Time-Dependent Switching of Constitutional Dynamic Libraries and Networks from Kinetic to Thermodynamic Distributions

Article abstract of DOI:10.1021/jacs.9b09395

The distribution of the constituents of a constitutional dynamic library (CDL) may undergo time-dependent changes as a function of the kinetics of the processes generating the CDL from its components. Thus, the constitutional dynamic network (CDN) representing the connections between the constituents changes from a kinetic distribution to the thermodynamic one as a function of time. We investigated the behavior of dynamic covalent libraries (DCLs) of four constituents generated by reversible formation of C═N bonds between four components, 2 aldehydes and 2 amino compounds, both in absence and in the presence of metal cations. The associated [2 × 2] networks underwent time-dependent changes from the initial kinetic distribution to the final thermodynamic one, involving an orthogonal switch from one diagonal to the other diagonal of the square [2 × 2] network leading to a very large change in distribution. The DCL constituents could be switched from kinetic products (imines) to thermodynamic products (oximes or acylhydrazones) based on the reactivities of the components and the thermodynamic stabilities of the constituents without addition of any external effector, solely on the basis of the intrinsic properties of the self-contained system. Such processes were achieved for purely organic DCLs/CDNs as well as for inorganic ones containing two metal cations, the latter changing from the silver(I) complex of an imine (kinetic product) to the zinc(II) complex of a hydrazone (thermodynamic product). The results bear relationship to out-of-equilibrium systems concerning kinetic behavior in adaptive chemistry.

Full text of DOI:10.1021/jacs.9b09395

Article
pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Time-Dependent Switching of Constitutional Dynamic Libraries and
Networks from Kinetic to Thermodynamic Distributions
Meixia He and Jean-Marie Lehn*
́
́
́
́
Laboratoire de Chimie Supramoleculaire, Institut de Science et d’Ingenierie Supramoleculaires (ISIS), Universite de Strasbourg, 8
́
Allee Gaspard Monge, 67000 Strasbourg, France
S
* Supporting Information
ABSTRACT: The distribution of the constituents of a
constitutional dynamic library (CDL) may undergo time-
dependent changes as a function of the kinetics of the
processes generating the CDL from its components. Thus, the
constitutional dynamic network (CDN) representing the
connections between the constituents changes from a kinetic
distribution to the thermodynamic one as a function of time.
We investigated the behavior of dynamic covalent libraries
(DCLs) of four constituents generated by reversible formation
of CN bonds between four components, 2 aldehydes and 2 amino compounds, both in absence and in the presence of metal
cations. The associated [2 × 2] networks underwent time-dependent changes from the initial kinetic distribution to the final
thermodynamic one, involving an orthogonal switch from one diagonal to the other diagonal of the square [2 × 2] network
leading to a very large change in distribution. The DCL constituents could be switched from kinetic products (imines) to
thermodynamic products (oximes or acylhydrazones) based on the reactivities of the components and the thermodynamic
stabilities of the constituents without addition of any external effector, solely on the basis of the intrinsic properties of the self-
contained system. Such processes were achieved for purely organic DCLs/CDNs as well as for inorganic ones containing two
metal cations, the latter changing from the silver(I) complex of an imine (kinetic product) to the zinc(II) complex of a
hydrazone (thermodynamic product). The results bear relationship to out-of-equilibrium systems concerning kinetic behavior
in adaptive chemistry.
either agonistic or antagonistic nature and lead to either
simultaneous amplification/up-regulation or down-regulation of
the responsive constituent(s), respectively.9c
1. INTRODUCTION
The design, construction, and study of networks of inter-
connected entities bear great significance for the understanding
and control of the behavior of complex systems1 in chemistry
and biology involving for instance the regulation of interde-
pendent reactions. A special case of particular interest is that of
dynamic networks connecting chemical entities of different
constitutions on molecular and supramolecular levels as
presented in constitutional dynamic chemistry (CDC).1c,2
Such constitutional dynamic networks (CDNs)3 link the sets
of chemical entities forming constitutional dynamic libraries
(CDLs),4 that undergo constitutional variation either by
internal rearrangement or by exchange, incorporation, and
extrusion of components through reversible formation of
covalent and noncovalent linkages. As a consequence of this
dynamicity, the dynamic system may be shifted from one
constitutional state to another one in response to physical
stimuli (such as light,5 temperature,3d,5g,6 pressure,7 and various
other factors3e,8) or chemical effectors (metal cations,9
protons,3d,10 reactive molecules,11 and self-sorting12), thus
achieving adaptation. A particular case is that where the system
adapts to a shape switching resulting from a change of the
molecular geometry of a component.13 The variations in the
distribution of the members of a CDL are determined by the
nature of the relationship between constituents which may be of
On the molecular level, the reversible formation of CN
bonds from a carbonyl group and an amino group to give imine-
type compounds (imines,14 hydrazones,15 acylhydrazones,16
and oximes15a,17) has been extensively used to set up dynamic
covalent libraries (DCLs) of various types. It presents a wide
range of rates18 and thermodynamic features19 and is a reaction
of fundamental importance.2e Relative rates of CN formation
will lead to time-dependent changes in the composition of a
DCL representing the kinetic evolution of the system.3f,20 From
previous studies, imines are known to form rapidly, whereas
oximes and acylhydrazones are formed more slowly but have
higher thermodynamic stabilities at equilibrium.20 Relevant
cases in line with these results are found in the experiments
shown below.
Unlike the variations induced by an external agent,5−12 the
internal features of a DCL may lead to a time-dependent change
in the distribution of the constituents according to its kinetic and
thermodynamic properties.3f,20
Received: August 30, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
A particularly interesting case is that where the DCL
undergoes network switching from an initial kinetic distribution
of constituents determined by reaction rates, to the final
thermodynamic distribution at equilibrium, in a sort of kinetic
adaptation or adaptation to time.
The above considerations prompted us to design systems
based on multiple components possessing different structural
and electronic properties to selectively adjust the kinetic and
thermodynamic products within the DCL. We thus designed
libraries of four constituents generated from two aldehydes A, A′
and two amino compounds B, B′ and investigated their kinetic
and thermodynamic properties which lead to component
selection and amplification of the different constituents as a
function of time (Scheme 1). Such a DCL is represented by a
than benzaldehyde and can also generate an NN bidentate or
NNO tridentate coordination site for metal cation binding by
reaction with amines or hydrazides.23 As mentioned above,
amines, together with hydroxylamines or hydrazides, were
chosen because of the differences in the thermodynamic
stabilities of their imino products.19,20
First, we investigated competitive imine and oxime formation
from the three components A1, B1, B2, to check their relative
rates. Equal amounts of each component were mixed in CD3CN
at room temperature. After less than 5 min, only the imine A1B1
was detected as the kinetic product and no oxime was observed
at this moment. After 24 h, the imine was largely converted into
the oxime A1B2 as the thermodynamic product (see Table S1 in
the Supporting Information, SI). The evolution of the 1H NMR
spectra is presented in Figure 1a. The corresponding kinetic
traces (Figure 1b) show the crossing of formation curves of
A1B1 and A1B2. Similar experiments were conducted for the set
[A2 + B1 + B2] containing the much less reactive aldehyde A2
to further explore the design and the kinetics of such DCLs. The
reactions were now much slower and allowed for a better
observation of the beginning of the process (Figure 1c, d and
Table S2). The corresponding separate experiments of imine
A1B1, A2B1 and oxime A1B2, A2B2 formation are shown in
the SI (Figures S26−S33, Tables S3−S7).
Scheme 1. Evolution of a DCL of Four Constituents
Undergoing a Time-Dependent Orthogonal Switching of Its
Associated [2 × 2] CDN from One Diagonal to the Other
One
The reaction in the set of [A1 + A2 + B1] was also studied, but
the exchange reaction was too slow to follow due to the low
reactivity of A2 (see Figures S34 and S35 of the SI). The set [A1
+ A2 + B2] may be expected to behave similarly. The results
above (Figure 1) confirm that, as expected, in both cases, the
imine is the kinetic product while the oxime is the
thermodynamic one.
square [2 × 2] CDN with the four constituents AB, AB′, A′B,
and A′B′ at the corners, linking antagonistic constituents (e.g.,
AB and A′B) along the edges and agonistic ones (AB and A′B′;
AB′ and A′B) through the diagonals as presented earlier.3d The
antagonists (e.g., AB and AB′) share one common component
so that when the amount of one of them increases, that of the
other one will decrease. In contrast, the agonists (e.g., AB and
A′B′) share no component, so that an increase in the amount of
one constituent will enforce the enhancement of the other one,
in an agonist amplification manner. When all of the constituents
have similar energy, a near-statistical distribution is obtained.9d
By selecting aldehydes and amino compounds with appropriate
properties, the library may be made to contain initially almost
exclusively one pair of diagonally located constituents as kinetic
products AB and A′+B′ (here initially unreacted but amplified
agonistic components) and evolve with time (as observed by 1H
NMR spectroscopy21 toward the orthogonal pair of constituents
as thermodynamic products AB′ and its agonist A′B, thus
achieving kinetic network switching of the system (Scheme 1) as
observed.22
2.1.2. Kinetic and Thermodynamic Features of Three DCLs
([1], [2], and [3]) of Four Constituents Generated from Four
Components. 2.1.2.1. DCL[1]. After mixing the four compo-
nents A1, A2, B1, B2 in CD3CN (30 mM each), the
composition of the DCL[1] was monitored by 1H NMR
spectroscopy21,22 as a function of time. As seen in the 1H NMR
spectra shown in Figure 2, A1B1 was detected as the main
component within 25 min together with A1B2, A2B1, and
A2B2 formed in a biased distribution of 36%, 13%, 6%, and 1%,
respectively, as well as 44% free A2 and 37% free B2 and 10%
free B1 (Scheme 2). A1B1 was the overwhelming kinetic
product of the DCL. As such, it should amplify its agonist
constituent A2B2, which was however not detected at this stage,
due to the low reactivities of A2 and B2, which stayed in the free
form. The reason for this biased kinetic distribution may be
attributed to the comparatively fast formation of imine A1B1
with respect to that of the oxime A1B2. As seen above, the
competitive reactions performed for (i) A1 with B1 and B2
(Figure 1a) (ii) A2 with B1 and B2 (Figure 1c) as well as (iii) B1
with A1 and A2 (Figures S34 and S35) showed that A1 and B1
are as expected more reactive than A2 and B2, respectively. Over
time, the oxime A1B2 progressively formed together with its
agonist A2B1, generated concomitantly from the initially
unreacted A2 and B2, whereas at the same time the imine
A1B1 decreased. After about 15 h, the initial components had
almost fully reacted and only the four expected constituents
were present. The composition of the DCL was finally 14.0%/
36%/35%/13% for the constituents A1B1, A1B2, A2B1, A2B2,
respectively (Scheme 2, Table S8). The process thus realizes a
kinetic switching of the CDN associated with DCL[1] from the
diagonal [A1B1, A2 + B2] to the orthogonal agonists diagonal
[A1B2, A2B1]. The kinetic traces show the corresponding
2. RESULTS AND DISCUSSION
2.1. Kinetic Switching of a Constitutional Dynamic
Network―Kinetic vs Thermodynamic Distributions in
Dynamic Covalent Libraries of Imino Compounds.
2.1.1. Component Selection in Competitive Reactions. With
the goal of designing four-component systems suitable for
achieving kinetic switching of constitutional dynamic networks,
three-component experiments were first performed to determine
their kinetic behaviors in a competitive setup. Two constituents
were generated by the reaction of pyridine-2-carboxaldehyde or
benzaldehyde derivatives with amines and hydroxylamines or
hydrazides. The pyridine aldehyde is known to be more reactive
B
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. (Top) Evolution of the 1H NMR (400 MHz) spectra of the mixtures generated from equal amounts of components (a) A1 + B1 + B2 (three
bottom traces) and (c) A2 + B1 + B2 (three bottom traces) after respectively 5 min, 5 h, and 24 h (from the bottom). The two top traces are the spectra
of the isolated constituents A1B1 and A1B2 (left), A2B1 and A2B2 (right), respectively. The arrows indicate the aldehyde CHO proton (9.5−10.0
ppm region) and imine CHN proton NMR signals of the compounds. (Bottom) Kinetic plots of the evolution as a function of time of the
compounds in the mixtures generated from equal amounts of components (b) A1 + B1 + B2 and (d) A2 + B1 + B2, corresponding to the NMR spectra
above. The composition % data have been obtained by integration of the imine CHN and aldehyde CHO proton signals in 1H NMR spectra (30 mM
each, CD3CN, r.t.). For clarity, the kinetic curves are shown for only some of the compounds (see also notes S21 and S22, Tables S1 and S2).
evolution with line crossings (Figure 3). Similar results were
obtained when A2 was replaced by p-chlorobenzaldehyde A4 or
by p-fluorobenzaldehyde A5, which however were more reactive
(as shown in Schemes S1 and S2, Figures S36−S39, and Tables
S9 and S10).
reactions. As shown in the evolution of the 1H NMR spectra and
the kinetic traces (Figures S40 and S41), this initial DCL[2]
underwent reorganization with time and shifted to a
composition where the oxime A3B2 was the thermodynamic
product, resulting also in the amplification of its diagonal agonist
A4B1. The composition of the library became then 9%, 41%,
43%, and 7% for the constituents A3B1, A3B2, A4B1, and
A4B2, respectively. Like in case DCL[1] above, the process
realizes again a kinetic switching of the CDN associated with
DCL[2] from the diagonal [A3B1, A4 + B2] to the orthogonal
agonist diagonal [A3B2, A4B1].
2.1.2.2. DCL[2]. In order to explore the requirements for
obtaining a kinetic network switching, an experiment similar to
that in DCL[1] was performed with DCL[2] formed by
1
components A3, A4, B1, and B2 (Scheme S3). From the H
NMR spectra (Figure S40), the behavior of this DCL[2] was
similar to that of DCL[1] described above. After 4 min, four
components A3B1, A3B2, A4B1, and A4B2 were obtained,
giving a biased distribution of 32%, 14%, 6%, < 1% (Scheme S3,
Figure S41, and Table S11) with the percentage of free A4 and
B2 being 44% and 36%. The reason for this bias may again be
attributed to the fast formation of imine A3B1 rather than that of
the oxime A3B2 due to the low reactivity of A4 compared to A3
as well as of B2 compared to B1 (shown Figures S42−S47,
Tables S12−S13) as confirmed by the separate competitive
2.1.2.3. DCL[3]. Finally a third four component DCL[3] was
studied in the same conditions as for DCL[1] and DCL[2]
above, based on the components A3, A4, B1, and the N-
methylbenzohydrazide B3 (Scheme 3). The formation of the
four constituents A3B1, A3B3, A4B1, and A4B3 was again
1
monitored by H NMR as a function of time. Similarly to
oximes, acylhydrazones are known to be thermodynamically
favored products relative to simple imines.20 B3 was selected to
C
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. Evolution of 1H NMR (400 MHz) spectra of the compounds
generated from the DCL[1] mixture of equal amounts of A1, A2, B1,
B2 (30 mM each in CD3CN, r.t.) after 5 min, 25 min and 15 h (three
bottom traces). The four top traces correspond to the isolated
constituents A1B1, A1B2, A2B1, and A2B2. The arrows indicate the
aldehyde CHO proton (9.5−10.0 ppm region) and imine CHN
proton NMR signals of the compounds.
Figure 3. Kinetic plots of the evolution of the compounds generated
from a mixture of equal amounts of components A1 + A2 + B1 + B2 as a
function of time over 16 h. The composition % data have been obtained
by integration of the imine CHN and aldehyde CHO proton signals
in the 400 MHz 1H NMR spectra (30 mM each, CD3CN, r.t.). The lines
are not calculated but just introduced to guide the eye (see also notes
S21 and S22).
Scheme 2. Kinetic Switching of the [2 × 2] CDN Formed by
the DCL[1] Set of Components [A1 + A2 + B1 + B2] (top)
from the [A1B1, A2 + B2] State (Middle Left) to the
Orthogonal State [A1B2, A2B1] (Middle Right) in CD3CN at
Scheme 3. Kinetic Switching of the [2 × 2] CDN Formed by
the DCL[3] Set of Components [A3 + A4 + B1 + B3] (top)
from the [A3B1, A4 + B3] State (Middle Left) To the
Orthogonal State [A3B3, A4B1] (Middle Right) in CD3CN at
a
Room Temperature )
a
Room Temperature
a
The indicated composition % values of the different compounds
present correspond to reaction times of 25 min (left) and 15 h (right),
see also Table S8 in SI. Data obtained from the 400 MHz H NMR
a
1
The indicated composition % values of the different compounds
present correspond to reaction times of 5 min (left) and 96 h (right),
see also Table S14 in SI. Data obtained from the 400 MHz 1H NMR
spectra (see Figures 4 and 5).
spectra (see Figures 2 and 3).
replace B2 to investigate how this new DCL would behave. After
5 min, only two products were detected, giving a composition
A3B1 48%, A4B1 6%, together with 43% of A4 and 50% of B3
left unreacted in the solution. After 96 h at room temperature,
the reaction reached an equilibrium with a constituent
composition of A3B1 9%, A3B3 41%, A4B1 38%, and A4B3
12%. As shown in the evolution of the 1H NMR spectra and the
kinetic traces (Figures 4, 5, and Table S14), the DCL[3] showed
again an evolution from an initial kinetic distribution to the final
thermodynamic distribution with an orthogonal switching of the
associated CDN from the diagonal axis [A3B1, A4 + B3] to the
orthogonal [A3B3, A4B1] one. Interestingly, the initial
distribution showed <1% of the acylhydrazone A3B3 compared
to 14% for the corresponding oxime A3B2 in the DCL[2] above,
indicating an appreciably lower reactivity of the (N-methylated)
hydrazide B3 compared to the alkoxy amine B2. This feature is
also reflected in the fact that the evolution of the kinetic traces
(Figure 5) are more gradual than in the previous cases above.
Indeed, a separate competition experiment between B2 and B3
reacting with A3 in 1:1:1 ratio yielded 21% of A3B2 and 3% of
A3B3 after 8 days showing that A3B2 is thermodynamically
preferred over A3B3 (Figures S48, S49, and Table S15). On the
kinetic level, the same reaction in 1:1:2 stoichiometry (Figures
S50, S51, and Table S16) gave after 7 days 17% of A3B2 and 4%
D
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
amplification from a statistical distribution, as described
previously.9c−e Such large changes resulting from kinetic factors
represent a very significant feature of the behavior of networks.
2.2. Kinetic Switching of CDNs of Ligand Constituents
Driven by Metal Cations of Different Coordination
Geometries. 2.2.1. Effects of Metal Cations on the Behavior
of DCLs of Ligands. The behavior of a DCL of ligands may be
expected to be strongly influenced by the presence of metal
cations in a cation specific fashion. We thus investigated whether
a switching from a kinetic distribution to a thermodynamic one
could also be achieved in a DCL by competition for
coordination of two different metal cation effectors M1 and
M2, that is, switching the CDN from that favored by one metal
ion to that favored by the other one on the basis of the difference
in both the rate of formation of the respective ligands in the
presence of the cations and relative stabilities of the two
generated complexes.25,26 The evolution of a system undergoing
such a CDN switching is represented in Scheme 4. To achieve
Figure 4. Evolution of 1H NMR (400 MHz) spectra of the compounds
generated from the DCL[3] mixture of equal amounts of A3, A4, B1,
and B3 (30 mM each in CD3CN, r.t.) after 5 min and 96 h (two bottom
traces). The four top traces correspond to the isolated constituents
A3B1, A3B3, A4B1, and A4B3. The arrows indicate the aldehyde CHO
proton (9.5−10.0 ppm region) and imine CHN proton NMR signals
of the compounds.
Scheme 4. Evolution of a DCL of Four Ligand Constituents
Undergoing a Time-Dependent Switching of Its Associated
[2 × 2] CDN Driven by the Formation of the Complexes of
Two Different Metal Cations M1 and M2
this goal requires to select components that will generate ligand
constituents corresponding to the coordination geometries of
different metal ions. It is known that Ag(I) forms preferentially
tetracoordinated metal complexes with bidentate ligands
whereas Zn(II) forms hexacoordinated octahedral complexes
with tridentate ligands. Taking into account these coordination
features together with the stabilities of metal complexes,27,28 our
aim was to produce kinetic switching from a tetrahedral complex
to an octahedral one under the action of the two metal cation
effectors, Ag(I) and Zn(II). To this end, the behavior of the
DCL[4] (A3 + A6 + B4 + B5) in the presence of both Ag(I) and
Zn(II) cations was investigated.
2.2.2. Kinetic and Thermodynamic Features of Three DCLs
([4], [5], and [6]) of Four Ligand Constituents Generated from
Four Components in the Presence of Metal Cations.
2.2.2.1. DCL[4]. After exploration of different component
combinations, the DCL[4] based on A3 + A6 + B4 + B5 was
selected to generate four constituents, comprising in particular
the imine-based ligands A3B4 and A3B5 presenting different
coordination features. The time dependence of its composition
was studied in the presence of both AgOTf and Zn(OTf)2.
One may note that in the process, cation binding by a
component may influence the kinetic behavior of that
component. Thus, it was found in a separate experiment that
Zn(II) binds to the pyridyl-hydrazine B5 to give [Zn(B5)2]2+
(see Figures S54−S56 in SI). Furthermore, one also notes that
the metal cations may catalyze component exchange between
the ligand constituents and thus facilitate the recombination
kinetics.29,30
Figure 5. Kinetic plots of the evolution of a mixture of equal amounts of
components A3 + A4 + B1 + B3 as a function of time. The composition
% data have been obtained by integration of the imine −CH2− (of the
benzyl group), hydrazone N−CH3 and aldehyde CHO proton signals
1
in the 400 MHz H NMR spectra (30 mM each, CD3CN, r.t.). For
clarity, the kinetic curves are shown for only some of the compounds
(see also notes S21 and S22).
of A3B3 with large amount of unreacted A3, indicating also that
the alkoxy amine B2 reacted faster than the hydrazide B3. In the
case of the set of components A1 + A3 + B1 + B3 (Scheme S4),
the similar reactivities of A1 and A3 (6-phenyl derivative of A1)
led to two imines A3B1 (40%) and A1B1 (29%) as kinetic
products (5 min) with some A3 (10%) and A1 (20%) unreacted
in the solution. At equilibrium they finally reached a near-
statistical distribution for A3B1 (19%), A1B1 (25%), A3B3
(32%), and A1B3 (21%) (see data in Figures S52, S53, and
Table S17).
The present DCLs all undergo a network switching from a
kinetic distribution to the thermodynamic distribution as a
function of time. This process that may be considered as an
adaptation to time via component exchange, caused by the
internal kinetic and thermodynamic properties of the con-
stituents.
It is important to note that network switching achieves a much
larger change in distribution24 (by factors of 3 to 6) than simple
E
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
In order to be able to identify the entities present in the rather
complicated mixtures, the H NMR spectra of all individual
distribution of 10% [Ag(A3B4)2]+, 41% [Zn(A3B5)2]2+, 38%
A6B4, 6% A6B5 as well as 5% free A6, 3% B4 and 2% B5
(Figures S71, S72, and Table S18). Thus, the [2 × 2] CDN
associated with the DCL[4] had undergone an orthogonal
kinetic switching between two metal cation complexes of
different coordination geometries from the kinetic product
[Ag(A3B4)2]+ together with [A6+B5] to the thermodynamic
one [Zn(A3B5)2]2+ and A6B4. The switching process is
represented in Scheme 5. In the course of this experiment one
1
components and constituents were measured in the absence and
in the presence of the same metal salts AgOTf and Zn(OTf)2
separately as well as together (see Figures S57−S70 in SI).
A3 + A6 + B4 + B5 (10 mM, 1 equiv each), AgOTf (5 mM,
0.5 equiv), and Zn(OTf)2 (5 mM, 0.5 equiv) were mixed
together in CD3CN at room temperature and the behavior of the
mixture was followed by 1H NMR as a function of time (Figures
6 and S71). As shown in Figures 7 and S72, after 1 h, the silver
Scheme 5. Kinetic Switching of the [2 × 2] CDN Formed by
the DCL[4] Set of Components [A3 + A6 + B4 + B5] (Top)
from the {[Ag(A3B4)2]+, [A6 + B5]} State (Middle Left) To
the Orthogonal State {[Zn(A3B5)2]2+, A6B4} (Middle Right)
a
(in CD3CN at r.t.) after 1 h (Left) and after 70 h (Right)
Figure 6. Evolution of 1H NMR (400 MHz) spectra of the compounds
generated from the DCL[4] mixture of equal amounts of A3, A6, B4,
and B5 (10 mM each in CD3CN, r.t.) + 0.5 equiv. AgOTf+0.5 equiv.
Zn(OTf)2 after 1 h, 24 h, and 70 h (three bottom traces). The four top
traces correspond to the isolated constituents [Ag(A3B4)2]+,
[Zn(A3B5)2]2+, A6B4, and A6B5. The arrows indicate the aldehyde
CHO proton (9.5−10.0 ppm region) and imine CHN proton NMR
signals of the compounds.
a
The indicated composition % values of the different compounds
present correspond to reaction times of 1 h (left) and 70 h (right), see
also Table S18 and Figure S72 in the SI. Free anions are not
1
indicated. Data obtained from the 500 MHz H NMR spectra (see
Figures 6 and 7).
observes that the 1H NMR signals become broad after about 1 h.
To test the possibility that this could be due to the interaction of
ligand A3B4 with both silver and zinc cations, an experiment was
conducted where a solution of [Ag(A3B4)2]+ was titrated with
Zn(OTf)2 (Figure S73). The 1H NMR imine peak of the silver
complex first became broad when 0.1 equiv. Zn(OTf)2 was
added and became broader as more zinc was added. Finally, with
1.0 equiv. Zn(OTf)2, the spectrum was the same as that of
separately prepared [Zn(A3B4)2]2+, showing that initial broad-
ening was due to interaction with the zinc cation which displaces
progressively the silver cation and takes up all the A3B4 ligand.
One also notes that in the absence of metal cations no such
kinetic switching is observed, indicating that the switching
behavior is enforced by the metal cations.
2.2.2.2. DCL[5]. To test whether switching may also be
achieved with 2-pyridinecarboxaldehyde bearing different
substituents, A3 was replaced by the methyl-bearing aldehyde
A7 and the behavior of DCL[5] generated from A6, A7, B4, and
B5 was studied in the presence of both AgOTf and Zn(OTf)2
(Scheme S5). From the 1H NMR spectra (Figure S74), after 1 h
Figure 7. Kinetic plots of the evolution of a mixture of equal amounts of
components A3 + A6 + B4 + B5 (10 mM each) together with 0.5 equiv.
AgOTf + 0.5 equiv. Zn(OTf)2 as a function of time. The composition %
data have been obtained by integration of the imine CHN and
aldehyde CHO proton signals in the 500 MHz 1H NMR spectra
(CD3CN, r.t.) (see Figures S71, S72 and Table S18 in SI). For clarity,
the kinetic curves are shown for only some of the compounds (see notes
S21 and S22).
complex [Ag(A3B4)2]+ of A3B4 had fully formed together with
48% unreacted A6 and 50% B5 as its zinc complex. The solution
was then left at r.t. for 3 days to reach equilibrium (2 days at 60
°C). The silver complex [Ag(A3B4)2]+ strongly decreased and
the formation of the zinc complex [Zn(A3B5)2]2+ (see crystal
structure in SI) was observed together with amplification of its
agonist A6B4. At equilibrium, the switched library gave a
F
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
the silver complex [Ag(A7B4)2]+ (see crystal structure in SI) of
A7B4 was almost fully formed (49%) together with unreacted
A6 and B5 due to the slow rate of hydrazone formation. After 1
day, the silver complex [Ag(A7B4)2]+ had decreased and the
zinc complex [Zn(A7B5)2]2+ (see crystal structure in the SI)
had appeared and then increased with simultaneous amplifica-
tion of its agonist compound A6B4. The kinetic curves are
shown in Figure S75. The equilibrium was reached after heating
at 60 °C for 2 days. The final distribution of the DCL was
[Ag(A7B4)2]+ (11%), [Zn(A7B5)2]2+ (39%), and A6B4 (40%)
together with 10% of free A6 and B5 (Table S19 and Figures
S74−S76).
eventual ancillary interactions between metal cations and
components.
(4) Importantly, for both cases, in the absence as well as in the
presence of metal cations, the switching behavior was
obtained by simply mixing the components of the DCL,
indicating that it resulted solely from the internal kinetic
and thermodynamic properties of the system.
(5) On the basis of the above general considerations, the
adaptation of the CDNs to the kinetics of CN bond
formation as well as to the relative stabilities of the
constituents generated time-depended distributions
resulting in a network switching.
Similar results were obtained when A7 was replaced by 6-
bromopyridine aldehyde A8 (see DCL[6] in SI Scheme S6,
Figures S77−S79, and Table S20 and crystal structure for
[Ag(A8B4)2]+).
Taken together, the operation of the switching process may be
attributed to the following features of the present systems:
The present work explicitly implements network switching
driven by internal kinetic factors and thermodynamic properties
in the self-contained CDNs. The very large constituent
amplification (from 3 to 20 times or more) represents an
adaptation to time, a significant step in the adaptive behavior of
dynamic covalent systems. It paves the way to further
exploration of time-dependent adaptive behavior in constitu-
tional and reactional3f dynamic networks in constitutional
dynamic chemistry. The results described also bear relationship
to the behavior of out-of-equilibrium systems opening toward
higher level kinetic behavior (for instance in training
processes)9d in adaptive chemistry.
(i) the formation of the hydrazone-based zinc complex
[Zn(A3B5)2]2+ is much slower than the formation of
imine-based silver complex[Ag(A3B4)2]+;
(ii) the zinc complex [Zn(A3B5)2]2+ of the tridentate
hydrazone A3B5 is thermodynamically more stable than
the silver complex [Ag(A3B4)2]+ of the bidentate imine
A3B4;
(iii) trapping of Zn(II) by binding to the pyridyl-hydrazine B5
to give [Zn(B5)2]2+ (see above and Figures S54−S56 in
SI) may be expected to slow down the formation of the
[Zn(A3B5)2]2+ complex thus allowing the silver complex
with the imine constituent [Ag(A3B4)2 ]+ to form first.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.9b09395.
■
S
Notes on instrument and experimental details, synthetic
procedures, and characterization of compounds, quanti-
tative NMR spectra, kinetic traces, composition tables,
crystal structures, and data (PDF)
The three metallo-DCLs described here present orthogonal
network switching of the corresponding CDNs from one
diagonal to the other diagonal of the [2 × 2] square, from a
nonequilibrium state to an equilibrium state under the action of
the two metal cation effectors, Ag(I) and Zn(II) as a function of
time. The kinetic up-regulation amounts to a factor of at least 20
for the constituents A6B4 and [Zn(A3B5)2]2+.24
AUTHOR INFORMATION
Corresponding Author
*lehn@unistra.fr
■
ORCID
3. CONCLUSIONS
Jean-Marie Lehn: 0000-0001-8981-4593
The present results demonstrate that the constitutional dynamic
libraries (CDLs) investigated here achieve a time-dependent
switching of the underlying [2 × 2] CDNs from kinetic to
thermodynamic distributions of constituents by selective
screening of the appropriate aldehyde and amino components
(amine, hydroxylamine, hydrazine, hydrazide). Such a kinetic
switching was realized in the absence (section 2.1) as well as in
the presence of metal cations (section 2.2). The design of these
CDLs rests on the selection of components displaying the
appropriate interplay between kinetic and thermodynamic
properties. It involves several determining factors.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the ERC (Advanced Research Grant SUPRADAPT
290585), the USIAS, and the University of Strasbourg for
financial support. M.H. gratefully acknowledges a doctoral
fellowship from the China Scholarship Council. M.H. thanks
Prof. Jack Harrowfield, Dr. Jean-Franco̧ is Ayme, Dr. Artem
Osypenko, Dr. Youssef Atoini for extensive helpful discussions
of the experimental results, as well as Dr. Jean-Louis Schmitt and
Cyril Antheaume for discussion on NMR experiments.
(1) The formation rates usually decrease and the thermody-
namic stabilities increase along the sequence of (aliphatic)
imine, acylhydrazone, hydrazone, and oxime.31
(2) In view of the difference in reactivities of the aldehyde and
amino components, the formation kinetics and distribu-
tion of the imine constituents of the DCL can be
modulated over a large range of reaction rates and
thermodynamic stabilities.
REFERENCES
■
(1) (a) Mainzer, K. Thinking in Complexity; Springer-Verlag: Berlin/
Heidelberg, 1994; pp 1−347. (b) Strogatz, S. H. Exploring Complex
Networks. Nature 2001, 410, 268−276. (c) Lehn, J. M. From
Supramolecular Chemistry towards Constitutional Dynamic Chemistry
and Adaptive Chemistry. Chem. Soc. Rev. 2007, 36, 151−160.
(d) Cohen, R.; Havlin, S. Complex Networks; Cambridge University
Press: Cambridge, 2010; pp 1−238. (e) Nicolis, G.; Nicolis, C.
Foundations of Complex Systems; World Scientific: Singapore, 2012; pp
1−328. (f) Cougnon, F. B. L.; Sanders, J. K. M. Evolution of Dynamic
(3) In the presence of metal cations, as summarized above, the
network switching results from the kinetics of complex
formation, the relative stabilities of the complexes and
G
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Combinatorial Chemistry. Acc. Chem. Res. 2012, 45, 2211−2221.
(g) Lutz, J.-F.; Lehn, J.-M.; Meijer, E. W.; Matyjaszewski, K. From
Precision Polymers to Complex Materials and Systems. Nat. Rev. Mater.
2016, 1, 16024. (h) Ashkenasy, G.; Hermans, T. M.; Otto, S.; Taylor, A.
F. Systems Chemistry. Chem. Soc. Rev. 2017, 46, 2543−2554.
Structural and Functional Evolution of a Library of Constitutional
Dynamic Polymers Driven by Alkali Metal Ion Recognition. Angew.
Chem., Int. Ed. 2009, 48, 7635−7638. (f) Barboiu, M.; Dumitru, F.;
Legrand, Y.-M.; Petit, E.; van der Lee, A. Self-Sorting of Equilibrating
Metallosupramolecular DCLs via Constitutional Crystallization. Chem.
Commun. 2009, 2192−2194. (g) Barboiu, M. Dynamic Interactive
Systems: Dynamic Selection in Hybrid Organic−Inorganic Constitu-
tional Networks. Chem. Commun. 2010, 46, 7466−7476. (h) Miller, B.
L. Dynamic Covalent Chemistry: Catalysing Dynamic Libraries. Nat.
Chem. 2010, 2, 433−434. (i) del Amo, V.; Philp, D. Integrating
Replication-Based Selection Strategies in Dynamic Covalent Systems.
Chem. - Eur. J. 2010, 16, 13304−13318. (j) Hunt, R. A. R.; Otto, S.
Dynamic Combinatorial Libraries: New Opportunities in Systems
Chemistry. Chem. Commun. 2011, 47, 847−858. (k) Rancan, M.;
Tessarolo, J.; Casarin, M.; Zanonato, P. L.; Quici, S.; Armelao, L.
Double Level Selection in a Constitutional Dynamic Library of
Coordination Driven Supramolecular Polygons. Inorg. Chem. 2014, 53,
7276−7287.
(2) (a) Lehn, J.-M.; Eliseev, A. V. Dynamic Combinatorial Chemistry.
Science 2001, 291, 2331−2332. (b) Rowan, S. J.; Cantrill, S. J.; Cousins,
G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Dynamic Covalent Chemistry.
Angew. Chem., Int. Ed. 2002, 41, 898−952. (c) Corbett, P. T.; Leclaire,
J.; Vial, L.; West, K. R.; Wietor, J.-L.; Sanders, J. K. M.; Otto, S. Dynamic
Combinatorial Chemistry. Chem. Rev. 2006, 106, 3652−3711.
(d) Sadownik, J. W.; Ulijn, R. V. Dynamic Covalent Chemistry in
Aid of Peptide Self-Assembly. Curr. Opin. Biotechnol. 2010, 21, 401−
411. (e) Belowich, M. E.; Stoddart, J. F. Dynamic Imine Chemistry.
Chem. Soc. Rev. 2012, 41, 2003−2024. (f) Gasparini, G.; Dal Molin, M.;
Lovato, A.; Prins, J. P. Dynamic Covalent Chemistry. Supramol. Chem.
From Mol. to Nanomater.; Gale, P. A., Steed, J. W., Eds.; John Wiley &
Sons Ltd: Chichester, U.K., 2012; 41, pp 1−29. (g) Luisier, N.; Schenk,
K.; Severin, K. A Four-Component Organogel Based on Orthogonal
Chemical Interactions. Chem. Commun. 2014, 50, 10233−10236.
(h) Jin, Y.; Yu, C.; Denman, R. J.; Zhang, W. Recent Advances in
Dynamic Covalent Chemistry. Chem. Soc. Rev. 2013, 42, 6634−6654.
(i) Li, J.; Nowak, P.; Otto, S. Dynamic Combinatorial Libraries: From
Exploring Molecular Recognition to Systems Chemistry. J. Am. Chem.
Soc. 2013, 135, 9222−9239.
(3) (a) Constitutional Dynamic Chemistry; Barboiu, M., Ed.; Topics
in Current Chemistry; Springer: Berlin/Heidelberg, 2012, 322, pp 1−
314. (b) Herrmann, A. Dynamic Combinatorial/Covalent Chemistry: a
Tool to Read, Generate and Modulate the Bioactivity of Compounds
and Compound Mixtures. Chem. Soc. Rev. 2014, 43, 1899−1933.
(c) Zhang, W.; Jin, Y. Dynamic Covalent Chemistry; Zhang, W., Jin, Y.,
Eds.; John Wiley & Sons Ltd: Chichester, U.K., 2017; pp 1−464.
(d) Giuseppone, N.; Lehn, J. M. Protonic and Temperature
Modulation of Constituent Expression by Component Selection in a
Dynamic Combinatorial Library of Imines. Chem. - Eur. J. 2006, 12,
1715−1722. (e) Hafezi, N.; Lehn, J. M. Adaptation of Dynamic
Covalent Systems of Imine Constituents to Medium Change by
Component Redistribution under Reversible Phase Separation. J. Am.
Chem. Soc. 2012, 134, 12861−12868. (f) Armao, J. J.; Lehn, J. M.
Nonlinear Kinetic Behavior in Constitutional Dynamic Reaction
Networks. J. Am. Chem. Soc. 2016, 138, 16809−16814. (g) Dhers, S.;
Holub, J.; Lehn, J.-M. Coevolution and Ratiometric Behaviour in Metal
Cation-Driven Dynamic Covalent Systems. Chem. Sci. 2017, 8, 2125−
2130. (h) Wang, S.; Yue, L.; Shpilt, Z.; Cecconello, A.; Kahn, J. S.; Lehn,
J.-M.; Willner, I. Controlling the Catalytic Functions of DNAzymes
within Constitutional Dynamic Networks of DNA Nanostructures. J.
Am. Chem. Soc. 2017, 139, 9662−9671. (i) Yue, L.; Wang, S.;
Cecconello, A.; Lehn, J. M.; Willner, I. Orthogonal Operation of
Constitutional Dynamic Networks Consisting of DNA-Tweezer
Machines. ACS Nano 2017, 11, 12027−12036. (j) Yue, L.; Wang, S.;
Lilienthal, S.; Wulf, V.; Remacle, F.; Levine, R. D.; Willner, I.
Intercommunication of DNA-Based Constitutional Dynamic Net-
works. J. Am. Chem. Soc. 2018, 140, 8721−8731. (k) Zhou, Z.; Yue, L.;
Wang, S.; Lehn, J.-M.; Willner, I. DNA-Based Multiconstituent
Dynamic Networks: Hierarchical Adaptive Control over the
Composition and Cooperative Catalytic Functions of the Systems. J.
Am. Chem. Soc. 2018, 140, 12077−12089. (l) Hewitt, S. H.; Wilson, A.
J. Generation of Dynamic Combinatorial Libraries Using Hydrazone-
Functionalized Surface Mimetics. Eur. J. Org. Chem. 2018, 2018, 1872−
1879.
(5) (a) Vantomme, G.; Lehn, J.-M. Reversible Adaptation to
Photoinduced Shape Switching by Oligomer-Macrocycle Interconver-
sion with Component Selection in a Three-State Constitutional
Dynamic System. Chem. - Eur. J. 2014, 20, 16188−16193.
(b) Vantomme, G.; Hafezi, N.; Lehn, J.-M. A Light-Induced Reversible
Phase Separation and Its Coupling to a Dynamic Library of Imines.
Chem. Sci. 2014, 5, 1475−1483. (c) Ji, S.; Cao, W.; Yu, Y.; Xu, H.
Dynamic Diselenide Bonds: Exchange Reaction Induced by Visible
Light without Catalysis. Angew. Chem., Int. Ed. 2014, 53, 6781−6785.
̌
̌
(d) Kathan, M.; Kovarícek, P.; Jurissek, C.; Senf, A.; Dallmann, A.;
Thunemann, A. F.; Hecht, S. Control of Imine Exchange Kinetics with
Photoswitches to Modulate Self-Healing in Polysiloxane Networks by
Light Illumination. Angew. Chem., Int. Ed. 2016, 55, 13882−13886.
̈
̀
(e) Kassem, S.; Lee, A. T. L.; Leigh, D. A.; Markevicius, A.; Sola, J. Pick-
up, Transport and Release of a Molecular Cargo Using a Small-
molecule Robotic Arm. Nat. Chem. 2016, 8, 138−143. (f) Cvrtila, I.;
́
Fanlo-Virgos, H.; Schaeffer, G.; Monreal Santiago, G.; Otto, S. Redox
Control over Acyl Hydrazone Photoswitches. J. Am. Chem. Soc. 2017,
139, 12459−12465. (g) Herder, M.; Lehn, J.-M. The Photodynamic
Covalent Bond: Sensitized Alkoxyamines as a Tool to Shift Reaction
Networks Out-of-Equilibrium Using Light Energy. J. Am. Chem. Soc.
2018, 140, 7647−7657.
(6) Mukherjee, S.; Cash, J. J.; Sumerlin, B. S. Responsive Dynamic
Covalent Polymers; Zhang, W., Jin, Y., Eds.; John Wiley & Sons, Ltd:
Chichester, U.K., 2018; pp 321−358.
(7) Sobczak, S.; Drozdz, W.; Lampronti, G. I.; Belenguer, A. M.;
̇ ̇
Katrusiak, A.; Stefankiewicz, A. R. Dynamic Covalent Chemistry under
High-Pressure: a New Route to Disulfide Metathesis. Chem. - Eur. J.
2018, 24, 8769−8773.
(8) (a) Giuseppone, N.; Fuks, G.; Lehn, J.-M. Tunable Fluorene-
Based Dynamers through Constitutional Dynamic Chemistry. Chem. -
Eur. J. 2006, 12, 1723−1735. (b) Giuseppone, N.; Lehn, J.-M. Electric-
Field Modulation of Component Exchange in Constitutional Dynamic
Liquid Crystals. Angew. Chem., Int. Ed. 2006, 45, 4619−4624.
(c) Zhang, Y.; Barboiu, M. Mechanism Insight into the Constitutional
Phase Change Selection of Dynameric Framework Libraries. ACS
Omega 2018, 3, 329−333.
(9) (a) Giuseppone, N.; Schmitt, J. L.; Lehn, J. M. Driven Evolution of
a Constitutional Dynamic Library of Molecular Helices toward the
Selective Generation of [2 × 2] Gridlike Arrays under the Pressure of
Metal Ion Coordination. J. Am. Chem. Soc. 2006, 128, 16748−16763.
(b) Klein, J. M.; Saggiomo, V.; Reck, L.; Lu
̈
ning, U.; Sanders, J. K. M.
(4) (a) Lehn, J.-M. Dynamic Combinatorial Chemistry and Virtual
Combinatorial Libraries. Chem. - Eur. J. 1999, 5, 2455−2463. (b) Brisig,
B.; Sanders, J. K. M.; Otto, S. Selection and Amplification of a Catalyst
from a Dynamic Combinatorial Library. Angew. Chem., Int. Ed. 2003,
42, 1270−1273. (c) Turega, S. M.; Lorenz, C.; Sadownik, J. W.; Philp,
D. Target-Driven Selection in a Dynamic Nitrone Library. Chem.
Commun. 2008, 4076−4078. (d) Xu, S.; Giuseppone, N. Self-
Duplicating Amplification in a Dynamic Combinatorial Library. J.
Am. Chem. Soc. 2008, 130, 1826−1827. (e) Fujii, S.; Lehn, J.-M.
Dynamic Combinatorial Libraries for the Recognition of Heavy Metal
Ions. Org. Biomol. Chem. 2012, 10, 60−66. (c) Vantomme, G.; Jiang, S.;
Lehn, J.-M. Adaptation in Constitutional Dynamic Libraries and
Networks, Switching between Orthogonal Metalloselection and
Photoselection Processes. J. Am. Chem. Soc. 2014, 136, 9509−9518.
(d) Holub, J.; Vantomme, G.; Lehn, J.-M. Training a Constitutional
Dynamic Network for Effector Recognition: Storage, Recall, and
Erasing of Information. J. Am. Chem. Soc. 2016, 138, 11783−11791.
(e) Men, G.; Lehn, J.-M. Higher Order Constitutional Dynamic
H
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Networks: [2 × 3] and [3 × 3] Networks Displaying Multiple,
Synergistic and Competitive Hierarchical Adaptation. J. Am. Chem. Soc.
2017, 139, 2474−2483. (f) Osypenko, A.; Dhers, S.; Lehn, J.-M.
Pattern Generation and Information Transfer through a Liquid/Liquid
Interface in 3D Constitutional Dynamic Networks of Imine Ligands in
Response to Metal Cation Effectors. J. Am. Chem. Soc. 2019, 141,
12724−12737.
(n) Liang, C.; Kulchat, S.; Jiang, S.; Lehn, J.-M. Gelation-Driven
Selection in Dynamic Covalent CC/CN Exchange. Chem. Sci.
̌
́
2017, 8, 6822−6828. (o) Hsu, C. W.; Miljanic, O. S. Self-Sorting through
Dynamic Covalent Chemistry; Zhang, W., Jin, Y., Eds.; John Wiley &
Sons Ltd: Chichester, UK, 2018; pp 253−284.
(13) Lehn, J.; Ulrich, S. Adaptation to Shape Switching by
Component Selection in a Constitutional Dynamic System. J. Am.
Chem. Soc. 2009, 131, 5546−5559.
̈
(10) (a) Tauk, L.; Schroder, A. P.; Decher, G.; Giuseppone, N.
Hierarchical Functional Gradients of pH-Responsive Self-Assembled
Monolayers Using Dynamic Covalent Chemistry on Surfaces. Nat.
Chem. 2009, 1, 649−656. (b) Bracchi, M. E.; Fulton, D. A. Orthogonal
Breaking and Forming of Dynamic Covalent Imine and Disulfide Bonds
in Aqueous Solution. Chem. Commun. 2015, 51, 11052−11055.
(c) Wang, Y.; Xing, P.; An, W.; Ma, M.; Yang, M.; Luan, T.; Tang,
R.; Wang, B.; Hao, A. pH-Responsive Dipeptide-Based Dynamic
Covalent Chemistry Systems Whose Products and Self-Assemblies
Depend on the Structure of Isomeric Aromatic Dialdehydes. Langmuir
(14) (a) Layer, R. W. The Chemistry of Imines. Chem. Rev. 1963, 63,
489−510. (b) The Chemistry of Carbon-Nitrogen Double Bond; Patai, S.,
Eds.; John Wiley & Sons Ltd, 1970; pp 1−794. (c) Zhou, Y.; Li, L.; Ye,
H.; Zhang, L.; You, L. Quantitative Reactivity Scales for Dynamic
Covalent and Systems Chemistry. J. Am. Chem. Soc. 2016, 138, 381−
389.
(15) (a) Dirksen, A.; Dawson, P. E. Rapid Oxime and Hydrazone
Ligations with Aromatic Aldehydes for Biomolecular Labeling.
Bioconjugate Chem. 2008, 19, 2543−2548. (b) Blanden, A. R.;
Mukherjee, K.; Dilek, O.; Loew, M.; Bane, S. L. 4-Aminophenylalanine
as a Biocompatible Nucleophilic Catalyst for Hydrazone Ligations at
Low Temperature and Neutral pH. Bioconjugate Chem. 2011, 22,
1954−1961. (c) Kool, E. T.; Park, D.-H.; Crisalli, P. Fast Hydrazone
Reactants: Electronic and Acid/Base Effects Strongly Influence Rate at
Biological pH. J. Am. Chem. Soc. 2013, 135, 17663−17666. (d) Crisalli,
P.; Kool, E. T. Importance of Ortho Proton Donors in Catalysis of
Hydrazone Formation. Org. Lett. 2013, 15, 1646−1649. (e) Tatum, L.
A.; Su, X.; Aprahamian, I. Simple Hydrazone Building Blocks for
Complicated Functional Materials. Acc. Chem. Res. 2014, 47, 2141−
2149. (f) Su, X.; Aprahamian, I. Hydrazone-based Switches, Metallo-
Assemblies and Sensors. Chem. Soc. Rev. 2014, 43, 1963−1981.
(g) Pramanik, S.; Aprahamian, I. Hydrazone Switch-Based Negative
Feedback Loop. J. Am. Chem. Soc. 2016, 138, 15142−15145.
̈
2018, 34, 13725−13734. (d) Ren, Y.; Svensson, P. H.; Ramstrom, O. A
Multi-Controlled Enamine Configurational Switch Undergoing Dy-
namic Constitutional Exchange. Angew. Chem., Int. Ed. 2018, 57, 6256−
6260.
̈
(11) (a) Ramstrom, O.; Lehn, J.-M. Drug Discovery by Dynamic
Combinatorial Libraries. Nat. Rev. Drug Discovery 2002, 1, 26−36.
(b) Mondal, M.; Hirsch, A. K. H. Dynamic Combinatorial Chemistry: a
Tool to Facilitate the Identification of Inhibitors for Protein Targets.
Chem. Soc. Rev. 2015, 44, 2455−2488. (c) Ren, Y.; You, L. Dynamic
Signaling Cascades: Reversible Covalent Reaction-Coupled Molecular
Switches. J. Am. Chem. Soc. 2015, 137, 14220−14228. (d) Frei, P.;
Hevey, R.; Ernst, B. Dynamic Combinatorial Chemistry: A New
Methodology Comes of Age. Chem. - Eur. J. 2019, 25, 60−73. (e) Zou,
H.; Hai, Y.; Ye, H.; You, L. Dynamic Covalent Switches and
Communicating Networks for Tunable Multicolor Luminescent
Systems and Vapor Responsive Materials. J. Am. Chem. Soc. 2019,
141, 16344.
̌
̌
(16) (a) Van Dijken, D. J.; Kovarícek, P.; Ihrig, S. P.; Hecht, S.
Acylhydrazones as Widely Tunable Photoswitches. J. Am. Chem. Soc.
2015, 137, 14982−14991. (b) Gordillo Varela, M. A.; Zuluaga, F.;
Chaur Valencia, M. N. Acylhydrazone-based Dynamic Combinatorial
Libraries: Study of the Thermodynamic/Kinetic Evolution, Configura-
tional and Coordination Dynamics. Rev. Colomb. Quim. 2016, 45, 39−
50.
(12) (a) Nitschke, J. R.; Lehn, J.-M. Self-Organization by Selection:
Generation of a Metallosupramolecular Grid Architecture by Selection
of Components in a Dynamic Library of Ligands. Proc. Natl. Acad. Sci.
U. S. A. 2003, 100, 11970−11974. (b) Sreenivasachary, N.; Lehn, J.-M.
Gelation-Driven Component Selection in the Generation of Constitu-
tional Dynamic Hydrogels Based on Guanine-Quartet Formation. Proc.
Natl. Acad. Sci. U. S. A. 2005, 102, 5938−5943. (c) Saur, I.; Scopelliti,
R.; Severin, K. Utilization of Self-Sorting Processes to Generate
Dynamic Combinatorial Libraries with New Network Topologies.
Chem. - Eur. J. 2006, 12, 1058−1066. (d) Schultz, D.; Nitschke, J. R.
Kinetic and Thermodynamic Selectivity in Subcomponent Substitu-
tion. Chem. - Eur. J. 2007, 13, 3660−3665. (e) Sreenivasachary, N.;
Lehn, J.-M. Structural Selection in G-Quartet-Based Hydrogels and
Controlled Release of Bioactive Molecules. Chem. - Asian J. 2008, 3,
134−139. (f) Sadownik, J. W.; Philp, D. A Simple Synthetic Replicator
Amplifies Itself from a Dynamic Reagent Pool. Angew. Chem., Int. Ed.
2008, 47, 9965−9970. (g) Lao, L. L.; Schmitt, J.-L.; Lehn, J.-M.
Evolution of a Constitutional Dynamic Library Driven by Self-
Organisation of a Helically Folded Molecular Strand. Chem. - Eur. J.
2010, 16, 4903−4910. (h) Orrillo, A. G.; Furlan, R. L. E.
Supramolecular Interactions between Library Members Modulate the
Behavior of Dynamic Combinatorial Libraries. J. Org. Chem. 2010, 75,
211−214. (i) Ghosh, S.; Isaacs, L. Complex Self-Sorting Systems. Miller,
B. L., Ed.; John Wiley & Sons, Inc: Hoboken, 2010; pp 118−154.
(j) Simpson, M. G.; Pittelkow, M.; Watson, S. P.; Sanders, J. K. M.
Dynamic combinatorial chemistry with hydrazones: cholate-based
building blocks and libraries. Org. Biomol. Chem. 2010, 8, 1173−1180.
(17) (a) Jencks, W. P. Studies on the Mechanism of Oxime and
Semicarbazone Formation. J. Am. Chem. Soc. 1959, 81, 475−481.
(b) Polyakov, V. A.; Nelen, M. I.; Nazarpack-Kandlousy, N.; Ryabov, A.
D.; Eliseev, A. V. Imine Exchange in O-aryl and O-alkyl Oximes As a
Base Reaction for Aqueous Dynamic Combinatorial Libraries. A
Kinetic and Thermodynamic Study. J. Phys. Org. Chem. 1999, 12, 357−
363. (c) Dirksen, A.; Hackeng, T. M.; Dawson, P. E. Nucleophilic
Catalysis of Oxime Ligation. Angew. Chem., Int. Ed. 2006, 45, 7581−
́
7584. (d) Thygesen, M. B.; Munch, H.; Sauer, J.; Clo, E.; Jørgensen, M.
R.; Hindsgaul, O.; Jensen, K. J. Nucleophilic Catalysis of Carbohydrate
Oxime Formation by Anilines. J. Org. Chem. 2010, 75, 1752−1755.
(e) Wendeler, M.; Grinberg, L.; Wang, X.; Dawson, P. E.; Baca, M.
Enhanced Catalysis of Oxime-Based Bioconjugations by Substituted
Anilines. Bioconjugate Chem. 2014, 25, 93−101. (f) Wang, S.; Gurav,
D.; Oommen, O. P.; Varghese, O. P. Insights into the Mechanism and
Catalysis of Oxime Coupling Chemistry at Physiological pH. Chem. -
Eur. J. 2015, 21, 5980−5985. (g) Schmidt, P.; Stress, C.; Gillingham, D.
Boronic Acids Facilitate Rapid Oxime Condensations at Neutral PH.
Chem. Sci. 2015, 6, 3329−3333. (h) Agten, S. M.; Suylen, D. P. L.;
Hackeng, T. M. Oxime Catalysis by Freezing. Bioconjugate Chem. 2016,
27, 42−46.
(18) (a) Dirksen, A.; Dirksen, S.; Hackeng, T. M.; Dawson, P. E.
Nucleophilic Catalysis of Hydrazone Formation and Transimination:
Implications for Dynamic Covalent Chemistry. J. Am. Chem. Soc. 2006,
128, 15602−15603. (b) Crisalli, P.; Kool, E. T. Water-Soluble
Organocatalysts for Hydrazone and Oxime Formation. J. Org. Chem.
2013, 78, 1184−1189. (c) Rashidian, M.; Mahmoodi, M. M.; Shah, R.;
Dozier, J. K.; Wagner, C. R.; Distefano, M. D. A Highly Efficient
Catalyst for Oxime Ligation and Hydrazone-Oxime Exchange Suitable
for Bioconjugation. Bioconjugate Chem. 2013, 24, 333−342. (d) Kool,
E. T.; Crisalli, P.; Chan, K. M. Fast Alpha Nucleophiles: Structures That
̌
́
(k) Ji, Q.; Lirag, R. C.; Miljanic, O. S. Kinetically Controlled
Phenomena in Dynamic Combinatorial Libraries. Chem. Soc. Rev.
̀
2014, 43, 1873−1884. (l) Sola, J.; Lafuente, M.; Atcher, J.; Alfonso, I.
Constitutional Self-Selection from Dynamic Combinatorial Libraries in
Aqueous Solution through Supramolecular Interactions. Chem.
̈
Commun. 2014, 50, 4564−4566. (m) Schaufelberger, F.; Ramstrom,
O. Kinetic Self-Sorting of Dynamic Covalent Catalysts with Systemic
Feedback Regulation. J. Am. Chem. Soc. 2016, 138, 7836−7839.
I
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Undergo Rapid Hydrazone/Oxime Formation at Neutral pH. Org. Lett.
(31) These sequences of relative rates and stabilities are only
qualitative and may change depending on the structures of the reagents
and other parameters such as solvent, pH, and so forth.
̈
2014, 16, 1454−1457. (e) Kolmel, D. K.; Kool, E. T. Oximes and
Hydrazones in Bioconjugation: Mechanism and Catalysis. Chem. Rev.
2017, 117, 10358−10376. (f) Larsen, D.; Kietrys, A. M.; Clark, S. A.;
Park, H. S.; Ekebergh, A.; Kool, E. T. Exceptionally Rapid Oxime and
Hydrazone Formation Promoted by Catalytic Amine Buffers with Low
́
Toxicity. Chem. Sci. 2018, 9, 5252−5259. (g) Drienovska, I.; Mayer, C.;
Dulson, C.; Roelfes, G. A Designer Enzyme for Hydrazone and oxime
Formation Featuring an Unnatural Catalytic Aniline Residue. Nat.
Chem. 2018, 10, 946−952.
(19) Kalia, J.; Raines, R. T. Hydrolytic Stability of Hydrazones and
Oximes. Angew. Chem., Int. Ed. 2008, 47, 7523−7526.
(20) Kulchat, S.; Chaur, M. N.; Lehn, J.-M. Kinetic Selectivity and
Thermodynamic Features of Competitive Imine Formation in
Dynamic Covalent Chemistry. Chem. - Eur. J. 2017, 23, 11108−11118.
(21) The compositions of the DCLs given below have been
determined by integration of characteristic 1H NMR signals (−CHO,
−CHN, aromatic −H, −CH2, −CH3) with respect to the proton
signal (at 0.06 ppm) of hexamethyldisiloxane (HMDSO) as internal
standard. The sum of the % for each component in the different entities
present in the DCL is equal to 50%, the maximum any constituent can
reach. The error in 1H NMR integration amounts to about 5% (see also
SI for details).
(22) In view of the complicated behavior of competitive reactions in
the present systems, a determination of the rate constants and a full
kinetic analysis were not attempted. Furthermore, the goal was to
demonstrate the occurrence of kinetic network switching, as is clearly
shown by the crossing of the distribution curves in Figures 1, 3, 5, and 7.
Tables S1, S2, S8, S14, and S18 in the SI provide quantitative data on
the distributions at different time points.
(23) Men, G.; Lehn, J.-M. Multiple Adaptation of Constitutional
Dynamic Networks and Information Storage in Constitutional
Distributions of Acylhydrazones. Chem. Sci. 2019, 10, 90−98.
(24) In the previous studies, starting from a statistical distribution
(about 25% of each of the four constituents) of the initial equilibrated
state of the DCL, the maximum amplification achievable for a given
constituent was at most twice that in the initial distribution, i.e., from
25% to a maximum of 50% (together with 50% of its agonist). The
processes described here display much larger changes in distributions:
they amount to a network switching from an initial state where one pair
of diagonally linked agonistic constituents is almost absent to a final
state where this same pair is strongly dominant and conversely for the
other pair of constituents. For example, in DCL[4], the constituents
[Zn(A3B5)2]2+ and A6B4 are amplified from 1% and 2%, respectively,
in the initial kinetic distribution to 41% and 38% in the final
thermodynamic distribution, a change by more than a factor of 20.
(25) Ayme, J.-F.; Lehn, J.-M. From Coordination Chemistry to
Adaptive Chemistry. In Adv. Inorg. Chem.; van Eldik, R., Puchta, R.,
Eds.; Academic Press: Cambridge, MA, 2018; 71, pp 3−78.
(26) (a) Uruska, I., Libus, W. The Relative Stability of Octahedral and
Tetrahedral Complexes of Transition Metal Ions. In Proceedings of the
8th International Conference on Coordination Chemistry; Gutmann, V.,
Eds.; Springer: Vienna, 1964; pp 340−343. (b) Campbell, V. E.; de
Hatten, X.; Delsuc, N.; Kauffmann, B.; Huc, I.; Nitschke, J. R.
Cascading Transformations within a Dynamic Self-assembled System.
Nat. Chem. 2010, 2, 684−687.
(27) Roberts, D. A.; Pilgrim, B. S.; Nitschke, J. R. Covalent Post-
assembly Modification in MetalloSupramolecular Chemistry. Chem.
Soc. Rev. 2018, 47, 626−644.
(28) Irving, H.; Williams, R. J. P. The Stability of Transition-Metal
Complexes. J. Chem. Soc. 1953, 3192−3210.
(29) Giuseppone, N.; Schmitt, J.-L.; Schwartz, E.; Lehn, J.-M.
Scandium(III) Catalysis of Transimination Reactions. Independent
and Constitutionally Coupled Reversible Processes. J. Am. Chem. Soc.
2005, 127, 5528−5539.
̈
(30) Zhang, Y.; Xie, S.; Yan, M.; Ramstrom, O. Dynamic Covalent
Chemistry of Aldehyde Enamines: BiIII- and ScIII-Catalysis of Amine−
Enamine Exchange. Chem. - Eur. J. 2017, 23, 11908−11912.
J
DOI: 10.1021/jacs.9b09395
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX