Generic postfunctionalization route from amino-derived metal-organic frameworks

Article abstract of DOI:10.1021/ja909613e

This study deals with the development of a soft, generic, one-pot postfunctionalization method for metal-organic frameworks (MOFs) starting from compounds with an amino group on the linker. The first step consists of transforming the amino group into azid

Full text of DOI:10.1021/ja909613e

Published on Web 03/16/2010
Generic Postfunctionalization Route from Amino-Derived Metal-Organic
Frameworks
Marie Savonnet,†,‡ Delphine Bazer-Bachi,‡ Nicolas Bats,‡ Javier Perez-Pellitero,‡ Erwann Jeanneau,§
Vincent Lecocq,‡ Catherine Pinel,† and David Farrusseng*,†
IRCELYON, Institut de recherches sur la catalyse et l’enVironnement de Lyon; UniVersite´ Lyon 1 - UMR 5256
CNRS, 2 aVenue Albert Einstein, F-69626 Villeurbanne Cedex, France, IFP-Lyon, BP n°3, 69360 Solaize, France,
and UniVersite´ Lyon 1, Centre de Diffractome´trie, 69629 Villeurbanne Cedex, France
Received November 12, 2009; E-mail: david.farrusseng@ircelyon.univ-lyon1.fr
The secondary building unit (SBU) approach for engineering of
metal-organic frameworks (MOFs) with tunable pore sizes is very
attractive for designing practical properties such as separation by
molecular sieving and shape-selective catalysis.1 The conceptual
approach used to increase the pore sizes of MOFs through the use
of longer ligands has also been extended to the design of
multifunctional MOFs bearing a functional group on the organic
moiety. This is the case for IRMOF-3,2 MOF-101(-Br),3 and MIL-
53(Al)-NH2,4 to name a few. However, this extension is not
straightforward in practice.5 Indeed, the chemistry of MOF network
formation is very sensitive to the chemical reactivity and solubility
of the functionalized linkers.6 This is particularly the case for
functions such as -OH, -COOH, and N-donating groups, which
can interfere with the coordination chemistry associated with the
assembly of the nodes.
derived MOF compounds.14 The first step involves an original
method to convert a MOF in its amino form to the corresponding
azide compound. Next, the desired functionalized material is
obtained by “clicking” a synthon onto the azide without isolation
of the intermediate.
The starting materials DMOF-NH2 [Zn(bdc-NH2)(DABCO)]
(section S1 in the Supporting Information) and MIL-68(In)-NH2
[In(OH)(bdc-NH2)]15 (section S5) were selected as representative
compounds. The former is a zero-dimensional-type MOF with a
layered structure made of Zn carboxylate sheets supported by 1,4-
diazabicyclo(2.2.2)octane (DABCO) pillars. It is representative of
a large class of porous coordination polymers (PCPs).16 On the
other hand, MIL-68(In)-NH2 belongs to another important class of
MOFs characterized by a one-dimensional rod-shaped structure.
This class has been reviewed by Yaghi et al.17
When self-assembly fails in the synthesis of an MOF with
functionalized linkers, the postfunctionalization of a parent MOF
appears to be a very valuable alternative.7 In fact, postsynthesis
opens the door to advanced porous solid engineering by multiple
synthesis steps8 and, as a consequence, to the design of new types
of adsorbents and catalysts. Postfunctionalization consists of
modifying the organic part of the MOF by a chemical reaction that
takes place within the porous framework. In this case, the parent
MOF must possess accessible reactive functions. Similar issues have
been resolved for MCM-like materials, for which various func-
tionalization methods have been developed.9 In a similar fashion
to alkylamino-functionalized MCMs,10 amino-derived MOFs such
It should be noted that the usual route for preparing azide
compounds from the corresponding amines via their diazonium salts
cannot be applied here because DMOF-NH2 dissolves under acidic
conditions. Instead, we investigated another pathway that uses mild
conditions and involves stable, nonexplosive compounds.18
Scheme 1. One-Pot, Two-Step Functionalization of DMOF-NH2
11
as IRMOF-3 and DMOF-NH2 are excellent platforms for the
grafting of various synthons such as aldehydes, isocyanates, and
acid anhydrides. That said, the suitability, diversity and availability
of such synthons can appear limited when one considers the ever-
increasing demands for different functionalities and the ambitious
projects imagined by chemists.
Valuable alternatives lie in the development of all kinds of
generic postfunctionalization methods that are soft, do not liberate
byproducts that may react or remain blocked in the pores, and enable
grafting of a wide variety of chemical functions with high efficiency
and selectivity. The Sharpless “click” reaction using CuI-catalyzed
Huisgen cycloaddition of azides to alkynes fulfills all of these
criteria.12,13 The corresponding azide linkers are, however, highly
unstable and not commercially available; this significantly limits
the synthesis of MOFs bearing azide groups by self-assembly
methods.
In a typical synthesis (Scheme 1), the freshly dried DMOF-NH2
was treated with tBuONO and TMSN3 in THF overnight at room
temperature to produce the corresponding azide intermediate
compound DMOF-N3.19 In the same vessel, the functionalized
DMOF (DMOF-fun) was obtained by addition of excess pheny-
lacetylene in the presence of CuI(CH3CN)4PF6 followed by continu-
ous stirring for 24 h (section S3). For characterization purposes,
the synthesis was stopped after formation of the azide intermediate
DMOF-N3.
MIL-68(In)-NH2 was modified by applying a similar procedure.20
For sake of brevity, details are given in section S6 in the Supporting
Information.
The objective of this work was to develop a one-pot, two-step
functionalization method starting from already available amino-
Clear proof of azide formation and the subsequent (3 + 2)
cycloaddition was obtained by IR spectroscopy. The absorption
band of DMOF-N3 at 2123 cm-1 is characteristic of the N3
asymmetric stretching vibration (section S3). In addition, unam-
biguous characterization and quantification were provided by liquid
† IRCELYON UMR 5256 CNRS.
‡ IFP-Lyon.
§ Universite´ Lyon 1.
9
4518 J. AM. CHEM. SOC. 2010, 132, 4518–4519
10.1021/ja909613e  2010 American Chemical Society

C O M M U N I C A T I O N S
1H NMR analysis. Samples of DMOF-N3 were digested and
dissolved in dilute DCl/D2O/DMSO-d6 solution (Figure 1) and then
analyzed by 1H NMR spectroscopy, which confirmed the formation
of the azide compound by the appearance of new aromatic signals
(7.74-7.85 ppm, m, 3H, ArH). This coincided with the complete
disappearance of the aromatic signals of the amine (7 ppm, d, 1H,
J ) 8.3 Hz; 7.4 ppm, s, 1H; 7.74 ppm, d, 1H, J ) 8.3 Hz), thus
indicating full conversion to the azide form. On the other hand,
the aliphatic moiety was not affected by the reaction (section S3).
DMOF-fun obtained after the one-pot synthesis was characterized
using identical techniques. Here again, the conversion to the final
compound was complete after 24 h. IR analysis revealed the
complete disappearance of the azide stretching band at 2123 cm-1
(section S3). The liquid 1H NMR spectrum of DMOF-fun illustrates
that the corresponding triazole derivative was formed as the unique
product; aromatic shifts of the grafted compound were assigned
by 1H-1H correlation spectroscopy (COSY) experiments (δ ) 7.38,
7.49, 7.93, 8.04, 8.19, 9.15 ppm), and no azido or amino compound
was detected (Figure 1). In addition, positive-mode mass spec-
trometry performed after digestion clearly showed a base peak at
m/z 310 corresponding to the functionalized linker [2-(4-phenyl-
1,2,3-triazol-1-yl)terephthalic acid]. The powder X-ray diffraction
patterns indicated that the two-step reaction proceeded without loss
of long-range order (section S3). We stress that the crystallinity of
the parent DMOF-NH2 is strongly affected by post-treatments such
as solvent removal and/or solvent exchange (section S2).
byproducts such as water, acids, or bases that could damage the
structure by hydrolysis.22 Finally, in contrast to the anhydride
condensation method, which has
a limited grafting yield
(30-50%),23 we have shown that this approach allows complete
functionalization even for a bulky group. This is in line with
molecular modeling results showing weak steric demand (section
S3). Notably, the crystallite sizes of the two MOFs are ∼1 µm
(section S1), whereas Wang et al.11 used DMOF-NH2 crystallites
with diameters of 100 µm. We believe that the accessibility of
reactants to the centers of the crystals arises from the very small
size of the MOF crystallites.
However, this small size comes at the expense of a decrease in
microporous volume. Thanks to the efficiency of the azide
formation, the grafting rate can be controlled by adding pheny-
lacetylene in default with respect to -NH2 functions. For a grafting
rate of 50% on MIL-68(In)-NH2, the surface area decreased only
by 55% (SBET ) 571 m2/g), which is in line with other methods.8
In this study, controlled functionalization was performed with
phenylacetylene as a proof of concept. A systematic study dealing
with the grafting of moieties exhibiting different functions on
diverse amino-MOF platforms will be reported soon. We believe
that this method will allow the design of tailor-made catalysts with
more complex functional groups.
Acknowledgment. We thank A. Camarata for the technical
work, along with IRCELYON and IFP scientific services.
Nitrogen physisorption experiments performed on the parent
DMOF-NH2 and DMOF-fun at 77 K revealed a decreases in the
porous surface area (1320 to 244 m2/g) and the microporous volume
(0.54 to 0.08 cm3/g) due to pore blocking (section S3).
Supporting Information Available: Synthetic procedures, charac-
terization data, and molecular modeling results. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Fe´rey, G. Chem. Soc. ReV. 2008, 37, 191. (b) Farrusseng, D.; Aguado,
S.; Pinel, C. Angew. Chem., Int. Ed. 2009, 48, 7502.
(2) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.; O’Keefe, M.;
Yaghi, O. M. Science 2002, 295, 469.
(3) Eddaoudi, M.; Kim, J.; O’Keeffe, M.; Yaghi, O. M. J. Am. Chem. Soc.
2002, 124, 376.
(4) Bauer, S.; Serre, C.; Devic, T.; Horcajada, P.; Marrot, J.; Fe´rey, G.; Stock,
N. Inorg. Chem. 2008, 47, 7568.
(5) (a) Yang, C.; Wang, X. P.; Omary, M. A. J. Am. Chem. Soc. 2007, 129,
15454. (b) Kawano, M.; Kawamichi, T.; Haneda, T.; Kojima, T.; Fujita,
M. J. Am. Chem. Soc. 2007, 129, 15418.
(6) Fischer, R. A.; Woll, C. Angew. Chem., Int. Ed. 2008, 47, 8164.
(7) Wang, Z. Q.; Cohen, S. M. Chem. Soc. ReV. 2009, 38, 1315.
(8) Wang, Z. Q.; Cohen, S. M. Angew. Chem., Int. Ed. 2008, 47, 4699.
(9) Wright, A. P.; Davis, M. E. Chem. ReV. 2002, 102, 3589.
(10) (a) Brunel, D. Microporous Mesoporous Mater. 1999, 27, 329. (b) Brunel,
D.; Blanc, A. C.; Galarneau, A.; Fajula, F. Catal. Today 2002, 73, 139.
(11) Wang, Z. Q.; Tanabe, K. K.; Cohen, S. M. Inorg. Chem. 2009, 48, 296.
(12) Goto, Y.; Sato, H.; Shinkai, S.; Sada, K. J. Am. Chem. Soc. 2008, 130,
14354.
(13) (a) Gadzikwa, T.; Lu, G.; Stern, C. L.; Wilson, S. R.; Hupp, J. T.; Nguyen,
S. T. Chem. Commun. 2008, 5493. (b) Gadzikwa, T.; Farha, O. K.;
Malliakas, C. D.; Kanatzidis, M. G.; Hupp, J. T.; Nguyen, S. T. J. Am.
Chem. Soc. 2009, 131, 13613.
(14) Savonnet, M.; Bazer-Bachi, D.; Pinel, C.; Lecocq, V.; Bats, N.; Farrusseng,
D. FR Patent 09/05.107, 2009.
(15) Savonnet, M.; Bazer-Bachi, D.; Pinel, C.; Lecocq, V.; Bats, N.; Farrusseng,
D. 2009, FR Patent 09/05.101, 2009.
(16) (a) Kitagawa, S.; Tanaka, D. Chem. Mater. 2008, 20, 922. (b) Seki, K.
Langmuir 2002, 18, 2441. (c) Kitagawa, S.; Kitaura, R.; Noro, S. Angew.
Chem., Int. Ed. 2004, 43, 2334.
(17) Yaghi, O. M.; O’Keeffe, M.; Rosi, N. L.; Kim, J.; Eddaoudi, M.; Chen,
B. L. J. Am. Chem. Soc. 2005, 127, 1504.
(18) Barral, K.; Moorhouse, A. D.; Moses, J. E. Org. Lett. 2007, 9, 1809.
(19) Savonnet, M.; Bazer-Bachi, D.; Pinel, C.; Lecocq, V.; Bats, N.; Farrusseng,
D. FR Patent 09/04.521, 2009.
(20) Savonnet, M.; Bazer-Bachi, D.; Pinel, C.; Lecocq, V.; Bats, N.; Farrusseng,
D. FR Patent 09/05.102, 2009.
(21) Ahnfeldt, T.; Guillou, N.; Gunzelmann, D.; Margiolaki, I.; Loiseau, T.;
Fe´rey, G.; Senker, J.; Stock, N. Angew. Chem., Int. Ed. 2009, 48, 5163.
(22) Savonnet, M.; Aguado, S.; Ravon, U.; Bazer-Bachi, D.; Lecocq, V.; Bats,
N.; Pinel, C.; Farrusseng, D. Green Chem. 2009, 11, 1729.
(23) Tanabe, K. K.; Cohen, S. M. Angew. Chem., Int. Ed. 2009, 48, 7424.
Figure 1. Liquid 1H NMR spectra of digested materials: (top) DMOF-
NH2; (middle) DMOF-N3; (bottom) DMOF-fun.
These investigations were also undertaken for the postfunction-
alization of MIL-68(In)-NH2, and the same results were obtained
in terms of grafting rate (>90%), decrease in surface area (1260 to
120 m2/g) and microporous volume (0.48 to 0.03 cm3/g), and
preservation of crystallinity (section S6).
Cycloaddition in azide-functionalized MOFs has recently been
demonstrated for IRMOF-type compounds.12 In this work, we have
shown that direct functionalization from amino-derived MOFs is
possible, provided that the pore cavity is large enough to accom-
modate a C5 ring (section S3). The main advantage of this one-pot
method is the ease of preparing amino-functionalized MOFs. To
the best of our knowledge, five different structures based on
2-aminoterephthalic acid have been reported to date: IRMOF-3,2
MIL-101-NH2,4 CAU-1,21 DMOF-NH2, and MIL-53(Al)-NH2. The
large library of amino acids also opens promising perspectives for
new potential amino-functionalized MOFs as synthetic platforms.
A second advantage is the softness of the method. Indeed, both
reaction steps proceed at room temperature and do not liberate
JA909613E
9
J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010 4519