Chiral Phosphoric Acids in Metal–Organic Frameworks with Enhanced Acidity and Tunable Catalytic Selectivity

Article abstract of DOI:10.1002/anie.201908959

Chiral phosphoric acids are incorporated into indium-based metal–organic frameworks (In-MOFs) by sterically preventing them from coordination. This concept leads to the synthesis of three chiral porous 3D In-MOFs with different network topologies constructed from three enantiopure 1,1′-biphenol-phosphoric acid derived tetracarboxylate linkers. More importantly, all the uncoordinated phosphoric acid groups are periodically aligned within the channels and display significantly enhanced acidity compared to the non-immobilized acids. This facilitates the Br?nsted acid catalysis of asymmetric condensation/amine addition and imine reduction. The enantioselectivities can be tuned (up to '99 % ee) by varying the substituents to achieve a nearly linear correlation with the concentrations of steric bulky groups in the MOFs. DFT calculations suggest that the framework provides a chiral confined microenvironment that dictates both selectivity and reactivity of chiral MOFs.

Full text of DOI:10.1002/anie.201908959

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Chiral Phosphoric Acids in Metal-Organic Frameworks with
Enhanced Acidity and Tunable Catalytic Selectivity
Authors: Xu Chen, Hong Jiang, Yong Cui, Jianwen Jiang, Yan Liu,
Wei Gong, Xing Han, Xu Li, Bang Hou, Fanfan Zheng, and
Xiaowei Wu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201908959
Angew. Chem. 10.1002/ange.201908959
Link to VoR: http://dx.doi.org/10.1002/anie.201908959
http://dx.doi.org/10.1002/ange.201908959

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
Chiral Phosphoric Acids in Metal-Organic Frameworks with
Enhanced Acidity and Tunable Catalytic Selectivity**
Xu Chen,[a]+ Hong Jiang,[a]+ Xu Li,[b] Bang Hou,[a] Wei Gong,[a] Xiaowei Wu,[a] Xing Han,[a] Fanfan
Zheng,[a] Yan Liu,[a]* Jianwen Jiang,[b]* and Yong Cui[a]*
Abstract: New heterogeneous Brønsted acid catalysts with high
activity and selectivity are critically needed to substitute precious acid-
based catalysts for the sustainable production of fine and commodity
chemicals. However, the advance in synthetic methodology for the
preparation of such highly desirable catalysts remains a huge
scientific challenge. Herein, we report the incorporation of chiral
phosphoric acids into indium-based metal-organic frameworks (In-
MOFs) by sterically preventing them from coordination. This concept
leads to the synthesis of three chiral porous 3D In-MOFs with different
network topologies constructed from three enantiopure 1,1’-biphenol-
phosphoric acid derived tetracarboxylate linkers. More importantly, all
the uncoordinated phosphoric acid groups are periodically aligned
within the channels and display significantly enhanced acidity
compared to the non-immobilized acids. This facilitates the Brønsted
acid catalysis of asymmetric condensation/amine addition and imine
reduction. The enantioselectivities can be tuned (up to > 99% ee) by
varying the substituents to achieve a nearly linear correlation with the
concentrations of steric bulky groups in the MOFs. Density-functional
theory calculations suggest that the framework provides a chiral
confined microenvironment that dictates both selectivity and reactivity
of chiral MOFs.
homogeneous catalysts.[3] The requirement for the development
of new chiral solid acid catalysts may be realized by MOFs for
their structural diversity, uniformity and permanent porosity.[4]
MOFs have been extensively explored as catalysts for
asymmetric catalysis.[5,6] Among them MOFs adopting the
concept of using privileged chiral ligands/catalysts provide an
efficient and straightforward strategy for enantioselective
processes.[5d,5g,6,7a-7c] The applications, however, have been
mainly focused on Lewis acidic catalysts and rarely for Brønsted
acid catalysis.[5e-5h,6] To date, only a few catalytically active chiral
MOF-based Brønsted acid catalysts have been reported albeit
with low to moderate enantioselectivity (5-84% ee).[7b-7f] In general,
it is very difficult to retain a strong Brønsted acid site such as
phosphoric acid in MOFs, mainly due to spontaneous
deprotonation of the acids in solution, which leads to direct metal-
acid coordination. [8] Deprotonation of the acids could be avoided
by using a strong acid solution for MOF synthesis.[9] Unfortunately,
with a few notable exceptions such as MIL-100 and MIL-101,[10]
most of the reported MOFs are decomposed under such harsh
conditions. It remains a challenge to design chiral Brønsted acid
type of MOF catalysts. To address this issue, here we show that
phosphoric acids can be introduced into MOFs by sterically
protecting them from coordination.
Introduction
Chiral phosphoric acids derived from axially biaryls such as
1,1’-binaphthol (BINOL) and 1,1’-biphenol have been recognized
as one of the most versatile organocatalysts capable of affording
numerous enantioselective transformations.[10] Generally, steric
bulky substituents at 3,3'-positions of the BINOL or biphenol
skeleton exhibit distinct advantages for enantioselectivity, which
have a dual role of providing bulky as well as tuning the
electronics of the catalysts.[12] We have synthesized several
dicarboxylate ligands based on 1,1'-biphenol-2,2'-phosphoric acid
with steric bulky groups at the 3,3'-position, having their metal
ions coordinated with both carboxylate and phosphonate groups
forming MOFs that contain the conjugate base form of phosphoric
acid.[13] We surmise that the introduction of steric bulky bridging
carboxylate groups at the 3,3'-position of 1,1'-biphenol may
protect the phosphoric acids from coordinating with metal ions,
whereas the primary functional groups are linked by metal-
connecting units to produce extended networks. Herein, we
demonstrate as a proof-of-concept of the design of three porous
3D In-MOFs from three newly designed 3,3’,5,5’-tetracarboxylate
ligands of chiral 1,1’-biphenol-2,2’-phosphoric acid. Despite of
their different topological structures, all phosphoric acids are not
coordinated with metal ions in the MOFs and are periodically
aligned within the channels that endow the frameworks catalytic
activity and enantioselectivity in condensation/amine addition and
imine reduction. The stereoselectivity is modulated by varying
3,3’-substituents of
As one versatile class of eco-friendly heterogeneous catalysts,
porous solid Brønsted acids have been extensively explored for
many challenging and practically important reactions in chemical
and petrochemical processes.[1] However, presumably due to the
lacking of structural uniformity and designed catalytic sites, the
traditional solid acid systems normally show poor selectivity,[2]
especially in asymmetric organic reactions. Consequently, most
high value-added enantiopure chemicals are still catalyzed by
[a]
X, Chen+, H, Jiang+, B, Hou, W, Gong, X. Wu, X. Han, F. Zheng,
Prof. Y. Liu, Prof. Y. Cui
School of Chemistry and Chemical Engineering and State Key
Laboratory of Metal Matrix Composites,
Shanghai Jiao Tong University
Shanghai 200240 (China)
E-mail: yongcui@sjtu.edu.cn, liuy@sjtu.edu.cn
X, Li, Prof. J. Jiang
[b]
Department of Chemical and Biomolecular Engineering,
National University of Singapore
Singapore 117576, Singapore
E-mail: chejj@nus.edu.sg
[+]
[**]
These authors contributed equally to this work
This work was financially supported by the National Science
Foundation of China (Grants, 21522104, 21620102001, 21875136,
and 91856204), the National Key Basic Research Program of China
(2016YFA0203400), Key Project of Basic Research of Shanghai
(17JC1403100 and 18JC1413200) and Shanghai Rising-Star Program
(19QA1404300).
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
(a)
In(OOC)4
H5L1
MOF 1
(b)
O
O
HO
HO
OH
O
O
O
P
OH
In(OOC)4
O
OH
O
H5L2
MOF 2
(c)
In3(OOC)8
H5L3
MOF 3
Figure 1. Schematic view of the preparation for MOFs (a) 1, (b) 2 and (c) 3 (In, blue; O, red; P, green; C, gray. H atoms are omitted for clarity).
Synthesis and Characterization. The ligands H5L1, H5L2 and
H5L3 were prepared from (R)- or (S)-6,6'-dimethyl-3,3'-di-tert-
butyl-5,5'-di(4-methoxycarbonyl)phenyl-1,1'-biphenyl-2,2'-diol in
six steps in 24%, 36% and 47% overall yields, respectively.
Heating In(NO3)3·6H2O and H5L1, H5L2 or H5L3 in a mixed solvent
containing dimethylformamide (DMF) at 100 °C produced single
biphenol and that the ee values show a nearly linear correlation
with the concentrations of bulky groups in the MOFs. A model
rationalizing the observed difference is also presented.
crystals
of
[Me2NH2][In(L1)]·H2O·2DMF
(1),
[Me2NH2]2[In2(L2)2]·2H2O·DMF (2), or
Results and Discussion
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
(b)
(a)
(d)
(c)
Figure 2. (a) PXRD patterns; (b) N2 adsorption (filled symbols) and desorption (open symbols) isotherms at 77 K; (c) Conversions and ee values of kinetic experiment
with 2-thenaldehyde as substrate and (d) Recycling test of MOF 1 with 2-thenaldehyde as substrate.
[In3(L3)2][NO3]·6.7H2O·4DMF (3), respectively. The obtained In-
MOFs were stable in air and insoluble in water and common
organic solvents. They were characterized using elemental
analysis, IR spectra, and thermogravimetric analysis (TGA). The
phase purity of the bulky samples was established by comparing
the observed and simulated powder X-ray diffraction (PXRD)
patterns.
ligands to form 4-connected [In(COO)4] SBU (Figure S4b), and
each L2 connects to four In ions. Along the c-axis, adjacent In ions
are linked by four L2 to generate a 31 helix, leading to an irregular
tubule. Each of the helical chain is connected to six adjacent
chains by sharing In ions to give a 3D network with 1D hexagonal
channel (Figure 1b and S4b). The channels have an opening size
of about 1.10 × 1.10 nm2 and are periodically decorated with
uncoordinated phosphoric acids pointing outward (Figures 1b and
S5b). With respect to the topology, 2 exhibits a 4-c net with a qtz
network (Figure S3b), with the presence of two different types of
4-connected nodes, simplified by the ligand and metal node.
Single-crystal X-ray diffraction showed that 1 crystallizes in the
orthorhombic chiral space group P222. The asymmetric unit
contains L1 of 1/2 occupancy and two In ions of 1/4 occupancy.
Each In ion is octahedrally coordinated by four carboxylate groups
from four different L1 ligands to form a 4-connected [In(COO)4]
building unit, and each L1 connects to four [In(COO)4] nodes
(Figures 1a and S4a). Thus, the 3D structure of 1 is built by
[In(COO)4] SBU and tetracarboxylate-phosphoric acid L1 ligand,
which displays a 4-connected topology network with the Schlafli
symbol of [86] (Figure S3a),. The framework consists of three
types of open helical channels along the a-axis (1.20 × 1.24 nm2,
1.05 × 1.26 nm2 and 0.98 × 1.09 nm2, measured from van der
Waals surfaces), with the largest channel being periodically
decorated with free phosphoric acids (Figures 1a and S5a).
3 crystallizes in the monoclinic chiral space group C2, with the
asymmetric unit containing two independent In ions, one In ion of
1/2 occupancy and one independent L3 ligand. The basic building
block is a trinuclear [In3] unit, which is linked by eight bidentate
carboxylate groups of eight L3 ligands (Figures 1c and S4c). The
central In is octahedrally coordinated by six oxygen atoms from
six bidentate carboxylate groups, and each terminal In is
coordinated by seven oxygen atoms from bidentate carboxylate
groups. Each ligand binds to four In3 units via four bidentate
carboxylate groups leaving the free phosphoric acid pointing to
the channel. If we do not consider the central metal in the [In3]
units, the structure of 3 presents a 2-fold interpenetrated network
with an uninodal topology denoted as dia, with the presence of 4-
connected nodes, as simplified by the L3 ligand and In3 cluster
(Figure S3c). On the other hand, a
Although 2 shares the same [In(COO)4] building unit and
similar 4-connected L2 ligand as 1, 2 crystallizes in the hexagonal
chiral space group P6422. The asymmetric unit contains one L2
and one In ion, both of which are in 1/4 occupancy. Each In is
coordinated by four bidentate carboxylate groups for four L2
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
MeOOC
COOMe
O
O
O
P
OH
MeOOC
COOMe
Me4L3
(2.8 < H0 < 3.3)
Me4L1
(1.5 < H0 < 2.8)
Me4L2
(4.0 < H0 < 4.8)
Dithio-biphenol
(-4.4 ≤ H0 ≤ -3.2)
Thio-biphenol
(-2.4 ≤ H0 ≤ -1.5)
MOF 3
(1.5 < H0 < 2.8)
MOF 1
(-3.0 < H0 < -2.4)
MOF 2
(3.3 < H0 < 4.0)
Figure 3. The Hammett acidity of MOFs 1-3 and related molecular acids.
different topological simplification could be obtained, if the central
In ion is taken into consideration. This simplification results in a
3D 4,8-connected network with the Schlafli symbol of [44624] [46]2
(Figure S3d), which contains two types of irregular 1D channels
with opening sizes of about 1.40 × 1.12 and 1.32 × 0.78 nm2
(measured from van der Waals surfaces) along the a-axis (Figure
1c).The enantiomeric nature of them formed from R- and S-
enantiomers of H5L was confirmed by solid-state circular
dichroism (CD) spectra, which exhibit nearly mirror images of
each other (Figure S8). Thermogravimetric analysis (TGA)
showed the frameworks of 1-3 are thermally stable up to around
nitrodiphenylamine, indicating H0 ≤ -2.40, which is higher than that
of the biphenol thiophosphoric acid (-2.4 ≤ H0 ≤ -1.5), and even
close to that of the strong biphenol dithiophosphoric acid (-4.4 ≤
H0 ≤ -3.2) (Table S14). The above results suggest that the metal-
directed assembly of phosphoric acid catalysts into MOFs is
capable of effectively enhancing their Brønsted acidity, giving rise
to porous heterogeneous catalysts for fine chemical synthesis.
Further study is in progress to understand the origin of the acidity
enhancement of the immobilized phosphoric acids.
Heterogeneous Asymmetric Catalysis. Chiral BINOL-derived
Brønsted acids have been proved to be highly active for
catalyzing a broad range of reactions involving C−C and C−N
bonds formation in an enantioselective fashion.[11] The presence
of uniformly distributed phosphoric acids within the 1D channel of
those present MOFs prompted us to evaluate their catalytic
activities in asymmetric addition and reduction of imines.
o
400 C (Figure S9). PXRD showed all three frameworks remain
highly crystalline and intact after removal of the guest solvent
molecules (Figures 2a, S10a and S10b). Calculations using
PLATON indicated that MOFs 1-3 have about 51%, 79% and 68%
void volume available for guest inclusion.[14] The permanent
porosity was examined by N2 adsorption measurements at 77 K.
The Brunauer-Emmett-Teller (BET) surface areas were estimated
to be 699, 847 and 473 m2/g for 1-3, respectively (Figures 2b and
S11). Dye uptake measurements showed that the open channels
of them were well retained in solution (Figures S1, S10a-b and
Table S1).
We first studied the catalytic activity of the MOFs in promoting
the asymmetric condensation/amine addition reactions, which
provide direct access to optical active 2,3-dihydroquinazolinones
with important pharmacological activities.[16] All three MOFs were
found to be capable of catalyzing the cascade reaction, whereas
1 gave much higher reactivity and stereoselectivity than 2 and 3.
Specially, after screening a variety of reaction conditions such as
catalyst loading, reaction time, temperature and solvents, it was
found that, with 5 mol% loading, 1 could catalyze the reaction of
benzaldehyde and 2-aminobenzamide to the desired 2,3-
dihydroquinazolinone in 96% yield and 99% ee in CHCl3 at 80 oC
after 20 h (Table 1, 4a). Under the optimized conditions, a variety
of aromatic aldehydes (with substituents at different positions and
heteroaromatic aldehydes) were evacuated. In all cases, the
catalytic reactions provided the targeted products in 91-97%
yields and 94-99.5% ee (Table 1, 4a-4i).
Considering the importance of the catalyst acidity for catalytic
applications, the Brønsted acidity of the three In-MOFs and
several related organic phosphoric acids was determined by
means of the Hammett indicator method.[15] As shown in Figure 3,
the molecular catalysts Me4L1, Me4L2 and Me4L3 were found to
have moderate acidity (1.5 < H0 < 2.8, 4.0 < H0 < 4.8 and 2.8 < H0
< 3.3, respectively), with the order of acidity being consistent with
installation of different electron-withdrawing carboxylate groups
on biphenol. After assembly into MOFs, the acidity was greatly
enhanced (-3.0 < H0 < -2.4, 3.3 < H0 < 4.0 and 1.5 < H0 < 2.8 for
MOFs 1-3, respectively), as shown in Figure 3. Specifically, MOF
1 showed a color change even in a benzene solution of 4-
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
Table 1. Asymmetric condensation/amine addition reactiona
values than 1 (Table 1, 4a-4d). To further understand the reaction
detail, kinetic experiments were performed using 2-
aminobenzamide and 2-thenaldehyde as the substrates (Figure
2c). It was found that the reaction catalyzed by Me4L1 was faster
than that by 1, but with much low enantioselectivity in the whole
process. Although both 1 and Me4L1 share the same bulky
substituents, the resulting ee values are clearly different for the
same reactions under identical conditions. The observed
difference in ee values indicates that chiral phosphoric acids of
1,1’-biphenol that together with the In ions, phenyl rings and bulky
anthracenyl substituents create a chiral microenvironment in the
porous structure, which is believed to be responsible for the
observed high enantioselectivity, by concentrating substrates and
generating additional steric and electronic effects around the acid
active sites.[13]
To further understand the influence of the steric bulky groups
on the MOF-base catalysis, we prepared a series of multivariate
(MTV) MOFs containing different molar ratios of L1 and L3 through
a solvent-assistant ligand-exchange approach. The MOFs 1-L1
1-
x-L3x (X stands for the exchanged ratio) that are unable to access
de novo were prepared by heating 1 in a MeOH solution of H5L3
at 100 oC for a period of time. The ratio of L1 to L3 decreases with
increasing reaction time, which were determined by 1H NMR after
digesting the samples with dilute hydrochloric acid. As shown in
Figure 4b, L3 reaches its equilibrium after 36 h at 53% (Figure S1).
Further prolongation of reaction time and exchange of reaction
solution with fresh one do not further increase the content of L3,
indicating that a dynamic equilibrium was reached. PXRD showed
the resulting MTV-MOFs are all isostructural to the parent MOF
crystals (Figures 2a and S10c). Subsequently, we examined their
catalytic activities for the reaction between benzaldehyde and 2-
aminobenzamide. As shown in Figure 4, as the ratio of L3
increases, the yields of the product remain constant but the ee
values decrease smoothly from 99% for MOF 1 to 58% for 1-
L10.47-L30.53. The ee values show a nearly linear correlation with
the concentration of steric bulky groups in the MOF catalyst. This
result demonstrates that the bulky anthracenyl groups of the
biphenol ligand are indeed responsible for the observed high
stereoselectivity for the heterogeneous catalyst system.
[a] Reaction conditions: Cat. (5 mol %), 2-aminobenzamide (0.15
mmol) and aldehyde/ketone (0.30 mmol), CHCl3 (1 mL), 80 °C /
60 °C. The absolute configurations of the products 4a, 4b, and 4f
were assigned as S by comparing their HPLC profiles with those
reported in literature.[17] 4c, 4d, 4e, 4g, 4h and 4i were assigned
were assigned by analogy. The R configuration of 4l was defined
by single crystal x-ray diffraction. 4k and 4m were assigned by
analogy. [b] Isolated yield. [c] Determined by HPLC analysis.
Under identical conditions, MOFs 2 and 3 with different aryl
substituents at the 3,3’-positions can also catalyze these
reactions, affording 90-94% and 89-96% yields with 17-63% and
3-25% ee, respectively. The ee values are much lower than those
obtained with MOF 1, as presented in Table 1. Interestingly,
despite sharing the same less bulky substituents phenyl groups
at 3,3'-positions, the ee values obtained with 2 were 8-60% higher
than those with 3, probably due to the different shortest distances
between phosphoric acid sites and the framework walls (0.49 nm
and 0.92 nm) that lead to different levels of asymmetric induction.
The above results indicat that both the local environmental and
the position of the phosphoric acids are critically important for
enantioselectivities of the MOF catalysts.
Multiple experiments were performed to prove the
heterogeneity and recyclability of the MOF 1 catalyst. Upon
completion of the reaction of 2-aminobenzamide and 2-
thenaldehyde, 1 could be recovered by simple filtration and
reused at least ten times without any loss of its activity and
enantioselectivity (Figures 2a and 2b). After ten cycles, both the
PXRD pattern and BET surface area (660 m2g-1) of 1 remained
almost unchanged as compared to the pristine sample (Figures
2b). Secondly, a hot filtration test showed no indication of
catalysis by leached homogeneous species. Inductively coupled
plasma optical emission spectrometry (ICP-OES) analysis of the
product solution indicated almost no loss of In ions (~0.002%)
from the structure per cycle.
To assess the contribution of the MOF structure to the
asymmetric catalysis, we examined the catalytic activities of the
homogeneous analogue Me4L1. Control experiments showed
Me4L1 (5 mol% loading) gave high yields of the products from the
condensation/amine addition reactions, but with much lower ee
It should be noted that numerous Brønsted acid catalysts have
been explored for the enantioselective synthesis of 2,3-
dihydroquinazolinone,[17] but, in most cases, only moderate to
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
O
O
O
O
O
O
O
O
O
O
O
In
In
O
(a)
O
O
O
O
H5L3
HO
O
HO
O
O
O
O
O
O
O
O
O
O
O
P
P
P
P
OH
OH
H5L1
O
O
O
O
O
O
O
O
O
In
OO
In
OO
O
O
O
MOF 1-L11-x-L3
MOF 1
x
(b)
(c)
Figure 4. (a) Illustration of the post-synthetic ligand exchange; (b) Exchange ratio depended on the reaction time; (c) Yields and ee values at different exchange
ratio in addition reaction with 2-aminobenzamide and benzaldehyde as substrates.
excellent enantioselectivities were observed.[17b-17g] High stereo-
and regioselectivity are noteworthy features of the present
protocol based on MOF 1, which is even among the highest
values reported for homogeneous phosphoric acid catalysts
(Table S8). The synthetic utility of this MOF 1-based methodology
was shown by the synthesis of enantiopure 2-biphenyl substituted
2,3-dihydroquinazolinone (DHQZ) analogue 4i (Table 1, 4i), which
was the potent tubulin inhibitor against a number of human cancer
cell lines.[16] It was found that 1 loading at 5 mol% catalyzed the
asymmetric condensation/amine addition reaction between 2-
give comparable yields to MOF 1, their enantioselectivities were
significantly reduced (39-67% and 10-20% ee) (Table 1, 4k-4m).
Furthermore, when Me4L1 was used as a catalyst, 91-95% yields
and 61-80% ee were achieved for aliphatic ketones (Table 1, 4k
and 4l), but only 11% yield and 56% ee were detected for the
aromatic ketone (Table 1, 4m). Again, this is presumably due to
that immobilizing the molecular catalysts into the MOF could
generate additional steric effects around active sites, thereby
increasing the level of asymmetric induction.
Chiral phosphoric acids have been shown to be highly efficient
catalysts for the enantioselective reduction of C=N bonds with an
organic hydride source.[19] Recently, microporous polymers and
MOFs have been used to catalyze the asymmetric transfer
hydrogenation of benzoxazines and ketimines, but none of them
could compromise balance between clear polymeric structure and
high enantioselectivities (Tables S9 and S11).[7c, 20] This promoted
us to evaluate the catalytic activities of MOFs 1-3 in the reduction
of imines. The reduction of ketimines offers a straightforward
approach to produce chiral amines, which constitutes privileged
building blocks for pharmaceuticals and agrochemicals.[21] To our
delight, all three MOFs were found to be able to promote the
ketimine reduction reactions, whereas 1 give the highest
enantioselectivity. With 5 mol% loading, 1 catalyzed the reduction
of (E)-N-(4-methoxyphenyl)-1-phenylethan-1-imine, giving the
product 6a in 84% yield with 99% ee in toluene at 50 oC for 36 h
(Table 2, 6a). The substrates with both electron-donating and
withdrawing groups were compatible with this transformation,
affording 86-99.9% ee. Imines with electron-withdrawing and
heteroaromatic groups were also subjected to the reactions,
amino-5-nitrobenzamide
and
2'-methylbiphenyl-4-
carboxaldehyde in 95% yield and 95% ee. Although the obtained
ee was slightly lower than those obtained with homogeneous
catalysts,[16a] this new synthetic method only requires a short
enatioselective pathway with improved synthesis yield (from 7%
to 63%) (Scheme S5).
Despite great efforts, the asymmetric condensation/amine
addition of 2-aminobenzamides and ketones has not been well
studied, mainly due to the intrinsically low reactivity of ketones.[18]
Encouraged by the outstanding catalytic performance of the
MOFs, we decided to evaluate their catalytic activities in the
condensation/amine addition reactions involving ketones.
Although both MOFs and Me4L1 were proved to be efficient, only
MOF 1 turned out to be promising in term of yields and
enantioselectivities for the tested substrates. Specially, with 5
mol% catalytic loading, 1 provided > 90% yields and 83% and
97% ee for two aliphatic ketones (Table 1, 4k and 4l). However,
for the substrate with phenyl group, reduced yield and ee value
were obtained (Table 1, 4m). Despite that MOFs 2 and 3 could
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
providing the amines in 86-93% ee (Table 2, 6b, 6d, 6e and 6f).
In contrast, only 63-80% and 19-34% ee were obtained with
MOFs 2 and 3 for the tested substrates, respectively, although the
yields were comparable with those of 1 (Table 2, 6a-6d). Under
identical conditions, the homogeneous analogue Me4L1 produced
the products in 0-39% ee values, which are substantially lower
than those obtained with 1 (Table 2, 6a-6d). This further proves
that bulky substituents at the 3,3'-positions of the biphenyl unit
combined with the confined space of MOFs could highly enable
the asymmetric activation of the substrate
dibenzyloxy phenyl groups were introduced to the asymmetric
condensation/amine addition and imine reduction reactions
(Table 1, 4j, Table 2, 6j and Figure S7). Only trace mount of the
desired products were detected by 1 even after 48 h, much lower
than 45% and 61% isolated yield obtained with Me4L1 (5 mol%
loading) as the homogeneous catalyst, indicating that catalysis
does not occur mainly at surface sites. This point was further
supported by the fact that grounded and unground particles of 1
exhibited similar catalytic performance in asymmetric
condensation/amine addition reaction.
Table 2 Asymmetric transfer hydrogenation reactiona.
(a)
EtO2C
CO2Et
N
O
N
O
H
1
5 mol% (R)-MOF
CHCl3, 50 °C, 20 h
R = Ph, 4-Br-Ph, 4-MeO-Ph
R
N
H
R
93-97% yield, 98-99% ee
(b)
EtO2C
CO2Et
R1
N
R1
N
N
O
O
H
1
5 mol% (R)-MOF
N
R2
N
H
R2
THF, 80 °C, 36 h
R1/R2 = H/Ph, Me/Ph, H/4-Br-Ph
58-65% conv., 99.9% ee
Scheme 1. Enantioselective transfer hydrogenation of (a) benzoxazines and (b)
quinoxalinones.
The capability of the MOF catalyst was further demonstrated
by enantioselective reduction reaction of other types of imines.
For example, by using Hantzsch ester as the stoichiometric
hydride source, MOF 1 was catalytically active for asymmetric
transfer hydrogenation of 1,4-benzoxazines with both electron-
donating and -withdrawing groups on the phenyl rings, affording
the products in 93-97% yields and 98-99% ee (Scheme 1a). In
addition, in the presence of using Hantzsch ester, MOF 1 could
also catalyze reduction of quinoxalinones with varying
substituents on the phenyl ring into the corresponding
dihydroquinoxalinones in 58-65% conversions and 99.9% ee
(Scheme 1b). In both cases, the ee values are comparable to
those obtained with Me4L1 (Tables S12 and S13) and other
reported molecular catalysts (Tables S9 and S10),[23] but are
much higher than those with MOFs 2 and 3 (80-91%, 13-51% and
11-33%, 7-24% ee). In particular, although MOFs 2 and 3 share
the same less bulky substituents –phenyl at 3,3'-positions, the ee
values obtained with 2 are 57-71% and 6-27% higher than those
with 3, probably due to contributions of the frameworks and the
short distances between phosphoric acid sites and the
frameworks.
[a] Reaction conditions: Cat. (5 mol %), ketimine (0.15 mmol) and
benzothiazoline (0.30 mmol) in toluene (1.0 mL), 50 °C, 36 h. The
absolute configurations of the products were assigned as R by
comparing their HPLC profiles with those reported in literature.[22]
PMP, p-methoxyphenyl. [b] Isolated yield. [c] Determined by
HPLC analysis. [d] The ketimines were generated in situ.
The major drawback of this reaction was the need for isolated
ketimines, which is unstable and diffcult to synthesiz and purify.
The most convenient way is to generate imines in situ for the
hydrogenation reaction. As shown in Table 2, with 5 mol% loading
of 1, the reactions of different aromatic/heterocyclic ketones and
aromatic amines proceeded smoothly and afforded the products
in 82% yields and 91-99.7% ee in toluene at 60 oC after 48 h. The
resulting yields and ee values are comparable with those obtained
from the direct reduction of ketimines catalyzed by 1. Besides,
several tests also prove that 1 is a heterogeneous and recyclable
catalyst in catalyzing the imine reduction reaction (Figures 2a and
S74).
DFT Calculations. We carried out DFT calculations to elucidate
how the framework impacts on the catalytic activities. As shown
in Figure S12a, one repeating unit of 1 was used in the
calculations. For comparison, the homogeneous catalyst Me4L1
was also examined. According to previously reported
literatures,[24] we chose the asymmetric hydrogenation of
ketimines with the imine of
In order to explore whether catalysis by
1 occurs
predominantly within the channels or instead on the external
surface, the more sterically demanding substituent 3,5-
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
O
O
O
O
O
O
In
(a)
O
O
O
HO
O
O
O
OH
O
OH
O
O
O
O
P
O
P
P
P
OH
O
O
O
O
O
In
O
O
O O
MOF 1
HOMO
(b)
(c)
N
N
N
N
H
N
N
H
N
H
H
H
H
H
S
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
H
P
P
P
H
P
P
P
O
O
O
O
O
H
O
O
O
H
H
H
H
H
H
N
N
N
N
S
S
S
N
S
S
(
)-TS'
R
(R)-product / product'
1A/1A'
(S)-TS
(S)-product / product'
1B/1B'
Scheme 2. (a) Simplified Me4L1 and MOF 1. (b) Free energies (kcal mol-1) of intermediates, transition states and products in the heterogeneous (left) and
homogeneous (right) system with 1 and Me4L1. (The pink and blue pathways are for (S)- and (R)-stereoisomers, respectively, and the reaction barrier of transition
state is shown in parentheses.) (c) Schematic structures of intermediates and transition states.
acetophenone and (R)-benzothiazoline (simplified chemical
model) as a representative example and proposed a possible
catalytic cycle (Figure S12b).
stronger than configurations 1A and 1B. The two pathways to (S)
and (R)-product’ are also exothermic with free energy changes of
-19.7 and -20.3 kcal mol−1, respectively. However, the reaction
barriers for transition states (S)-TS’ (9.3 kcal mol−1) and (R)-TS’
(5.5 kcal mol−1) are much lower compared to (S)-TS and (R)-TS
in the homogeneous system. Thus, the reaction activity is
enhanced due to confinement effect in 1, which stabilizes the
intermediates and transition states. The other striking result is that
the transition state (R)-TS’ (Figures S17 and S18) has a relatively
smaller steric effect along the channel of 1; consequently, it has a
lower reaction barrier enantioselectively leading to (R)-product’ as
the major product. These calculation results are in good
agreement with experimental observation and provide
microscopic insight into the confinement and steric effects of MOF
1 on the activity and selectivity of asymmetric hydrogenation.
In a homogeneous system with Me4L1 (Figure S13), imine and
benzothiazoline initially form intermediates with different
configurations 1A and 1B and then undergo a proton transfer step
to generate transition states (S)-TS and (R)-TS leading to (S) and
(R)-product, respectively. As shown in Scheme 2, all the
intermediates and transition states are bound on the chiral 1,1’-
biphenol-2,2’-phosphoric acid through two hydrogen-bonds. Both
reactions yielding (S) and (R)-product are exothermic within free
energy changes of -26.0 and -25.0 kcal mol−1, respectively. The
reaction barrier of transition state (S)-TS (17.4 kcal mol−1) is very
close to that of (R)-TS (17.7 kcal mol−1) (Figures S15 and S16).
Therefore, both reactions have similar thermodynamics and
kinetics with a low enantioselectivity. This is consistent with the
experimentally observed low values of ee.
Conclusion
In MOF 1 (Figure S14), all the intermediates and transition
states are also bound on the chiral 1,1’-biphenol-2,2’-phosphoric
acid (Scheme 2). The binding energies of configurations 1A’ and
1B’ are -26.3 and - 28.4 kcal mol−1, respectively, which are slightly
In summary, we have demonstrated steric protection of
catalytically active sites as an efficient strategy to design MOF
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
[9]
a) N. Stock, S. Biswas, Chem. Rev. 2012, 112, 933-969; b) M. J. Van
Vleet, T. Weng, X. Li, J. R. Schmidt, Chem. Rev. 2018, 118, 3681-3721.
a) G. Ferey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surble, J.
Dutour, I. Margiolaki, Angew. Chem. Int. Ed. 2004, 43, 6296-6301;
Angew. Chem. 2004, 116, 6456-6461;b) G. Ferey, C. Mellot-Draznieks,
C. Serre, F. Millange, J. Dutour, S. Surble, I. Margiolaki, Science 2005,
309, 2040-2041.
with strong Brønsted acids. Incorporation of chiral phosphoric
acids into MOFs leads to significant enhancement of their
Brønsted acidity and endows the frameworks with high catalytic
activity in the addition and reduction of imines, in which their
enantioselectivities can be modulated by varying the 3,3’-
substituents. DFT calculations suggest that the achieved high
enantioselectivity is attributed to the confined microenvironment
of catalytically active phosphoric acids in the framework.
Therefore, this work lifts the constraints placed on traditional solid
Brønsted acid catalysis and can lead to a new type of
heterogeneous acid catalysts for the eco-friendly synthesis of fine
chemicals.
[10]
[11]
[12]
D. Parmar, E. Sugiono, S. Raja, M. Rueping, Chem. Rev. 2014, 114,
9047-9153.
a) J. P. Reid, J. M. Goodman, Chem. Eur.J. 2017, 23, 14248-14260; b)
L. W. Qi, J. H. Mao, J. Zhang, B. Tan, Nat. Chem. 2018, 10, 58-64; c) W.
Kashikura, K. Mori, T. Akiyama, Org. Lett. 2011, 13, 1860-1863.
a) X. Chen, Y. Peng, X. Han, Y. Liu, X. Lin, Y. Cui, Nat. Commun. 2017,
8, 2171; b) X. Chen, H. Jiang, B. Hou, W. Gong, Y. Liu, Y. Cui, J. Am.
Chem. Soc. 2017, 139, 13476-13482.
[13]
[14]
[15]
[16]
A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7-13.
C. Walling, J. Am. Chem. Soc. 1950, 72, 1164-1168.
Experimental Section
a) G. M. Chinigo, M. Paige, S. Grindrod, E. Hamel, S. Dakshanamurthy,
M. Chruszcz, W. Minor, M. L. Brown, J. Med. Chem. 2008, 51, 4620-
4631; b) M. Badolato, F. Aiello, N. Neamati, RSC Adv. 2018, 8, 20894-
20921.
Full experimental procedures and DFT calculations are provided in the
Supplemental Information.
[17]
a) S. Ayyanar, P. K. Vijaya, M. Mariyappan, V. Ashokkumar, V.
Sadhasivam, S. Balakrishnan, C. Chinnadurai, S. Murugesan, New J.
Chem. 2017, 41, 7980-7986; b) D. Huang, X. Li, F. Xu, L. Li, X. Lin, ACS
Catal. 2013, 3, 2244-2247; c) M. Rueping, A. P. Antonchick, E. Sugiono,
K. Grenader, Angew. Chem. Int. Ed. 2009, 48, 908-910; Angew. Chem.
2009, 121, 925-927; d) P. Du, H. Zhou, G. Shen, K. Zou, Chin. J. Org.
Chem. 2015, 35,1641-1649; e) X. Cheng, S. Vellalath, R. Goddard, B.
List, J. Am. Chem. Soc. 2008, 130, 15786-15787; f) D.-J. Cheng, Y. Tian,
S.-K. Tian, Adv. Syn. Catal. 2012, 354, 995-999; g) B. Zhang, L. Shi, R.
Guo, Catal. Lett. 2015, 145, 1718-1723.
Keywords: metal-organic frameworks • phosphoric acid •
Brønsted acid• asymmetric catalysis • porosity
[1]
M. E. Davis, Nature 2002, 417, 813-821.
[2]
a) F. Su, Y. Guo, Green Chem. 2014, 16, 2934-2957; b) J. Liang, Z. Liang,
R. Zou, Y. Zhao, Adv. Mater. 2017, 29,.1701139.
[3]
H‐U. Blaser, H-J. Federsel, Asymmetric Catalysis on Industrial Scale:
Challenges, Approaches and Solutions, Second Edition, Wiley-VCH,
Weinheim, 2010.
[18]
[19]
[20]
Y. Jiang, Y. Liu, S.-J. Tu, F. Shi, Tetrahedron: Asymmetry 2013, 24,
1286-1296.
[4]
[5]
a) Y. Liu, W. Xuan, Y. Cui, Adv. Mater. 2010, 22, 4112-4135; b) M. Yoon,
R. Srirambalaji, K. Kim, Chem. Rev. 2012, 112, 11961231.
C. Zhu, K. Saito, M. Yamanaka, T. Akiyama, Acc. Chem. Res. 2015, 48,
388-398.
a) P. Wu, C. He, J. Wang, X. Peng, X. Li, Y. An, C. Duan, J. Am. Chem.
Soc. 2012, 134, 14991-14999; b) Q. Han, C. He, M. Zhao, B. Qi, J. Niu,
C. Duan, J. Am. Chem. Soc. 2013, 135, 10186-10189; c) K. Mo, Y. Yang,
Y. Cui, J. Am. Chem. Soc. 2014, 136, 1746-1749; d) T. Sawano, P. Ji, A.
R. McIsaac, Z. Lin, C. W. Abney, W. Lin, Chem. Sci. 2015, 6, 7163-7168;
e) T. Shi, Z. Guo, H. Yu, J. Xie, Y. Zhong, W. Zhu, Adv. Synth. Catal.
2013, 355, 2538-2543; f) Y. Zhang, J. Guo, L. Shi, Y. Zhu, K. Hou, Y.
Zheng, Z. Tang, Sci. Adv. 2017;3: e1701162; g) J. Li, Y. Ren, C. Qi, H.
Jiang, Chem. Commun. 2017, 53, 8223-8226; h) Z. Li, Y. Liu, X. Kang,
Y. Cui, Inorg. Chem. 2018, 57, 9786-9789.
a) M. Rueping, E. Sugiono, A. Steck, T. Theissmann, Adv. Syn. Catal.
2010, 352, 281-287; b) C. Bleschke, J. Schmidt, D. S. Kundu, S. Blechert,
A. Thomas, Adv. Syn. Catal. 2011, 353, 3101-3106; c) D. S. Kundu, J.
Schmidt, C. Bleschke, A. Thomas, S. Blechert, Angew. Chem. Int. Ed.
2012, 51, 5456; Angew. Chem. 2012, 124, 5552-5555. d) A. V. Malkov,
M. Figlus, P. Kocovsky, J. Org. Chem. 2008, 73, 3985-3995.
T. C. Nugent, Chiral Amine Synthesis: Methods, Developments and
Applications, Wiley-VCH: Weinheim, 2010.
[21]
[22]
a) S. Hoffmann, A. M. Seayad, B. List, Angew. Chem. Int. Ed. 2005, 44,
7424-7427; Angew. Chem. 2005, 117, 7590-7593; b) R. I. Storer, D. E.
Carrera, Y. Ni, D. W. MacMillan, J. Am. Chem. Soc .2006, 128, 84-86; c)
C. Zhu, T. Akiyama, Org. Lett. 2009, 11, 4180-4183; d) K. Saito, H.
Miyashita, T. Akiyama, Org. Lett. 2014, 16, 5312-5315.
[6]
a) C. Tan, X. Han, Z. Li, Y. Liu, Y. Cui, J. Am. Chem. Soc. 2018, 140,
16229-16236; b) T. Sawano, N. C. Thacker, Z. Lin, A. R. McIsaac, W.
Lin, J. Am. Chem. Soc. 2015, 137, 12241-12248; c) M. Banerjee, S. Das,
M. Yoon, H. J. Choi, M. H. Hyun, S. M. Park, G. Seo, K. Kim, J. Am.
Chem. Soc. 2009, 131, 7524-7525; d) D. Dang, P. Wu, C. He, Z. Xie, C.
Duan, J. Am. Chem. Soc. 2010, 132, 14321-14323; e) J. M. Falkowski,
T. Sawano, T. Zhang, G. Tsun, Y. Chen, J. V. Lockard, W. Lin, J. Am.
Chem. Soc. 2014, 136, 5213-5216; f) L. Ma, J. M. Falkowski, C. Abney,
W. Lin, Nat. Chem. 2010, 2, 838-846; g) Q. Xia, Z. Li, C. Tan, Y. Liu, W.
Gong, Y. Cui, J. Am. Chem. Soc. 2017, 139, 8259-8266;
[23]
[24]
a) M. Rueping, A. P. Antonchick, T. Theissmann, Angew. Chem. Int. Ed.
2006, 45, 6751-6755; Angew. Chem. 2006, 118, 6903; b) Y. Zhang, R.
Zhao, R. L.-Y. Bao, L. Shi, Eur. J. Org. Chem. 2015, 3344-3351; c) M.
Rueping, F. Tato, F. R. Schoepke, Chem. Eur. J. 2010, 16, 2688-2691;
d) J. L. Nunez-Rico, A. Vidal-Ferran, Org. Lett. 2013, 15, 2066-2069.
a) L. Simon, J. M. Goodman, J. Am. Chem. Soc. 2008, 130, 8741-8747;
b) Y. Shibata, M. Yamanaka, J. Org. Chem. 2013, 78, 3731-3736.
[7]
a) K. Tanaka, K. Sakuragi, H. Ozaki, Y. Takada, Chem. Commun. 2018,
54, 6328-6331; b) M. Zheng, Y. Liu, C. Wang, S. Liu, W. Lin, Chem. Sci.
2012, 3. 2623-2627; c) Z. Zhang, Y. R. Ji, L. Wojtas, W. Y. Gao, S. Ma,
M. J. Zaworotko, J. C. Antilla, Chem. Commun. 2013, 49, 7693-7695; d)
D. J. Lun, G. I. Waterhouse, S. G. Telfer, J. Am. Chem. Soc. 2011, 133,
5806-5809; e) M. J. Ingleson, J. P. Barrio, J. Bacsa, C. Dickinson, H.
Park, M. J. Rosseinsky, Chem. Commun. 2008, 1287-1289; f) L. Liu, T.
Y. Zhou, S. G. Telfer, J. Am. Chem. Soc. 2017, 139, 13936-13943.
a) G. K. Shimizu, R. Vaidhyanathan, J. M. Taylor, Chem. Soc. Rev. 2009,
38, 1430-1449; b) M. Taddei, F. Costantino, F. Marmottini, A. Comotti, P.
Sozzani, R. Vivani, Chem. Commun. 2014, 50, 14831-14834; c) S. Kim,
B. Joarder, J. A. Hurd, J. Zhang, K. W. Dawson, B. S. Gelfand, N. E.
Wong, G. K. H. Shimizu, J. Am. Chem. Soc. 2018, 140, 1077-1082.
[8]
This article is protected by copyright. All rights reserved.

10.1002/anie.201908959
Angewandte Chemie International Edition
RESEARCHARTICLE
RESEARCH ARTICLE
Text for Table of Contents
Xu Chen,[a]+ Hong Jiang,[a]+ Xu Li,[b] Bang
Hou,[a] Wei Gong,[a] Xiaowei Wu,[a] Xing
Han,[a] Fanfan Zheng,[a] Yan Liu,[a]*
Jianwen Jiang,[b]* and Yong Cui[a]*
Page No. – Page No.
Text for Table of Contents
Chiral Phosphoric Acids in Metal-
Organic Frameworks with Enhanced
Acidity
and
Tunable
Catalytic
Selectivity
New heterogeneous Brønsted acid catalysts with high activity and selectivity are
critically needed to replace precious acid-based catalysts for the sustainable
production of fine and commodity chemicals, but synthetically they still face a big
challenge. We report here that chiral phosphoric acids can be constructed into
metal-organic frameworks (MOFs) by sterically preventing them from coordination.
This leads to the synthesis of three chiral porous 3D In-MOFs with different network
topology from three tetracarboxylate linkers derived from enantiopure 1,1’-biphenol-
phosphoric acid. All phosphoric acid groups are uncoordinated and periodically
aligned within the channels that display significantly enhanced acidity compared to
non-immobilized acids. This facilitates the Brønsted acid catalysis of asymmetric
condensation/amine addition and imine reduction. The enantioselectivities can be
tuned (up to >99% ee) by varying substituents and have a nearly linear correlation
with the concentrations of steric bulky groups in the MOFs. Density-functional
theory calculations suggest that the framework provides a chiral confined
microenvironment that dictates both selectivity and reactivity of the chiral MOFs.
This article is protected by copyright. All rights reserved.