MOFs as acid catalysts with shape selectivity properties

Article abstract of DOI:10.1039/b803953b

MOFs permit the paraalkylation of large polyaromatic compounds with nearly 100% of regioselectivity, offering new alternatives to standard zeolite catalysts. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/b803953b

View Online / Journal Homepage / Table of Contents for this issue
LETTER
www.rsc.org/njc | New Journal of Chemistry
MOFs as acid catalysts with shape selectivity properties
`
Ugo Ravon, Marcelo E. Domine, Cyril Gaudillere, Arnold Desmartin-Chomel and
David Farrusseng*
Received (in Montpellier, France) 6th March 2008, Accepted 10th April 2008
First published as an Advance Article on the web 7th May 2008
DOI: 10.1039/b803953b
MOFs permit the paraalkylation of large polyaromatic
compounds with nearly 100% of regioselectivity, offering new
alternatives to standard zeolite catalysts.
the preparation of a series of MOFs based on aromatic
dicarboxylic acids coordinated to Zn O subunits. These struc-
tures, referred to as iso-reticular metal–organic frameworks
IRMOFs), have demonstrate that control over the pore size
4
(
Metal–organic frameworks (MOFs) are a class of crystalline
hybrid material whose crystal structures are made up of
extended 3D networks of metal ions or small discrete clusters
and functionality can be achieved. For instance, IRMOF-8
and IRMOF-10 are isoreticular compounds, with 2,6-
0
naphthalenedicarboxylic acid (2,6-NDC) and 4,4 -biphenyl-
0
1
connected through multidentate organic spacers. The devel-
dicarboxylic acid (4,4 -BPDC) as linkers, respectively.
opment of these porous coordination polymers has already
opened up new perspectives in applications such as gas
The catalyst IRMOF-1 was prepared according to the
method developed by Huang et al., which consists of adding
pure triethylamine to a solution of zinc nitrate and BDC in
2
separation and storage. The ability to tune pore size on an
8
Angstrom scale while allowing the design of accessible metallic
nanoclusters in a highly porous structure makes these com-
3
pounds very attractive for catalysis. The catalytic properties
DMF. This method was extended to the synthesis IRMOF-8
by using 2,6-NDC as the ligand.
Structural (X-ray diffraction, IR) and thermal (TDA, TGA)
characterisation data for IRMOF-1 are in very good agree-
of HKUST-1, which uses Cu paddle wheel clusters as Lewis
4
centres, is an outstanding proof of the concept. However, we
8
ment with the results reported by Huang et al. and Havizovic
9
et al. (Fig. 1 and Fig. 2). The sharp band at 3602 cm can be
ꢀ1
consider that published catalytic results are not in line with the
generally acknowledged potential of MOFs. Indeed, only a
limited number of successful catalytic studies have been
1
0
assigned to Zn–OH species. We also measured a 10% excess
of Zn with respect to the theoretical formulae (elemental
analysis and TGA), which leads to a global composition of
5
reported. Furthermore, the chosen catalytic reactions were
usually model reactions of little industrial interest and/or
reactions that did not need sophisticated catalysts in order
to take place.
Zn
4
O(BDC)ꢁ2H
2 2
O + Zn(OH) . According to Havizovic
9
et al., zinc hydroxide nanoclusters that partly occupy the
cavities are responsible for the structure distortion evidenced
by the splitting of the diffraction peak at 9.71 (see inset to
In contrast, the catalytic alkylation of aromatics is under-
6
taken on a large scale in the chemical industry. Among para-
1
Fig. 1). On the other hand, the H NMR spectra of different
dialkylated aromatics, para-xylene, para-diisopropylbenzene,
para-ethyltoluene, para-diethylbenzene, para- and meta-cym-
enes, and 4-tert-butyltoluene are very important in the fine
chemical and petrochemical industries. Medium and large
pore acidic zeolites are outstanding catalysts, since they com-
bine high activity (acidic sites) and shape selectivity thanks to
their micropore structures, which favour para-alkylation (less
bulky products). Recently, the benzylation of toluene using the
IRMOF-1 samples indicated the presence of Zn–OH–Zn
species (very small peak at d = ꢀ0.4), like those encountered
1
1,12
in Zn
(OH)
3
(BDC)
2
ꢁ2DEF (MOF-69c),
which contains
2
7
coordination polymer MIL-100(Fe) was reported. However,
shape selectivity properties were not investigated.
For the first time, we report herein MOFs that undertake an
outstanding shape-selective alkylation of aromatics that out-
perform reference zeolites. The MOF catalysts are derived
1
c,3b
from well known IRMOF materials.
of IRMOF-1 consists of Zn O subunits that are linked in
The cubic framework
4
octahedral arrays of 1,4-benzenedicarboxylic acidic (BDC)
groups to form a porous material with a channel window of
˚
about 8 A. The topological simplicity of IRMOF-1 has lead to
Fig. 1 Powder X-ray diffraction pattern of IRMOF-1: Top: Experi-
mental result with zoom-in inset for the angles 8–11 (2y). Bottom:
Simulated pattern from MOF5-Zn.
IRCELYON, UMR UCBL-CNRS 5256, 2, Avenue Albert Einstein,
69626 Villeurbanne, France. E-mail: david.farrusseng@ircelyon.univ-
lyon1.fr; Fax: +33 (0) 4 72445365
9
This journal is ꢂc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
New J. Chem., 2008, 32, 937–940 | 937

View Online
Table 1
2
N physisorption results and comparison with literature data
Pore
volume/
Micropore
volume/
cm g
Surface
area/m g
Pore
diameter/A
2
ꢀ1a
3
cm g
ꢀ1
3
ꢀ1b
c
˚
Sample
IRMOF-1
IRMOF-8
H-BEA
a
450
730
430
0.32
0.46
0.62
0.18
0.29
0.10
c
8.8
7.3/11.6
7.7
16 b
From BET analysis.
From t-plot analysis. From DFT analysis.
In order to complete the picture of the IRMOF samples
with regard to their porosity properties, the calculated sizes of
their cages and channel windows from the structural data of
pure crystals are reported in Table 2. For example, the window
between the cages of IRMOF-1 have a van der Waals space
˚
that can accommodate a sphere of approximately 7.8 A
diameter, herein referred to as the ‘‘free-space’’ diameter. In
contrast to zeolite pore sizes, these values shall be interpreted
with care for MOF materials, especially when correlating
structural features and properties. Indeed, in addition to their
general flexibility, the aromatic rings of IRMOF can rotate,
1
b,7,17
thus making feasible the diffusion of larger guests.
Because of size matching between alkylbiphenyl compounds
and large pore zeolites, the alkylation of biphenyl is an
appropriate reaction to highlight the pore shape selectivity
1
9
properties in acidic catalysis. The activity and selectivity of
the MOF samples were tested in the alkylation of toluene and
biphenyl with tert-butylchloride in a batch reactor. The acidic
1
Fig. 2 IRMOF-1 short range characterization: Top: H NMR.
Bottom: In situ DRIFT after desorption at 220 1C.
form of beta zeolite (denoted hereafter H-BEA), AlCl
ZnCl were used as reference catalysts. As a result, IRMOF
samples were as active as H-BEA and AlCl , and reactions
3
and
2
Zn–(m
3
-OH)–Zn chains. Peak integration gave a ratio of one
3
m
3
-OH for every 25 BDC linkers. The later phase can be
were complete within 2 h at 170 1C (Table 3). For toluene
alkylation, IRMOF samples exhibited a high selectivity for the
formed during the synthesis of IRMOF-1, and is kinetically
favoured in moisture conditions, as was the case in this
3
less bulky para-oriented products, whereas AlCl produced
1
study. Therefore, in addition to zinc hydroxide clusters,
3
mixtures of ortho and para compounds. On the other hand, the
alkylation of biphenyl revealed an exceptional selectivity for
para-oriented 4-tert-butylbiphenyl with IRMOF samples. In-
deed, only traces of dialkylated products (about 1%) were
detected when using the IRMOFs. In contrast, H-BEA and
IRMOF-1 samples also contain Zn–(m -OH)–Zn chains in
3
minor quantities, which could either be a crystalline default
of the cubic structure or non-XRD detectable MOF-69c
microcrystallites. Finally, it was recently demonstrated that
pure IRMOF-1 undergoes a rapid phase transformation under
moist conditions that is accompanied by a drastic decrease in
3
AlCl showed much higher selectivities for the ortho-oriented
product, and about 15% for the dialkylated products. In
conclusion, these results reveal the outstanding pore shape
selectivity properties of IRMOFs for large polyaromatics.
Measurements of conversion at different reaction times
indicate that there are similar kinetics for IRMOF-1,
IRMOF-8 and H-BEA at 100 1C (Fig. 3). In contrast to
MIL-101(Cr), IRMOFs are active in the first minutes of
1
4
accessible porous volume.
From N physisorption analysis at 77 K, the IRMOFs were
2
determined to be microporous, and the isotherms were of type
I. Their surface areas, micropore volumes and pore sizes
correspond to those reported elsewhere that had similar pre-
8
,9
paration procedures to those used for IRMOF-1
and
1
5
IRMOF-8 (Table 1). A logarithmic plot (not shown) reveals
that for IRMOF-8, N adsorption occurs in two steps, reveal-
2
Table 2 Theoretical cage and window sizes from structural data
ing two distinct pore sizes. The actual calculation of the pore
size using the Horvath–Kawazoe or Density Functional The-
ory (DFT) method is questionable, since the principal inter-
action potentials of the adsorbate and the adsorbent are
unknown, and assumptions have to be made about pore
morphology. The calculated pore diameters from the DFT
method (Table 2) are therefore not fully accurate but, more
importantly, allow comparison among members of a class of
similar materials, which presumably have the same interaction
potentials.
Cage free-space
˚
diameter/A
Window free-space
˚
diameter/A
a
a
IRMOF-1
IRMOF-8
H-BEA
15.1
18.0
6.6
7.8
9.2
6.1
b
a
Free-space diameter refers to the diameter of the largest sphere that
can fit into the cage or through the window when the van der Waals
b
radii from ref. 3b are assumed for all the framework atoms. From
ref. 18.
9
38 | New J. Chem., 2008, 32, 937–940
This journal is ꢂc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Online
a
Table 3 Selectivity of tert-butylation reaction at 170 1C
soluble in decane, indicating that the reactions did not proceed
in a homogeneous phase.
Toluene
Biphenyl
The exact catalytic centre is still unknown. As described
earlier, acidic centres may arise either from zinc hydroxide
clusters, from the presence of MOF-69c, or from structural
Zn–OH defects formed at the synthesis step or upon water
adsorption. Further experiments are ongoing in order to
identify and assess these acidic sites.
b
d
c
d
Catalyst
para
ortho
di
para
ortho
di
IRMOF-1
IRMOF-8
H-BEA
82
84
72
46
18
16
28
54
0
0
0
0
96
95
55
51
3
3
22
38
1
2
23
11
AlCl
3
a
b
Conversions ranged from 60 to 80%. para-tert-Butyltoluene.
d
-tert-Butylbiphenyl. Sum of dialkylated products.
From the experimental and theoretical data, there is, in
principle, no steric hindrance to monoalkylated products
diffusing out of the porous solids, including H-BEA. Even
c
4
0
very bulky products, such as the dialkylated product 4,4 -
˚
operation. These findings suggest that strong structural mod-
di-tert-butylbiphenylene (12.5 ꢃ 5.0 A), can be formed in large
ifications are not required to create catalytic centers such as
7
Zn–Cl . Furthermore, the poor activity of ZnCl
1
9
amounts in H-BEA. Therefore, the shape selectivity proper-
ties observed for the IRMOF samples cannot arise from
diffusion hindrance of the most bulky products, since the pore
windows are even larger. For the reaction to take place, we
suggest that the biphenyl is absorbed in a specific manner,
allowing the formation of a transition state for the para-
oriented product. Once alkylated, it can not be activated in
the same manner because of steric hindrance, and double
alkylation cannot proceed. Because of the porous structure
similarities of IRMOF-1 and IRMOF-8, it is not surprising
that similar selectivities are obtained.
2
tested as a
2
powder at 170 1C (only 8% conversion) confirms that ZnCl is
not the catalytic specy.
After filtration, the catalysts could be re-used without any
loss of activity or selectivity. XRD patterns of used IRMOF-1
1
did not show major modifications, whereas H NMR revealed
two significant changes (not shown). Firstly, integration of the
broad peak at d = 14 revealed that about 20% of the
carboxylate groups had been hydrolyzed to their acid form.
Secondly, a very intense peak at d = 3.4 indicated the presence
of a guest in the structure, which could be assigned as HCl.
1
In conclusion, we have demonstrated for the first time that
MOFs can be used to undertake very shape-selective (almost
Indeed, very similar H NMR spectra were obtained when
concentrated HCl (36%) was added to fresh IRMOF-1. We
suggest that the HCl, which is produced during the course of
the reaction, is trapped in the framework, most likely at the
oxygen containing nodes, provoking the partial hydrolysis of
the host.
1
00%) catalytic alkylations of large molecules. This discovery
may open new perspectives for the C–C coupling of poly-
aromatics or biomolecules that are too large to be addressed
2
0
by zeolites.
Saturation of the metal ion coordination sphere in coordi-
nation polymers is often argued to be a limitation of MOFs for
In order to check that the reaction takes place under
heterogeneous conditions, issues of leaching were investigated.
Firstly, tests in absence of solid catalysts were carried out
under the same conditions, and any conversion was measured.
Secondly, reaction mixtures were filtered-off after completion
of the catalytic reactions, both solid catalyst and reaction
solution being recovered separately. When a new reaction
was started by adding fresh reactants to the recovered solu-
tion, any additional conversion was detected. This is consistent
with the fact that HCl is trapped in the framework, making the
filtrate HCl-free. Finally, we have observed that the reaction
takes place at 50 1C with IRMOF-1. Under these conditions of
pressure and temperature, the IRMOF samples were not
2
1
catalytic applications. We believe that the engineering of
structural defects in MOF materials may lead to the genera-
tion of a new class of catalysts with different acidities and
hydrophobic properties than are encountered in zeolites.
Original MOFs exhibiting M–OH functions and 100%
regioselectivity have been recently discovered in our labora-
tory, and our findings will be published shortly. The investiga-
tion of other alkylating agents is also in progress.
We thank Dr C. Mirodatos for his support and
IRCELYON Scientific Services.
Experimental
Synthesis
All chemicals were used as received: DMF (Aldrich, 99.8%),
Zn(NO ) ꢁ6H O (Riedel-deHae
¨
n, pure), triethylamine (Riedel-
n, pure), BDC (Sigma Aldrich, 98%), 2,6-NDC (Alfa
Aesar, 98%), AlCl (Sigma-Aldrich, 99.99%) and anhydrous
ZnCl (Alfa Aesar). IRMOF-1 was prepared according to a
3
2
2
deHae
¨
3
2
8
procedure published by Huang et al. A pure solution of 4.4 ml
triethylamine (31.6 mmol) was added with stirring to a mixture
of 660 mg (4.0 mmol) BDC and 2.4 mg (8.1 mmol) of
Zn(NO
3
)ꢁ6H
2
O dissolved in DMF. The white precipitate was
recovered by filtration, washed with DMF and heated at
130 1C in air. H-BEA samples were obtained from Na-form
(ENI Technology, Si/Al = 23 : 1) by ammonium exchange
Fig. 3 Kinetic studies of toluene alkylation at 100 1C using IRMOF-1
(
K), IRMOF-8 (’) and H-BEA (B ).
This journal is ꢂc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008
New J. Chem., 2008, 32, 937–940 | 939

View Online
(
10 wt% NH
4
COOCH
3
dissolved in water) at 80 1C for 4–5 h,
1998, 31, 474–484; (f) M. Higuchi, S. Horike and S. Kitagawa,
Supramol. Chem., 2007, 19, 75–78.
followed by calcination at 550 1C for 3 h. This procedure was
repeated twice.
2
(a) C. Sanchez, B. Julian, P. Belleville and M. Popall, J. Mater.
Chem., 2005, 15, 3559–3592; (b) S. Kitagawa, R. Kitaura and S.
Noro, Angew. Chem., Int. Ed., 2004, 43, 2334–2375; (c) C. Janiak,
Dalton Trans., 2003, 2781–2804; (d) D. Maspoch, D. Ruiz-Molina
and J. Veciana, Chem. Soc. Rev., 2007, 36, 770–818; (e) C. Serre, C.
Mellot-Draznieks, S. Surble, N. Audebrand, Y. Filinchuk and G.
Ferey, Science, 2007, 315, 1828–1831.
Characterization
Samples were analyzed by X-ray diffraction (Bruker D5005),
DRIFT (Magna 550 Nicolet, SpectraTech cell, ZnSe windows,
MCT detector), N
2
absorption measurements (Micrometrics
3 (a) G. Ferey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour,
S. Surble and I. Margiolaki, Science, 2005, 309, 2040–2042; (b) M.
Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keefe and
O. M. Yaghi, Science, 2002, 295, 469–472.
4 (a) K. Schlichte, T. Kratzke and S. Kaskel, Microporous Mesopor-
ous Mater., 2004, 73, 81–88; (b) L. Alaerts, E. Seguin, H. Poelman,
F. Thibault-Starzyk, P. A. Jacobs and D. E. De Vos, Chem.–Eur.
J., 2006, 12, 7353–7363.
5 For recent reviews on catalytic MOFs, see: (a) G. Ferey, Chem.
Soc. Rev., 2008, 37, 191–214; (b) U. Mueller, M. Schubert, F.
Teich, H. Puetter, K. Schierle-Arndt and J. Pastre, J. Mater.
Chem., 2006, 16, 626–636.
ASAP 2010 M) and chemical analysis (Spectroflame ICP-
OES). IRMOF samples were further characterized by TG/
1
TD analysis (Setaram-type SETSYS Evolution 12). For H
NMR (DX 400BRUCKER) MAS technique: 30 kHz spinning
rate, pulse duration 3 ms and repetition time 16 s. Peaks: d ꢀ0.4
(
(
Zn–(m -OH)–Zn), d 1–3 (DMF), d 4–5.5 (H O) and d 6–10
3
2
C
6
6
H ).
Catalytic alkylation
6
C. Perego, S. Amarilli, A. Carati, C. Flego, G. Pazzuconi, C. Rizzo
and G. Bellussi, Microporous Mesoporous Mater., 1999, 27, 345–354.
Toluene (Chimie-Plus, 99%), biphenyl (Alfa Aesar, 99%),
decane (Alfa Aesar, 99%), tert-butylchloride (Alfa Aesar
7 P. Horcajada, S. Surble, C. Serre, D. Y. Hong, Y. K. Seo, J. S.
Chang, J. M. Greneche, I. Margiolaki and G. Ferey, Chem.
Commun., 2007, 2820–2822.
9
8+%) and mesitylene (Aldrich, 98%) were used as received.
The reactants were dissolved in decane. The reaction was
carried out with an aryl : tert-butylchloride ratio of 2 : 1.
Typically, a mixture of 2 ml (18.8 mmol) of toluene, 1 ml of
tert-butylchloride (9.2 mmol) and 7 ml of decane were placed
in a 48 ml Teflon-lined autoclave (Top Industrie). 30 mg of
8
L. M. Huang, H. T. Wang, J. X. Chen, Z. B. Wang, J. Y. Sun, D.
Y. Zhao and Y. S. Yan, Microporous Mesoporous Mater., 2003, 58,
105–114.
J. Hafizovic, M. Bjorgen, U. Olsbye, P. D. C. Dietzel, S. Bordiga,
C. Prestipino, C. Lamberti and K. P. Lillerud, J. Am. Chem. Soc.,
9
2
007, 129, 3612–3620.
10 J. H. Liao, T. J. Lee and C. T. Su, Inorg. Chem. Commun., 2006, 9,
01–204.
ꢀ
5
catalysts were desorbed at 130 1C and 10 bar. After 2 h of
stirring at 170 1C, the solid was recovered by room tempera-
ture filtration, and the filtrate was analysed by gas chromato-
graphy (HP 6890N equipped with a 30 m HP5 column), using
2
1
1 N. L. Rosi, M. Eddaoudi, J. Kim, M. O’Keeffe and O. M. Yaghi,
Angew. Chem., Int. Ed., 2001, 41, 284–287.
12 T. Loiseau, H. Muguerra, G. Ferey, M. Haouas and F. Taulelle, J.
Solid State Chem., 2005, 178, 621–628.
5
% mesitylene as the internal standard.
1
3 S. Hausdorf, F. Baitalow, J. Seidel and F. Mertens, J. Phys. Chem.
A, 2007, 111, 4259–4266.
Catalyst re-use and leaching tests
1
4 M. Sabo, A. Henschel, H. Froede, E. Klemm and S. Kaskel, J.
Mater. Chem., 2007, 17, 3827–3832.
After the completion of reactions, the solid catalysts were first
filtered off. The filtrate and recovered solid were then treated
separately so as to investigate leaching issues and undertake
reuse, respectively. Fresh reactants in the same ratio as those
used in the previous tests were then added to the filtrate. The
solution mixture (in absence of solid) was then heated to 170 1C
for 2 h. Any conversion was then measured. On the other hand,
the recovered solid was washed with water, ethanol and
acetone prior to post-reaction characterization and reuse.
1
1
5 Y. W. Li and R. T. Yang, J. Am. Chem. Soc., 2006, 128, 726–727.
6 K. S. Walton and R. Q. Snurr, J. Am. Chem. Soc., 2007, 129,
8
552–8556.
17 A. J. Fletcher, K. M. Thomas and M. J. Rosseinsky, J. Solid State
Chem., 2005, 178, 2491–2510.
1
8 M. D. Foster, I. Rivin, M. M. J. Treacy and O. D. Friedrichs,
Microporous Mesoporous Mater., 2006, 90, 32–38.
1
9 (a) D. Mravec, P. Zavadan, A. Kaszonyi, J. Joffre and P. Moreau,
Appl. Catal., A, 2004, 257, 49–55; (b) R. Millini, F. Frigerio, G.
Bellussi, G. Pazzuconi, C. Perego, P. Pollesel and U. Romano, J.
Catal., 2003, 217, 298–309; (c) J. Horniakova, D. Mravec, J. Joffre
and P. Moreau, J. Mol. Catal. A: Chem., 2002, 185, 249–257; (d) Y.
Sugi, S. Tawada, T. Sugimura, Y. Kubota, T. Hanaoka, T.
Matsuzaki, K. Nakajima and K. Kunimori, Appl. Catal., A,
1999, 189, 251–261.
References
1
(a) J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous
Mater., 2004, 73, 3–14; (b) M. J. Rosseinsky, Microporous Meso-
porous Mater., 2004, 73, 15–30; (c) O. M. Yaghi, M. O’Keeffe, N.
W. Ockwig, H. K. Chae, M. Eddaoudi and J. Kim, Nature, 2003,
20 (a) E. Armengol, M. L. Cano, A. Corma, H. Garcia and M. T.
Navarro, J. Chem. Soc., Chem. Commun., 1995, 519–520; (b) S. B.
Pu, J. B. Kim, M. Seno and T. Inui, Microporous Mater., 1997, 10,
25–33.
423, 705–714; (d) A. K. Cheetham, G. Ferey and T. Loiseau,
Angew. Chem., Int. Ed., 1999, 38, 3268–3292; (e) O. M. Yaghi, H.
L. Li, C. Davis, D. Richardson and T. L. Groy, Acc. Chem. Res.,
21 F. X. Llabres i Xamena, A. Corma and H. Garcia, J. Phys. Chem.
´
C, 2007, 111, 80–85.
9
40 | New J. Chem., 2008, 32, 937–940
This journal is ꢂc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008