Polymer Meets Frustrated Lewis Pair: Second-Generation CO2-Responsive Nanosystem for Sustainable CO2 Conversion

Article abstract of DOI:10.1002/anie.201804034

Frustrated Lewis pairs (FLP), a couple comprising a sterically encumbered Lewis acid and Lewis base, can offer latent reactivity for activating inert gas molecules. However, their use as a platform for fabricating gas-responsive materials has not yet developed. Merging the FLP concept with polymers, we report a new generation CO₂-responsive system, differing from the first-generation ones based on an acid–base equilibrium mechanism. Two complementary Lewis acidic and basic block copolymers, installing bulky borane- and phosphine-containing blocks, were built as the macromolecular FLP. They can bind CO₂ to drive micellar formation, in which CO₂ as a cross-linker bridges the block chains. This dative bonding endows the assembly with ultrafast response (<20 s), thermal reversibility, and excellent reproducibility. Moreover, such micelles bound highly active CO₂ can function as nanocatalysts for recyclable C₁ catalysis, opening a new direction of sustainable CO₂ conversion.

Full text of DOI:10.1002/anie.201804034

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Polymer Meets Frustrated Lewis Pair: Second-Generation CO2-
Responsive Nanosystem for Sustainable CO2 Conversion
Authors: Qiang Yan, Liang Chen, and Renjie Liu
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.201804034
Angew. Chem. 10.1002/ange.201804034
Link to VoR: http://dx.doi.org/10.1002/anie.201804034
http://dx.doi.org/10.1002/ange.201804034

10.1002/anie.201804034
Angewandte Chemie International Edition
COMMUNICATION
Polymer Meets Frustrated Lewis Pair: Second-Generation CO2-
Responsive Nanosystem for Sustainable CO2 Conversion
Liang Chen,[†] Renji Liu,[†] and Qiang Yan*
Abstract: Frustrated Lewis pair (FLP), as a couple of sterically
encumbered Lewis acid and Lewis base, can offer latent reactivity for
activating inert gas molecules. However, their use as a platform for
fabricating gas-responsive materials has yet developed. Merging FLP
concept and polymers, we report a new generation of CO2-responsive
system, differing from the first-generation ones based on acid-base
equilibrium mechanism. Two complementary Lewis acidic and basic
block copolymers, installing bulky borane- and phosphine-containing
blocks, were built as the macromolecular FLP. They can bind CO2 to
drive the micellar formation, where CO2 as a cross-linker bridges the
block chains. This dative bonding endows the assembly with ultrafast
response (< 20 s), thermal reversibility, and excellent reproducibility.
Moreover, such micelles bound highly active CO2 can function as
nanocatalysts for recyclable C1 catalysis, opening a new direction of
sustainable CO2 conversion.
workers pioneered the first FLP-based self-healing gel material.[10]
Inspired by these, we wondered if the unique reactivity of FLPs
could be used to construct next-generation CO2-responsive
polymer systems, where CO2 would chemically bridge the FLP-
containing polymers and offer dynamic bonding to drive a
reversible polymer micellization. Moreover, we postulated that
CO2 species bound in the micelles would be highly active, which
could make the micelles function as recyclable nanocatalysts to
realize C1 feedstock conversion, as shown in Scheme 1b.
Since the advent of carbon dioxide (CO2)-responsive polymer, it
has continued to attract much attention of chemists because CO2
is a unique gas trigger with mild, non-invasive and contamination-
free features.[1] Compare to other chemically responsive polymers,
CO2-responsive polymers manifest irreplaceable advantages in
many fields, including cell mimics,[2] latex,[3] switchable surfaces,[4]
and gas-guided nanodelivery.[5] Such gas-triggered systems have
been increasingly considered as promising candidates for future
smart materials.
A seminal work in this theme was derived from Yuan and co-
workers.[2a] They have reported an amidine-containing polymer in
response to CO2. In this specific polymer, the basic amidine
groups (proton-acceptor) can react with acidic CO2 gas (proton-
donor). Up to now, all the known CO2-sensitive systems work
through this Brønsted acid-base pair principle (Scheme 1a).[1-5,6]
The essence of this principle is acid-base equilibrium; however,
the innate weak acidity of CO2 causes low responsivity. To solve
the bottleneck, seeking a new polymer system with distinct CO2-
sensitive mechanism is pressing.
A fundamental dilemma in enhancing CO2 responsiveness is its
chemical inertness and limited reactivity. In this respect, frustrated
Lewis pair (FLP) chemistry is possible to become an ideal way for
CO2 activation.[7] FLPs refer to Lewis acids and bases that are
sterically prevented from forming Lewis adduct and yet offer latent
Lewis acidity and basicity for activation with small molecules,[8] for
example, used for H2 heterolysis.[9] Most recently, Shaver and co-
Scheme 1. a) The working principle of first-generation CO2-responsive polymer
based on Brønsted acid-base pair and a typical example of amidine-containing
polymer. b) Frustrated Lewis pair (FLP) and the designed FLP-containing block
copolymers (P1 and P2) as a second-generation CO2-responsive system for
realizing CO2-activated micellization and serving as recyclable nanocatalysts for
CO2 catalytic conversion.
To fulfil this goal, we design and synthesize two styrene-based
monomers with bulky substituent groups, 4-styryl-di(pentafluoro-
phenyl)borane (1) and 4-styryl-dimesitylphosphine (2), as the FLP
acceptor and donor. Two diblock copolymers consisting of the
complementary FLP blocks and common polystyrene block (P1
and P2) are obtained via RAFT polymerizations (the details of
copolymer synthesis and characterization are in the Supporting
Information, Scheme S1-S2 and Figure S10‒S17). Using the
[a]
[†]
L. Chen, R. J. Liu, Prof. Dr. Q. Yan
State Key Laboratory of Molecular Engineering of Polymers
Department of Macromolecular Science, Fudan University
Shanghai 200433 (China)
E-mail: yanq@fudan.edu.cn
The authors contributed equally to this work
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201804034
Angewandte Chemie International Edition
COMMUNICATION
Gutmann-Beckett method,[11] the Lewis acidity (acceptor number,
AN) of the monomer 1 was determined to be 72.4 in toluene-d8
(Figure S1), close to that of standard reference B(C6F5)3 (AN =
75.7).[11b] It suggests that the formation of polymer-type Lewis
acid-base pair between P1 and P2 can be achievable.
in reported small-molecule FLPs.[14] Moreover, this CO2-binding is
highly efficient, since neither free 31P nor 11B resonance was
observed. Conductivity experiments support the above results
and demonstrate that this CO2 reactivity is thermally reversible
(Figure S5). After treatment with CO2, the solution conductivity
raised remarkably from 7.5 to 680 nS cm-1, indicating that the
bridging structure is zwitterionic. As heating up to 60 oC, the
conductivity returned back, indicating the cleavage of bridging
bonds. This procedure can be repeated at least three cycles,
which means that this FLP polymer can reversibly capture and
release CO2.
Figure 1. NMR studies of CO2 binding to complementary FLP polymer (P1 and
P2) before and after treatment with CO2 stimulus: a) 31P NMR spectra change
(red line); b) 11B NMR spectra change (blue line). Inset, the structure of CO2
bridging bonds between the boron-containing frustrated Lewis acidic polymers
(P1) and the phosphorus-containing frustrated Lewis basic polymers (P2)
Figure 2. CO2-responsiveness of FLP polymer: a) Hydrodynamic diameter of
FLP polymer solution in the absence (pink solid line) and presence (green solid
line) of CO2 stimulus, and the reversible response upon heating (pink dash line)
and re-addition of CO2 (green dash line) by DLS. b) TEM image illustrating the
micellization of FLP polymer upon addition of CO2 (inset, the detailed core-
corona structure). c) Gas response rate of the FLP polymer in various gas partial
pressure (1.0 atm, black open circle; 0.5 atm, blue open circle; 0.1 atm, green
open circle). d) CO2 response repeatability monitored by applying an alternating
CO2/thermal cycle: each cycle contains 0.5 min of CO2 stimulus period (blue
circle) and 9.5 min of thermal recovery period (red circle). The concentration of
FLP polymer solution (P1 and P2) in all experiments is fixed at 0.6 mM in toluene.
Mixing equimolar amounts of P1 and P2 in toluene (0.6 mM)
showed no significant reaction. However, when CO2 gas (pCO2
=
1.0 atm, flow rate = 0.1 mL s-1) was passed through the mixture,
a rapid binding reaction occurred, as evidenced by the solution
altering from transparent to semiopaque (Figure S2). To elucidate
this CO2-binding mechanism, NMR experiments were carried out
to provide detailed information on the polymer structural changes.
In the absence of gas stimulus, 31P NMR spectrum of the polymer
mixture of P1 and P2 gave a single sharp peak, positioning at -
22.4 ppm. However, upon addition of CO2, the 31P NMR signal
showed a large downfield shift from -22.4 to 49.2 ppm (∆δ = 71.6
ppm, Figure 1a), suggesting the bonding between phosphorus
atom of P2 and the carbon atom of CO2. Furthermore, 11B NMR
spectral analysis exhibited that P1 and P2 mixture had a broad
resonance around 74 ppm ascribed to the boron signal of typical
three-coordination triphenylborane,[12] whereas this resonance
produced a great opposite shift to upfield region from 74 to 2 ppm
(∆δ = -72 ppm, Figure 1b) in the presence of CO2, indicating the
formation of a four-coordination borane compound and confirming
that CO2 is covalently bridged with the boron in P1 and
phosphorus in P2 (Figure 1, inset). The diagnostic 19F NMR
spectrum of P1•CO2•P2 adduct corroborate this CO2-bridged
chemical structure (Figure S3).[13] The IR spectrum of product
showed a strong vibration at 1692 cm-1 ascribed to a characteristic
C=O stretch (Figure S4). These data coincide well with the cases
After understanding the CO2-responsive mechanism, we next
aimed to study the responsive self-assembly behavior of the FLP
polymer. Prior to CO2 addition, the hydrodynamic diameter of P1
and P2 mixture was 4.2 nm (PDI = 0.14) measured by dynamic
light scattering (DLS), consistent with the chain lengths of their
unimers (Figure 2a, pink solid line). When we aerated the solution
with CO2 gas, interestingly, their size gave an 11-fold increase up
to 46.8 nm (PDI = 0.08, Figure 2a, green solid line), implying the
formation of polymer aggregates. Through TEM observation,
spherical nanoparticles with a mean size of 40 nm appeared in
solution (Figure 2b). Close inspection found that they are typical
core-corona micelles (Figure 2, inset): the dark core, accounting
for two-third of the total micellar volume, is comprised of the cross-
linked FLP-bearing block chains, which induces the micellization;
while the polystyrene block chains surround outside as the corona.
DLS and TEM results disclosed that these micelles can entirely
dissociate upon heating, but rebuild in CO2 atmosphere (Figure
This article is protected by copyright. All rights reserved.

10.1002/anie.201804034
Angewandte Chemie International Edition
COMMUNICATION
2a, pink and green dashed lines; Figure S6). This confirms the
assumption that the dative bonding between FLP polymer and
CO2 is dynamically switchable.
FLP polymers, we thus speculated that these micelles could be
used as recyclable nanocatalysts for C1 catalytic conversion. Our
design is as follows: i) P1 and P2 dissolve in toluene, and purge
with CO2 to induce the micellization; ii) Upon addition of reactants,
they can react with the highly active CO2 buried in the micelles to
yield products, while the micelles that lose the CO2 cross-linking
disassemble into free unimers; iii) After the reaction finished, we
can trigger the micellar reconstruction by a new round of CO2
stimulus; and iv) Finally, a new catalytic cycle can restart after
separating the products (Scheme 2 and Figure S9).
As a new CO2-responsive system based on FLP chemistry, we
wanted to assess its responsive parameters, such as sensitivity,
repeatability and chemical tolerance. P2 showed a strong UV-Vis
peak at 267 nm assigned to the absorption band of neutral
triphenylphosphine side groups.[15] When the P1∙CO2∙P2 adducts
formed, the positively charged triphenylphosphonium suppressed
this absorption (Figure S7). Thus, monitoring the UV-Vis spectra
change upon CO2 stimulus can determine the above parameters.
Keeping the gas flow rate constant at 0.1 mL s-1, we found that
the CO2-binding rate of FLP polymer is dependent on the CO2
partial pressure. In 1.0 atm, it is clear that the response rate was
very fast (18 s, Figure 2c, black open circle), which is 65-fold
superior to that of traditional CO2-responsive systems based on
Brønsted acid-base pair principle (>20 min).[2,4,5] Moreover, even
though reducing CO2 partial pressure to 0.1 atm, the response
rate remained below 120 s (Figure 2c, blue and green open
circles). On the other hand, this CO2-responsive behavior
exhibited satisfactory switchability. After undergoing alternating
CO2/thermal cycle over 8 times, the FLP polymer system can still
resist the responsive attenuation (Figure 2d). This, to some extent,
addresses the issue of poor reproducibility in the traditional CO2-
responsive systems. A main reason is that this FLP mechanism
arises from the covalent interactions between the FLP moieties in
polymer and CO2, circumventing the limit of acid-base chemical
equilibrium. In addition, we inspected the applicability of this FLP
polymer in different solvents. As expected, it is insusceptible to
environmental polarity, allowing to rapidly bind CO2 regardless if
in low polar tetrachloromethane, moderate polar chloroform and
high polarity of dimethyl formamide (Figure S8).
Table 1. Catalytic formylation of N-H bonds using CO2-bound FLP polymer
nanocatalysts toward a variety of amine substrates.
Scheme 2. Schematic illustration of FLP polymer nanoparticles as recyclable
nanocatalysts: Step i) FLP polymer micellization triggered by CO2; Step ii) highly
active CO2 bound nanoparticles catalyse the reactant for C1 catalytic utilization;
Step iii) applying a new round of CO2 gas stimulus to induce re-micellization of
FLP polymer; Step iv) separating products and re-dispersing the FLP polymer
nanoparticles for a new catalytic cycle.
To validate this, we investigated a typical CO2 hydrogenation
as a model catalytic reaction. Conversion of amines to their
formamides (R1R2NH→R1R2NHCHO) using CO2 and H2 is a well-
known process.[17] However, this kind of reaction requires metal
catalysts working at high temperature and pressure, and the metal
catalysts suffer from the problems of deactivation and poisoning.
Having in hand the CO2-bound FLP polymer nanoparticle system,
we then surveyed their catalytic ability for the formylation of a
Utilizing CO2 as a C1 feedstock to produce other chemicals is
one of the most crucial frontiers in sustainable chemistry.[16]
A
prerequisite to achieve this goal is the efficient activation of inert
CO2 gas. In our micelles, the CO2 species are highly activated by
This article is protected by copyright. All rights reserved.

10.1002/anie.201804034
Angewandte Chemie International Edition
COMMUNICATION
variety of N-H bonds with phenylsilane (PhSiH3) as a reductant.
Aliphatic secondary amines, such as diethyl amine, morpholine,
and piperidine, can be quantitaively converted to their formamides
with a turnover number (TON) over 12000 at room temperature
within only 3 h, as evaluated by 1H NMR and mass spectral
analysis (Entry 1-3, Table 1 and Supporting Information). Notably,
increasing the steric hindrance at nitrogen atom does not shut
down the catalytic activity. Diisopropylamine (i-Pr2NH) and tert-
butylamine (t-BuNH2) had 75% formylated yields under the same
conditions (Entry 4-5, Table 1). For the cases of aniline derivatives,
interestingly, both N-H bonds proved reactive but the bis-
formylated products accounted for relatively small. Electron-
donating group in aniline (e.g. p-methoxy) has a positive influence
on the catalytic conversion (in a shorter time, 1 h). On the contrary,
electron-withdrawing substituent group (e.g. p-nitro) impeded the
reaction (Entry 6-8, Table 1). Other less reactive and fragile N-H
bonds was also explored. The formylation of benzophenone imine
was efficient (65% yield) and the reduction of C=N bond was not
observed (Entry 9, Table 1). Similarly, utilizing the nanocatalysts
to catalyze N-heterocycles provided delightful results. Imidazole
and triazole displayed over 40% yields with a longer reactive time
(~8 h), but avoiding the hydrogenation of heterocycles (Entry 10-
11, Table 1). On the other hand, we focused on the reusability of
the nanocatalysts. By applying CO2 stimulus to regenerate the
micelles for at least 8 times, their catalytic activity towards various
categories of amines (diethyl amine, morpholine, aniline, imine
and imidazole) retained over 80% (Figure 3). Overall, this CO2-
bound FLP polymer nanoparticles can be regarded as a universal
catalytic nanoplatform for recyclable C1 conversion.
for recyclable CO2 catalytic conversion. Although there still exists
defects, merging of FLP chemistry and polymer will open a new
avenue to construct gas-responsive materials, and provide a new
outlook in sustainable C1 chemistry.
Acknowledgements
This work was finacially supported by National Natural Science
Foundation of China (21674022 and 51703034).
Keywords: carbon dioxide conversion • frustrated Lewis pair •
block copolymer • responsive polymer • nanocatalyst
[1]
[2]
a) A. Darabi, P. G. Jessop, M. F. Cunningham, Chem. Soc. Rev. 2016,
45, 4391-4436; b) S. J. Lin, P. Theato, Macromol. Rapid Commun. 2013,
34, 1118-1133.
a) Q. Yan, R. Zhou, C. K. Fu, H. J. Zhang, Y. W. Yin, J. Y. Yuan, Angew.
Chem. Int. Ed. 2011, 50, 4923-4927; Angew. Chem. 2011, 123, 5025-
5029; b) Q. Yan, Y. Zhao, Angew. Chem. Int. Ed. 2013, 52, 9948-9951;
Angew. Chem. 2013, 125, 01132-10135; c) Q. Yan, Y. Zhao, J. Am.
Chem. Soc. 2013, 135, 16300-16303.
[3]
a) T. Fang, Z. Y. Wan, M. Huo, J. Y. Yuan, Adv. Sustainable Syst. 2017,
201700051; b) P. G. Jessop, S. M. Mercer, D. J. Heldebrant, Energ.
Environ. Sci. 2012, 5, 7240-7253; c) Y. Qian, Q. Zhang, X. Q. Qiu, S. P.
Zhu, Green Chem. 2014, 16, 4963-4968; d) Y. M. Zhang, Y. D. Zhang,
C. Wang, X. F. Liu, Y. Fang, Y. J. Feng, Green Chem. 2016, 18, 392-
396.
[4]
[5]
a) H. L. Che, M. Huo, L. Peng, T. Fang, N. Liu, L. Feng, W. Yen, J. Y.
Yuan, Angew. Chem. Int. Ed. 2015, 54, 8934-8938; Angew. Chem. 2015,
127, 9062-9066; b) S. Kumar, X. Tong, Y. Dory, M. Lepage, Y. Zhao,
Chem. Commun. 2013, 49, 90-92; c) S. J. Lin, J. J. Shang, P. Theato,
ACS Macro Lett. 2018, 7, 431-436.
a) Q. Yan, H. J. Zhang, Y. Zhao, ACS Macro Lett. 2014, 3, 472-476; b)
S. J. Lin, J. J. Shang, X. X. Zhang, P. Theato, Macromol. Rapid Commun.
2018, 39, 1700313; c) L. Lei, Q. Zhang, S. X. Shi, S. P. Zhu, ACS Macro
Lett. 2016, 5, 828-832; d) A. C. Feng, J. Liang, J. Z. Ji, J. B. Dou, S. F.
Wang, J. Y. Yuan, Sci. Rep. 2016, 6, 23624.
[6]
[7]
Q. Yan, Y. Zhao, Chem. Commun. 2014, 50, 11631-11641.
a) D. W. Stephan, G. Erker, Angew. Chem. Int. Ed. 2015, 54, 6400-6441;
Angew. Chem. 2015, 127, 6498-6541; b) D .W. Stephan, Science 2016,
354, aaf7229.
[8]
[9]
D. W. Stephan, J. Am. Chem. Soc. 2015, 137, 10018-10032.
G. C. Welch, R. R. S. Juan, J. D. Masuda, D. W. Stephan, Science 2006,
314, 1124-1126.
[10] M. Wang, F. Nudelman, R. R. Matthes, M. P. Shaver, J. Am. Chem. Soc.
2017, 139, 14232-14236.
[11] a) Gutmann, V. Coord. Chem. Rev. 1976, 18, 225-255; b) I. B. Sivaev,
V. I. Bregadze, Coord. Chem. Rev. 2014, 270-271, 75-88.
Figure 3. Catalytic activities of the CO2-bound FLP polymer nanocatalysts after
many times of catalytic cycle (black column: the first catalytic cycle; red column:
the eighth catalytic cycle) with a various categories of amine substrates.
[12] a) B. Wrackmeyer, Annu. Rep. NMR Spectro. 1988, 20, 61-200; b) F.
Focante, R. Leardini, A. Mazzanti, P. Mercandelli, D. Nanni,
Organometallics 2006, 25, 2166-2172.
[13] T. Beringhelli, D. Donghi, D. Maggioni, G. D’Alfonso, Coord. Chem. Rev.
2008, 252, 2292-2313.
Conventional CO2-sensitive systems have to rely on Brønsted
acid-base equilibrium principle. This work has broken through this
limit and demonstrated a new generation of CO2-responsive
system based on frustrated Lewis pair (FLP) principle. Introducing
complementary FLP units into polymer, CO2 gas can bridge the
polymer chains to achieve CO2-triggered micellization. This dative
bonding between CO2 and FLP units enables the polymers to
respond to CO2 in ultrafast speed, excellent reversibility and wide
applicability. Moreover, since CO2 bound in the micelles are highly
active, the FLP polymer nanoparticles can act as nanocatalysts
[14] C. M. Mömming, E. Otten, G. Kehr, R. Fröhlich, S. Grimme, D. W.
Stephan, G. Erker, Angew. Chem. Int. Ed. 2009, 48, 6643-6646; Angew.
Chem. 2009, 121, 6770-6773.
[15] S. M. Habibi Khorassani, M. T. Maghsoodlou, A. Ebrahimi, M.
Zakarianejad, M. Fattahi, J. Solution Chem. 2007, 36, 1117-1127.
[16] a) J. Artz, T. E. Müller, K. Thenert, Chem. Rev. 2018, 118, 434-504; b)
Q. Liu, L. P. Wu, R. Jackstell, M. Beller, Nat. Commun. 2015, 6, 5933.
[17] a) S. Schreiner, J. Y. Yu, L. Vaska, J. Chem. Soc., Chem. Commun. 1988,
602; b) F. C. Liu, M. B. Abrams, R. T. Baker, W. Tumas, Chem. Commun.
This article is protected by copyright. All rights reserved.

10.1002/anie.201804034
Angewandte Chemie International Edition
COMMUNICATION
2001, 5, 433-434; c) Y. Kayaki, Y. Shimokawatoko, T. Ikariya, Adv. Synth.
Catal. 2003, 345, 175-179; d) C. Federesel, A. Boddien, R. Jackstell, R.
Jennerjahn, P. J. Dyson, R. Scopelliti, G. Laurenczy, M. Beller, Angew.
Chem. Int. Ed. 2010, 49, 9777-9780; Angew. Chem. 2010, 122, 9971-
9974.
This article is protected by copyright. All rights reserved.

10.1002/anie.201804034
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 2:
COMMUNICATION
L. Chen, R. J. Liu, Q. Yan*
Page No. – Page No.
Polymer Meets Frustrated Lewis Pair:
Second-Generation CO2-Responsive
Nanosystem for Sustainable CO2
Conversion
CO2 bridged micelles: A second-generation CO2-responsive polymer system was
developed based on frustrate Lewis pair principle. Borane-containing Lewis acidic
polymer and phosphine-containing Lewis basic polymer can dynamically bind CO2,
inducing CO2-triggered micellization. The micelles bound highly active CO2 species
can function as universal nanocatalysts for recyclable C1 conversion.
This article is protected by copyright. All rights reserved.