Catalytic covalent organic frameworks via pore surface engineering

Article abstract of DOI:10.1039/c3cc48813f

We report a synthetic strategy for construction of the first example of organocatalytic covalent organic frameworks via pore surface engineering. The COF catalyst combines a number of striking features, including enhanced activity, broad applicability, go

Full text of DOI:10.1039/c3cc48813f

Featuring research from Donglin Jiang’s
laboratory / Department of Materials Molecular Science,
Institute for Molecular Science, Okazaki, Japan
As featured in:
Catalytic covalent organic frameworks via pore surface
engineering
We report a synthetic strategy for construction of the first
example of organocatalytic covalent organic frameworks via pore
surface engineering. The COF catalyst combines a number of
striking features, including enhanced activity, broad applicability,
good recyclability, and high capability to perform catalytic
transformation under continuous flow.
See Donglin Jiang et al.,
Chem. Commun., 2014, 50, 1292.
www.rsc.org/chemcomm
Registered charity number: 207890

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Catalytic covalent organic frameworks via pore
surface engineering†
Cite this: Chem. Commun., 2014,
50, 1292
Hong Xu,a Xiong Chen,a Jia Gao,a Jianbin Lin,a Matthew Addicoat,b Stephan Irleb
and Donglin Jiang*a
Received 19th November 2013,
Accepted 29th November 2013
DOI: 10.1039/c3cc48813f
www.rsc.org/chemcomm
We report a synthetic strategy for construction of the first example cross-coupling reaction.10 The construction of a covalently linked, yet
of organocatalytic covalent organic frameworks via pore surface highly active catalyst remains a synthetic challenge in the field.11
engineering. The COF catalyst combines a number of striking
Here we report the pore surface engineering strategy for the
features, including enhanced activity, broad applicability, good construction of covalently linked and highly active organocatalytic
recyclability, and high capability, to perform catalytic transforma- COFs (Scheme 1). Pore surface engineering has been demonstrated
tion under continuous flow conditions.
using boronate-linked COFs based on alkyne–azide5 and thiol–ene7
reactions. Recently, we explored a novel strategy for engineering the
Covalent organic frameworks (COFs) are a class of crystalline porous pore surface of stable imine-linked COFs and developed the first
polymers that enable the atomically precise integration of building example of a covalently linked COF catalyst. We demonstrate this
blocks into two- or three-dimensional (2D or 3D, respectively) strategy by highlighting the controlled integration of organocatalytic
periodicities.1 In 2D COFs, the building blocks for the vertices and sites into the pore walls to synthesize organocatalytic COFs that
edges are covalently linked to form extended 2D polygon sheets that exhibit enhanced activity in asymmetric Michael addition reactions
stack to constitute layered frameworks, giving rise to two structural while retaining stereoselectivity. This COF catalyst combines a
characters, i.e., periodic p arrays and ordered one-dimensional (1D) number of striking features, including broad applicability, good
channels. The size and shape of 1D pores can be tailored and they recyclability, and high capability to perform catalytic transformation
are useful for exploring gas adsorption and guest encapsulation.1–10 under continuous flow conditions.
If catalytic sites can be successfully integrated into the frameworks,
We utilized a mesoporous imine-linked porphyrin COF as a
the highly ordered skeletal alignment and open-channel structure of scaffold in which the porphyrin units are located at the vertices
2D COFs provide an intriguing motif for exploring well-defined and the phenyl groups occupy the edges of tetragonal polygon
nanoreactors. Two strategies can be used for constructing catalytic frameworks.6 A three-component reaction system consisting of
COFs. Incorporating building blocks that possess catalytic sites con- 5,10,15,20-tetrakis(40-tetraphenylamino) porphyrin as the vertices
stitutes the direct method but requires a tedious solvothermal synth- and a mixture of 2,5-bis(2-propynyloxy)terephthalaldehyde (BPTA)
esis. Particularly, if the catalytic site is bulky, forming a crystalline COF and 2,5-dihydroxytere-phthalaldehyde (DHTA) at varying molar ratios
structure becomes difficult. Another methodology involves the post- (X = [BPTA]/([BPTA] + [DHTA]) ꢀ 100 = 0, 25, 50, 75, 100) as the edge
synthetic integration of catalytic sites into a crystalline COF skeleton. units was developed to synthesize the COFs, by allowing the integra-
This approach can reduce the influence of the bulky catalytic sites on tion of ethynyl units of varying content into the edges (Scheme 1,
the COF crystallinity and the undesired effect of harsh solvothermal [HCRC]X-H2P-COFs, X = 25, 50, 75, and 100 (X = 0: H2P-COF)). These
conditions on the catalytic sites. The COF-based catalyst is limited to reactions exhibited similar isolated yields to each other, indicating
only one example that is based on a Pd-loaded COF for the Suzuki that the reactivity of BPTA and DHTA is similar under the solvo-
thermal conditions (Table S1, ESI†). The click reaction of the ethynyl
units with the azide compounds takes place quantitatively and
a Department of Materials Molecular Science, Institute for Molecular Science,
National Institutes of Natural Sciences, 5-1 Higashiyama, Myodaiji,
Okazaki 444-8787, Japan. E-mail: jiang@ims.ac.jp
b WPI-Research Initiative-Institute of Transformative Bio-Molecules and Department
of Chemistry, Graduate School of Science, Nagoya University, Furo-cho,
Chikusa-ku, Nagoya 464-8602, Japan
anchors the desired groups to the pore walls at the desirable contents
(Scheme 1, [Pyr]X-H2P-COFs). Therefore, this engineering methodology
features catalytic sites on the pore walls whose density can be system-
atically designed and synthetically controlled (Scheme 1B).
Infrared (IR) spectroscopy provides direct evidence for the
presence of ethynyl units in [HCRC]X-H2P-COFs with characteristic
† Electronic supplementary information (ESI) available: Synthetic methods, IR and gas
sorption analysis, computational results, and cycle test. See DOI: 10.1039/c3cc48813f
CRC and H–CRC vibration bands at 2120 and 3290 cmꢁ1
,
1292 | Chem. Commun., 2014, 50, 1292--1294
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Scheme 1 (A) The general strategy for the pore surface engineering of imine-linked COFs via a condensation reaction and click chemistries (the case for
X = 50 was exemplified). (B) A graphical representation of [Pyr]X-H2P-COF with different densities of catalytic sites on the pore walls (gray: carbon, red:
nitrogen, green: oxygen, purple: carbon atoms of the pyrrolidine unit; hydrogen is omitted for clarity).
respectively, which were absent in the spectrum for H2P-COF (Fig. S1, coordinated COFs.10 The presence of ethynyl groups does not create
ESI†).5 The intensities of these two bands increased with increasing new XRD peaks. Because the [HCRC]X-H2P-COFs exhibit similar
ethynyl content, indicating the successful integration of ethynyl units XRD patterns as H2P-COF, [HCRC]X-H2P-COFs possess the same
at different contents onto the pore walls of [HCRC]X-H2P-COFs. crystalline structure as the H2P-COF.6
Elemental analysis reveals that the content of ethynyl units is close to
the theoretical value (Table S2, ESI†).
As more pyrrolidine units were integrated into the pore walls, the
XRD intensity decreased (Fig. 1B); a similar phenomenon was
Pyrrolidine derivatives are well-known organocatalysts for Michael observed for [HCRC]X-H2P-COFs (Fig. 1A). Similarly, the XRD peak
addition reactions, while the derivatives with bulky substituents give positions remained unaffected, indicating that the crystalline
high stereoselectivity.12,13 In this study, we utilized the simplest skeleton was retained. For example, both [Pyr]50-H2P-COF (Fig. 1B,
pyrrolidine unit for demonstrating the effectiveness of a click reaction blue) and H2P-COF (Fig. 1A, black) exhibited peaks at 3.51, 4.91, 7.01,
in synthesizing heterogeneous organocatalysts with active sites on the and 231, which are assignable to 100, 110, 200, and 001 facets,
1D channel walls; the simplest pyrrolidine unit has a relatively low respectively. The Pawley refined XRD pattern (Fig. 1B, orange)
steroselectivity. The click reaction of [HCRC]X-H2P-COFs with pyrro- reproduced the experimental curve (blue), confirming the aforemen-
lidine azide in the presence of a CuI catalyst in toluene–tert-butanol at tioned peak assignment, as is evident from their negligible difference
25 1C quantitatively yielded the corresponding [Pyr]X-H2P-COFs (black). The density-functional tight-binding method including
(Scheme 1A, and ESI†).5 After the click reaction, the IR spectra Lennard-Jones dispersion was employed to determine the optimal
demonstrated the disappearance of the vibration bands at 2120 stacking isomer structures and revealed that these COFs adopt a
and 3290 cmꢁ1 that were attributed to the H–CRC units (Fig. S1, 0.8 Å-slipped AA stacking structure (Table S3, ESI†).
ESI†), indicating that all the CRC units are clicked to bear the
As the ethynyl content increased, [HCRC]X-H2P-COFs exhibited a
catalytic sites.
decrease in the Brunauer–Emmett–Teller (BET) surface area (Fig. S2A,
The X-ray diffraction (XRD) intensity tended to decrease with ESI†). The BET surface area was 1126, 1092, 859, 324, and 206 m2 gꢁ1
increasing ethynyl content (Fig. 1A). The H2P-COF exhibited the (Table S4, ESI†) as the ethynyl content increased from 0 to 25, 50, 75,
highest XRD intensity of 1.47 ꢀ 104 counts, with 1.05 ꢀ 103 counts and 100, respectively, whereas their pore sizes decreased from 2.2 to
for the [HCRC]100-H2P-COF. This decrease in intensity is caused by 2.0, 1.9, 1.5, and 1.5 nm (Fig. S3, ESI†). After the click reactions, much
the large number of amorphous ethynyl chains on the pore walls. A bulky groups were introduced into the pore walls of the COFs. Thus,
similar trend was previously reported for boronate-linked COFs upon [Pyr]X-H2P-COFs exhibited a distinct decrease in their BET surface
pore surface engineering with amorphous units5 and Pd-imine areas (Fig. S2B, ESI†). The [Pyr]25-H2P-COF had a surface area of
960 m2
g
ꢁ1, which decreased to 675, 86, and 63 m2 gꢁ1 for
[Pyr]50-H2P-COF, [Pyr]75-H2P-COF, and [Pyr]100-H2P-COF, respectively
(Table S4, ESI†). Their pore sizes decreased from 1.9 to 1.6, 1.4, and
1.4 nm, respectively. The pore size distribution profile revealed that
only one pore type existed (Fig. S3, ESI†), indicating that the catalytic
units are homogeneously engineered onto the walls. The similar pore
sizes of COFs with 75% and 100% engineered surfaces are likely
related to the slipped AA structure.
We investigated the catalytic activities of [Pyr]X-H2P-COFs in a
Michael addition reaction in aqueous solutions, using (S)-4-(phenoxy-
methyl)-1-(pyrrolidin-2-ylmethyl)-1H-1H-1,2,3-triazole as
a control
(ESI†), which has an active structure identical to the pyrrolidine
catalytic site of [Pyr]X-H2P-COFs. The pyrrolidine control yields a
homogeneous system that required 3.3 h to achieve 100% conversion
Fig. 1 XRD patterns of (A) H2P-COF (black) and [HCRC]X-H2P-COFs,
and (B) [Pyr]X-H2P-COFs (orange: Pawley refined XRD curve of [Pyr]50
-
H2P-COF, black: their difference).
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 1292--1294 | 1293

View Article Online
ChemComm
Communication
Table 1 Comparison of the pyrrolidine control, amorphous nonporous
polymers, and COFs as catalysts for a Michael addition reaction
Time for 100%
conversion (h)
dr
ee (%)
Control
3.3
1
2.5
5
9
43
65
60/40
70/30
70/30
70/30
65/35
70/30
65/35
49
49
50
51
44
48
46
[Pyr]25-H2P-COF
[Pyr]50-H2P-COF
[Pyr]75-H2P-COF
[Pyr]100-H2P-COF
Amorphous polymer 1
Amorphous polymer 2
Fig. 2 Representative chart for the flow reaction system based on the
organocatalytic COF column.
to prevent COFs from flowing out) at the bottom and [Pyr]25
-
H2P-COF (10 mg, 10 mm bed height) atop the silica gel (Fig. 2).
After optimizing the parameters for the continuous-flow pro-
cess, the device was found to work well at room temperature
and yielded an optimal conversion when a solution of trans-4-
chloro-b-nitrostyrene (8.3 mM) and propionaldehyde (83 mM)
in a mixture of water–EtOH (1/1 v/v) was passed through at a
flow rate of 18 mL minꢁ1. The column maintained a 100%
conversion and its stereoselectivities (44% ee, 65/35 dr) for
more than 48 h under flow conditions.
In summary, we have developed a pore surface engineering
strategy that allows the molecular design of COF skeletons, controls
the density and composition of the functional groups on the pore
walls, and offers a general principle for designing catalytic COFs.
Engineering pyrrolidine units onto the pore walls creates aqueous
organocatalytic COFs, which combine a number of striking catalytic
features, including significantly enhanced activity, good recyclabil-
ity, and high capability to perform transformation under contin-
uous flow conditions while retaining stereoselectivity. With this
major advancement, we envisage that this work will initiate the
exploration of various catalytic systems based on COFs via the
structural engineering of both pores and skeletons.
with ee and dr values of 49% and 60/40, respectively (Table 1). The
steroselectivity of pyrrolidine derivatives is highly dependent on their
substitutions; a large substituent gives a high stereoselectivity, while
the simplest pyrrolidine unit that we employed in this study yields
moderate steroselectivity.12,13 Remarkably, the activity was signifi-
cantly enhanced when the pyrrolidine units were integrated to the
pore walls of COFs. For example, the reaction time was shortened to
only 1 h when [Pyr]25-H2P-COF was used as a heterogeneous catalyst,
yielding ee and dr values of 49% and 70/30, respectively. Similarly, the
reaction in the presence of [Pyr]50-H2P-COF required 2.5 h for 100%
conversion and resulted in 50% ee and 70/30 dr. Therefore, the
organocatalytic COFs have significantly higher catalytic activity than
the monomeric catalyst while retaining the stereoselectivity. The 1D
channels of the COFs could accommodate the reactant and substrate,
which cannot be dissolved in a water–ethanol mixture.
The catalytic activity depends upon the density of the active
sites on the pore walls. [Pyr]75-H2P-COF with 75% active sites on
the walls requires 5 h to complete the reaction. When each edge
is anchored with catalytic sites, [Pyr]100-H2P-COF requires 9 h to
reach 100% conversion. Therefore, highly dense pyrrolidine
units in the pores cause a steric congestion and impede the
mass transport through the channels.
Notes and references
The crystallinity and porosity of COFs play a vital role in
determining their catalytic activities. Amorphous and nonporous
polymers 1 and 2 (Fig. S2C, ESI†), analogues to [Pyr]25-H2P-COF and
[Pyr]50-H2P-COF, exhibited sluggish reactions, requiring 43 and 65 h
to complete, respectively. The amorphous and nonporous polymers
can drastically decrease the catalytic activity because only catalytic
sites exposed to the particle surface are effective for the reaction.
We further investigated seven different substrates to confirm
the generality of the reaction system. As indicated in Table S5
(ESI†), the COF catalysts are widely effective for different substrates
with similar activities and ee and dr values. [Pyr]X-H2P-COFs are
easily separated from the reaction mixture via centrifugation. Thus,
[Pyr]25-H2P-COFs can be reused at least four times (Table S6, ESI†).
The slight decrement in activity could be attributed to the channels
becoming blocked upon repetitive use, as evidenced by the
decreased BET surface area after recycling (Fig. S4, ESI†).
1 X. Feng, X. Ding and D. Jiang, Chem. Soc. Rev., 2012, 41, 6010–6022.
¨
2 F. J. Uribe-Romo, J. R. Hunt, H. Furukawa, C. Klock, M. O’Keeffe and
O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 4570–4751.
3 R. W. Tilford, S. J. Mugavero III, P. J. Pellechia and J. J. Lavigne, Adv.
Mater., 2008, 20, 2741–2746.
4 S. Wan, J. Guo, J. Kim, H. Ihee and D. Jiang, Angew. Chem., Int. Ed.,
2008, 47, 8826–8830.
5 A. Nagai, Z. Guo, X. Feng, S. Jin, X. Chen, X. Ding and D. Jiang, Nat.
Commun., 2011, 2, 536.
6 X. Chen, M. Addicoat, S. Irle, A. Nagai and D. Jiang, J. Am. Chem.
Soc., 2013, 135, 546–549.
7 D. N. Bunck and W. R. Dichtel, Chem. Commun., 2013, 49,
2457–2459.
8 N. L. Campbell, R. Clowes, L. K. Ritchie and A. I. Cooper, Chem.
Mater., 2009, 21, 204–206.
9 P. Kuhn, M. Antonietti and A. Thomas, Angew. Chem., Int. Ed., 2008,
47, 3450–3453.
10 S. Y. Ding, J. Gao, Q. Wang, Y. Zhang, W. G. Song, C. Y. Su and
W. Wang, J. Am. Chem. Soc., 2011, 133, 19816–19822.
11 Z. Xie, C. Wang, K. E. deKrafft and W. Lin, J. Am. Chem. Soc., 2011,
133, 2056–2059.
We explored the possibility of a continuous flow reaction
using a columnar setup. We prepared a column consisting of a
vertically mounted Teflon pipe loaded with silica gel (as a plug
12 D. W. MacMillan, Nature, 2008, 455, 304–308.
13 C. A. Wang, Z. K. Zhang, T. Yue, Y. L. Sun, L. Wang, W. D. Wang,
Y. Zhang, C. Liu and W. Wang, Chem.–Eur. J., 2012, 18, 6718–6723.
1294 | Chem. Commun., 2014, 50, 1292--1294
This journal is ©The Royal Society of Chemistry 2014