Covalent Adaptable Networks with Tunable Exchange Rates Based on Reversible Thiol–yne Cross-Linking

Article abstract of DOI:10.1002/anie.201912902

The design of covalent adaptable networks (CANs) relies on the ability to trigger the rearrangement of bonds within a polymer network. Simple activated alkynes are now used as versatile reversible cross-linkers for thiols. The click-like thiol–yne cross-linking reaction readily enables network synthesis from polythiols through a double Michael addition with a reversible and tunable second addition step. The resulting thioacetal cross-linking moieties are robust but dynamic linkages. A series of different activated alkynes have been synthesized and systematically probed for their ability to produce dynamic thioacetal linkages, both in kinetic studies of small molecule models, as well as in stress relaxation and creep measurements on thiol–yne-based CANs. The results are further rationalized by DFT calculations, showing that the bond exchange rates can be significantly influenced by the choice of the activated alkyne cross-linker.

Full text of DOI:10.1002/anie.201912902

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Covalent adaptable networks with tunable exchange rates based
on reversible thiol-yne cross-linking
Authors: Niels Van Herck, Diederick Maes, Kamil Unal, Marc Guerre,
Johan M. Winne, and Filip Eduard Du Prez
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.201912902
Angew. Chem. 10.1002/ange.201912902
Link to VoR: http://dx.doi.org/10.1002/anie.201912902
http://dx.doi.org/10.1002/ange.201912902

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
Covalent adaptable networks with tunable exchange rates based
on reversible thiol-yne cross-linking
Niels Van Herck,[a] Diederick Maes,[a] Kamil Unal,[a,b] Marc Guerre,[a,c] Johan M. Winne,*[b] and Filip E.
Du Prez*[a]
Abstract: The design of covalent adaptable networks (CANs) relies
on the ability to trigger the rearrangement of bonds within a polymer
network. Here, we introduce simple activated alkynes as versatile
reversible cross-linkers for thiol functionalities. The click-like thiol-yne
cross-linking reaction readily enables network synthesis from
polythiols, through a double Michael addition with a reversible and
tunable second addition step. The resulting thioacetal cross-linking
moieties serve as robust but dynamic linkages, offering a versatile
platform for the development of CANs. A series of different activated
alkynes have been synthesized and systematically probed for their
ability to produce dynamic thioacetal linkages, both in kinetic studies
of small molecule models, as well as in stress relaxation and creep
measurements on thiol-yne-based CANs. The results are further
rationalized by DFT calculations and show that the bond exchange
rates can be significantly influenced by the choice of the activated
alkyne cross-linker.
bond-breaking event happens prior to the formation of a new bond
(i.e., an elimination-addition mechanism), whereas bond
exchange in permanent dynamic networks, also called vitrimers
or associative CANs, is achieved through a bond-forming/bond-
breaking sequence (i.e., an addition-elimination reaction).[5,6] As a
result, dissociative CANs can show a temporary decrease in
network connectivity during the stimulus application, which may
result in a sudden decrease of viscosity, while associative CANs
or vitrimers, preserve their network integrity in all conditions and
cannot undergo a gel-to-sol transition.[7] In this regard, many
different chemistries have now been applied or purposely
developed for CANs. For example, typical dissociative reversible
systems are based on furan-maleimide chemistry,[8] alkoxyamine
chemistry,[9] triazolinedione-indole chemistry,[10–12] phthalate
monoester transesterification,[13] etc. while typical associative
reversible systems are based on transesterification,[14] vinylogous
urethane transamidation,[15–18] transalkylation,[19,20] allyl sulfide
exchange,[21] boroxine exchange,[22] etc.
Introduction
As part of our group’s continuing interest towards CANs and the
exploration and development of new reversible chemistry
platforms for their design, we were inspired by the base-catalyzed
dynamic equilibria of thiols and activated alkenes. Although thiol-
ene reactions are long-known in polymer chemistry as click-like
bond forming reactions, their reversible nature has only recently
been explored in polymer chemistry by Konkolewicz and
coworkers (Scheme 1a).[23] Further inspired by recent work by
Kalow and coworkers using thioacetals as versatile dynamic
cross-links (Scheme 1b),[24] we became intrigued by the possibility
of using simple activated alkynes, such as alkynones, directly as
cross-linkers for thiol-based monomers and polymers, in a
straightforward way resulting in possibly highly dynamic thioacetal
linkages, starting from very simple precursors (Scheme 1c).
Covalent adaptable networks (CANs) are polymer materials that
combine the strength of covalently crosslinked materials, such as
high chemical resistance and dimensional stability, with the
reprocessability of thermoplastic polymers.[1–4] These responsive
polymer networks have the ability to reversibly rearrange their
network topology through the action of dynamic exchanges. Upon
application of an external stimulus (e.g., heat, light, etc.), the
dynamic chemistry is activated to allow bond exchanges on a
molecular level, which enables network plasticity and allows
reshaping, reprocessing, stress relaxation, self-healing and
imprinting.
An important characteristic of CANs is the underlying bond
exchange mechanism as it ultimately determines the mechanical
properties of a CAN during the applied stimulus.[3] Reversible
polymer networks or so-called dissociative CANs imply that a
[a]
N. Van Herck, D. Maes, K. Unal, Dr. M. Guerre, Prof. Dr. F. E. Du
Prez
Polymer Chemistry Research Group, Center of Macromolecular
Chemistry (CMaC), Departement of Organic and Macromolecular
Chemistry, Faculty of Sciences, Ghent University
Krijgslaan 281 (S4-bis), 9000 Ghent, Belgium
E-mail: Filip.DuPrez@UGent.be
[b]
[c]
K. Unal, Prof. Dr. J. M. Winne
Laboratory for Organic Synthesis, Department of Organic and
Macromolecular Chemistry, Faculty of Sciences, Ghent University
Krijgslaan 281 (S4-bis), 9000 Ghent, Belgium
E-mail: Johan.Winne@UGent.be
Scheme 1. Overview of dynamic thiol exchange chemistries for polymer
networks: (a) Reversible thiol-ene Michael reaction of thiols and activated
alkenes, (b) Ketene thioacetal exchanges through reversible Michael additions
to a Meldrum’s acid derivative, (c) Investigated reversible thiol-yne Michael
reactions of thiols and activated alkynes.
Dr. M. Guerre
Laboratoire des IMRCP, Université de Toulouse, CNRS UMR 5623,
Université Paul Sabatier, 118 route de Narbonne, 31062 Toulouse
Cedex 9, France
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.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
In fact, the base-promoted dynamics of alkynone-derived
thioacetals (cf. Scheme 1c) was investigated on small molecule
level by Anslyn and coworkers in 2012.[25] In this encouraging
prior study, it was shown that the dynamic thiol exchange of
thioacetals via retro-Michael reactions (vinylogous thioesters)
was possible, and can thus likely be tuned for interesting
applications in the context of dynamic combinatorial libraries.[26]
Recently, there has been a surge in reports on CANs that rely on
can then result in a thioacetal derivative of X (TA-X, Scheme 2).
The basic catalyst promotes deprotonation of the thiol to its more
nucleophilic thiolate form, which performs the first Michael
addition ((1), Scheme 2) on the electron-deficient triple bond of
the alkynone. This can yield both isomers (E/Z) of the b-sulfido-
a,b-unsaturated carbonyl or vinylogous thioester derivative of the
alkynone X (VTE-X).[25,34] Subsequently, this initial Michael adduct
can act again as a Michael acceptor to allow a second thiol
addition ((2), Scheme 2). In previous studies, this second addition
has been shown to be up to 1000 times slower than the first
addition.[35] The resulting final product, the thioacetal derivative
TA-X, can nevertheless be obtained in very high overall yields
(Table 1), depending on the used catalyst.[32,35–37] This double-
addition reaction pathway was first verified and monitored by
1H NMR spectra for the reaction between a simple alkynone, but-
3-yn-2-one (X = 1, Table 1), and 2 equivalents of butane-1-thiol
(Figure S1). Using only 1 mol% of 1,5,7-triazabicyclo[4,4,0]dec-5-
ene (TBD) as a catalyst, the desired thioacetal adduct (TA-1)
could be synthesized within 10 minutes at room temperature in
chloroform with 95% yield. In these studies, TBD was chosen as
a non-nucleophilic, strong, non-volatile and lipophilic base that is
compatible with polymer formulations and their envisaged thermal
reprocessing, and also immediately gave very good results on
small molecule level.[36]
dynamic
thiol
exchange
via
reversible
thiol-Michael
chemistry,[23,27–29]
transthioesterification[30] and reversible
exchange at a Meldrum’s acid alkylidene.[24] As the currently
available thiol-based dynamic chemistries either require exotic
reagents, free thiols that are prone to side reactions (e.g.,
oxidation), long relaxation times or aqueous basic conditions to
allow exchange, we envisaged that the thiol-yne-derived dynamic
thioacetal links could thus provide
a quite valuable and
complementary addition to the growing toolbox of dynamic ‘thiol-
click’ chemistries.[31–33] The very specific two-step mechanism of
the ‘double’ Michael addition on an ynone moiety studied in this
work offers two important advantages over existing thiol-based
dynamic cross-linking chemistries. From a fundamental or pure
polymer chemistry point of view, it can be expected that both the
kinetics and the thermodynamics of bonding and debonding can
be controlled (i.e., tunability) through introduction of substituents
on key position of the activated yne, offering options to control the
rate of exchange and the degree of connectivity, thereby bridging
the standard situation of permanent versus non-permanent
dynamic networks. This chemistry thus presents a good testing
ground for new ideas, with many handles to control reactivity.
Furthermore, from a very practical or applied polymer chemistry
point of view, monofunctional ynone- or ynoate-type compounds
can be directly used as bivalent cross-linkers for thiol moieties,
making their practical application and synthesis cost less
challenging.
Scheme 2. Synthesis of a thioacetal adduct (TA-X) via the base-catalyzed
double thiol-Michael addition on alkynone precursors (X). (VTE-X) represents
the b-sulfido-a,b-unsaturated carbonyl intermediate. Reaction (2)’ corresponds
to the dynamic thiol exchange occurring at high temperature.
In this work, we thus aimed for the exploration of dynamic thiol-
yne cross-linking as a new platform for the dynamic exchange of
thioacetal linkages. The most desirable reactivity profiles in terms
of CANs were first explored using kinetic studies on model
compounds, followed by stress relaxation measurements on
material level. It was found that the exchange rate could be
controlled through the judicious choice of the alkynone (or
activated alkyne) cross-linker. The observed influence of the
alkynone structure on the exchange rate was further evaluated by
density functional theory (DFT) calculations, which confirmed and
could rationalize the experimental observations. The obtained
polymer networks were subjected to extensive mechanical,
thermal, stress relaxation and reprocessing tests to demonstrate
the desired dynamic behavior.
Even though the forward reactions to yield the thioacetal adduct
are spontaneous and thermodynamically favored, Anslyn
reported that dynamic thiol exchange can happen upon the
addition of alternative thiols and triethylamine in solution via the
suspected retro-Michael reaction ((2)’, Scheme 2).[25] As our
objective was to explore the utility of this reversible reaction in the
context of CANs with dynamic thiol-yne derived thioacetal
linkages, we first synthesized a series of small molecule TA-X
compounds following the synthetic procedure depicted above
(1 mol% TBD in chloroform). With these thioacetals in hand, the
dynamic thiol exchange was evaluated at elevated temperature
by means of kinetic studies through the monitoring of reaction
progress via 1H NMR spectroscopy.
Synthesis of model compounds
Results and Discussion
To explore the scope and possible limitations of the thiol-yne
dynamic chemistry platform for CANs, a wide series of alkynone-
type double Michael acceptors (X) was selected or prepared
(Table 1) and then evaluated for their click-like ability to reliably
form thioacetal cross-linkages (TA-X) (cf. Scheme 2). Only limited
examples of double Michael additions to alkynones can be found
in the literature, and the success of these reactions can be
The key functional monomers and cross-linkers used in this work
are activated alkyne moieties (X, Scheme 2) like conjugated
alkynones, that can serve as double Michael acceptors. Such
moieties are readily available or can be easily prepared from
simple precursors. A base-catalyzed click-reaction with two thiols
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
expected to be dependent on steric and electronic factors, where
the second Michael reaction could theoretically favor either the
VTE-X or the TA-X side. A high yield for this step is certainly
derivatives of 1-phenylprop-2-yn-1-one, which allowed the
exploration of the hypothetical control on the exchange through
introduction of electronically (de)activating para-substituents.
possible, but is thus not guaranteed (e.g., see entry where X = 12). Alkynones 10-12 were chosen to evaluate the effect of
The substituents directly bound to the alkyne function (Scheme 2,
R1), as well as the substituents directly bound to the carbonyl
function (R2) were varied with the intent to probe both steric and
electronic effects on the dynamic exchange. In this series of
alkynones, but-3-yn-2-one (X = 1) was arbitrarily selected as the
benchmark to which all other alkynones will be compared.
Alkynones 2, 13-15 were chosen to evaluate the effect of
replacing the ketone functionality with an ester, amide,
sulfonamide or sulfone, respectively, as these are also common
moieties in polymer chemistry. Alkynones 3-9 represent
substituents directly bound to the alkyne function. All alkynones
were synthesized according to known literature procedures[38–43]
(see SI-file), except for alkynones 1, 2 and 12, which are
commercially available.
In a next step, the dynamics of the most readily obtained model
compounds (TA-X) were evaluated via kinetic exchange studies.
Table 1 shows overall very good yields for the first thiol-Michael
addition to yield VTE-X compounds. For VTE-4, VTE-8 and VTE-9,
the isolated yield was lower than expected only because of the
very swift second Michael addition reaction, directly generating
the double thiol-Michael product TA-X. However, the activated
alkyne substrates 10-14 showed an incomplete second addition
and were not further considered during the kinetic model studies
or network synthesis, as they cannot serve as a thiol-yne-type
crosslinker. This lower efficiency of the second thiol Michael
reaction can be related to electronic and steric considerations.
Thus, only the alkynones that gave highly efficient double thiol-
Michael additions with butane-1-thiol were further considered for
our current investigations (see SI-file).
Table 1. Overview of investigated alkynones (X) with their corresponding
experimental data on model compound and material level. ꢀꢁ: activation energy.
Entry
X
VTE-X[a]
TA-X[b]
Model study
Rheology
Yield (%)
Yield (%)
ꢀꢁ (kJ/mol)
ꢀꢁ (kJ/mol)
O
1
2
99
99
95
94
76.8 ± 4.4
137.4 ± 1.8
197.8 ± 2.5
O
[d]
O
/
O
O
O
3
4
99
81
92
71.0 ± 4.9
76.2 ± 2.1
128.4 ± 0.3
134.1 ± 0.6
Kinetic Model Studies
To evaluate the dynamic behavior of the synthesized thioacetal
adducts (TA-X), stock solutions containing TA-X, 1 mol% TBD
and an internal standard in deuterated benzene were prepared,
followed by the addition of a fivefold excess of benzyl mercaptan
as a competitive thiol to approach pseudo first-order conditions
for the initial exchange rates (see SI for experimental details).
During the exchange reaction, the butanethiol groups of the
respective model compound were found to be gradually replaced
by benzyl mercaptan groups, leading to the formation of a mixture
of three different thioacetals: TA-XAA, TA-XAB, TA-XBB, where AA,
AB and BB describe the alkyl groups present in the thioacetal
adduct: butane/butane, butane/benzyl and benzyl/benzyl
respectively (Figure 1a for TA-1). The exchange reaction was
63[c]
5
6
7
8
99
99
81
93
88
95
74.4 ± 3.8
75.4 ± 2.6
66.2 ± 14.9
66.2 ± 1.3
178.6 ± 1.1
186.5 ± 6.8
133.5 ± 1.0
124.3 ± 1.1
OCH3
N(CH3
Cl
O
)
2
O
O
O
92
83[c]
CF3
9
35[c]
99
97
25
0
75.1[e]
128.0 ± 1.1
1
NO2
monitored by following the characteristic H NMR signal of the
O
original thioacetal motif (TA-XAA, triplet at 4.35-4.50 ppm), which
decreased in intensity, while the new thioacetal signals of the
10
11
-
-
-
-
O
exchanged products (TA-XAB
,
TA-XBB
)
emerged upfield
93
(Figure 1b). The remaining molar fraction of TA-XAA was followed
over time to determine the rate constant ꢂ. This was repeated at
different temperatures comprised between 40 and 70 °C, which
allowed creating an Arrhenius plot from which the activation
energy ꢀꢁ could be deduced as the slope of the linear fit (see SI-
file).
O
12
99
34
-
-
O
N
H
13
14
99
99
27
6
-
-
-
-
O
S
O
N
H
O
O
S
[d]
15
99
85
/
-
[a] Yield after addition of 1 eq thiol. [b] Yield after addition of 2 eq thiol. [c]
Isolated yield of VTE-X was lower than expected as a result of overreaction,
generating the double thiol-Michael product TA-X. [d] ꢀꢁ could not be
determined since no exchange was observed in the investigated temperature
range. [e] ꢀꢁ could not be determined in a reliable way since exchange was too
fast to collect kinetic data above 40 °C.
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
TA-1BB
TA-1AB
a)
b)
significantly, whereas electron donating substituents showed the
reverse effect (i.e., lower rate constants). The fastest exchange
was observed for TA-9, where the introduction of a nitro group (-
NO2) allowed complete exchange within 2 minutes at 40 °C, even
causing problematic sampling at higher temperatures. Conversely,
the slowest exchange was observed for the analogous compound
TA-6, where the only difference is the introduction of an electron-
donating dimethylamino group (-N(CH3)2), showing only 50%
conversion after 4 hours at 40 °C. The influence of para-
substituents on the exchange rate was also confirmed through
Hammett analysis (Figure S2), which showed a strong positive
linear correlation indicating an exchange mechanism via anionic
intermediates.[44,45]
S
O
SH
+
TA-1AA
180 min
97 min
40 min
15 min
7 min
4 min
2 min
1 min
0 min
S
(5 eq)
TA-1AA
C6D6
TBD
ꢀT
S
O
S
O
S
S
TA-1AA
TA-1AB
S
O
S
ꢀꢁꢆ
ꢀꢁꢅ
ꢀꢁꢀ
ꢀꢁꢄ
ꢀꢁꢃ
ꢀꢁꢂ
TA-1BB
ꢇꢈꢉꢊꢋꢌꢍꢎꢏꢐꢈꢋꢑꢒꢏꢓꢔꢔꢊꢕ
c)
d)
Another interesting observation was that model compounds TA-2
and TA-15 did not show exchange at the investigated temperature
and time scale. Hence, they could be considered as practically
irreversible, owing to the presence of the intrinsically less electron
withdrawing ester or sulfone functional group. As intermediate
conclusion, these results already indicated that the choice of the
alkynone precursor is quite important with respect to the
reversible nature of the thioacetal motif. Moreover, these
experiments showed that the exchange rate could be significantly
manipulated by means of the electronic substituents on the
alkynone precursor.
Figure 1. (a) Dynamic thiol exchange of TA-1 at elevated temperature with
benzyl mercaptan, leading to 3 different exchange products: TA-1AA, TA-1AB,
TA-1BB. (b) Evolution of characteristic 1H NMR signals of the thioacetal motifs
over time. (c) Overlay of remaining molar fraction of TA-X as a function of time
during the exchange experiment at 40 °C. Dashed lines represent a fit with
exponential decay. (d) Arrhenius plots of the reversible model compounds.
Dashed lines represent a linear fit (the same color legend as in (c) applies).
Chemical formulas represent the para-substituents of 1-phenylprop-2-yn-1-one-
derived compounds (TA-X, with X = 3-9).
For the design of CANs, the nature of the bond exchange
mechanism is a key parameter since it dictates the viscoelastic
profile of the resulting materials. The mechanism will also rely on
the conditions and stoichiometry, and the reaction conditions for
the model experiments described above may thus be a poor
substitute for those found in thiol-yne-based polymer networks,
where there may not be large amounts of free thiols present. Thus,
to probe the thioacetal exchange in a more realistic setting, we
also investigated whether the bond exchange can happen in the
absence of free thiols. Therefore, two different thioacetal adducts
were synthesized by reacting alkynone 1 with either butane-1-
thiol (TA-1AA) or benzyl mercaptan (TA-1BB). In a next step, these
model compounds were mixed in a 1:1 stoichiometric ratio in
deuterated benzene in the presence of TBD and heated to 70 °C.
1H NMR analysis revealed the gradual appearance of the mixed
thioacetal moiety (i.e., TA-1AB) over time, which indicates that no
free thiols are required to allow exchange (Figure S3). However,
the exchange was found to proceed slower compared to the
model experiments where an excess of free thiols was present.
This result indicates that, while exchange rates are already fast at
room temperature in our small molecule models, the situation
inside a polymer network may be quite different as the exchange
will also rely on statistics and diffusion limitations, therefore not
directly reflecting the intrinsic chemical reaction rate.
Figure 1c shows an overlay of the remaining molar fractions of
TA-X as a function of time during the exchange experiments at
40 °C. Interestingly, the exchange kinetics were significantly
influenced by the used model compound. Compared to the
benchmark TA-1, some model compounds showed a remarkably
faster dynamic thiol exchange, whereas others showed slower or
even no exchange. From the Arrhenius plots in Figure 1d, similar
activation energies ꢀꢁ within the range of 66-77 kJ/mol were
obtained for all reactions (Table 1). This could be expected since
the exchange mechanism should be unchanged regardless of the
used model compound. Nevertheless, a small trend could be
observed in the data for the series of phenylpropynone-based
model compounds (TA-X, with X = 3-9). When the activation
energy of TA-3 (no substituent on phenyl group) was taken as a
reference (ꢀꢁ = 71 kJ/mol), the introduction of electron donating
para-substituents (X = 4-6) added a penalty of 3-5 kJ/mol to the
energy barrier, whereas electron withdrawing para-substituents
(X = 7-9) reduced this barrier by 5 kJ/mol. As can be expected,
the anionic transition state, i.e., when a thiolate anion is attacking
the Michael acceptor, will be selectively stabilized over the neutral
ground states, by the presence of electron withdrawing
substituents on the aromatic ring conjugated to the receiving
carbonyl function. However, considering that the small difference
in ꢀꢁ is within experimental error for some model compounds, this
interpretation of the temperature dependencies of the exchange
rates remains speculative.
DFT Calculations
Intrigued by the remarkable effect of the used alkynone precursor
on the exchange kinetics, also density functional theory
calculations were performed to confirm these observations. The
Gaussian 16 package was used with the M06-2X/6-311+G(d,p)
level of theory, which has been shown successful to describe the
thermodynamics of the thiol-Michael addition (see SI for
computational details).[46] Using this theoretical method, the
Nevertheless, the same trend in electronic activation is found to
have a major effect on the actual exchange rates, spanning
several orders of magnitude, as can be clearly seen in the
Arrhenius plot (Figure 1d). It is very clear from these data that
electron withdrawing substituents accelerate the exchange
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
calculated Gibbs free energy profile of the double thiol-Michael
addition on the alkynone compound (X) could be modeled (see
Figure S24-S37 and Table S2). The most crucial parameter
coming out of this calculational approach is the relative Gibbs free
energy of the transition state of the rate limiting step of the
pathway (∆ꢃ°,‡), referenced to the energetic ground state of the
reactants and products.
Our DFT model was unable to provide a rationalization for the
distinct difference in reactivity observed for the seemingly non-
reversible adducts TA-2 and TA-15, as comparable energy
profiles were in fact obtained here (see SI). Thus, more advanced
models may be required here.
Synthesis of CANs
Based on the outcome of the small molecule model study, the
most interesting alkynones were selected for further investigation
on a material’s level. Alkynones 1 and 3-9 were expected to result
in reversible cross-links in a dynamic polymer network, whereas
alkynone 2 was assumed to give an irreversible material. Since
alkynones have a functionality (ꢄ) of 2, the combination of an
activated alkyne with any multifunctional thiol with ꢄ ≥ 3 should
directly result in a crosslinked polymer network. Originally,
reaction of the alkynones with the commercially available
pentaerythritol tetrakis(3-mercaptopropionate) was considered
for the synthesis of polymer networks. However, this
tetrafunctional thiol contains ester bonds, which are known to
exchange via transesterification reactions in the presence of
TBD.[47] Therefore, this thiol was not suited for unambiguous
design of dynamic materials that are solely relying on dynamic
thiol exchange. Hence, an aliphatic trifunctional thiol (trithiol)[48]
was synthesized and used as a reference building block to
produce dynamic polymer materials.
In the graphical summary of these calculations (Figure 2), it can
be seen that all thiol-Michael additions to the VTE-X intermediates
are exergonic, with a thermodynamic driving force between only
15-20 kJ/mol. This readily indicates and confirms the reversible
nature of this second bond formation step. The thermodynamic
profiling of the individual reaction pathways is more challenging
and subject to interpretation, but the relative Gibbs free energy
difference for the modeled transition states, with reference to the
ground states of the thioacetal adducts, indeed closely matches
the trends observed in our kinetic experiments. Indeed, thiol
exchange could only occur when the energetic barrier of ∆ꢃ°,‡ is
overcome to allow dissociation of the thioacetal (TA-X) into the
vinylogous thioester (VTE-X) and thiolate anion. Therefore, this
parameter is closely related to the activation energy ꢀꢁ as
experimentally determined earlier during kinetic model studies.
‡
S
O
259.66
-N(CH3)2
S
R
TS
Thiol-yne polymer networks were prepared through reaction of
the alkynone of interest with trithiol in the presence of TBD (see
Scheme 3 and SI for experimental details). First, a vinylogous
thioester prepolymer (VTEp-X) was prepared using 1 equivalent
of trithiol with 3 equivalents of alkynone. In a second step, another
equivalent of trithiol was added to the prepolymer to produce and
fully cure the polymer network (CAN-X). As a result of this two-
phased protocol, also the relatively high exotherm associated with
the double thiol-Michael addition could be better controlled,
resulting in well-defined and reproducible network formation at
completely stoichiometric conditions, theoretically containing no
free thiols. The prepared networks were transparent and
displayed a slight coloration depending on the used alkynone
(Figure S4).
-OCH3
-CH3
-H
246.12
245.44
242.29
233.92
220.82
-Cl
-NO2
ΔG°,‡
SH
+
O
S
R
S
O
VTE-X + MeSH
21.21
S
R
0.0 TA-X
16.23
ΔΔG°
Figure 2. Calculated free energy diagram for the second thiol-Michael addition
of methanethiol to VTE-X (with X = 3-9) with the formation of the TA-X product
via the corresponding transition state (TS). The zoomed section shows the
difference in ∆G°,‡
.
Interestingly, these calculations also provided a means to
rationalize the incomplete formation of the desired thioacetal
adducts from alkynones 10 and 13. In contrast to all other
modeled VTE-X, the DFT-derived reaction Gibbs free energy for
thiol addition to VTE-10 was found to favor the debonded state
(see Table S2). Figure S33 shows that the reaction Gibbs free
energy for 10 is calculated as slightly endergonic (∆∆ꢃ1°0 = 0.61
kJ/mol), indeed predicting an equilibrium mixture of VTE-10 and
TA-10, as is experimentally observed. The same reasoning could
not be applied to explain the incomplete formation of TA-13, as
the reaction itself was found to be clearly exergonic by our
calculation. However, the low yield for TA-13 could be attributed
to the higher forward energetic barrier ( ∆ꢃ°,‡ ), which may
substantially slow down the second Michael reaction. The latter
was also observed experimentally, as only 27% conversion of
TA-13 was reached after 1 week at room temperature.
Scheme 3. Network synthesis via the reaction of trithiol (1 eq) with alkynone (3
eq) in the presence of TBD to yield the vinylogous thioester prepolymer (VTEp-
X), followed by addition of trithiol (1 eq) to produce a thioacetal cross-linked
network (CAN-X).
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
High-resolution magic angle spinning (HRMAS) 1H NMR was
used to confirm the presence of the thioacetal motif in the
prepared polymer networks (Figure S5). Deuterated chloroform
was selected as a non-reactive solvent providing very good
swelling. Using this technique, a conversion of 85-90% could be
estimated (residual VTE-signals integrating for ~10%), which is
presumably caused by diffusion limitations in these highly cross-
linked networks. Thus, the degree of polymerization should be
well above the theoretical gel point (i.e., 71%) according to the
Flory-Stockmayer theory for a trifunctional (A3) and difunctional
monomer (B2).[49,50] In order to further confirm network formation,
swelling experiments were performed in tetrahydrofuran (THF) at
room temperature to determine the soluble fraction for CAN-1 and
CAN-3 as representative networks (Table S1). Those networks
showed a swelling ratio between 120-300% and a soluble fraction
below 3%.
temperatures. Typical stress relaxation curves are presented in
Figure 4a for CAN-1 and in Figures S6-S13 for all other CANs. As
a result of dynamic bond rearrangement at high temperatures, full
stress relaxation of the networks could be obtained. Next,
relaxation times ( ꢊ ) were determined at 37% (1/e) of the
normalized relaxation modulus (ꢃ/ꢃꢋ) according to the Maxwell
model for viscoelastic fluids and plotted in an Arrhenius plot to
obtain the corresponding viscous flow activation energy ꢀꢁ of 137
kJ/mol for CAN-1 (Figure 4b).[51]
The relaxation measurements were performed below 150 °C for
most networks (except CAN-1 and CAN-2) to prevent
degradation of the TBD catalyst.[52] In the case of CAN-1, an
upper temperature of 170 °C was used to increase the number of
data points in the Arrhenius plot. For CAN-2, which was actually
prepared as a reference material with the expectation of being an
irreversible network, a slight stress relaxation was observed at
150 °C (Figure S6), which encouraged us to explore the limits of
this material. Therefore, relaxation measurements were
performed until an upper temperature as high as 240 °C to probe
the possible bond exchange via VTE-2-type moieties (vide supra).
Indeed, full stress relaxation could be observed if enough energy
was added to the system (Figure S6).
Thermal analysis of CANs
Thermogravimetric analysis (TGA) of the prepared polymer
networks demonstrated a single decomposition step with a
maximum degradation rate at ±330 °C for most of the networks
(Figure 3a). CAN-4 and CAN-6 also showed an earlier
decomposition step with an onset (5% weight loss) degradation
temperature of ꢅꢆꢇ% = 176 and 149 °C, respectively.
Differential scanning calorimetry (DSC) revealed the presence of
b)
d)
O
a)
CAN-1
S
a glass transition temperature (ꢅ ) that varied within the range of
ꢈ
O
-16 to 52 °C depending on the used alkynone precursor
(Figure 3b). Since all networks were prepared in the same way,
this variable rigidity was attributed to the presence of freely
rotating groups and conjugated ꢉ -systems in the alkynone
S
1
#
"
compound. An increased ꢅ was observed when a larger
ꢈ
conjugated system was present in the matrix (cf. CAN-3 vs
CAN-9). Interestingly, the simple presence of a (freely rotating)
methoxy group in CAN-5 resulted in a very pronounced decrease
c)
of ꢅ . Besides the glass transition, an endotherm starting at
ꢈ
120 °C was observed, which was attributed to the activation of the
retro-Michael reaction at this temperature.
a)
b)
e)
O
O
O
O
O
O
O
O
O
<
<
<
<
<
<
<
<
Cl
CF3
NO2
N(CH3
)
OCH3
2
7
8
9
4
3
6
5
1
2
Figure 4. (a) Stress relaxation experiment of CAN-1 between 130 and 170 °C.
Relaxation times (ꢊ) were determined at G/G0 = 1/e. (b) Arrhenius plot of the
obtained relaxation times with linear fit according to Arrhenius equation and the
corresponding viscous flow activation energy (ꢀꢁ). (c) Overview of Arrhenius
plots with linear fits for all networks. Values in brackets represent the activation
energy in kJ/mol. (d) Overview of stress relaxation curves at 130 °C for all
networks. Values in brackets represent the relaxation time (at 1/e) in seconds
at 130 °C. For CAN-6 a Maxwell fit (dashed line) was created for clear
comparison. (e) Ranking of CANs with regard to relaxation time at 130 °C.
Figure 3. Thermal analysis of CANs. (a) Overlay of TGA thermograms. Values
in brackets are the onset degradation temperatures at 5% weight loss (ꢅꢆꢇ%).
(b) Overlay of DSC thermograms (2nd heating). The onset ꢅ is displayed above
ꢈ
each trace.
Rheological analysis of CANs
An overview of all Arrhenius plots is presented in Figure 4c. With
the exception of CAN-6, all networks exhibited a linear correlation
of ln( ꢊ ) with 1000/T in the observed temperature window,
indicating Arrhenius flow characteristics. The unusual stress
relaxation curves of CAN-6 (Figure S10) were attributed to the
To examine the viscoelastic properties of the dynamic networks,
stress relaxation experiments were performed. During these
measurements, a constant strain of 1% was applied and the
relaxation modulus was followed as a function of time at different
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
absence of a rubbery plateau within this temperature range, which
was confirmed by temperature sweep measurements (Figure
S14). The Arrhenius plot of CAN-2 is situated on the far left of
Figure 4c, as very high temperatures were required to allow
relaxation within a reasonable time frame. CAN-1 and CAN-3-9
showed relaxation in the same temperature range, however, the
Arrhenius plot of CAN-1 was shifted to significantly longer
relaxation times at a certain temperature compared to the
phenylpropynone-derived materials. Similar activation energies in
the range of 124-137 kJ/mol were observed for CAN-1,3,4,7-9,
whereas higher activation energies in the range of 179-198 kJ/mol
were observed for CAN-2,5,6. This difference is expected to
originate from the presence of electronically deactivating groups
(ester functionality in CAN-2 or electron donating para-
substituents in CAN-5,6). Figure 4d displays an overview of stress
relaxation curves at 130 °C of the networks to visualize the
difference in relaxation times. Besides CAN-2 which did not show
relaxation at 130 °C, CAN-1 showed the slowest relaxation at
130 °C (5030 s), whereas the fastest relaxation was observed for
CAN-7 (64 s).
dependent decreasing plateau modulus was observed, nicely
confirming the dissociative nature of the dynamic bond exchange.
It is clear that in spite of this decrosslinking, none of the CANs,
with exception of CAN-6 (Figure S10), go through a gel-to-sol
transition within the probed temperature range, and there is thus
no sudden drop in viscosity in the stress relaxation experiments.
a)
b)
CAN-1
CAN-3
100°C
110°C
120°C
130°C
100°C
110°C
120°C
130°C
140°C
140°C
150°C
160°C
170°C
150°C
Figure 5. Frequency sweep measurements of (a) CAN-1 and (b) CAN-3. Full
circles: G’, empty circles: G”.
Our small molecule models showed fast bond exchanges already
happening at room temperature for most alkynones. At the other
hand, the macroscopic exchange kinetics in the thiol-yne
networks was much slower, with a much more pronounced
temperature dependence (i.e., activation energy). To study the
network integrity of the prepared CANs at lower to ambient
temperatures, creep experiments are most suitable. These were
performed on CAN-1 and CAN-3 as representative networks
(Figure S15a,b). Both networks showed full shape recovery after
deformations at 60 and 80°C. However, a permanent deformation
of both materials became prominent at temperatures above
100 °C, indicating activation of the dynamic chemistry. In line with
previous observations in the stress relaxation experiments, the
creep resistance of CAN-1 was better than the one of CAN-3 as
a result of slower exchange. Also here, the relation between the
The trends in exchange rate for different alkynone crosslinkers,
as determined from kinetic model studies and DFT calculations,
were not exactly reproduced on material level (Figure 4e).
Moreover, the difference in stress relaxation times between
phenylpropynone-derived materials (CAN-3-9) was not as
pronounced as could be anticipated based on the remarkable
differences in chemical relaxation times (Figure 1d). Nevertheless,
compared to the unsubstituted reference (CAN-3), electron
withdrawing para-substituents still showed faster relaxation
(CAN-7,8,9), whereas electron donating para-substituents
exhibited slower relaxation (CAN-5,6), conserving the overall
trends. Only CAN-4 did not follow this trend and showed a faster
relaxation than the unsubstituted reference (CAN-3), but this
effect can be attributed to an increased mobility in the matrix as
also shown during DSC analysis (ꢅꢈ, CAN-3 > ꢅꢈ, CAN-4). Similarly,
creep rate (ꢍ̇) and temperature (ꢅ) was fitted to the Arrhenius
the difference in ꢅ could also explain the discrepancy in the
ꢈ
equation to retrieve the corresponding activation energy (ꢀꢁ) from
the slope of the linear fit (Figure S16).[16] The determined ꢀꢁ was
similar for both CANs, but ca. 30 kJ/mol lower than the ꢀꢁ
obtained during stress relaxation experiments.
ranking of CAN-7,8,9, where the high ꢅ of CAN-9 indicates a
ꢈ
restricted mobility, resulting in a slower relaxation time compared
to CAN-7,8. These observations highlighted again the fact that
exchange on a material level is not only controlled by the kinetics
of dynamic bond exchange, but will also be substantially
influenced by matrix effects, that are not captured by small
molecule models.[3]
Reprocessability of CANs
Following the study on the viscoelastic properties of the dynamic
networks, the mechanical properties of these materials upon
reprocessing were analyzed. For this purpose, CAN-1 and CAN-3
were subjected to 5 cutting/reprocessing cycles using thermal
compression molding at 140 °C for 30 minutes while monitoring
their properties. Figure 6a,c shows the stress-strain curves of the
original materials and after each reprocessing step. Since CAN-1
It is important to note that in the absence of an associative bond
exchange mechanism, the prepared CANs cannot be classified
as vitrimers, despite the observed linear relationship of relaxation
times and temperature in the Arrhenius plots (Figure 4c). To
illustrate the dissociative bond exchange mechanism operating in
dynamic thiol-yne networks, frequency sweep measurements
were performed at different temperatures. Whereas associative
CANs or vitrimers are characterized by a constant plateau
modulus ꢃꢌ at different temperatures as a result of their overall
constant and temperature-independent crosslink density,
dissociative CANs should show a decreasing ꢃꢌ with increasing
temperature as a result of a net de-crosslinking.[5,53] Figure 5
shows the frequency sweep measurements for CAN-1 and
CAN-3 as representative networks. As expected, a temperature
has a ꢅ below room temperature, the material showed low stress
ꢈ
and high strain, in contrast to CAN-3, with a ꢅ above room
ꢈ
temperature. Upon reprocessing, the elongation, stress at break,
and Young modulus remained very similar for CAN-1, proving
adequate restoration of the original properties (Figure 7b). For
CAN-3, a slight increase of the elongation and stress at break was
observed (Figure 7d). The reprocessed materials also showed
similar properties when monitored using swelling experiments
(Table S1), DSC analysis (Figure S17), Fourier transform infrared
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
spectroscopy (Figure S18) and Raman spectroscopy
(Figure S19). Although the restoration of the original properties
was successful in these recycling experiments, it was observed
that CAN-1 required slightly longer remolding times after the 3rd
reprocessing cycle in order to regenerate a sample with the same
properties, which implied some type of ageing. In an attempt to
accelerate and exaggerate thermal ageing effects upon
reprocessing, consecutive stress relaxation measurements were
performed at 170 °C on CAN-1 and CAN-3 (Figure S20-S21).
Even though full relaxation was attained for both networks, an
increased modulus upon each reprocessing cycle was observed,
which was accompanied with an increased relaxation time. This
effect can be readily rationalized by two factors, the first one being
catalyst ageing (thermal degradation) leading to slower/less bond
exchanges. Another explanation is that the networks become
more perfect after reprocessing, as the virgin materials were
found to still contain significant amounts of uncured VTE-
functions (~10%, vide supra).
phenylpropynone scaffold showed a clear trend in rate effects,
showing faster exchange rates for electron withdrawing
substituents, and slower rates for electron donating groups. This
behavior is in line with mechanistic expectations and was also
further rationalized by density functional theory calculations.
The dynamic thiol-yne platform was subsequently implemented in
materials to produce CANs. An extensive rheological analysis
was performed to probe the interdependence of chemical
structure and the dynamic network properties. A remarkable
difference was observed in macroscopic network relaxation times,
compared to the ‘chemical’ exchange rates observed in small
molecule models. Whereas the chemical exchange is already
quite dynamic at or just above room temperature in small
molecule models, the corresponding thiol-yne networks proved to
be much less reactive, also confirmed by creep experiments. This
demonstrates that overreliance on chemical model reactions for
the design of CANs may be a misleading approach. Indeed, the
thermodynamic landscape of bond exchange pathways can be
quite dramatically altered by the macromolecular matrix and
context.
a)
b)
CAN-1
E
The
prepared
dynamic
networks
showed
excellent
reprocessability and overall network integrity as proven by
mechanical and thermal characterization, showing that this new
chemistry platform could be further explored in many application
areas for which CANs are being investigated. Not only can this
reversible chemistry be used for reversible cross-linking in the
context of CANs, but also in other dynamic applications and
concepts, such as reversible functionalization, reversible
polymerization, dynamic combinatorial libraries, etc. Especially
the double functionality of the activated yne can lead to interesting
dynamic applications and new concepts. In conclusion, this
platform opens up many new avenues for further research within
the field of CANs.
ε
σ
c)
d)
CAN-3
E
ε
σ
Acknowledgements
N.V.H. and M.G. acknowledge the Research Foundation -
Flanders (Fonds Wetenschappelijk Onderzoek - Vlaanderen) for
Ph.D. and Postdoctoral fellowships. K.U. acknowledges UGent
for funding. F.D.P acknowledges UGent funding (BOF-GOA). B.
De Meyer, J. Goeman and T. Courtin are thanked for technical
support. The computational resources (Stevin Supercomputer
Infrastructure) and services used in this work were provided by
the VSC (Flemish Supercomputer Center), funded by Ghent
University, FWO and the Flemish Government - department EWI.
Figure 6. Overview of tensile strength experiments. (a) Representative stress-
strain curves of reprocessed CAN-1. (b) Young modulus (E), strain (ε) and
stress values (σ) after each reprocessing cycle of CAN-1. (c) Representative
stress-strain curves of reprocessed CAN-3. (d) Young modulus, strain and
stress values after each reprocessing cycle of CAN-3.
Conclusion
This work demonstrates the use of simple activated alkyne
species as versatile dynamic cross-linkers for thiol-yne based
polymer networks, resulting in a promising new chemical platform
for the design of covalent adaptable networks. A wide series of
alkynone precursors was explored for their click-like cross-linking
reactivity that can join two thiols into a dynamic thioacetal linkage.
Kinetic model studies revealed that exchange rates were
significantly influenced by the steric and electronic nature of the
alkyne cross-linker, with exchange rate constants spanning
Keywords: Thiol-yne • Retro reactions • Dynamic chemistry •
Covalent adaptable network • Thioacetal
several orders of magnitude. Diverse substituents on
a
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
[1]
[2]
C. J. Kloxin, C. N. Bowman, Chem. Soc. Rev. 2013, 42, 7161–7173.
G. M. Scheutz, J. J. Lessard, M. B. Sims, B. S. Sumerlin, J. Am. Chem.
Soc. 2019, jacs.9b07922.
[38] M. Zheng, F. Wu, K. Chen, S. Zhu, Org. Lett. 2016, 18, 3554–3557.
[39] G. Ieronimo, G. Palmisano, A. Maspero, A. Marzorati, L. Scapinello, N.
Masciocchi, G. Cravotto, A. Barge, M. Simonetti, K. L. Ameta, et al., Org.
Biomol. Chem. 2018, 16, 6853–6859.
[3]
[4]
J. Winne, L. Leibler, F. E. Du Prez, Polym. Chem. 2019, Accepted.
M. K. McBride, B. T. Worrell, T. Brown, L. M. Cox, N. Sowan, C. Wang,
M. Podgorski, A. M. Martinez, C. N. Bowman, Annu. Rev. Chem. Biomol.
Eng. 2019, 10, 175–198.
[40] J. M. Wilbur, B. A. Bonner, J. Polym. Sci. Part A Polym. Chem. 1990, 28,
3747–3759.
[41] V. X. Truong, A. P. Dove, Angew. Chemie Int. Ed. 2013, 52, 4132–4136.
[42] D.-J. Zhang, M.-S. Xie, G.-R. Qu, Y.-W. Gao, H.-M. Guo, Org. Lett. 2016,
18, 820–823.
[5]
[6]
C. J. Kloxin, C. N. Bowman, Chem. Soc. Rev. 2013, 42, 7161–73.
P. Chakma, D. Konkolewicz, Angew. Chemie Int. Ed. 2019, 58, 9682–
9695.
[43] Y. S. Lee, K. H. Chung, Y. H. Kim, Heteroat. Chem. 1999, 10, 461–464.
[44] Y. Zhong, Y. Xu, E. V. Anslyn, European J. Org. Chem. 2013, 5017–
5021.
[7]
[8]
W. Denissen, J. M. Winne, F. E. Du Prez, Chem. Sci. 2016, 7, 30–38.
X. Chen, M. A. Dam, K. Ono, A. Mal, H. Shen, S. R. Nutt, K. Sheran, F.
Wudl, Science 2002, 295, 1698–1702.
[45] B. Zhang, P. Chakma, M. P. Shulman, J. Ke, Z. A. Digby, D. Konkolewicz,
Org. Biomol. Chem. 2018, 16, 2725–2734.
[9]
Y. Higaki, H. Otsuka, A. Takahara, Macromolecules 2006, 39, 2121–
2125.
[46] E. H. Krenske, R. C. Petter, K. N. Houk, J. Org. Chem. 2016, 81, 11726–
11733.
[10] S. Billiet, K. De Bruycker, F. Driessen, H. Goossens, V. Van Speybroeck,
J. M. Winne, F. E. Du Prez, Nat. Chem. 2014, 6, 815–821.
[11] H. A. Houck, K. De Bruycker, C. Barner-Kowollik, J. M. Winne, F. E. Du
Prez, Macromolecules 2018, 51, 3156–3164.
[47] M. Capelot, M. M. Unterlass, F. Tournilhac, L. Leibler, ACS Macro Lett.
2012, 1, 789–792.
[48] J. Van Damme, O. van den Berg, J. Brancart, L. Vlaminck, C. Huyck, G.
Van Assche, B. Van Mele, F. Du Prez, Macromolecules 2017, 50, 1930–
1938.
[12] N. Van Herck, F. E. Du Prez, Macromolecules 2018, 51, 3405–3414.
[13] M. Delahaye, J. M. Winne, F. E. Du Prez, J. Am. Chem. Soc. 2019, 141,
15277–15287.
[49] P. J. Flory, J. Am. Chem. Soc. 1941, 63, 3083–3090.
[50] W. H. Stockmayer, J. Chem. Phys. 1943, 11, 45–55.
[51] L. H. Sperling, Introduction to Physical Polymer Science, John Wiley &
Sons, Inc., Hoboken, NJ, USA, 2005.
[14] D. Montarnal, M. Capelot, F. Tournilhac, L. Leibler, Science 2011, 334,
965–968.
[15] W. Denissen, G. Rivero, R. Nicolaÿ, L. Leibler, J. M. Winne, F. E. Du
Prez, Adv. Funct. Mater. 2015, 25, 2451–2457.
[52] C. Jehanno, I. Flores, A. P. Dove, A. J. Müller, F. Ruipérez, H. Sardon,
[16] M. Guerre, C. Taplan, R. Nicolaÿ, J. M. Winne, F. E. Du Prez, J. Am.
Chem. Soc. 2018, 140, 13272–13284.
Green Chem. 2018, 20, 1205–1212.
[53] P. J. Flory, Chem. Rev. 1944, 35, 51–75.
[17] Y. Spiesschaert, M. Guerre, L. Imbernon, J. M. Winne, F. Du Prez,
Polymer 2019, 172, 239–246.
[18] C. Taplan, M. Guerre, J. M. Winne, F. E. Du Prez, Mater. Horizons 2019,
10.1039/C9MH01062A.
[19] B. Hendriks, J. Waelkens, J. M. Winne, F. E. Du Prez, ACS Macro Lett.
2017, 6, 930–934.
[20] M. M. Obadia, A. Jourdain, P. Cassagnau, D. Montarnal, E.
Drockenmuller, Adv. Funct. Mater. 2017, 27, 1703258.
[21] T. F. Scott, A. D. Schneider, W. D. Cook, C. N. Bowman, Science 2005,
308, 1615–1617.
[22] W. A. Ogden, Z. Guan, J. Am. Chem. Soc. 2018, 140, 6217–6220.
[23] B. Zhang, Z. A. Digby, J. A. Flum, P. Chakma, J. M. Saul, J. L. Sparks,
D. Konkolewicz, Macromolecules 2016, 49, 6871–6878.
[24] J. S. A. Ishibashi, J. A. Kalow, ACS Macro Lett. 2018, 7, 482–486.
[25] G. Joshi, E. V Anslyn, Org. Lett. 2012, 14, 4714–4717.
[26] B. M. Matysiak, P. Nowak, I. Cvrtila, C. G. Pappas, B. Liu, D. Komaromy,
S. Otto, J. Am. Chem. Soc. 2017, 139, 6744–6751.
[27] P. Chakma, L. H. Rodrigues Possarle, Z. A. Digby, B. Zhang, J. L. Sparks,
D. Konkolewicz, Polym. Chem. 2017, 8, 6534–6543.
[28] S. P. Daymon, K. M. Miller, Polymer 2018, 145, 286–293.
[29] N. Kuhl, R. Geitner, R. K. Bose, S. Bode, B. Dietzek, M. Schmitt, J. Popp,
S. J. Garcia, S. van der Zwaag, U. S. Schubert, et al., Macromol. Chem.
Phys. 2016, 217, 2541–2550.
[30] B. T. Worrell, S. Mavila, C. Wang, T. M. Kontour, C.-H. Lim, M. K.
McBride, C. B. Musgrave, R. Shoemaker, C. N. Bowman, Polym. Chem.
2018, 9, 4523–4534.
[31] S.-H. Shim, M. Ham, J. Huh, Y.-K. Kwon, Y.-J. Kwark, Polym. Chem.
2013, 4, 5449.
[32] L. J. Macdougall, V. X. Truong, A. P. Dove, ACS Macro Lett. 2017, 6,
93–97.
[33] A. B. Lowe, Polymer 2014, 55, 5517–5549.
[34] D. P. Nair, M. Podgórski, S. Chatani, T. Gong, W. Xi, C. R. Fenoli, C. N.
Bowman, Chem. Mater. 2014, 26, 724–744.
[35] H. Kuroda, I. Tomita, T. Endo, Polymer 1997, 38, 6049–6054.
[36] O. Daglar, U. S. Gunay, G. Hizal, U. Tunca, H. Durmaz, Macromolecules
2019, 52, 3558–3572.
[37] C. Xu, J. K. Bartley, D. I. Enache, D. W. Knight, M. Lunn, M. Lok, G. J.
Hutchings, Tetrahedron Lett. 2008, 49, 2454–2456.
This article is protected by copyright. All rights reserved.

10.1002/anie.201912902
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
RESEARCH ARTICLE
Controlling reactivity: Reversible
thiol-yne cross-linking is enabled via
the double Michael addition of thiols
on activated alkynes. The rate of
dynamic thiol exchange is
Niels Van Herck, Diederick Maes, Kamil
Unal, Marc Guerre, Johan M. Winne*,
Filip E. Du Prez*
Page No. – Page No.
manipulated by means of electronic
substituents on the activated alkyne,
offering a new versatile dynamic
platform for the development of
covalent adaptable networks.
Covalent adaptable networks with
tunable exchange rates based on
reversible thiol-yne cross-linking
This article is protected by copyright. All rights reserved.