study was IRMOF-3 [4b],
since the presence of NH2 groups allows for the easy covalent mod-
ification. The material was synthesized at room temperature, fol-
lowing a modification of the procedure described by Huang et al.
[8] using 2-aminoterephthalic acid (H2ATA) instead of terephthalic
acid. The material was washed with chloroform prior to post-syn-
thesis modification to remove occluded DMF solvent molecules. A
freshly prepared sample was contacted for a short time at room
temperature with a chloroform solution containing salicylalde-
hyde, which causes a color change from cream to light or deep yel-
low (depending on the loading of salicylaldehyde). The color
change is accompanied by a new UV–Vis absorption band at
450 nm (Fig. S1) and a new 13C NMR band at 167 ppm due to the
C = N (Fig. S2), which are indicative of the formation of the corre-
sponding salicylideneimine. This sample will be hereafter referred
to as IRMOF-3-SI. The contact time of this reaction is important,
since it was observed that for contact time longer than ca. 0.5 h
the material obtained had a very low crystallinity (Fig. S3). The fi-
nal step to prepare the Au(III)-containing material, IRMOF-3-SI-Au,
consisted in reacting a suitable gold precursor, NaAuCl4, with the
2. Materials and methods
The detailed preparation and characterization of IRMOF-3-SI-
Au, Au/ZrO2, and the homogeneous Au(III) Schiff complex catalysts
is given in the Supporting Information.
2.1. Catalytic reactions
Three-component coupling and cyclization of N-protected ethyny-
laniline, amine, and aldehyde: All reactions were performed in a
closed glass reactor (2.0 mL, SUPELCO) with vigorous stirring (ca.
1000 rpm) at 40 °C (or 80 °C). The gold catalyst of desired weight
was added to a mixture of N-protected ethynylaniline (0.20 mmol),
O
H
OH
O
Au3+
NH2
N
N
NaAuCl
4
OH
Cl
Cl
IRMOF-3
IRMOF-3-SI
IRMOF-3-SI-Au
Scheme 1. Post-synthesis modification procedure for obtaining MOFs containing Au(III) Schiff base complex. Zn: green; O: red; C: light blue; N: deep blue; Au: yellow; Cl:
white. H atoms are omitted. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

X. Zhang et al. / Journal of Catalysis 265 (2009) 155–160
157
fore, according to the relatively high reduction temperature
needed for reducing AuIII in IRMOF-3-SI-Au, we can conclude that
cationic gold species are highly stabilized in the Schiff base com-
plex within the framework of the MOF (Scheme 1).
A
B
c
3.2. The gold MOF catalyst for three component coupling and
cyclization: comparison with homogeneous catalysts
b
a
b
c
The new class of heterogeneous IRMOF-3-SI-Au should then be
of catalytic interest due to the presence of well-defined isolated
Au(III) stable active sites, fully accessible to reactant molecules.
Therefore we expect that this type of Au(III)-MOF catalyst could
be a useful heterogeneous counterpart of the Au(III) salt catalysts
used up to now as well as a new way for heterogenizing homoge-
neous gold complexes [3a,10]. We will show here that the IRMOF-
3-SI-Au catalyst gives TOF as high as (or even higher than) Au(III)
salt catalysts but, unlike the salts, without reduction/decomposi-
tion during the reaction. This is firstly shown by applying the IR-
MOF-3-SI-Au catalyst for multicomponent domino coupling and
cyclization of N-protected ethylaniline, aldehyde, and amine, a
reaction that is known to be catalyzed by Au(III) species [11].
The indole nucleus is one of the most ubiquitous heterocycle in
nature and synthetic compounds with biological activities, and
thus the development of new, efficient, and sustainable synthesis
protocols is highly desired [12]. Recent developments on synthetic
strategies are mainly based on transition-metal (e.g., Pd and Cu)
working in homogenous phase [12]. Very recently we have re-
ported an atom-economy approach for three-component coupling
of alkyne, amine, and aldehyde through C–H activation [11], and
one example of indole synthesis based on domino coupling and
cyclization catalyzed by supported gold. It was shown there that
only a fraction of the total gold species, i.e. only the Au(III), are ac-
tive for these reactions. If this is true, the IRMOF-3-SI-Au catalyst
should then give a higher activity per total amount of gold than
any of the previously reported heterogeneous gold catalysts, and
certainly more active than AuCl3 that is easily and rapidly re-
duced/decomposed during reactions [3b,3c,3d,3e].
0
100 200 300 400 500 600 700
T/oC
5
10 15 20 25 30 35 40
2θ / degree
Fig. 1. (A) X-ray diffraction patterns and (B) hydrogen temperature programmed
reduction of samples IRMOF-3 (a), IRMOF-3-SI (b), and IRMOF-3-SI-Au (c).
imine-modified material. The preparative procedure is summa-
rized in Scheme 1, while a detailed description of each step is given
as Supporting Information.
According to the elemental analysis of the samples, using this
preparative method a maximum functionalization of about 3% of
the total –NH2 groups was produced without losing the framework
integrity, while gold complexation by the Schiff base complex was
almost quantitative. Increasing the degree of functionalization
with salycilideneimine required longer contact times, but this
caused a severe diminution of the crystallinity of the final material,
as commented above. Thus, 3% functionalization is the best com-
promise we found between degree of functionalization and crystal-
linity. Nevertheless, since the ensuing gold complexation is
quantitative, a 3% functionalization of the Schiff base allows intro-
ducing up to 2 wt% of gold in the material, which is a typical value
for the metal loading of heterogeneous catalysts.
The N2 adsorption-desorption experiments show a decrease in
BET surface area from 750 to 400 m2 gÀ1 after IRMOF-3 was mod-
ified by salicyladehyde, while the surface area remained practically
unchanged after Au(III) complexation (380 m2 gÀ1).
The XRD pattern (Fig. 1A) of IRMOF-3 corresponds to that of a
crystalline material with the expected structure [4b]. After func-
tionalization with the imine and ensuing complexation with
Au(III), the structure integrity of the solid is preserved. XRD pattern
of IRMOF-3-SI-Au remains unaltered when the material was
heated up to 300 °C in air, demonstrating that IRMOF-3-SI-Au is
thermally stable. TG-DTA analysis shows a first weight loss of
about 9 wt% between room temperature and 150 °C, due to re-
moval of chloroform and a small amount of remaining DMF, and
a second weight loss due to combustion of the organic part of
the framework of ca. 56 wt% between 340 and 550 °C (Fig. S4).
The absence of a diffraction peak at 2h = 38.2° (Au [111]), and
examination of the samples by TEM (Fig. S5) both exclude the
occurrence of metallic gold particles within our detection limits.
Fig. 1B shows hydrogen temperature-programmed reduction
(TPR) profiles of IRMOF-3-SI and IRMOF-3-SI-Au. The TPR profile
of IRMOF-3-SI-Au features two hydrogen consumption peaks:
one at low temperature (190–280 °C) and another at high temper-
ature (300–550 °C), while only one peak associated to hydrogen
consumption at high temperature (370–610 °C) is found for IR-
MOF-3-SI. The high temperature peak can be assigned to decompo-
sition of the IRMOF-3-SI support. The low temperature peak is due
to reduction of AuIII ions into Au0 atoms. Since the moles of hydro-
gen-consumption is exactly 1.5 times the molar number of gold in
the IRMOF-3-SI-Au (as determined by ICP-AES), this result indi-
cates that all gold atoms are in a +3 oxidation state, with no detect-
able amounts of Au0 species, as already deduced from the XRD and
TEM analysis. It is very important to point out that the reduction
temperature (ca. 239 °C) of AuIII ions in IRMOF-3-SI-Au is higher
than that found for gold on nanocrystalline CeO2 or on ZrO2
(Fig. S6), which stabilize surface cationic gold species [9]. There-
Fig. 2 shows the kinetic curves for IRMOF-3-SI-Au, Au/ZrO2,
AuCl3, and a homogeneous Au(III) Schiff base complex used as
reference [10c], for the domino coupling and cyclization reaction
of N-tosyl protected ethynylaniline (E1), piperidine, and parafor-
maldehyde at 40 °C. It can be seen there that E1 conversion in-
creases continuously up to 93%, 36%, and 38% over IRMOF-3-SI-
Au, Au/ZrO2, and Au(III) Schiff base complex, respectively, within
14 h, while a maximum conversion of ca. 60% was obtained with
AuCl3 over 6 h.. If one calculates the turnover frequency (TOF)
from the initial reaction rates from Fig. 2 taking into account
the total gold content of the catalyst, the values obtained are
100
80
60
40
20
0
0
2
4
6
8
10 12 14
Reaction Time /h
Fig. 2. Comparison of the catalytic activity over IRMOF-3-SI-Au (j, gold:
0.0008 mmol), Au/ZrO2 (N, gold: 0.0014 mmol), homogeneous Au(III) Schiff com-
plex (e, gold: 0.0008 mmol), and AuCl3 (s, gold: 0.025 mmol) for domino coupling
and cyclization of ethynylaniline (E1, 0.10 mmol), piperidine (0.12 mmol), and
paraformaldehyde (0.20 mmol) with 1,4-dioxane (1.0 mL) as solvent at 40 °C.

158
X. Zhang et al. / Journal of Catalysis 265 (2009) 155–160
52, 12, 40, and 3 hÀ1 for IRMOF-3-SI-Au, Au/ZrO2, Au(III) Schiff
base complex, and AuCl3, respectively. The first conclusion from
these results is that the IRMOF-3-SI-Au gives a much higher
TOF than AuCl3. Compared with the homogeneous gold(III) Schiff
base complex, IRMOF-3-SI-Au gives a similar initial rate (52 vs
40 hÀ1) but it is more stable, thus allowing to achieve higher con-
versions, and therefore higher turnover numbers (TONs). On the
other hand, it is not surprising that the TOF of IRMOF-3-SI-Au
is much larger than that of Au/ZrO2 since it has been found that
in the latter catalyst the fraction of AuIII with respect to total gold
is only 0.25 [11]. However, if we consider the fraction of the cat-
ionic gold in Au/ZrO2 to recalculate the TOF then the value ob-
Table 1
Coupling of N-protected ethynylaniline, amine, and aldehyde catalyzed by IRMOF-3-
SI-Au in dioxane.a
Entry
Ethynyl-aniline
R1-CHO
R2R3NHb
T (h)
Yield (%)c
1
2
3
4
5
6
7
8
E1
E1
E1
E1
E1
E2
E2
E2
E2
E3
E3
E3
(HCOH)n
Cyclohexyl
Heptyl
Piperidine
Piperidine
Piperidine
M2
16
4
6
12
6
16
12
4
90
80
95
83
91
94
83
85
92
78
70
80
tained is 48 hÀ1
, which is not too different from the TOF
measured with the MOF in which all AuIII species should be
accessible and active. This comparison allows us to reach the fol-
lowing conclusions: (a) The cationic gold Au(III) species are the
active sites for this reaction; (b) During reaction, cationic gold
species are indeed stabilized in the IRMOF-3-SI-Au and Au/ZrO2
catalyst, but they are less stabilized in the homogeneous Schiff
base complex catalyst and AuCl3 salt; (c) IRMOF-3-SI-Au with
well-defined and accessible sites can, in practice, be a bridge be-
tween the homogeneous and heterogeneous gold catalysts, and
can be used to bring additional light in the debate on the nature
of gold species that are active on solid catalysts, as will be dem-
onstrated later for the selective hydrogenation of 1,3-butadiene.
As we said before, and in contrast to the homogeneous Au(III)
and Cu(I) salts [12c], the IRMOF-3-SI-Au catalyst can be success-
fully reused (100%, 99%, 95%, and 97% conversions were obtained
after 3 h for four successive runs at 80 °C). Furthermore, the struc-
tural integrity of the material is basically preserved after the cata-
lytic use (368 m2 gÀ1), in spite of some apparent crystallinity loss
(Fig. S7). Both the absence of the [111] peak of metallic gold and
the H2-TPR on the reused catalyst show that reduction of Au(III)
ions does not occur during the reaction, and the H2 consumption
of the reused catalyst still corresponds to exactly 1.5 times the to-
tal number of gold mols. The reusability of the material, together
with the fact that the gold loading (as determined by ICP-AES) re-
mained unchanged after the second cycle, excludes the occurrence
of gold leaching from the solid to the liquid. Nevertheless, we have
also carried out the so-called hot filtration test [4e,4f] to further
demonstrate this point. The reaction in the presence of the IR-
MOF-3-SI-Au catalyst was allowed to proceed only to partial con-
version (40% in 2 h) at 40 °C. At this point, the solution was
quickly filtrated and the solids were removed. Then, when the fil-
trated was allowed to further react for 13 h, only an increase of
5.0% conversion was observed, which is due to the thermal reac-
tion. (Fig. S8).
The IRMOF-3-SI-Au is a general catalyst for the three-compo-
nent coupling and cyclization, as can be deduced by reacting vari-
ous amines, aldehydes, and N-protected ethylanilines. The results
summarized in Table 1 show that 90% of isolated yield can be ob-
tained at 40 °C for the three-component coupling and cyclization of
E1, paraformaldehyde, and piperidine (entry 1). Although the reac-
tivity is a little lower when changing paraformaldehyde by cyclo-
hexylcarboxaldehyde or octylaldehyde, good to excellent yields
can still be obtained when the reaction was performed at 80 °C
(entries 2 and 3). Remarkably the reaction with chiral amines
was also successful, giving yields of more than 83% (entries 4
and 5). Ethynylaniline E2 bearing an electron-donating methyl
group reacts smoothly with paraformaldehyde and cyclohexyl-
carboxaldehyde and gives yields of 94% and 83%, respectively (en-
tries 6 and 7), while 78% and 70% yields are found with E3 bearing
one electron-withdrawing methoxycarbonyl group (entries 10 and
11). Both E2 and E3 are allowed to react with chiral amines to af-
ford accordingly coupling and cyclization products with good to
excellent yields (entries 8, 9, and 12).
(HCOH)n
(HCOH)n
(HCOH)n
Cyclohexyl
(HCOH)n
(HCOH)n
(HCOH)n
Cyclohexyl
(HCOH)n
M3
Piperidine
Piperidine
M2
9
M3
4
10
11
12
Piperidine
Piperidine
M2
12
12
6
a
Reaction conditions: ethynylaniline (0.2 mmol), aldehyde (0.4 mmol), and
amine (0.24 mmol), IRMOF-3-SI-Au (14 mg, gold: 0.001 mmol), 1,4-dioxane
(1.0 mL), 40 °C (80 °C for entries 2–5, 7–9, 11, 12).
b
Chiral amines denoted as M2, M3 are (s)-(+)-2-(methoxymethyl)-pyrrolidine
and (s)-(+)-2-methylpiperdine, respectively.
c
Isolated yields based on ethynylaniline.
3.3. The gold MOF as catalyst for 1,3-butadiene hydrogenation:
Elucidation of the active sites
As we said before, highly stabilized Au(III) species on MOF
can help to elucidate the nature of the actives sites for particular
reactions, especially in those cases where the reaction is per-
formed under reducing atmosphere, and cationic gold species
are claimed as the active sites. This is, for instance, the case
for the hydrogenation of alkenes with gold catalysts. Indeed, it
has been shown that supported gold catalysts are active for
hydrogenation of alkenes and small metallic gold nanoparticles
were assumed to be the active sites [13]. However, the sugges-
tion was unconfirmed, and recently oxide-supported (MgO,
ZrO2) mononuclear Au(III) species [14] and isolated Au(III) ions
[9c,9d,9e] have been claimed as the actual active sites for ethene
and 1,3-butadiene hydrogenations, respectively. If this last
hypothesis is true, i.e. isolated Au(III) ions are the active species
for the selective hydrogenation of butadiene [9c], then our IR-
MOF-3-SI-Au catalyst with single, isolated Au(III) species should
be active and selective for carrying out the hydrogenation of
butadiene to butenes. If this was so, not only the hypothesis
on the cationic nature of the active sites would be demonstrated,
but also a new highly active heterogeneous IRMOF-3-SI-Au cata-
lyst with well-defined and controlled active sites would be
available.
Activity and selectivity over IRMOF-3-SI-Au for hydrogenation
of 1,3-butadiene are shown in Fig. 3. For comparison, the results
obtained over Au/TiO2 catalyst (1.5 wt% gold) provided by World
Gold Council pretreated either in flowing Ar at 130 °C, or in H2 at
250 °C, have also been included. 1% 1,3-butadiene conversion
was obtained in a blank test using gold-free IRMOF-3-SI.
The overall conversions for Au/TiO2 catalyst pretreated in Ar
and in H2 were very similar (ca. 9%), and much lower than the con-
version obtained over IRMOF-3-SI-Au (ca. 96%). The TOF (540 hÀ1
)
for IRMOF-3-SI-Au calculated on the bases of total gold weight, is
one order of magnitude higher than that of Au/TiO2 catalysts
(50.4 hÀ1). Considering that Au/TiO2 catalyst pretreated in Ar and

X. Zhang et al. / Journal of Catalysis 265 (2009) 155–160
159
A
B
100
80
60
40
20
0
50
SE2
S1
40
30
20
10
0
SZ2
Sn
Ar
Ar
Au/TiO2
H2
0
1
2
3
4
5
6
Time on Stream /h
IRMOF-3-SI-Au
Fig. 3. (A) 1,3-Butadiene conversion versus time on stream and (B) product selectivity over IRMOF-3-SI (M), IRMOF-3-SI-Au (j), and Au/TiO2 pretreated in flowing H2 at
250 °C (s) and Ar 130 °C (.) at 130 °C. S1, SE2, SZ2, and Sn represent the selectivity of 1-butene, E-2-butene, Z-2-butene, and n-butane, respectively.
H2 both contain metallic gold, our results show that the oxidation
state of gold is proven to be important for hydrogenation of 1,3-
butadiene. The MOF-stabilized Au(III) species is more active than
TiO2-supported metallic gold. Interestingly, IRMOF-3-SI-Au shows
very high selectivity (up to 97%) for butenes with 1-butene and
E-2-butene as the main products, and only 3% of butane selectivity
is detected at almost completely consumption of 1,3-butadiene. On
the contrary, a significant amount of butane (ca. 8%) was formed on
Au/TiO2 pretreated in H2 even at low conversion of 1,3-butadiene,
and the butane selectivity sharply decreased for Ar pretreated Au/
TiO2 catalyst (Fig. 3b).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2009.04.021.
References
[1] (a) For selected reviews, see: G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng. 41
(1999) 319;
(b) M. Haruta, Chem. Rec. 3 (2003) 75;
(c) A.S.K. Hashmi, G.J. Hutchings, Angew Chem. 118 (2006) 8064 (Angew.
Chem. Int. Ed. 45 (2006) 7896);
Since gold remains as Au(III) in IRMOF-3-SI-Au after the hydro-
genation reaction, we confirm the previous suggestion that mono-
nuclear Au(III) species [14] and isolated Au(III) [9c,9d,9e] are the
active sites for ethane and 1,3-butadiene hydrogenation. This con-
clusion would be in agreement with the fact that homogeneous
Au(III) Schiff base complex can activate H2, and catalyzes the li-
quid-phase hydrogenation of diethyl itaconate in a batch reactor
[10]. In the case of the MOF, we have now a solid catalyst with ac-
tive surface-type Au(III) Schiff base complex that can activate H2
and selectively perform the gas-phase hydrogenation of 1,3-buta-
diene. The clear advantage of the gold MOF catalyst with respect
to MgO-supported mononuclear Au(III) species is that the latter
are not stable in air and must be kept under anaerobic and anhy-
drous conditions [14], while IRMOF-3-SI-Au is stable in air and
shows very stable activity for the selective hydrogenation of 1,3-
butadiene (Fig. S9).
(d) D.J. Gorin, F.D. Toste, Nature 446 (2007) 395;
(e) A. Corma, P. Serna, Science 313 (2006) 332.
[2] (a) For selected references, see: G.C. Bond, C. Louis, D.T. Thompson, in:
Catalysis Science Series, vol. 6, Catalysis by Gold, 2006;
(b) X. Zhang, H. Wang, B.-Q. Xu, J. Phys. Chem. B 109 (2005) 9678;
(c) A.S.K. Hashmi, Angew Chem. 117 (2005) 7150 (Angew. Chem. Int. Ed. 44
(2005) 6990);
(d) A. Hoffmann-Röder, N. Krause, Org. Biomol. Chem. 3 (2005) 387.
[3] (a) A. Corma, Catal. Rev. 46 (2004) 319;
(b) X. Zhang, A. Corma, Chem. Commun. 3080 (2007);
(c) X. Zhang, A. Corma, Dalton Trans. 397 (2008);
(d) Y. Fukuda, K. Utimoto, J. Org. Chem. 56 (1991) 3729;
(e) T.E. Müller, Tetrahedron Lett. 39 (1998) 5961.
[4] (a) For selected references, see: M. Eddaoudi, D.B. Moler, H. Li, B. Chen, T.M.
Reineke, M. O’Keeffe, O.M. Yaghi, Acc. Chem. Res. 34 (2001) 319;
(b) M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe, O.M. Yaghi,
Science 295 (2002) 469;
(c) G. Ferey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surble, I.
Margiolaki, Science 309 (2005) 2040;
(d) M. Alvaro, E. Carbonell, B. Ferrer, F.X. Llabres i Xamena, H. Garcia, Chem.
Eur. J 13 (2007) 5106;
In conclusion a novel catalyst has been prepared by modifica-
tion of a metal organic framework (IRMOF-3) with coordinative
unsaturated Au(III) species based on a post-synthesis strategy.
The resulting heterogeneous IRMOF-3-SI-Au catalyst exhibits
high ability to stabilize cationic Au(III) species and 100% utility
of gold active sites, which has emulated the catalytic properties
of homogeneous gold catalysis. The IRMOF-3-SI-Au gives much
higher catalytic performance in catalyzed domino coupling and
cyclization of N-protected ethylanilines, amine, and aldehyde in
comparison with oxide-supported gold and homogeneous gold
salt/complex, and it has also demonstrated that Au(III) species
are highly active and selective for the selective hydrogenation
of 1,3-butadiene.
(e) F.X. Llabres i Xamena, O. Casanova, R. Galiasso Tailleur, H. Garcia, A. Corma,
J. Catal. 255 (2008) 220;
(f) F.X. Llabres i Xamena, A. Abad, A. Corma, H. Garcia, J. Catal. 250 (2007) 294;
(g) B.V. Harbuzaru, A. Corma, F. Rey, P. Atienzar, J.L. Jorda, H. Garcia, D.
Ananias, L.D. Carlos, J. Rocha, Angew. Chem. Int. Ed. 47 (2008) 1080.
[5] (a) S.S.Y. Chui, S.M.F. Lo, J.P.H. Charmant, A.G. Orpen, I.D. Williams, Science 283
(1999) 1148;
(b) S. Horike, M. Dinca, K. Tamaki, J.R. Long, J. Am. Chem. Soc. 130 (2008) 5854;
(c) S. Noro, S. Kitagawa, M. Yamashita, T. Wada, Chem. Commun. 222 (2002);
(d) D.N. Dybtsev, A.L. Nuzhdin, H. Chun, K.P. Bryliakov, E.P. Talsi, V.P. Fedin, K.
Kim, Angew. Chem. Int. Ed. 45 (2006) 916;
(e) M. Kawano, T. Kawamichi, T. Haneda, T. Kojima, M. Fujita, J. Am. Chem. Soc.
129 (2007) 15418.
[6] (a) Z.Q. Wang, S.M. Cohen, J. Am. Chem. Soc. 129 (2007) 12368;
(b) K.K. Tanabe, Z. Wang, S.M. Cohen, J. Am. Chem. Soc. 130 (2008) 8508;
(c) J.S. Costa, P. Gamez, C.A. Black, O. Roubeau, S.J. Teat, J. Reedijk, Eur. J. Inorg.
Chem. 1551 (2008);
(d) M.J. Ingleson, J.P. Barrio, J. Bacsa, C. Dickinson, H. Park, M.J. Rosseinsky,
Chem. Commun. 1287 (2008);
(e) Y.K. Hwang, D.Y. Hong, J.S. Chang, S.H. Jhung, Y.K. Seo, J. Kim, A. Vimont, M.
Daturi, C. Serre, G. Ferey, Angew. Chem. Int. Ed. 47 (2008) 4144;
(f) W. Morris, C.J. Doonan, H. Furukawa, R. Banerjee, O.M. Yaghi, J. Am. Chem.
Soc. 130 (2008) 12626;
Acknowledgments
The authors thank the Dirección General de Investigación
´
Cientıfica y Técnica of Spain (Project MAT2006-14274-C02- 01)
(g) M.J. Ingleson, J.P. Barrio, J.B. Guilbaud, Y.Z. Khimyak, M.J. Rosseinsky, Chem.
Commun. 2680 (2008).
and the UPV (Project PAID-06-06-4726) for funding. The Ministerio
de Educación y Ciencia of Spain is gratefully acknowledged for a
‘‘Ramón y Cajal” research contract to FXLX. Experimental assis-
[7] (a) J.M. Notestein, A. Katz, Chem. Eur. J. 12 (2006) 3954;
(b) A. Corma, Catal. Rev. Sci. Eng. 46 (2004) 369.
[8] L.M. Huang, H.T. Wang, J.X. Chen, Z.B. Wang, J.Y. Sun, D.Y. Zhao, Y.S. Yan,
Micropor. Mesopor. Mater. 58 (2003) 105.
´
tance of Á. Cantın and E. Peris is acknowledged.

160
X. Zhang et al. / Journal of Catalysis 265 (2009) 155–160
[9] (a) P. Concepcion, S. Carrettin, A. Corma, Appl. Catal. A: General 307 (2006) 42;
[12] (a) G.R. Humphrey, J.T. Kuethe, Chem. Rev. 106 (2006) 2875;
(b) S. Cacchi, G. Fabrizi, Chem. Rev. 105 (2005) 2873;
(c) H. Ohno, Y. Ohta, S. Oishi, N. Fujii, Angew Chem. 119 (2007) 2345 (Angew.
Chem. Int. Ed. 46 (2007) 2295);
(b) J. Guzman, S. Carrettin, A. Corma, J. Am. Chem. Soc. 127 (2005) 3286;
(c) X. Zhang, H. Shi, B.-Q. Xu, Angew Chem. 117 (2005) 7294 (Angew. Chem.
Int. Ed. 44 (2005) 7132);
(d) X. Zhang, H. Shi, B.-Q. Xu, Catal. Today 122 (2007) 330;
(e) Z.-P. Liu, C.-M. Wang, K.-N. Fan, Angew Chem. 118 (2006) 7019 (Angew.
Chem. Int. Ed. 45 (2006) 6865).
(d) For another gold-catalyzed synthesis of the indole nucleus, see: A.S.K.
Hashmi, M. Rudolph, J.W. Bats, W. Frey, F. Rominger, T. Oeser, Chem. Eur. J. 14
(2008) 6672.
[10] (a) A. Comas-Vives, C. Gonzalez-Arellano, A. Corma, M. Iglesias, F. Sánchez, G.
Ujaque, J. Am. Chem. Soc. 128 (2006) 4756;
[13] (a) P.A. Sermon, G.C. Bond, P.B. Wells, J. Chem. Soc. Faraday Trans. 175 (1979)
385;
(b) C. Gonzalez-Arellano, A. Corma, M. Iglesias, F. Sánchez, Chem. Commun.
3451 (2005);
(b) M. Okumura, T. Akita, M. Haruta, Catal. Today 74 (2002) 265S;
(c) P. Claus, Appl. Catal. A 291 (2005) 222.
(c) C. Gonzalez-Arellano, A. Corma, M. Iglesias, F. Sánchez, Chem. Commun.
1990 (2005).
[14] (a) J. Guzman, B.C. Gates, Angew. Chem. 115 (2003) 714 (Angew. Chem. Int.
Ed. 42 (2003) 690);
[11] X. Zhang, A. Corma, Angew Chem. 120 (2008) 4430;
X. Zhang, A. Corma, Angew. Chem. Int. Ed. 47 (2008) 4358.
(b) J. Guzman, B.C. Gates, J. Catal. 226 (2004) 111.