Metal Cation-Driven Dynamic Covalent Formation of Imine and Hydrazone Ligands Displaying Synergistic Co-catalysis and Auxiliary Amine Effects

Article abstract of DOI:10.1002/chem.202100662

Optimizing C=N bond formation and C/N component exchange has major significance in Dynamic Covalent Chemistry (DCC). Imine and hydrazone generation from their aldehyde, amine and hydrazine components showed large accelerations in presence of AgOTf or Zn(O

Full text of DOI:10.1002/chem.202100662

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
Metal Cation-Driven Dynamic Covalent Formation of Imine
and Hydrazone Ligands Displaying Synergistic Co-catalysis
and Auxiliary Amine Effects
Meixia He[a] and Jean-Marie Lehn*[a]
Abstract: Optimizing C=N bond formation and C/N compo-
nent exchange has major significance in Dynamic Covalent
Chemistry (DCC). Imine and hydrazone generation from their
aldehyde, amine and hydrazine components showed large
accelerations in presence of AgOTf or Zn(OTf)2, up to 104 for
the Zn(II)-(p-anisidine)imine complex. Zn(OTf)2 and auxiliary p-
anisidine together accelerated 630 times the formation of the
vated towards reaction with the hydrazine component.
Reactions involving more entities showed kinetically faster
and thermodynamically simpler outputs due to dynamic
competition within a mixture of higher complexity. Catalytic
amounts of metal salts and auxiliary amine gave similar
marked rate accelerations and turnover, indicating true
catalysis. The synergistic effect achieved by combining metal-
lo- and organo-catalysis points to a powerful co-catalysis
strategy of bond-formation in DCC through interconnected
chemical transformations.
Zn(II)-hydrazone complex, revealing
a strong synergistic
effect, traced to very fast initial formation of the reactive
Zn(II)-imine complex presenting a C=N bond metallo-acti-
Introduction
have been established which are responsive to external
effectors by adaptation of the constitution of their members. As
a consequence, much interest and many efforts have been
devoted to the catalysis of the formation of dynamic covalent
bonds in order to accelerate component exchange and
constituent adaptation in systems based on imine[11] or
disulphide bonds,[12] metathesis,[13] transamination and
transimination,[14] and enamine exchange[15] either in aqueous
or in organic phase.
Simultaneous application of two or more chemical agents to
assist a chemical reaction may result in a synergistic effect to a
level unreachable for a single agent:[1] i) increasing the reactivity
of the reactants, thus significantly accelerating the reaction rate
and decreasing the reaction time; ii) increasing reaction
selectivity so as to generate cleaner outputs by way of more
inputs; iii) increasing reaction yield and efficiency of product
delivery. Similar behaviour manifests itself also in biological
processes.[2] The main challenges for such synergistic effects lie
in ensuring the compatibility of the different agents and in
analysing the kinetics of the key intermediate formation.[3] For
example, catalysis using a Lewis acid and a Lewis base together
may be prevented by their mutual interaction resulting in self-
quenching.[4] Such a detrimental outcome can be circumvented
by preventing the interaction as is the case in the frustrated
Lewis acid-base pairs.[5]
Reversible covalent reactions have been extensively ex-
plored in recent years in view of their importance for the
development of dynamic covalent chemistry (DCC)[6] and of its
potential to lead to numerous applications in a wide range of
fields. For example, a variety of responsive materials,[7] bio-
logical applications such as drug delivery,[8] dynamic receptors
and inhibitors[9] have been developed based on reversible
reactions. Dynamic covalent libraries (DCLs) and networks,[6,10]
The typical two-step imine C=N formation[6e,11,16] and
component exchange reactions cover a large range of rates
depending on differences in structural features and electronic
properties of both carbonyl and amino partners, such as
nucleophilic reactivity and basicity of amines.[17] Hydrazone,
acylhydrazone and oxime formation and exchange are much
slower than those of imines.[18] Thus, many catalysts, such as
aniline and some aniline derivatives,[19] protonic acid,[20] molec-
ular sieves,[21] metal oxides and salts[22] have been developed to
accelerate the rates of these condensations. Metal cations
especially of metal triflates[23] play an important role in organic
synthesis, not only serving as Lewis acids to facilitate
nucleophilic attack on bound ligands, but also acting as a
template[24] by coordination of the intermediates or products in
the reaction.
We report here investigations on the ability of metal cations
to drive C=N imine and hydrazone bond formation as well as
on the synergistic operation of metal salts and an auxiliary
amine on hydrazone formation between 6-phenyl pyridine-2-
carboxaldehyde (A1) and two amino compounds p-anisidine
(B1) or N-methyl pyridine hydrazine (B2). Similar results
obtained for the 6-methyl (A2) and 6-bromo (A3) pyridine-2-
carboxaldehydes as well as the experimental procedures for
data acquisition on rates and amounts of product formation are
given in the SI. The reactions were performed in stoichiometric
[a] Dr. M. He, Prof. Dr. J.-M. Lehn
Laboratoire de Chimie Supramoléculaire
Institut de Science et d’Ingénierie Supramoléculaires
Université de Strasbourg, 8 allée Gaspard Monge, 67000 Strasbourg
(France)
E-mail: lehn@unistra.fr
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202100662
Chem. Eur. J. 2021, 27, 7516–7524
7516
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
1
°
conditions in order to achieve highest efficiencies. However,
reactions conducted in sub-stoichiometric amounts of metal
cations and auxiliary amine indicated that the processes were
of true catalytic nature.
the reactions was followed by H NMR spectroscopy at 25 C in
CD3CN and the data obtained are plotted in Figure 1 and
Table 1.
The results indicated that the formation of A1B1 from the
condensation of A1 with B1 was very slow (92%, 250 h;
Figures 1b and S1), as characterized by an initial rate constant
k1 being 1.1×10À 3 MÀ 1sÀ 1 (entry 1 in Table 1; Figures 1c and S2).
This low reactivity is a reflection of the weak reactivity of the
aromatic amine B1 compared with aliphatic amines. In strong
contrast, in presence of 0.5 equiv. of AgOTf, the complex
[Ag(A1B1)2]+ was obtained much more rapidly (95%, 45 h;
Figures 1b and S3) meaning that the metal cation greatly
accelerates the A1+B1 condensation to give the imine ligand
in its coordinated form. The initial rate k1s of formation in
presence of AgOTf is ~18 times faster than that in absence of
AgOTf (entries 1 and 2 in Table 1; Figures 1c and S4). Interest-
Results and Discussion
The C=N formation in presence of metal salts
Imine formation driven by metal salts
To evaluate the efficiency of metal salts as Lewis acids in C=N
formation, we first studied the condensation between A1 and
B1, leading to the formation of the imine A1B1 (Figure 1a) both
in absence and presence of AgOTf or Zn(OTf)2. The evolution of
Figure 1. a) Imine and d) hydrazone formation from A1 and B1 or B2 in absence and in presence of metal salts [AgOTf or Zn(OTf)2]; kinetic plots of the
evolution of the corresponding b) imine and e) hydrazone formation as a function of time as obtained from integration of the imine CH=N proton signal in
1
°
the 500 MHz H NMR spectra (10 mM each, CD3CN, 25 C); Values of the initial rate constants (k) for the corresponding c) imine and f) hydrazone formation
process calculated from a fit to a second-order reaction over the first 10% of the reaction (error of about 5% in integration of the imine proton signal). The
numbers above the columns correspond to 103 k for ease of comparison. All the reactions also generated one molecule of water.
Table 1. Kinetic features of the reactions between 6-phenyl pyridine-2-carboxaldehyde (A1) and p-anisidine (B1) or N-methyl pyridine hydrazine (B2) in
absence and in presence of metal salts.
Entry[a]
Reaction
Catalyst 0.5 equiv.
Product
t50 [h][c]
t10 [h][d]
k
[e]/10À 3 MÀ 1sÀ 1
Acceleration Factor
[b]
1
2
3
4
5
6
A1+B1
–
A1B1
23
3
0.5
�0.08
1.94
0.35
0.55
1.1
20
�18200
0.97
14
n.a.[f]
18
�16500
n.a.[f]
14
A1+B1+Ag(I)
A1+B1+Zn(II)
A1+B2
A1+B2+Ag(I)
A1+B2+Zn(II)
Ag(I)
[Ag(A1B1)2]+
[Zn(A1B1)2]2+
A1B2
1.4
�0.08
21
2.4
4
Zn(II)
[b]
–
Ag(I)
Zn(II)
[Ag(A1B2)2]+
[Zn(A1B2)2]2+
7.5
8
1
°
[a] In all these entries, aldehyde (10 mM), amine (10 mM), AgOTf or Zn(OTf)2 (5 mM), at 25 C, in CD3CN, monitored by H NMR. [b] No catalyst was added. [c]
Time for 50% completion of the reaction. [d] Time for 10% completion of the reaction. [e] The rate constant k has been calculated from a fit to a second-order
reaction over the first 10% of the reaction (error of about 5% in integration of the imine proton signal). The values are listed as 103 k for ease of comparison.
[f] Not applicable.
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7517
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
ingly, the formation of A1B1 in presence of Zn(OTf)2 (entry 3 in
Table 1; Figures 1b and S5) is even more strongly accelerated,
being completed within 5 minutes, compared to its formation
in absence of metal cations or in presence of AgOTf (Figure 1;
Table 1) amounting to an acceleration of about 1.65 ×104
compared with that in absence of Zn(II) (Figures 1c and S6).
This strong facilitation of the reaction by Zn(OTf)2 may have a
double origin: being doubly charged and a strong Lewis acid
the cation binds and activates the pyridyl aldehyde towards
attack by the amine and it may also enhance the rate of
elimination of water from the hemiaminal. Titration experiments
of A1+B1 with Zn(OTf)2 (Figure 2) showed that, on addition of
only 0.05 equiv. of Zn(OTf)2, within 4 minutes, the imine had
formed almost fully both as free A1B1 (90%) and the remainder
Hydrazone formation driven by metal salts
In earlier studies, hydrazones were found to form much more
slowly in organic solvent than imines.[18] In order to verify
whether metal ions could also drive hydrazone formation, the
following reactions were conducted (entries 4 to 6 in Table 1).
Aldehyde A1 and hydrazine B2 were reacted in absence as well
as in presence of metal ions (Figure 1d). The hydrazone
complex [Ag(A1B2)2]+ formed faster from A1+B2+0.5 equiv.
AgOTf (95%, 18 h, Figures 1e, S25–S26, entry 5 in Table 1) than
the ligand A1B2 itself from A1+B2 (95%, 162 h, Figures 1e,
S23–S24, entry 4 in Table 1). The corresponding rate constants
are k2s =14×10À 3 MÀ 1sÀ 1 and k2 =0.97×10À 3 MÀ 1sÀ 1 in presence
and in absence of AgOTf respectively, with an acceleration
factor of 14 due to the presence of AgOTf (Figure 1f). While in
presence of Zn(OTf)2, the hydrazone complex [Zn(A1B2)2]2+ also
formed faster by a factor of 6 from A1+B2+0.5 equiv. Zn(OTf)2
(99%, 62 h, k2z =7.5×10À 3 MÀ 1sÀ 1; Figures 1e–1f, S27–S28, en-
try 6 in Table 1) than its ligand A1B2 itself from A1+B2.
1
as its [Zn(A1B1)2]2+ complex, identified by their H NMR imine
signals. On increasing progressively the amounts of zinc added
up to 0.5 equiv., the [Zn(A1B1)2]2+ complex was fully formed.
This result also indicated that the effect of the metal ion is
catalytic (see also below). By adding more zinc, the spectrum
1
°
did not change any more. Low temperature (À 35 C) H NMR
Similarly, results were obtained for the aldehydes A2 and
A3 (see Figures S29–S41 in Supporting Information).
spectra were measured for the mixture obtained by adding
either 1/2 equiv. or 1/3 equiv. of Zn(OTf)2 to the preformed
ligand A1B1. In the latter case [A1B1 to Zn(OTf)2 ratio=3:1],
two peaks were observed in the imine region (at 8.95 and
8.65 ppm) due to the complex [Zn(A1B1)2]2+ and free ligand
A1B1 respectively (as independently shown) in the ratio 2:1,
confirming the stoichiometry of the complex (Figure S7). It was
further verified by its single-crystal structure [Zn-
(A1B1)2(CF3SO3)2] (with deposition number 2052560 in CCDC)
where Zn(II) is coordinated by the two A1B1 ligands and two
triflate anions (Figures S8 and S9).
The results above indicate that metal salts efficiently drive
both imine and hydrazone formation, the effect being larger on
imine formation than on hydrazone formation. Taking into
account the fact that aniline derivatives significantly increase
the rate of hydrazone formation by acting as nucleophilic
catalysts[19] and on the other hand, that metal ions accelerate
the formation of hydrazone ligands in the course of concom-
itant cation complexation, it is worth investigating whether the
simultaneous application of a metal ion and the auxiliary amine
p-anisidine would exert a synergistic effect and increase even
further the rate of hydrazone formation. To this end a series of
five experiments were performed. The corresponding evolutions
in time of the compounds in these five dynamic mixtures are
shown in the curves displayed in Figure 3.
Similarly, reactivities were found using A2 and A3 instead of
A1 (see Figures S10–S22 in Supporting Information). In all three
cases, the metal cations may serve as Lewis acids to activate the
aldehyde group driving the formation of the imine ligand.
Synergistic effects of metal salts and an auxiliary amine on
hydrazone formation
Experiments involving four entities
To evaluate such a synergistic effect, the reaction of a mixture
of the two components A1+B2 (10 mM each) was followed by
1H NMR (Figure 3a) on addition of i) amine B1 (1.0 equiv.) or ii) a
metal salt (0.5 equiv.) or iii) both amine B1 and a metal salt (see
entries 1 to 5 in Table 2).
In presence of only amine B1, A1B2 was fully formed 3
times faster (k2a Figure 3a; entry 1 in Table 2, Figures 3b–3c;
Figures S42–S43) than in the reaction A1+B2 in absence of the
amine (entry 4 in Table 1, Figures 1e–1f). Separate experiments
in presence of AgOTf or Zn(OTf)2 gave an acceleration of the
same reaction by a factor of 14 and 8 (see rate constants in
entries 5 and 6 in Table 1; 18 h and 70 h, Figure 1f).
The addition of the amine B1 to the reaction in presence of
AgOTf had no visible effect (entry 2 in Table 2; Figures S44–
S45). In contrast, a strong effect is observed when B1 is
Figure 2. The titration experiment of A1 with B1 on addition of different
1
°
equivalents of Zn(OTf)2 in CD3CN at 25 C in the 500 MHz H NMR spectra
(The original concentration of A1 or B1 is 10 mM each).
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7518
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
Figure 3. a) Hydrazone formation from A1 and B2 in the presence of (i) B1 alone or together with either (ii) AgOTf or (iii) Zn(OTf)2 as well as hydrazone
formation from (iv) A1B1 and B2 and (v) [Zn(A1B2)2]2+ formation from [Zn(A1B1)2]2+ and B2. k2a, k2as, k2az, k3, k3z are the initial corresponding rate constants
calculated from the formation of 10% product using a second order kinetic equation. For the structures of A1B1 and [Zn(A1B1)2]2+ see Scheme 1 below; b)
1
°
Kinetic curves obtained as a function of time from integration of the imine CH=N proton signal in the 500 MHz H NMR spectra (10 mM each, CD3CN, 25 C); c)
initial rate constants (k) calculated from a fit to a second-order reaction over the first 10% of the reaction for the hydrazone formation process between A1
and B2 (Figure 1) in absence of catalyst and in presence of B1 or a metal salt, AgOTf or Zn(OTf)2 separately, as well as B1 and AgOTf or B1 and Zn(OTf)2
together as co-catalysts along with the additional control reactions of A1B1 with B2 and [Zn(A1B1)2]2+ with B2. Note that the curves for A1+B2+B1+Zn(II)
and for [Zn(A1B1)2]2+ +B2 are superimposed (error of about 5% in integration of the imine proton signal). The numbers above the columns correspond to
103 k for ease of comparison.
Table 2. Kinetic features of the reactions between 6-phenyl pyridine-2-carboxaldehyde (A1) and N-methyl pyridine hydrazine (B2) in presence of metal ions
2+
separately as well as in presence of both p-anisidine (B1) and metal ions together and control reactions of [Zn(A1B1)]2 with B2 and A1B1 with B2.
Entry[a]
Reaction
Catalyst 0.5 equiv.
Product
t50 [h][c]
t10 [h][d]
k[e]/10À 3 MÀ 1sÀ 1
Acceleration Factor
1
2
A1+B2+B1
A1+B2+B1+Ag(I)
B1 (1.0 equiv.)
B1 (1.0 equiv.)
+ Ag(I)
A1B2
8.8
2.7
0.97
0.44
2.4
16
3
16
[Ag(A1B2)2]+
3
A1+B2+B1+Zn(II)
B1 (1.0 equiv.)
[Zn(A1B2)2]2+
0.15
�0.08
�610
630
+ Zn(II)
[b]
4
5
[Zn(A1B1)2]2+ +B2
A1B1+B2
–
–
[Zn(A1B2)2]2+
A1B2
0.16
1.0
�0.08
�640
�660[b]
[b]
0.08
39
40[b]
1
°
[a] In all entries, aldehyde (10 mM), amine or hydrazine (10 mM), AgOTf or Zn(OTf)2 (5 mM), at 25 C, in CD3CN, monitored by H NMR. [b] Transimination
experiments; no catalyst added; the acceleration factors correspond to the reaction of B2 with the imine A1B1 free (entry 5) or complexed (entry 4) compared
to the reaction of B2 with the aldehyde A1 (entry 1). [c] Time for 50% completion of the reaction. [d] Time for 10% completion of the reaction. [e] The rate
constant k has been calculated from a fit to a second-order reaction over the first 10% of the reaction (error of about 5% in integration of the imine proton
signal). The values are listed as 103 k for ease of comparison.
combined with Zn(OTf)2 (entry 3 in Table 2; Figures S46–S47).
Indeed after just 4 minutes, the mixture contained about 25%
of free A1B1, 29% of free A1, 70% of free B1 and 64% B2 in its
coordinated form (Figures S48–S49) together with 26% [Zn-
(A1B2)2]2+; the remainder consisting of combinations of A1, B1,
B2 and zinc salt gave other signals in the spectrum but the
corresponding species could not be identified. The components
thereafter underwent recombination and after about 2 hours,
the zinc complex [Zn(A1B2)2]2+ was fully formed leaving free
B1 in the solution (Figures 3b and S46). Thus, the formation of
[Zn(A1B2)2]2+ (k2az Figures 3c and S47; 98%, 2 h, entry 3 in
Table 2) was markedly increased when both metal cation and
amine were added simultaneously to the reaction of A1 with
B2, with an acceleration factor of 630 compared to the
formation of A1B2 from A1+B2 alone. Further information
[Zn(A1B2)2]2+. They revealed that metal ions and auxiliary
amine together strongly enhance the hydrazone formation.
In order to gather more information about the mechanism
of such a synergistic catalysis effect, a number of control
experiments were conducted, in particular by modifying the
order of addition of the reagents: i) adding B2 to preformed
A1B1; ii) adding Zn(II) to preformed A1B1 followed by B2; iii)
and iv) adding Zn(II) together with B1 to the (A1+B2) mixture.
The results are given in the Supporting Information (Figur-
es S51–S56).
They confirmed the very marked synergistic co-catalytic
effect on addition of Zn(II) and B1 to the A1+B2 mixture which
leads to a much faster formation of the [Zn(A1B2)2]2+ complex
than when either Zn(II) or B1 is added separately (2 h versus
62 h or versus 50 h respectively; see entry 4 in Table 1 and
entries 1 and 3 in Table 2).
about the kinetic evolution of the reaction was obtained by
1
°
performing low temperature H NMR experiments from À 35 C
In order to establish the mechanism of such co-catalysis, the
features of the experiments conducted above taken together
can be summarized in the following basic facts:
°
to +25 C in steps of 10 degrees (for details see SI; Figure S50).
The results indicated that [Zn(A1B1)2]2+ formed in the very
beginning as activated species which reacted with B2 to form
1) The formation of the imine A1B1 from A1+B1 is very slow
(entry 1 in Table 1, k1 =1.1×10À 3 MÀ 1sÀ 1, Figure 1c).
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7519
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
2) The formation of the hydrazone A1B2 is very slow but
increases along the sequence: i) A1+B2 (entry 4 in Table 1,
k2 =0.97×10À 3 MÀ 1sÀ 1, Figure 3c); ii) A1+B2+B1 (entry 1 in
Table 2, k2a =2.4×10À 3 MÀ 1sÀ 1, Figure 3c) a little faster with
no A1B1 observed; iii) A1B1+B2 (entry 5 in Table 2, k3 =
3.9×10À 2 MÀ 1sÀ 1, Figure 3c).
formation of respectively the ligand A1B2 itself and of its
complex [Zn(A1B2)2]2+; iii) when BOTH are present together,
they exert a very strong synergistic catalysis effect towards the
formation of this same complex.
It means that i) complex [Zn(A1B1)2]2+ is the active species
in the process; ii) the C=N bond of imine A1B1 is more activated
than the C=O group of aldehyde A1 by Zn(II) binding; iii) the N
of imine A1B1 is a better binding site for Zn(II) than the O of
C=O in A1.
3) The formation of the imine-based complex [Zn(A1B1)2]2+
from A1+B1+Zn(II) (entry 3 in Table 1, k1z =18.2 MÀ 1sÀ 1,
Figure 1c) is much faster than that of the imine A1B1 in
absence of Zn(II), being completed when the first spectrum
is measured (about 5 min; >90%) giving an acceleration
factor at least 1.65×104.
4) The formation of the hydrazone-based zinc complex [Zn-
(A1B2)2]2+ from A1+B2+Zn(II) (entry 6 in Table 1, k2z =
7.5×10À 3 MÀ 1sÀ 1, Figure 1f) is comparable to that of the
hydrazone A1B2 from A1+B2+B1 (2ii above), both being
faster (about 8 times and 3 times) than that of the
hydrazone A1B2 from A1+B2 (see 2i).
5) The formation of the ligand A1B2 from A1+B2 is
accelerated by a factor of 3 by the addition of B1 (see 2i and
2ii) whereas the formation of the complex [Zn(A1B2)2]2+ is
accelerated by a factor of 8 by addition of Zn(II), see 4)
above.
Therefore, from these experiments and facts, the proposed
mechanism (Scheme 1) contains two main steps: 1) Zn(II)
strongly catalyses the formation of the imine-based two ligand
zinc complex [Zn(A1B1)2]2+, characterized by its single crystal
structure (see above and Figures S8); 2) B2 reacts with the
activated C=N bonds in this complex [Zn(A1B1)2]2+ to form
[Zn(A1B2)2]2+ and release the free B1. Furthermore, whatever
the different transient species forming in the two-component
mixture on addition of Zn(II) and B1, the observed overall
behaviour reveals reactional output from a mixture of higher
complexity (four entities: an interesting case of more efficient,
“simpler” and “faster” three molecular components A1+B2+
B1 and one metal ion), in line with the notion that an increase
in complexity may results in a simpler output, in simplexity.[10f,25]
Similar results as for A1 were obtained for reactions run
with A2 or A3: (A2+B2) or (A3+B2) in presence of B1 alone
(Figures S57–S60), 1.0 equiv. B1 and 0.5 equiv. of either AgOTf
(Figures S61–S64) or Zn(OTf)2 (Figures S65–S68). The corre-
sponding kinetic curves and rate constants are shown in
Figures S69–S70, Table S1 and Figures S71–S72, Table S2.
6) The formation of [Zn(A1B2)2]2+ from [Zn(A1B1)2]2+ +B2 is
very fast (entry 4 in Table 2, k3z =0.64 MÀ 1sÀ 1, Figures S53–
S54).
7) The formation of the hydrazone-based zinc complex [Zn-
(A1B2)2]2+ from A1+B2+B1+Zn(II) is very fast (entry 3 in
Table 2, k2az =0.61 MÀ 1sÀ 1, Figures 3b–3c, Figures S46–S47).
These facts indicate that i) the formation of the imine
complex [Zn(A1B1)2]2+ is by far the fastest reaction of all; ii) B1
and Zn(II) exert independently a moderate effect on the
Scheme 1. Proposed schematic mechanism for the formation of the zinc complex [Zn(A1B2)2]2+ from A1+B2 in presence of both Zn(OTf)2 and B1 via the
intermediate formation of the [Zn(A1B1)2]2+ complex with Zn(OTf)2 and B1 acting in a synergistic co-catalysis pathway. a) Overall reaction of [Zn(A1B2)2]2+
formation; b) Stepwise mechanism of [Zn(A1B2)2]2+ formation (see text section 2.1). The formation of [Zn(A1B2)2]2+ may proceed via an intermediate mixed-
ligand complex.
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7520
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
Experiments involving five entities
Control experiments involving five entities
When the same reaction A1+B2 (10 mM each) was run in
presence of the three co-factors, B1 (10 mM) together with both
AgOTf (5 mM) and Zn(OTf)2 (5 mM), the silver complex [Ag-
(A1B1)2]+ of the imine A1B1 was already formed (95%) after
only 10 minutes together with a similar amount of the Zn(II)
complex of B2, [Zn(B2)2]2+ in strong contrast to the very slow
formation of A1B1 in absence of metal cations. Thereafter, the
set rearranged by component exchange to form the zinc
complex [Zn(A1B2)2]2+, giving the final distribution of 8%
[Ag(A1B1)2]+, 92% [Zn(A1B2)2]2+ and free B1 (4 h; Figure S73).
Thus, in presence of both silver and zinc cations and of B1, the
silver complex of A1B1, [Ag(A1B1)2]+ is the kinetic product and
then the set slowly changes (38 hours) to the zinc complex of
A1B2, [Zn(A1B2)2]2+ as thermodynamic product (Scheme 2).
These results display a striking example of a faster and simpler
kinetic output (the silver complex) resulting from competition
within a mixture of higher complexity (five entities: three
molecular components A1+B2+B1 and two metal ions), thus
extending to the kinetic arena the behaviour simplification
noted above on the thermodynamic scene.
A number of other control experiments involving various
combinations of altogether five components and of addition
sequences were performed in order to obtain more information
on the reaction mechanism in the multicomponent mixture [A1
+B1+B2+AgOTf+Zn(OTf)2]. They are described in the Sup-
porting Information (Figures S74–S80). For instance, the fact
that the reaction of (A1+B1) with [B2+AgOTf+Zn(OTf)2] gives
first the [Ag(A1B1)2]+ complex whereas the reaction of (A1+
B1) with [AgOTf+Zn(OTf)2] gives the [Zn(A1B1)2]2+ complex at
about the same rate, indicates that Zn(OTf)2 is first bound by B2
which reduces its reactivity.
Experiments performed for the two sets of three compo-
nents A2+B2 and A3+B2 in presence of both B1 and two
metal cations [AgOTf and Zn(OTf)2] gave similar results (see
Figures S81–S82).
Catalysis of imine and hydrazone formation by metal salts
and acid
The conversion of [Ag(A1B1)2]+ into [Zn(A1B2)2]2+ amounts
to a kinetic switching process reminiscent of the network
switching reported earlier.[18b]
The large accelerations observed for the formation of imine and
hydrazone ligands by stoichiometric amounts of metal salts,
prompted us to explore the effect of catalytic amounts of the
salts (Figures 4 and S83). To this end the kinetics of the imine
A1B1 and hydrazone A1B2 formation were investigated in the
presence of only 0.05 equiv. (5 mol%) of AgOTf, Zn(OTf)2,
Sc(OTf)3 and CF3COOD (Table 3; Figure 4). Several parameters
Scheme 2. Proposed sequential mechanism for the formation of the zinc complex [Zn(A1B2)2]2+ from A1+B2 in presence of the auxiliary amine B1 and both
metal salts AgOTf and Zn(OTf)2 via the initial fast formation of the intermediate complex [Ag(A1B1)2]+, which then reacts much more slowly with the complex
[Zn(B2)2]2+ (see text section 2.2).
Table 3. Kinetic features of the reactions between 6-phenyl pyridine-2-carboxaldehyde (A1) and p-anisidine (B1) or N-methyl pyridine hydrazine (B2) in
absence and in presence of catalysts.
Entry[a]
Reaction
Catalyst 0.05 equiv.
t50 [h][c]
t10 [h][d]
k
[e]/MÀ 1sÀ 1
Acceleration Factor
TOF/hÀ 1
[b]
1
2
3
4
5
6
7
8
A1+B1
A1+B1
A1+B1
A1+B1
A1+B1
A1+B2
A1+B2
A1+B2
A1+B2
A1+B2
A1+B2
–
23
35
<3 min
<3 min
<2 min
21
6.2
6.4
0.15
0.17
7.7
3
2.2
<3 min
<3 min
<2 min
1.9
1.5
1.7
<4 min
<4 min
1.5
1.1×10À 3
1.6×10À 3
>5.0
1
1.5
n.a.[f]
0.005
430
Ag(I)
Zn(II)
Sc(III)
4.5×103
4.4×103
2.6×104
1
>4.8
>29
410
CF3COOD
>986
n.a.[f]
0.98
0.54
6.5
[b]
–
0.97×10À 3
3.1×10À 3
3.2×10À 3
>0.15
Ag(I)
Zn(II)
Sc(III)
CF3COOD
3.2
3.3
>155
>124
4.5
9
10
11
>0.12
3.5
0.60
Zn(II)+p-anisidine
4.4×10À 3
1
°
[a] Conditions for all these entries: aldehyde (10 mM), amine or hydrazine (10 mM), catalyst (0.5 mM), at 25 C, in CD3CN, monitored by H NMR. [b] No catalyst
was added. [c] Time for 50% completion of the reaction. [d] Time for 10% completion of the reaction. [e] The rate constant k has been calculated from a fit to
a second-order reaction over the first 10% of the reaction (error of about 5% in integration of the imine proton signal). [f] Not applicable.
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7521
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
Figure 4. Kinetic plots of the evolution of: a) imine A1B1 formation from a mixture of equal amounts of components A1+B1 in absence of catalyst or in
presence of catalysts as indicated; b) zoom of a); c) hydrazone A1B2 formation from a mixture of equal amounts of components A1+B2 in absence of catalyst
or in presence of catalysts as indicated; d) zoom of c). Data obtained as a function of time from integration of the CH=N proton signal in the 500 MHz 1H NMR
°
spectra (10 mM each, CD3CN, 25 C).
were determined including the initial rate constant, the
turnover frequency (TOF) and the acceleration factor.
On addition of 0.05 eq. of Ag(I), the formation of A1B1 from
its components A1 and B1 was accelerated by only a factor of
1.5 compared to the rate in the absence of the salt (entries 1
(entries 6 and 7 in Table 3; Figures S92–S93), an effect similar to
that observed on imine formation. In presence of 0.05 eq. of
Zn(II), a similar rate enhancement of 3.3 was obtained (entry 8
in Table 3; Figures S94–S95). On changing the aldehyde to A2,
the initial reaction rate was faster while the reaction conversion
time did not change much (Figures S96–S98). As in the case of
imine formation, Sc(III) or CF3COOD gave the hydrazone with a
significant acceleration by a factor of 155 and 124 respectively
(entries 9 and 10 in Table 3; Figures S99–S102). Finally, the
combined addition of both 0.05 equiv. of Zn(II) and of B1 made
only a very minor (if any) difference (entries 8 and 11 in Table 3,
Figures S103–S104), indicating that B1 had no significant effect
when present in just catalytic amount.
Taken together, the present results indicated that the
addition of a catalytic amount of Ag(I), Zn(II), Sc(III) metal salts
or of CF3COOD result in catalysis of both imine and hydrazone
formation. The effect displayed by these catalysts was much
more pronounced on the formation of imine than that of
hydrazone. It further demonstrated that the facilitation of the
hydrazone formation by the synergistic operation of both Zn(II)
and the auxiliary amine is due to a very fast initial imine
formation catalysed by Zn(II).
and
2 in Table 3, Figures S84–S85), indicating that Ag(I)
displayed a very weak catalytic effect on imine formation. In the
presence of the same amount of Zn(II), the formation of A1B1
was greatly accelerated by a factor of 4.5×103 (entry 3 in
Table 3, Figures S86–S87). The highest rates were found in the
presence of 0.05 equiv. of CF3COOD (entry 5 in Table 3,
Figures S88–S89), with a rate enhancement by a factor of 2.6×
104 over that in absence of added acid. The presence of 0.05 eq.
of Sc(III) triflate also displayed a remarkable catalytic effect on
A1B1 formation with an acceleration factor of 4.4×103 (entry 4
in Table 3; Figures S90–S91). These results are in line with a
study of the effect of different metal ions on the transamination
reaction.[14f]
A similar set of experiments were conducted for the
formation of hydrazone A1B2. Addition of 0.05 eq. of Ag(I) to
A1+B2 produced a catalytic effect with an acceleration by a
factor of 3.2 compared to the reaction in absence of salt
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7522
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
10018–10032; e) D. W. Stephan, G. Erker, Angew. Chem. Int. Ed. 2015, 54,
6400–6441; Angew. Chem. 2015, 127, 6498–6541.
Conclusion
[6] a) J.-M. Lehn, Chem. Eur. J. 1999, 5, 2455–2463; b) S. J. Rowan, S. J.
Cantrill, G. R. L. Cousins, J. K. M. Sanders, J. F. Stoddart, Angew. Chem.
Int. Ed. 2002, 41, 898–952; Angew. Chem. 2002, 114, 938–993; c) P. T.
Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor, J. K. M. Sanders, S.
Otto, Chem. Rev. 2006, 106, 3652–3711; d) J.-M. Lehn, Chem. Soc. Rev.
2007, 36, 151–160; e) M. E. Belowich, J. F. Stoddart, Chem. Soc. Rev.
2012, 41, 2003–2024; f) Y. Jin, C. Yu, R. J. Denman, W. Zhang, Chem. Soc.
Rev. 2013, 42, 6634–6654; g) C. W. Hsu, O. Š Miljanić, Dynamic Covalent
Chemistry: Principles, Reactions, and Applications (Eds.: W. Zhang, Y. Jin),
Wiley, Chichester, 2018, pp. 21–464.
[7] a) E. Moulin, G. Cormos, N. Giuseppone, Chem. Soc. Rev. 2012, 41, 1031–
1049; b) Z. Wei, J. H. Yang, J. Zhou, F. Xu, M. Zrínyi, P. H. Dussault, Y.
Osada, Y. M. Chen, Chem. Soc. Rev. 2014, 43, 8114–8131; c) N. Roy, B.
Bruchmann, J.-M. Lehn, Chem. Soc. Rev. 2015, 44, 3786–3807; d) D.
Zhang, T. K. Ronson, J. Nitschke, Acc. Chem. Res. 2018, 51, 2423–2436.
[8] a) S. Ulrich, Acc. Chem. Res. 2019, 52, 510–519; b) Y. Zhang, Y. Qi, S.
Ulrich, M. Barboiu, O. Ramström, Mater. Chem. Front. 2020, 4, 489–506.
[9] a) O. Ramström, J.-M. Lehn, Nat. Rev. Drug Discovery 2002, 1, 26–36;
b) L. A. Ingerman, M. E. Cuellar, M. L. Waters, Chem. Commun. 2010, 46,
1839–1841; c) M. Mondal, A. K. H. Hirsch, Chem. Soc. Rev. 2015, 44,
2455–2488.
[10] a) B. Brisig, J. K. M. Sanders, S. Otto, Angew. Chem. Int. Ed. 2003, 42,
1270–1273; Angew. Chem. 2003, 115, 1308–1311; b) N. Giuseppone,
J. M. Lehn, Chem. Eur. J. 2006, 12, 1715–1722; c) M. Barboiu, F. Dumitru,
Y.-M. Legrand, E. Petit, A. van der Lee, Chem. Commun. 2009, 2192–
2194; d) Q. Ji, R. C. Lirag, O. Š Miljanić, Chem. Soc. Rev. 2014, 43, 1873–
1884; e) F. Schaufelberger, O. Ramström, J. Am. Chem. Soc. 2016, 138,
7836–7839; f) J. Holub, G. Vantomme, J.-M. Lehn, J. Am. Chem. Soc.
2016, 138, 36, 11783–11791; g) G. Men, J.-M. Lehn, J. Am. Chem. Soc.
2017, 139, 2474–2483.
The results described herein demonstrate the marked
enhancement of the efficiency of imine and hydrazone
formation by metal cations in stoichiometric conditions as well
as the catalytic nature of the process. Moreover, simultaneous
addition of both a metal salt and an aniline derivative as
auxiliary amine to the reacting components of a DCL induced a
strong acceleration of the formation of the complex of the
hydrazone ligand by a synergistic fashion. Such effects are of
much interest for DCC systems based on chemical bond
formation by reversible reactions with differences in formation
rates and thermodynamic stabilities. In particular, simultaneous
addition of an auxiliary aniline as well as both a silver salt and a
zinc salt resulted in a synergistic co-catalysis whereby the DCC
system undergoes a switching from a kinetic product (the silver
complex of an imine) to the thermodynamic product (the zinc
complex of a hydrazone). This mode of co-catalysis, which
combines metallo- and organo-catalysis, represents a step in
the implementation in DCC of two or more catalytic species
acting together in a dynamic network through synergistic
interconnected chemical transformations. Such behaviours
point out that increasing system complexity may result in
kinetically faster and thermodynamically simpler outputs. Thus,
an increase in complexity leads to simplexity[10f,25] on both
thermodynamic and kinetic levels through advanced features
such as competition, feedback and synergy in constitutional
dynamic systems.
[11] a) R. W. Layer, Chem. Rev. 1963, 63, 489–510; b) R. D. Patil, S. Adimurthy,
Asian J. Org. Chem. 2013, 2, 726–744.
[12] a) J. Weissman, P. Kimt, Nature 1993, 365, 185–188; b) C. S. Sevier, C. A.
Kaiser, Nat. Rev. Mol. Cell Biol. 2002, 3, 836–847; c) B. Mandal, B. Basu,
RSC Adv. 2014, 4, 13854–13881.
[13] a) G. K. Cantrell, T. Y. Meyer, Organometallics 1997, 16, 5381–5383; b) M.
Ciaccia, R. Cacciapaglia, P. Mencarelli, L. Mandolinia, S. Di Stefano, Chem.
Sci. 2013, 4, 2253–2261; c) M. Ciaccia, S. Pilati, R. Cacciapaglia, L.
Mandolini, S. Di Stefano, Org. Biomol. Chem. 2014, 12, 3282–3287; d) R.
Gu, K. Flidrova, J.-M. Lehn, J. Am. Chem. Soc. 2018, 140, 5560–5568.
[14] a) D. L. Leussing, E. M. Hanna, J. Am. Chem. Soc. 1966, 88, 693–696;
b) E. E. Snell, W. T. Jenkins, Biochem. J. 1969, 114, 70–71; c) J. L. Hogg,
D. A. Jencks, W. P. Jencks, J. Am. Chem. Soc. 1977, 99, 4772–4778; d) H.
Fischer, F. X. DeCandis, S. D. Ogden, W. P. Jencks, J. Am. Chem. Soc.
1980, 102, 1340–1347; e) Y. Murakami, J.-I. Kikuchi, N. Shiratori, J. Phys.
Org. Chem. 1989, 2, 110–116; f) N. Giuseppone, J. L. Schmitt, E. Schwartz,
J.-M. Lehn, J. Am. Chem. Soc. 2005, 127, 5528–5539; g) F. Schaufelberger,
L. Hu, O. Ramström, Chem. Eur. J. 2015, 21, 9776–9783.
Acknowledgements
We acknowledge financial support from the ERC (Advanced
Research Grant SUPRADAPT 290585) and the University of
Strasbourg Institute for Advanced Study (USIAS). MH thanks
China Scholarship Council for a doctoral scholarship (No.
201606870040). She also acknowledges Prof. Jack Harrowfield,
Dr. Artem Osypenko, Dr. Jean-Louis Schmitt, and Cyril An-
theaume for extensive discussions.
[15] Y. Zhang, S. Xie, M. Yan, O. Ramström, Chem. Eur. J. 2017, 23, 11908–
11912.
[16] S. Patai, The Chemistry of Carbon-Nitrogen Double Bond (Ed.: S. Patai),
Wiley, 1970, pp. 1–794.
[17] a) B. Kallies, R. Mitzner, J. Phys. Chem. B 1997, 101, 2959–2967; b) V. S.
Bryantsev, M. S. Diallo, W. A. Goddard, III, J. Phys. Chem. A 2007, 111,
4422–4430.
Conflict of Interest
The authors declare no conflict of interest.
[18] a) S. Kulchat, M. N. Chaur, J. M. Lehn, Chem. Eur. J. 2017, 23, 11108–
11118; b) M. He, J.-M. Lehn, J. Am. Chem. Soc. 2019, 141, 18560–18569.
[19] a) E. H. Cordes, W. P. Jencks, J. Am. Chem. Soc. 1962, 84, 826–831; b) A.
Dirksen, S. Dirksen, T. M. Hackeng, P. E. Dawson, J. Am. Chem. Soc. 2006,
128, 15602–15603; c) A. Dirksen, T. M. Hackeng, P. E. Dawson, Angew.
Chem. Int. Ed. 2006, 45, 7581–7584; Angew. Chem. 2006, 118, 7743–
7746; d) M. Wendeler, L. Grinberg, X. Wang, P. E. Dawson, M. Baca,
Bioconjugate Chem. 2014, 25, 93–101; e) D. Larsen, A. M. Kietrys, S. A.
Clark, H. S. Park, A. Ekebergh, E. T. Kool, Chem. Sci. 2018, 9, 5252–5259.
[20] a) O. Temme, S. Laschat, J. Chem. Soc. Perkin Trans. 1. 1995, 125–131;
b) J. J. Armao, J.-M. Lehn, J. Am. Chem. Soc. 2016, 138, 16809–16814.
[21] a) A. Sayari, Chem. Mater. 1996, 8, 1840–1852 ; b) K. Taguchi, F. H.
Westheimer, J. Org. Chem. 1971, 36, 1570–1572.
Keywords: dynamic covalent chemistry · imine and hydrazone
ligand formation · synergistic catalysis
[1] A. E. Allen, D. W. C. MacMillan, Chem. Sci. 2012, 3, 633–658.
[2] a) K. A. Brown, J. Kraut, Faraday Discuss. 1992, 93, 217–224; b) N. Sträter,
W. N. Lipscomb, Angew. Chem. Int. Ed. 1996, 35, 2024–2055; Angew.
Chem. 1996, 108, 2158–2191; c) X. Ren, R. Fasan, Nat. Catal. 2020, 3,
184–185; d) Z. Zhou, G. Roelfes, Nat. Catal 2020, 3, 289–294.
[22] a) D. L. Leussing, B. E. Leach, J. Am. Chem. Soc. 1971, 93, 3377–3384 ;
b) D. Guo, G. Han, C. Y. Duan, K. L. Pang, Q. J. Meng, Chem. Commun.
2002, 1096–1097.
[23] S. Kobayashi, M. Sugiura, H. Kitagawa, W. W.-L. Lam, Chem. Rev. 2002,
102, 2227–2302.
[3] R. G. Pearson, J. Am. Chem. Soc. 1963, 85, 3533–3539.
[4] V. I. Polshakov, Russ. Chem. Bull. 2001, 50, 1733–1751.
[5] a) D. W. Stephan, Org. Biomol. Chem. 2008, 6, 1535–1539; b) S. R. Flynn,
D. F. Wass, ACS Catal. 2013, 3, 2574–2581; c) D. W. Stephan, Acc. Chem.
Res. 2015, 48, 306–316; d) D. W. Stephan, J. Am. Chem. Soc. 2015, 137,
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7523
© 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202100662
Chemistry—A European Journal
[24] a) M. C. Thompson, D. H. Busch, J. Am. Chem. Soc. 1964, 86, 213–217;
b) E. B. Fleischer, E. Klem, Inorg. Chem. 1965, 4, 637–642; c) N. F. Curtis,
Coord. Chem. Rev. 1968, 3, 3–47; d) G. W. Gokel, D. J. Cram, C. L. Liott,
H. P. Harris, F. L. Cook, Org. Synth. 1988, 6, 301; e) F. Ibukuro, M. Fujita,
K. Yamaguchi, J. P. Sauvage, J. Am. Chem. Soc. 1999, 121, 11014–11015;
f) P. G. Edwards, R. Haigh, D. Li, P. D. Newman, J. Am. Chem. Soc.
2006,128, 3818–3830.
2178–2188; c) P. Kovaříček, A. C. Meister, K. Flídrová, R. Cabot, K.
Kovaříčková, J.-M. Lehn, Chem. Sci. 2016, 7, 3215–3226; d) J.-F. Ayme, J.-
M. Lehn, Chem. Sci. 2019, 11, 1114–1121.
[25] a) B. H. Northrop, Y. R. Zheng, C. H. I. Ki-Whan, P. J. Stang, Acc. Chem.
Res. 2009, 42, 1554–1563; b) N. Giuseppone, Acc. Chem. Res. 2012, 45,
Manuscript received: February 22, 2021
Version of record online: April 28, 2021
Chem. Eur. J. 2021, 27, 7516–7524
www.chemeurj.org
7524
© 2021 Wiley-VCH GmbH