A general thermolabile protecting group strategy for organocatalytic metal-organic frameworks

Article abstract of DOI:10.1021/ja202223d

We present a general strategy for incorporating organocatalytic moieties into metal-organic frameworks (MOFs). The organocatalytic units are protected by a thermolabile protecting group during MOF synthesis and then unveiled by a simple postsynthetic heat

Full text of DOI:10.1021/ja202223d

COMMUNICATION
pubs.acs.org/JACS
A General Thermolabile Protecting Group Strategy for
Organocatalytic MetalꢀOrganic Frameworks
†
‡
,†
David J. Lun, Geoffrey I. N. Waterhouse, and Shane G. Telfer*
†
MacDiarmid Institute for Advanced Materials and Nanotechnology, Institute of Fundamental Sciences, Massey University, Palmerston
North, New Zealand
‡
Department of Chemistry, University of Auckland, Auckland, New Zealand
S Supporting Information
b
they can be chemically transformed under the conditions of
MOF synthesis, for example by (de)protonation or coordination
to metal ions. In addition to inhibiting catalysis, these changes
may also prevent the growth of MOF crystals. We herein present
a strategy designed to mitigate these undesired processes and
provide a general method for the incorporation of organocata-
lytic groups into MOFs. This strategy builds on our recent
demonstration of a new type of postsynthetic modification of
MOFs involving the transformation of a bulky tert-butoxycarbonyl
(Boc) group appended to a ligand skeleton into small, gaseous
ABSTRACT: We present a general strategy for incorporat-
ing organocatalytic moieties into metalꢀorganic frame-
works (MOFs). The organocatalytic units are protected
by a thermolabile protecting group during MOF synthesis
and then unveiled by a simple postsynthetic heating step.
The strategy is exemplified using a thermolabile tert-butox-
ycarbonyl (Boc) protecting group for a proline moiety,
the removal of which endows the resulting cubic zinc(II)
IRMOF with catalytic activity for asymmetric aldol reac-
tions. The bulky Boc groups also prevent framework inter-
penetration, producing open MOFs that can admit relatively
large substrates.
12
fragments by simple heating. The Boc unit suppresses framework
catenation (interpenetration) during the assembly of the MOF,
and its expulsion both increases the pore volume of the framework
and reveals an amino functional group.
Scheme 1 outlines how we envisaged the application of this
methodology to the surreptitious incorporation into MOFs of
proline groups designed to be specific sites for catalysis. Building
ecause of their porous and tunable nature, metalꢀorganic
0
on the fact that biphenyl-4,4 -dicarboxylic acid ligands, including
1
B
frameworks (MOFs) show promise as heterogeneous catalysts.
0
those substituted at the 2 and/or 2 positions, react with zinc(II)
A variety of structural features can engender catalytic activity in
12,13
to form MOFs with cubic topologies,
we designed H 1 as a
2
2
MOFs, such as open metal sites at framework nodes, which
building block for IRMOF-Pro-Boc. The designation IRMOF
14
often have unique coordination environments, and metal-
denotes the “isoreticular” nature of this family of cubic frame-
works. The proline unit of ligand 1 is protected by a Boc group,
which prevents its protonation or coordination to zinc(II) under
the conditions of MOF synthesis. Further, the bulky nature of the
Boc unit should discourage framework interpretation, resulting
in the production of an open MOF. In a subsequent step,
3,4
centered catalytic units that are integrated into the ligand struts.
We are interested in the development of a third mode of
catalytic activity, organocatalytic MOFs, where organic functional
groups appended to the framework provide specific sites for
catalysis. This approach is motivated by the observation that the
catalytic activity of many organocatalysts arises from small,
discrete moieties such as pyrrolidine, urea, or phosphonic acid
1
5
deprotection of the proline can be triggered by simple heating
to generate IRMOF-Pro in a single-crystal-to-single-crystal
(SCSC) fashion. Importantly, this approach ensures that (i)
IRMOF-Pro exhibits an open structure with contiguous channels
that can admit reasonably large substrates and (ii) there is
significant void space in the vicinity of the proline motifs, thus
freeing them to interact with incoming molecules.
5
groups. It has been demonstrated that these functional groups
can be immobilized on various supports (e.g., polymers, den-
drimers, or silica) to produce recoverable and recyclable
6
catalysts. In the case of MOFs, there have been surprisingly
7
few deliberate attempts to integrate organocatalytic moieties.
Recently, postsynthetic modification reactions have been em-
ployed to introduce carboxylic acid groups into MOFs, which
mimic known Brønsted acid catalysts for the methanolysis of
The synthesis of (S)-H 1 was achieved in two steps via the
reaction of N-Boc-L-proline and dimethyl 2-aminobiphenyl-4,
4 -dicarboxylate. Chiral HPLC established that the amide coupling
reaction resulted in only a minor degree of racemization, with the
2
0
8
epoxides. To exploit the well-recognized catalytic activity of
enantiomeric excess (ee) of (S)-H 1 found to be 94%. In accord
pyrrolidines, proline- and pyrrolidine-based ligands have been
2
12,13
9
with related literature results,
the solvothermal reaction of (S)-
incorporated into MOFs. A method of generating free car-
H 1 with Zn(NO ) in N,N-diethylformamide (DEF) produced
boxylic acid groups in aspartate-based MOFs has also been
2
3 2
1
0
well-faceted, colorless, cubic crystals of IRMOF-Pro-Boc
Scheme 1). Analysis by X-ray crystallography demonstrated that
reported. Frameworks derived from proline itself were also
investigated, but organocatalytic activity was precluded by pro-
(
11
tonation of the proline nitrogen atom.
As suggested by this last case, one reason for the paucity of
examples of organocatalytic units in MOFs may be the fact that
Received: March 10, 2011
Published: March 28, 2011
r 2011 American Chemical Society
5806
dx.doi.org/10.1021/ja202223d
|J. Am. Chem. Soc. 2011, 133, 5806–5809

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Conversion of Ligand H 1 to a Cubic Metalꢀ
2
Organic Framework, IRMOF-Pro-Boc ([Zn O(1) ]), Fol-
4
3
lowed by Thermolytic Expulsion of the Boc Moiety To
Generate IRMOF-Pro ([Zn O(2) ]); Isobutylene and CO
2
4
3
Are Produced as Side Products in This Step
Figure 2. Thermogravimetrograms of IRMOF-Pro-Boc under N . The
2
dashed curve was recorded after the sample was preheated to 250 °C.
The heating rate was 5 °C/min.
pattern closely matched that predicted from the single-crystal
structure (Figure S6 in the Supporting Information). To deter-
mine the temperature at which the Boc groups could be expelled
from IRMOF-Pro-Boc to generate IRMOF-Pro, we carried out
thermogravimetric analysis (TGA) of this material. Significant
weight loss was observed in the region 140ꢀ240 °C (Figure 2,
solid curve) as a consequence of the thermolytic reaction shown
in Scheme 1. Consistent with this analysis, when a sample was
pretreated at 250 °C, this weight loss was not observed (dashed
curve). Framework decomposition occurred above 350 °C.
With these results in hand, IRMOF-Pro-Boc was converted
to IRMOF-Pro in a SCSC reaction by heating to 165 °C in N,
N-dimethylformamide (DMF) usingmicrowave irradiation. Com-
plete expulsion of the Boc group was evidenced within a period of
1
4
h by H NMR spectroscopy of the MOF crystals digested in
dimethyl sulfoxide-d /DCl (Figure S4). Following thermolysis,
6
the enantiopurity of the proline groups of ligand (S)-2 in IRMOF-
Pro was ascertained by chiral HPLC to be 80% ee. It thus appears
that the thermolysis process results in the racemization of ∼7% of
the total number of proline groups. Lowering the thermolysis
temperature to 150 °C led to only a marginal improvement in the ee.
IRMOF-Pro was further characterized by microanalysis, IR
spectroscopy (see the Supporting Information), and XRD.
Powder XRD demonstrated that the framework maintained
crystallinity following the thermolysis reaction and that the cubic
lattice was retained (Figure S6). This was corroborated by single-
crystal XRD experiments, from which a lattice constant of
Figure 1. Structure of IRMOF-Pro-Boc as determined by X-ray crystal-
lography. (a) Ligand strut and Zn O nodes. Gray = C; red = O; blue = N;
turquoise = Zn. Hydrogen atoms have been omitted for clarity. (b)
Typical cubic repeating unit. One set of the statistically disordered N-
Boc-proline side chains is shown.
4
1
7.160(3) Å was deduced. A structural model could be satisfac-
the structure closely resembles other members of the IRMOF
torily refined against these data in the space group P43m
series; the stoichiometry is [Zn O((S)-1) ], and a framework
(Figure 3). The statistically disordered proline side chains were
4
3
1
5
with cubic topology is built up by linking of Zn O nodes by the
placed in calculated positions to complete the refinement.
4
0
divergent carboxyl groups of ligand 1 (Figure 1). Because of the
steric bulk of the N-Boc-proline side arm, the MOF adopts an
open, noninterpenetrated structure.
In previous work, we showed that 2-aminobiphenyl-4,4 -
dicarboxylic acid does not produce a MOF under standard synthesis
conditions, whereas MOF formation proceeds smoothly when
12
Strong diffraction from the rigid, highly ordered cubic lattice
the amino moiety is protected by a Boc group. In this light, we
comprising the Zn O nodes and the biphenyl ligand skeletons
attempted to prepare IRMOF-Pro directly from H 2, but all
4
2
generated a reciprocal lattice that was most consistent with the
cubic space group P43m [lattice constant = 17.1625(19) Å]. The
statistical disordering of the N-Boc-proline groups over four sites
mitigated against their identification in the Fourier difference
map, so they were placed in calculated positions to complete the
attempts met with failure. We suspect that the exposed proline
group interferes with MOF formation, perhaps as a consequence
of the coordination of its nitrogen atom to zinc(II). This experi-
ment also confirmed that the thermolysis of IRMOF-Pro-Boc to
IRMOF-Pro takes place via an SCSC pathway, as ligand 2 cannot
form IRMOF-Pro from solution.
1
6
refinement.
The phase purity of IRMOF-Pro-Boc was established by
powder X-ray diffraction (XRD), where the observed diffraction
The X-ray crystal structure of IRMOF-Pro established that this
material is amenable to heterogeneous catalysis, as the proline units
5
807
dx.doi.org/10.1021/ja202223d |J. Am. Chem. Soc. 2011, 133, 5806–5809

Journal of the American Chemical Society
COMMUNICATION
of the solid catalyst by filtration partway through the reaction
completely halted the consumption of the aldehyde. Second,
IRMOF-Pro-Boc was completely inactive toward this conver-
sion, demonstrating that the catalytic activity arises from the
presence of unprotected pyrrolidine groups. Third, there was no
evidence of dissolution of the crystals of IRMOF-Pro, and we
found that both aldol reactions were catalyzed only very slowly
by Zn(NO ) , even at high loadings. Although H 2 also func-
3
2
2
tioned as a catalyst for these reactions, different stereoselectivity
patterns in comparison with IRMOF-Pro were observed (see the
Supporting Information).
To test the recyclability of the IRMOF-Pro catalyst, we
subjected it to three cycles of the aldol reaction with cyclopenta-
none. The reaction was driven to completion in each cycle,
although diminished activity of the catalyst was evident from an
increase in the required reaction time, which was correlated with
a slight loss of long-range crystallinity (Figure S9).
Figure 3. Structure of IRMOF-Pro as determined by X-ray crystal-
lography. (a) Ligand strut and Zn O nodes. Gray = C; red = O; blue = N;
4
turquoise = Zn. Most of the hydrogen atoms have been omitted for
clarity. (b) Typical cubic repeating unit. One set of the statistically
disordered proline side chains is shown.
Although the initial focus of this work was to develop
thermolabile protecting groups for organocatalytic moieties
rather than to discover a highly enantioselective catalyst, we
did ponder why the enantioselectivities observed for the aldol
reactions catalyzed by IRMOF-Pro were modest. In addition to
the minor degree of racemization of the proline groups, this may
be ascribed to a lack of organization in the reaction transition
state as a consequence of several factors, including motion of the
catalytic unit with respect to the framework, the large intraframe-
work void space that allows the electrophile to approach either
side of the intermediate enamine, and the absence of accessible
Scheme 2. Aldol Reactions of Acetone and Cyclopentanone
with 4-Nitrobenzaldehyde Catalyzed by IRMOF-Pro
22
hydrogen-bond donors on the catalytic unit. Highly enantio-
selective small-molecule organocatalysts often allow for partici-
pation of both the proline nitrogen atom and its carboxyl group
in the catalytic cycle. Our current efforts to produce second-
generation catalysts that exhibit improved enantioselectivities are
focused on thermolabile groups that enable this to occur within
MOFs. Different MOF topologies are also being explored.
In summary, we have developed a general methodology for the
introduction of organocatalytic groups into MOFs that exhibits
several advantageous features: (1) it encompasses catalytically
active groups that would otherwise disrupt MOF formation or be
chemically altered under MOF synthesis conditions; (2) it
ensures that there is vacant space around the catalytic unit,
allowing it to interact with incoming substrates; and (3) bulky
protective groups can preclude framework interpenetration, pro-
ducing open networks. We expect that this strategy can be applied
to virtually any MOF and a substantial number of organocatalytic
line a continuous network of pores. The diameter of the windows
between these pores ranges from ∼5 to 10.5 Å (because of the
variable location of the proline units). Free diffusion of large
molecules into these pores was established experimentally using
4
,17
a merocyanine dye, which showed uniform penetration into
IRMOF-Pro crystals (Figure S7). Following activation by super-
1
8
critical CO2, N gas sorption experiments gave a Brunauerꢀ
2
2
ꢀ1
EmmettꢀTeller (BET) surface area of 138 m g (Figure S8),
which is lower than predicted on the basis of X-ray crystal-
lography and indicates a degree of pore collapse upon desolva-
tion. For optimal performance in catalysis, the above results led
to the practice of keeping the MOF crystals suspended in solvent
at all times.
To examine the catalytic ability of IRMOF-Pro, we employed
1
9
15
aldol reactions that are known to be catalyzed by proline
functional groups. A vast library of catalysts is thus potentially
20
21
and both unsupported and supported proline-based catalysts.
Crystals of IRMOF-Pro were suspended in acetone, and
accessible, and structureꢀactivity relationships may increase the
performance of these materials to the point where they are viable
and attractive catalysts for applications in synthetic chemistry.
4
-nitrobenzaldehyde was added. To our delight, the formation
of the aldol product of these reactants, 4-hydroxy-4-(4-nitrophe-
nyl)pentan-2-one (Scheme 2), was evident over the course of a
few hours, and the aldehyde was completely consumed within 24
h (Table S1 in the Supporting Information). The ee of the aldol
product was determined to be 29% by chiral HPLC analysis. We
also found that the aldol reaction of cyclopentanone and
’
ASSOCIATED CONTENT
S
Supporting Information. Full details of ligand and MOF
b
synthesis and characterization, crystallography (including a CIF),
dye uptake and gas sorption experiments, and catalysis reactions.
This material is available free of charge via the Internet at http://
pubs.acs.org.
4
-nitrobenzaldehyde to give hydroxy(4-nitrophenyl)methylcyclo-
pentanone was also catalyzed by IRMOF-Pro. The diastereo-
meric ratio (dr) of the anti to the syn products was determined to
be 75:25, with ee's of 14% for the anti isomer and 3% for the syn
isomer.
’
AUTHOR INFORMATION
A series of control experiments confirmed that intact IRMOF-Pro
was responsible for the observed catalytic activity. First, removal
Corresponding Author
s.telfer@massey.ac.nz
5
808
dx.doi.org/10.1021/ja202223d |J. Am. Chem. Soc. 2011, 133, 5806–5809

Journal of the American Chemical Society
COMMUNICATION
’
ACKNOWLEDGMENT
(20) Cobb, A. J. A.; Shaw, D. M.; Longbottom, D. A.; Gold, J. B.; Ley,
S. V. Org. Biol. Chem. 2005, 3, 84. Mase, N.; Nakai, Y.; Ohara, N.; Yoda,
H.; Takabe, K.; Tanaka, F.; Barbas, C. F., III. J. Am. Chem. Soc. 2006,
We thank the MacDiarmid Institute and the RSNZ Marsden
Fund for funding this work. Professor Geoffrey Jameson is
gratefully acknowledged for sage crystallographic advice, Dr.
Andrew Kay (Industrial Research Ltd.) for supplying a sample
of merocyanine dye, and Rajesh Deshpande for valuable techni-
cal assistance.
128, 734.
(21) Font, D.; Jimeno, C.; Pericas, M. A. Org. Lett. 2006, 8, 4653.
Gruttadauria, M.; Giacalone, F.; Marculescu, A. M.; Riela, S.; Noto, R.
Eur. J. Org. Chem. 2007, 4688.
(22) Mukherjee, S.; Yang, J. W.; Hoffmann, S.; List, B. Chem. Rev.
2
007, 107, 5471.
’
REFERENCES
(
1) Corma, A.; Garcia, H.; Llabr ꢀe s i Xamena, F. X. Chem. Rev. 2010,
10, 4606. Liu, Y.; Xuan, W.; Cui, Y. Adv. Mater. 2010, 22, 4112.
Farrusseng, D.; Aguado, S.; Pinel, C. Angew. Chem., Int. Ed. 2009,
8, 752. Lee, J.; Farha, O. K.; Roberts, J.; Scheidt, K. A.; Nguyen,
1
4
S. T.; Hupp, J. T. Chem. Soc. Rev. 2009, 38, 1450. Ma, L.; Abney, C.; Lin,
W. Chem. Soc. Rev. 2009, 38, 1248. Ma, L.; Lin, W. Top. Curr. Chem.
2
2
010, 293, 175.
2) Horike, S.; Dinc ꢁa , M.; Tamaki, K.; Long, J. R. J. Am. Chem. Soc.
008, 130, 5854. Zou, R.-Q.; Sakurai, H.; Han, S.; Zhong, R.-Q.; Xu, Q.
(
J. Am. Chem. Soc. 2007, 129, 8402. Corma, A.; Iglesias, M.; Llabr ꢀe s i
Xamena, F. X.; Sanchez, F. Chem.—Eur. J. 2010, 16, 9789.
(3) Cho, S.-H.; Ma, B.; Nguyen, S. T.; Hupp, J. T.; Albrecht-Schmitt,
T. E. Chem. Commun. 2006, 2563. Shultz, A. M.; Farha, O. K.; Hupp,
J. T.; Nguyen, S. T. J. Am. Chem. Soc. 2008, 131, 4204. Wu, C.-D.; Lin, W.
Angew. Chem., Int. Ed. 2007, 46, 1075. Tanabe, K. K.; Cohen, S. M. Inorg.
Chem. 2010, 49, 6766.
(
4) Song, F.; Wang, C.; Falkowski, J. M.; Ma, L.; Lin, W. J. Am. Chem.
Soc. 2010, 132, 15390.
(
(
5) MacMillan, D. W. C. Nature 2008, 455, 304.
6) Gruttadauria, M.; Giacalone, F.; Noto, R. Chem. Soc. Rev. 2008,
3
7, 1666. Handbook of Asymmetric Heterogeneous Catalysis; Wiley-VCH:
Weinheim, Germany, 2008.
7) Hasegawa, S.; Horike, S.; Matsuda, R.; Furukawa, S.; Mochizuki,
(
K.; Kinoshita, Y.; Kitagawa, S. J. Am. Chem. Soc. 2007, 129, 2607. Tanaka,
K.; Oda, S.; Shiro, M. Chem. Commun. 2008, 820. Hwang, Y. K.; Hong,
D.-Y.; Chang, J.-S.; Jhung, S. H.; Seo, Y.-K.; Kim, J.; Vimont, A.; Daturi,
M.; Serre, C.; F ꢀe rey, G. Angew. Chem., Int. Ed. 2008, 47, 4144. Wang, M.;
Xie, M.-H.; Wu, C.-D.; Wang, Y.-G. Chem. Commun. 2009, 2396.
(
8) Garibay, S. J.; Wang, Z.; Cohen, S. M. Inorg. Chem. 2010,
9, 8086.
9) Banerjee, M.; Das, S.; Yoon, M.; Choi, H. J.; Hyun, M. H.; Park,
4
(
S. M.; Seo, G.; Kim, K. J. Am. Chem. Soc. 2009, 131, 7524. Dang, D.; Wu,
P.; He, C.; Xie, Z.; Duan, C. J. Am. Chem. Soc. 2010, 132, 14321.
(
10) Ingleson, M. J.; Barrio, J. P.; Bacsa, J.; Dickinson, C.; Park, H.;
Rosseinsky, M. J. Chem. Commun. 2008, 1287.
11) Ingleson, M. J.; Bacsa, J.; Rosseinsky, M. J. Chem. Commun.
007, 3036.
12) Deshpande, R. K.; Minnaar, J. L.; Telfer, S. G. Angew. Chem., Int.
Ed. 2010, 47, 4598.
13) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.;
(
2
(
(
O’Keeffe, M.; Yaghi, O. M. Science 2002, 295, 469. Burrows, A. D.;
Frost, C. G.; Mahon, M. F.; Richardson, C. Chem. Commun. 2009, 4218.
Burrows, A. D.; Frost, C.; Mahon, M. F.; Richardson, C. Angew. Chem.,
Int. Ed. 2008, 47, 8482.
(
14) Yaghi, O. M.; O’Keeffe, M.; Ockwig, N. W.; Chae, H. K.;
Eddaoudi, M.; Kim, J. Nature 2003, 423, 705.
15) Wuts, P. G. M.; Greene, T. W. Greene’s Protective Groups in
Organic Synthesis, 4th ed.; Wiley: Hoboken, NJ, 2007.
16) Because of this lack of long-range order of the chiral side arms,
the structure conforms to a nonenantiomorphic space group.
17) Gedrich, K.; Heitbaum, M.; Notzon, A.; Senkovska, I.; Frohlich,
R.; Getzschmann, J.; Mueller, U.; Glorius, F.; Kaskel, S. Chem.—Eur. J.
011, 17, 2099.
18) Nelson, A. P.; Farha, O. K.; Mulfort, K. L.; Hupp, J. T. J. Am.
Chem. Soc. 2009, 131, 458.
19) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000,
22, 2395.
(
(
(
2
(
(
1
5
809
dx.doi.org/10.1021/ja202223d |J. Am. Chem. Soc. 2011, 133, 5806–5809