Adaptation of dynamic covalent systems of imine constituents to medium change by component redistribution under reversible phase separation

Article abstract of DOI:10.1021/ja305379c

A dynamic covalent library of interconverting imine constituents, dissolved in an acetonitrile/water mixture, undergoes constitutional reorganization upon phase separation induced by a physical stimulus (heat) or a chemical effector (inorganic salt, carbo

Full text of DOI:10.1021/ja305379c

Article
pubs.acs.org/JACS
Adaptation of Dynamic Covalent Systems of Imine Constituents to
Medium Change by Component Redistribution under Reversible
Phase Separation
Nema Hafezi and Jean-Marie Lehn*
Institut de Science et d’Ingen
́
́ ́ ́
ierie Supramoleculaires, ISIS Universite de Strasbourg, 8 allee Gaspard Monge, Strasbourg 67000, France
S
* Supporting Information
ABSTRACT: A dynamic covalent library of interconverting
imine constituents, dissolved in an acetonitrile/water mixture,
undergoes constitutional reorganization upon phase separation
induced by a physical stimulus (heat) or a chemical effector
(inorganic salt, carbohydrate, organic solvent). The process
has been made reversible, regenerating the initial library upon
phase reunification. It represents the behavior of a dynamic
covalent library upon reversible phase separation and its
adaptation to a phase change, with up-regulation in each phase of the fittest constituents by component selection. Finally, the
system exemplifies the splitting of a 2D (square) constitutional dynamic network into a 3D (cube) one.
of constituents to a change in medium.14 Phase changes may
INTRODUCTION
■
Constitutional dynamic chemistry (CDC)1 encompasses
chemical systems based on noncovalent interactions as well
as on reversible covalent reactions, the latter defining dynamic
covalent chemistry (DCC).2 Such systems may respond to the
application of physical stimuli or chemical effectors by
undergoing adaptation of their constituents through constitu-
tional variation via component exchange and selection. Physical
stimuli that have been applied to dynamic covalent systems
include temperature,3 crystallization,4 mechanical stress,5 light,6
and an electric field,7 while chemical effectors comprise
protons,3 metal cations,8 and medium/solvent9 as well as
molecular recognition interactions,10 such as hydrogen
bonding.10b These effectors induce the expression of the
‘fittest’ entity from a distribution of all possible constituents.
DCC is also making a profound impact on the development of
novel, dynamic entities, such as dynamic polymers (dy-
namers)11 as well as self-sensing devices7 and self-healing
materials,12 which can adapt in response to outside stimuli.
Recent work from our laboratory has in particular examined the
effects of gel formation, resulting from the self-assembly of
guanine quartets, on a dynamic library of acylhydrazones, which
demonstrated three constitutional dynamic features: supra-
molecular recognition interactions, dynamic covalent bond
connection, and gel/sol distribution.13 The latter represents a
response of the system to the formation of an organized
assembly, the gel. In general terms, it points to the behavior of a
dynamic system in response to a phase change, a feature that
deserves closer exploration, as it represents a physicochemical
transition of the system, an attractive step toward the design of
systems out of equilibrium. It relates to component selection
and constituent expression and adaptation1,3 under the pressure
of self-organization on one hand1b,13 as well as to phase transfer
and transport processes involving adaptation of a dynamic set
act on a constitutional dynamic system undergoing gas/liquid,
liquid/liquid, or liquid/solid transfer. Thus, the latter occurs on
crystallization of a specific entity from an interconverting set.4
Also along these lines, we have shown earlier that a change
from organo-aqueous to aqueous medium was associated with a
strong, hydrophobicity-driven monomer selection in the
generation of a dynamic polymer.15
Herein, we report on adaptation of a dynamic covalent
system to phase change, via component selection within sets of
interconverting imine constituents, caused by the separation of
a homogeneous liquid phase into two nonmiscible phases
through the action of a physical stimulus or the addition of a
chemical effector (Figure 1). It results in the reorganization of a
chemical effector (Figure 1). It results in the reorganization of a
dynamic library based on the medium preference of the library
members. Moreover, we demonstrate reversibility in these
phase separations by switching between the biphasic state and
the homogeneous solution. Finally, we note that such processes
may provide access to dynamic nonequilibrium systems.
RESULTS AND DISCUSSION
■
The separation of an organo-aqueous mixture into two separate
phases leads to the formation of two distinct solvent
environments from a single one. These two media differ
considerably from each other and from the homogeneous phase
in their physicochemical properties, such as polarity and
viscosity. As a consequence, they may be expected to affect the
distribution of the reversibly interconverting constituents of a
dynamic covalent system and drive the amplification/up-
Received: June 4, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Dynamic libraries of imines were selected because of their
facile formation and dissociation in neutral aqueous media.2
They consisted each of four components: a hydrophilic
aldehyde and amine and a hydrophobic aldehyde and amine.
Salicylaldehyde 1 was chosen in view of the high rate of
formation and exchange of its imines and its appreciable
hydrophobicity.26 In addition, for the first library (Scheme 1), a
charged ammonium analogue of salicylaldehyde 2 was selected
as the hydrophilic counterpart, while heptylamine 3 and the
sodium salt of taurine 4 were chosen as the organo-soluble and
hydrophilic amine components, respectively.
Effects of Reversible Phase Separation on the
Dynamic Covalent Library 5−8. Dissolution of compounds
1−4 (20 mM each) in 3:2 (v/v) AN/W generated a single
phase solution containing a given distribution of imines 5−8, as
observed by 1H NMR and quantified by integrating the distinct
CHN protons against an external standard (Figure 2). The
four imines, possessing either hydrophilic−hydrophilic (5),
hydrophobic−hydrophobic (8), or amphiphilic characteristics
(6 and 7), were present in almost statistical distribution (but
see also below, the case of 12−15).
Figure 1. Schematic representation of the component redistribution
occurring in a dynamic covalent library on phase separation and
recombination of a binary organo-aqueous solvent mixture. (Left)
Library constituents in the single homogeneous phase (green).
(Center) Distribution after phase separation and equilibration,
(yellow) organic phase, and (blue) aqueous phase. (Right)
regeneration of the initial library after phase reunification. Red and
blue spheres: lipophilic and hydrophilic components. In the present
case, a 3/2 (v/v) CH3CN/H2O mixture was used. Phase separation
was induced by treating the mixture with either NaCl, sucrose, diethyl
ether, KF, or by cooling, and phase recombination was achieved by
heating, fractional evaporation, LiClO4, or warming to room
temperature, respectively.
Phase separation was induced upon treatment with sodium
chloride, potassium fluoride, sucrose, or diethyl ether27 as well
as by cooling to 0 °C, leading to a marked reorganization of the
dynamic library, whereby the hydrophilic 5 was strongly
distributed into the aqueous phase, whereas the agonistic (see
Figure 6) hydrophobic constituent 8 was dominant in the
organic phase. As expected, this reorganization was more
pronounced when the separated phases had a more
pronounced difference in AN/W composition (see SI page
S27−S28).
Two parameters may be used to characterize the changes
introduced by the phase separation: the distribution of the
different constituents between the two phases and the
amplification (or conversely the reduction) of specific
constituents (summed over the two phases), with respect to
their fraction in the homogeneous phase. One notes that the
amplification cannot exceed 50% for any constituent; the other
50% being its agonist. Furthermore, one must take into account
the fact that some hydrolysis is present and may slightly change
between homogeneous and separated states, thus reducing the
total amount of library imine members.
regulation of the fittest compounds in their respective
environments. We herein describe such effects induced by
phase separation of acetonitrile/water (AN/W) mixtures on
two different dynamic covalent libraries of imines.
Medium, Phase Separation Procedures, and Compo-
nents of the Dynamic Covalent Library. We selected AN/
W mixtures (3/2, v/v) as a medium, in view of the miscibility of
the two solvents, their physicochemical properties, and their
extensive use.16−24 Phase separation was accomplished by
either a physical stimulus (temperature) or the addition of a
chemical effector of one of three different types: an inorganic
salt, such as sodium chloride or potassium fluoride, a
hydrophilic organic molecule, such as sucrose, or a water
immiscible organic liquid, such as diethyl ether.25 The AN/W
composition of the separated phases, which is expected to affect
phase distribution, varied depending on the procedure used for
causing separation. The strongly hydrophilic inorganic salts
caused higher differentiation in composition of the aqueous and
organic phases than agents which present less pronounced
solubility preferences between AN and W (see Supporting
Information (SI) for exact compositions, pages S27−S28). The
sets of imines investigated here were generated from the
aldehyde and amine components 1−4 and 1, 9−11 giving
imines 5−8 and 12−15, respectively.
Sodium chloride was found to induce the most bias in the
product distribution, whereas separations by temperature
change were the least biased. The distributions were similar
when other salts were used, such as CaCl2, KCl, and KF.
Scheme 1. Dynamic Covalent Library of Imines 5−8, Generated by Reaction of Salicylaldehydes 1 and 2 with Amines 3 and 4
B
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Phase reunification was achieved by different procedures
depending on the phase separation agents used. In the case of
phase separations induced by NaCl or sucrose, the biphase was
reunited by warming the solution to 70 °C, leading to the
reformation of imines 6 and 7. Although the imines were found
to give a statistical distribution, there was a slightly greater
degree of overall hydrolysis at the higher temperature.
Comparison of the NMR spectra of the homogeneous phase
before and after phase separation indicated that the systems
were reversible, with the initial and final phases having similar
compositions (see SI page S24−S25).
Whereas NaCl-induced phase-separated mixtures were re-
united using temperature, the recombination of a KF-induced
biphase was achieved by a chemical approach. Treatment of the
latter mixture with solid LiClO4 resulted in rapid precipitation
of both LiF (solubility product in water, Ksp[water] = 1.84 ×
10−3) and KClO4 (Ksp[water] = 1.02 × 10−2), which were
subsequently removed by filtration, to yield a single phase
solution. Remarkably, in this case the effector, the salt KF, used
to cause phase separation is eliminated by double precipitation,
so that the initial single phase is restored, therefore regenerating
also the initial imine distribution (within experimental
accuracy). This process represents repeatable switching
between single phase and phase-separated states.
For separations using diethyl ether, fractional distillation
caused reunification of both phases. The trace amount of ether
remaining in the newly formed homogeneous phase (less than
3%) had no observable effect on the product distribution of the
dynamic library.
Modulation of phase separation by cooling−heating cycles
represents a fully reversible procedure without use of any
chemical additive. Thus, cooling a solution of compounds 5−8
to 0 °C led to phase separation and to the distribution of imines
shown in Figure 6 of SI. Upon warming to 21 °C, the phases
reunited, and the initial distribution of imine constituents was
restored (Figure 2 of SI).
Figure 2. Distribution of a dynamic covalent library of imines 5−8
(Scheme 1) in a single phase of AN/W 3/2 (v/v). (Top) 400 MHz 1H
NMR spectrum of the mixture: −CHN− proton signals of imines
5−8 in the 8.3−8.5 ppm region; and −CHO signals of 1 and 2 around
10 ppm . (Bottom) Amounts and fractions of the four different imine
constituents 5−8, determined by integration of the corresponding
−CHN− proton signals against an external standard (see
Experimental Section).
Effects of Reversible Phase Separation on the
Dynamic Covalent Library 12−15. We extended our
dynamic libraries to include biogenic amines and aldehydes
(Scheme 2). The neurotransmitter tryptamine 10, an important
precursor to a myriad of biogenic alkaloids, was selected as the
organic amine component. While α-amino acids are an ideal
choice for an aqueous amine component, they form only low
amounts of imines. However, the sodium salt of the
parasympathetic neurotransmitter γ-aminobutyric acid (11)
gave satisfactory amounts of imines. Salicylaldehyde was again
selected as the organic aldehyde component.26 Whereas water-
soluble aldehydes have a propensity to form hydrates and
acetals, which interfere with imine formation,30 one notable
exception is pyridoxal-5-phosphate (9), which does not give a
Thus, when phase separation was performed with NaCl, the
compositions were 93% W for the aqueous phase and 85% AN
for the organic phase (see SI page S27−S28). The
corresponding distributions were 77% for 5 and 72% for 8 in
the aqueous and organic phases, respectively (Figure 3, bottom
left). The total fractions (summed over both phases) amounted
to 35% for 5 and 43% for 8 (Figure 3, bottom right), as
compared to relative values of 26% for 5 and 30% for 8 in the
homogeneous phase (Figure 2), indicating amplification of
these two constituents on phase separation.
The organic as well as the aqueous phases strongly disfavor
the formation of both amphiphilic 6 and 7, which are
antagonistically related to 5 and 8, and thus present in only
small amounts.
When one of the charged species was replaced with a water-
soluble neutral component, such as 2-pyridine carboxaldehyde
or ethanolamine, the selection driven by phase separation was
far less pronounced. Furthermore, when the homogeneous
mixture was treated with the chaotropic tetrabutylammonium
bromide, no phase separation occurred, and no perturbation of
the equilibrium distribution was observed.
1
hydrate detectable by H NMR and yields imines at a high
conversion and with fast rate.26
Thus, a solution of 1 and 9−11 (20 mM each) was prepared
in 3:2 (v/v) AN/W to give a distribution of the imines 12−15,
as shown in Figure 4. It should be noted that, in this case,
formation the AN/W medium influences the distribution of the
library even in a single phase. In line with this inference,
increasing the water content was found to decrease the
imbalance and approach to a statistical distribution of the four
imines (Figure 28, SI). Another factor could be the micro-
heterogeneity of AN/W mixtures,16 an aspect that is beyond
the present study but which, if confirmed, could be of much
C
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. (Top) 400 MHz 1H NMR spectrum of the aqueous (left) and organic (right) phases generated by NaCl-induced phase separation of an
AN/W 3/2 (v/v) mixture containing the imines 5−8 into two separate phases; −CHN− and −CHO proton signals in the 8.1−8.7 and 9.6−10.3
ppm regions, respectively. (Bottom left) Distributions of the four different imine constituents in the respective phases, aqueous (left) and organic
(right), determined by integration of the corresponding −CHN− proton signals against an external standard. (Bottom right) Integrated amounts
and fractions of the constituents 5−8 summed over both phases (see Experimental Section).
Scheme 2. Dynamic Covalent Library of Imines 12−15 Generated by Reaction of the Biogenic Amines 10 and 11 with the
Aldehydes 1 and 9
interest with respect to the general question of the
physicochemical structure of solutions.31
Upon phase separation with NaCl, sucrose, or diethyl ether, a
similar, even more pronounced recombination trend was
D
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
analyze by variable-temperature NMR. However, fractional
distillation of diethyl ether from the phase-separated mixture
successfully reunited the phases and resulted in the restoration
of the original product distribution. In this case again, phase
separation by KF was reversed with removal of all added salts
by double decomposition on treatment with LiClO4. The
changes in environment described above bear relation to
processes that may occur in a membrane system, with dynamic
redistribution taking place at the interface between the aqueous
medium and a lipid membrane.28 The analogy is even closer
when liquid membranes are considered, as represented here by
the organic (acetonitrile-rich) phase. In this respect, the present
set up may be considered as a switchable liquid membrane
system, that may be implemented for introducing reversible
modulation of membrane processes, such as (selective)
substrate pumping by coupling to a chemical effector (e.g.,
proton or redox gradient) or physical stimulus29 (e.g., light)29b
as well as establishing switched chemical potentials with
generation of out-of-equilibrium conditions. Thus, one may
imagine to drive transport systems29 by creating nonequili-
brium conditions via phase separation. Such work is being
pursued in this laboratory.
From 2D to 3D Constitutional Dynamic Networks. In
the general context of complex networks,32 the data reported
here can be represented within the framework of constitutional
dynamic networks.1b,3 The four constituents in the single,
homogeneous phase define a square, that splits, for the
separated two phase system, into two squares, whose corners
are connected through the interface, thus defining a cube
(Figure 6). The system exemplifies the splitting of a 2D
(square) into a 3D (cube) constitutional dynamic network. The
trans-phase edges of the cube connect identical constituents,
distributed in the two phases and made antagonistic by the
phase separation, an increase in one phase causing a depletion
in the other phase. The trans-phase diagonals connect the
agonists between the two phases, in particular those two
agonists that are amplified on phase separation and represent
the fittest constituent in each phase. Thus, the generation of the
fittest in one phase is linked to that of the fittest in the other
phase, each of them occupying their environmental “ecological”
niche.
Figure 4. Distribution of a dynamic covalent library of imines 12−15
(Scheme 2) in a single phase of AN/W 3/2 (v/v). (Top) 400 MHz 1H
NMR spectrum of the mixture: −CHN− proton signals of imines
12−15 in the 8.0−9.0 ppm region; and −CHO signals of 1 and 9
between 10.0 and 10.3 ppm. (Bottom) Amounts and fractions of the
four different imine constituents 12−15, determined by integration of
the corresponding −CHN− proton signals against an external
standard (see Experimental Section).
observed as found in the aforementioned system, giving the
hydrophilic−hydrophilic imine 12 as the largely dominant
species in the aqueous phase and the hydrophobic−hydro-
phobic imine 15 almost exclusively in the organic phase (Figure
5). Again, sodium chloride was found to be superior to sucrose
and diethyl ether in promoting the most bias in the product
distribution.
Indeed, phase separation using NaCl gave compositions of
93% W for the aqueous phase and 85% AN for the organic
phase (see SI page S27−S28). The distributions were
respectively 84% for 12 and 99% for 15 in the aqueous and
organic phases (Figure 5, bottom left). The total fractions
(summed over both phases) amounted to 40% for 12 and
about 50% for 15 (Figure 5, bottom right) as compared to
relative values of 37% for 12 and 35% for 15 in the
homogeneous phase (Figure 4), indicating here also
amplification of these two constituents on phase separation.
The organic as well as the aqueous phases strongly disfavor the
formation of both amphiphilic 13 and 14, which are
antagonistically related to 12 and 15, and thus present in
only small amounts.
CONCLUSIONS
■
The results described herein lead to the following conclusions:
(1) We have demonstrated that a dynamic library of reversibly
interconverting entities undergoes constitutional reorganization
when subjected to liquid−liquid phase separation of a binary
solvent system. In the present case, a set of imines in AN/W
solution displays component exchange upon phase separation
induced by temperature change as well as by the addition of a
salt, a hydrophilic agent (a carbohydrate), or a hydrophobic
solvent. (2) The phase separation leads to the up-regulation/
amplification of the best suited/fittest constituents for each
phase with down-regulation of the amphiphilic ones through
medium-induced component selection. (3) The process is
reversible as the biphasic mixture can be reunited, e.g., by gentle
heating, ion exchange, precipitation, or fractional distillation,
causing the reformation of the original single-phase library, in
particular of the amphiphilic components that were repressed
in the phase-separated conditions. It may also be considered as
self-sorting33 modulated by reversible phase change. (4) The
processes described couple the behavior of a constitutional
dynamic library to a phase change, a step in the induction of
Attempts to reunite the phases separated with either NaCl or
sucrose with heat were unsuccessful, as it was found that the
charged components raised the phase integration temperature
to near the boiling point of acetonitrile, rendering it difficult to
E
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. (Top) 400 MHz 1H NMR spectrum of the aqueous (left) and organic (right) phases generated by NaCl-induced phase separation of an
AN/W 3/2 (v/v) mixture containing the imines 12−15 into two separate phases; −CHN− and −CHO proton signals in the 8.0−8.7 and 9.6−
10.3 ppm regions, respectively. (Bottom left) Distributions of the four different imine constituents in the respective phases, aqueous (left) and
organic (right), determined by integration of the corresponding −CHN− proton signals against an external standard. (Bottom right) Integrated
amounts and fractions of the constituents 12−15 summed over both phases (see Experimental Section).
the fittest agonist constituents in the respective phase through
component selection under the pressure of the environment. In
terms of constitutional dynamic networks,1b,3 the system
described exemplifies the splitting of a 2D (square) network
into a 3D (cube) one. (7) Finally, phase change-induced
modulation of chemical constitution may also bear relation to
the selective evolution of biologically significant molecules,
whereby they adapt their constitution to specific interconnected
environmental niches in prebiotic conditions.
Figure 6. Splitting of a 2D (square) into a 3D (cube) constitutional
dynamic network by liquid/liquid phase separation. Phase transfer of
the interconverting constituents AB, A′B′, A′B, and AB′ across the
interface (horizontal plane) occurs with adaptation to each phase
through component exchange. The diagonals and the edges of the
square link constituents presenting respectively agonistic (+) and
antagonistic (−) relationships. The diagonals of the cube link agonistic
constituents, and the vertical edges link antagonistic constituents
across the interface. The larger size and bold letters of AB and A′B′ as
well as the thicker diagonal line linking them indicate simultaneous up-
regulation of these two agonistic constituents across the interface.
EXPERIMENTAL SECTION
■
General. All reagents were purchased from chemical suppliers
(Sigma-Aldrich, Merck, Fluka) and used without further purification
unless mentioned otherwise. Acetonitrile (CHROMOSOLV for
HPLC, gradient grade) was purchased from Sigma-Aldrich and used
as received. Water was purified using a Millipore Elix 10 filtration
1
system. Salicylaldehyde was freshly distilled prior to use. H NMR
spectra were recorded on a Bruker Avance 400 MHz NMR
spectrometer. These samples were equilibrated at 298 K and consisted
of nondeuterated solvents equipped with a sealed capillary charged
with DMSO-d6 and a small amount of DMSO-h6 as an external
standard. The preparation of 2 is described in the SI. (Caution:
Perchlorate salts are known to detonate on exposure to organic matter
and heat, thus great care must be taken in their handling.)
Phase Separation Experiments. An NMR tube containing a
capillary, as mentioned above, was charged with 250 μL of 80 mM 1−
4 in 3:2 (v/v) AN/W, resulting in a 1 mL solution of 20 mM of each
constitutional change by a phase transition. They allow in
principle to subject such libraries to nonequilibrium conditions.
They also touch upon the interesting question of the behavior
of a dynamic covalent library at an interface.14 Such behavior
represents an adaptation of the dynamic system to the medium
by constitutional variation, with simultaneous up-regulation of
F
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
component. The mixture was gently agitated for 5 min. Phase
separations were accomplished by treating the above solution with
either 25 mg NaCl, 25 mg KF, 100 mg of sucrose, or 250 μL of diethyl
ether. The biphase was gently agitated for 10 min, and the tube was
allowed to stand until clean phase separation was observed. The
aqueous phase could be measured by 1H NMR without any observable
interference from the top layer and gave the same results when the
phases were analyzed independently. The organic layer was carefully
separated using a syringe and transferred to a new NMR tube
containing the same external standard used previously.
For phase separation experiments involving imines 12−15, a
modified approach was used. An intimate stock solution of 80 mM 9
and 10 was prepared, as it was found that 9 alone was insoluble in 3:2
(v/v) AN/W. Thus, pyridoxal-5-phosphate (9) (79 mg, 0.320 mmol)
and tryptamine (10) hydrochloride (63 mg, 0.320 mmol) were
suspended in 2.4 mL of acetonitrile, diluted with 640 μL of water, and
treated with 960 μL of a 1.0 M NaOH volumetric standard. This
solution was stored at −20 °C to prevent phosphate hydrolysis.
An NMR tube containing an external standard was charged with
250 μL of 80 mM 1, 250 μL of 80 mM of 9 and 10, 250 μL of 80 mM
11, and 250 μL of 3:2 (v/v) AN/W, resulting in a 1 mL solution of 20
mM of each component. The mixture was gently agitated for 5 min.
Phase separation experiments were carried out in the same manner as
described for the library consisting of 5−8.
Phase Recombination Experiments. To demonstrate the
reversibility of these phase separations, an NMR tube containing a
phase-separated library (induced by addition of 25 mg NaCl or 100
mg sucrose) consisting of imines 5−8 and an external standard was
placed inside an NMR probe equilibrated at 70 °C. The sample was
allowed to stand for 10 min. Spinning of the sample was found to be
insufficient for mixing the organic and aqueous layers. Thus, the
sample was quickly removed from the probe and carefully agitated.
Once homogenization was achieved, the sample was quickly placed
into the probe and analyzed.
For phase separations induced by KF, a 1 mL solution containing
imines 5−8, phase separated by addition of 25 mg KF, was treated
with 46 mg of LiClO4 (1 equiv with respect to KF). This resulted in
precipitation of both LiF and KClO4, which were filtered off, leaving a
single phase filtrate.
(2) (a) Lehn, J.-M. Chem.Eur. J. 1999, 5, 2455−2463. (b) Rowan,
S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F.
Angew. Chem., Int. Ed. 2002, 41, 898−952. (b) Corbett, P. T.; Leclaire,
J.; Vial, L.; West, K. R.; Wietor, J.-L.; Sanders, J. K. M.; Otto, S. Chem.
Rev. 2006, 106, 3652−3711. (c) Ladame, S. Org. Biomol. Chem. 2008,
6, 219−226. (d) Dynamic Combinatorial Chemistry; Miller, B. L., Ed.;
Wiley: Chichester, U.K., 2010. (e) Dynamic Combinatorial Chemistry;
Reek, J. N. H., Otto, S., Eds.; Wiley-VCH: Weinheim, Germany, 2010.
(f) Hunt, R. A. R.; Otto, S. Chem. Commun. 2011, 47, 847−855.
(g) Belowich, M. E.; Stoddart, J. F. Chem. Soc. Rev. 2012, 41, 2003−
2024.
(3) Giuseppone, N.; Lehn, J.-M. Chem.Eur. J. 2006, 12, 1715−
1722.
(4) (a) Baxter, P. N. W.; Lehn, J.-M.; Rissanen, K. Chem. Commun.
1997, 1323−1324. (b) Chow, C.; Fujii, S.; Lehn, J.-M. Chem. Commun.
2007, 4363−4365. (c) Barboiu, M.; Dumitru, F.; Legrand, Y.-M.; Petit,
E.; van der Lee, A. Chem. Commun. 2009, 2192−2194.
(5) Belenguer, A. M.; Frisc
Sci. 2011, 2, 696−700.
̌ ̌ ́
ic, T.; Day, G. M.; Sanders, J. K. M. Chem.
(6) (a) Ingerman, L. A.; Waters, M. J. Org. Chem. 2009, 74, 111−117.
(b) Chaur, M. N.; Collado, D.; Lehn, J.-M. Chem.Eur. J. 2010, 17,
248−258.
(7) Giuseppone, N.; Lehn, J.-M. Angew. Chem. Int. Ed 2006, 45,
4619−4624.
(8) (a) Giuseppone, N.; Schmitt, J.-L.; Lehn, J.-M. J. Am. Chem. Soc.
2006, 128, 16748−16763. (b) Giuseppone, N.; Lehn, J.-M. J. Am.
Chem. Soc. 2004, 126, 11448−11489. (c) Giuseppone, N.; Fuks, G.;
Lehn, J.-M Chem.Eur. J. 2006, 12, 1723−1735. (d) Fujii, S.; Lehn, J.-
M. Angew. Chem., Int. Ed. 2009, 48, 7635−7638.
(9) (a) Baxter, P. N. W.; Khoury, R. G.; Lehn, J.-M.; Baum, G.;
Fenske, D. Chem.Eur. J. 2000, 6, 4140−4148. (b) Ramirez, J.;
Stadler, A.-M.; Kyritsakas, N.; Lehn, J.-M. Chem. Commun. 2007, 237−
239.
(10) See for instance: (a) Huc, I.; Lehn, J.-M. Proc. Natl. Acad. Sci.
U.S.A. 1997, 94, 2106−2110. (b) Berl, V.; Huc, I.; Lehn, J.-M.;
DeCian, A.; Fischer, J. Eur. J. Org. Chem. 1999, 3089−3094. (c) Furlan,
R. L. E.; Ng, Y.; Otto, S.; Sanders, J. K. M J. Am. Chem. Soc. 2001, 123,
8876−8883. (d) Au-Yeung, H. Y.; Cougnon, F. B. L.; Otto, S.; Pantos,
G. D.; Sanders, J. K. M. Chem. Sci. 2010, 1, 567−574. (e) Ghosh, S.;
Ingerman, L. A.; Frye, A.; Lee, S. J.; Gagne, M. R.; Waters, M. L. Org.
̧
For phase separations using diethyl ether, the reaction described
above was scaled up 4-fold and treated with 1 mL of diethyl ether,
upon which a phase separation occurred. The biphasic mixture was
then placed on a rotary evaporator, and the ether evaporated at 50 °C
at 1 atm until the phases had reunited.
́
Lett. 2010, 12, 1860−1863. (f) McNaughton, B. R.; Miller, B. L. Org.
Lett. 2006, 8, 1803−1806. (g) McNaughton, B. R.; Gareiss, P. C.;
Miller, B. L. J. Am. Chem. Soc. 2007, 129, 11306−11307.
(11) (a) Lehn, J.-M. Prog. Polym. Sci. 2005, 30, 814−831. (b) Lehn,
J.-M. Aust. J. Chem. 2010, 63, 611−623 and references therein..
(c) Maoulin, E.; Cormos, G.; Giuseponne, N. Chem. Soc. Rev. 2012,
41, 1031−1049.
ASSOCIATED CONTENT
■
S
* Supporting Information
Preparation of 2, tables describing product distributions, and
mole fractions of AN/W. This material is available free of
charge via the Internet at http://pubs.acs.org.
(12) (a) Chen, X.; Dam, M. A.; Ono, K.; Mal, A. K.; Shen, H.; Nutt,
S. R.; Sheran, K.; Wudl, F. Science 2002, 295, 1698−1702. (b) Chen,
X.; Wudl, F.; Mal, A. K.; Shen, H.; Nutt, S. R. Macromolecules 2006, 36,
1802−1807. (c) Reutenauer, P.; Buhler, E.; Boul, P. J.; Candau, S. J.;
Lehn, J.-M. Chem.Eur. J. 2009, 15, 1893−1900. (d) Roy, N; Lehn,
J.-M. Chem. Asian J. 2011, 6, 2419−2425.
AUTHOR INFORMATION
■
Corresponding Author
(13) Sreenivasachary, N.; Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A.
2005, 102, 5938−5948.
lehn@unistra.fr
(14) For dynamic processes in biphasic and transport systems, see:
(a) Perez-Fernandez, R.; Pittelkow, M.; Belenguer, A. M.; Sanders, J.
K. M. Chem. Commun. 2008, 1738−1740. (b) Perez-Fernandez, R.;
Pittelkow, M.; Belenguer, A. M.; Lane, L. A.; Robinson, C. V.; Sanders,
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENTS
J. K. M. Chem. Commun. 2009, 3708−3710. (c) Saggiomo, V.; Luning,
̈
■
U. Chem. Commun. 2009, 3711−3713.
N.H. thanks the University of Strasbourg and the ANR 2010
BLAN-717-1 project for postdoctoral fellowships. This work
has been supported financially by the University of Strasbourg,
the CNRS and the ANR 2010 BLAN-717-1 project.
(15) (a) Folmer-Andersen, J. F.; Lehn, J.-M. J. Am. Chem. Soc. 2011,
133, 10966−10973. (b) Folmer-Andersen, J. F.; Lehn, J.-M. Angew.
Chem., Int. Ed. 2009, 48, 7664−7667.
(16) Although acetonitrile/water mixtures have the appearance and
consistency of homogeneous solution and exhibit miscibility at any
volumetric ratio at ambient temperature, data from molecular
dynamics,17 IR,18 Raman,19 NMR,20 SANS,21 and XRD22 have
indicated that acetonitrile/water mixtures exhibit microheterogeneity
REFERENCES
■
(1) (a) Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4763−4768.
(b) Lehn, J.-M. Chem. Soc. Rev. 2007, 36, 151−160.
G
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
at the molecular scale and consist of water- and acetonitrile-rich
clusters. Treatment of these binary mixtures with ionic kosmotropes,
such as sodium chloride,23 or nonionic kosmotropes, like carbohy-
drates,24 can force water molecules to coalesce and thus induce a phase
separation. Conversely, a phase separation may be induced by the
addition of a water-immiscible organic solvent, which consolidates
acetonitrile clusters and expels water.
(33) See for instance: (a) Ghosh S.; Isaacs, L. Dynamic
Combinatorial Chemistry; Miller, B. L., Ed.; Wiley: Chichester, U.K.,
2010; Chpt. 4, pp 118−154. (b) Kramer, R.; Lehn, J.-M.; Marquis-
̈
Rigault, A. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 5394−5398. (c) Saur,
I.; Scopelliti, R.; Severin, K. Chem.Eur. J. 2006, 12, 1058−1066.
(d) Liu, T.; Langston, M. L. K.; Li, D.; Pigga, J. M.; Pichon, C.; Todea,
A. M.; Muller, A. Science 2011, 331, 1590−1592. (e) Northrop, B. N.;
̈
Zheng, Y.-R.; Chi, K.-W.; Stang, P. J. Acc. Chem. Res. 2009, 42, 1554−
(17) Kovacs, H.; Laaksonen, A. J. Am. Chem. Soc. 1991, 113, 5596−
5605 and references therein.
1563.
(18) (a) Loewenschuss, A.; Yellin, N. Spectrochim. Acta 1975, 37A,
207−212. (b) Jamroz, D.; Stangret, J.; Lindgren, J. J. Am. Chem. Soc.
1993, 115, 6165−6168.
(19) Rowlen, K. L.; Harris, J. M. Anal. Chem. 1991, 63, 964−969.
(20) Esteal, A. J. Aust. J. Chem. 1979, 1379−1384.
(21) Takamuku, T.; Matsuo, D.; Yamaguchi, A.; Tabata, M.; Yoshida,
K.; Yamaguchi, T.; Nagao, M.; Otomo, T.; Adachi, T. Chem. Lett.
2000, 29, 878−879.
(22) Takamuku, T.; Tabata, M.; Yamaguchi, A.; Nishimoto, J.;
Kumamoto, M.; Wakita, H.; Yamaguchi, T. J. Phys. Chem. B 1998, 102,
8880−8888.
(23) Takamuku, T.; Noguchi, Y.; Yoshikawa, E.; Kawaguchi, T.;
Matsugami, M.; Otomo, T. J. Mol. Liq. 2007, 131−132, 131−132.
(24) (a) Wang, B.; Ezejias, T.; Feng, H.; Blaschek, H. Chem. Eng. Sci.
2008, 63, 2595−2600. (b) Wang, B.; Feng, H.; Ezeji, T.; Blaschek, H.
Chem. Eng. Technol. 2008, 31, 1869−1874. (c) Dhamol, P. B.; Prafulla,
M.; Feng, H. J. Chem. Eng. Data 2010, 55, 3803−3806.
(25) (a) For reversible phase separation utilizing carbon dioxide, see:
(a) Jessop, P. G.; Phan, L.; Carrier, A.; Robinson, S.; Durr, C. J.;
̈
Harjani, J. R. Green Chem. 2010, 12, 809−814. (b) Mercer, S. M.;
Jessop, P. G. ChemSusChem 2010, 3, 467−470. (c) On another note,
the role of phase separation in a synthetic process has recently been
highlighted, see: Lennox, A. J. J.; Lloyd-Jones, G. C. J. Am. Chem. Soc.
2012, 134, 7431−7441.
(26) (a) Kovaricek, P.; Lehn, J.-M. J. Am. Chem. Soc. 2012, 134,
9446−9455. (b) Tobias, P. S.; Kallen, R. G. J. Am. Chem. Soc. 1975, 97,
6530−6539. (c) Weng, S.-H.; Leussing, D. L. J. Am. Chem. Soc. 1983,
105, 4082−4090.
(27) It must be noted that the ionic strength of the solution plays a
synergistic role in phase separations involving diethyl ether. When the
charged solutes are absent, phase separation with diethyl ether does
not occur.
(28) For a recent investigation of dynamic combinatorial chemistry at
a phospholipid bilayer interface, see: Mansfeld, F. M.; Au-Yeung, H.
Y.; Sanders, J. K. M.; Otto, S. J. Systems Chem. 2010, 1, 12.
(29) (a) Behr, J.-P.; Lehn, J.-M. J. Am. Chem. Soc. 1973, 95, 6108−
6110. (b) Grimaldi, J. J.; Boileau, S.; Lehn, J.-M. Nature 1977, 265,
229−230. (c) Hriciga, A.; Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A. 1983,
80, 6426−6428. (d) Lehn, J.-M. In Physical Chemistry of Trans-
membrane Ion Motions; Spach, G., Ed.; Elsevier: Amsterdam, The
Netherlands, 1983; pp 181−206.
́
(30) Godoy-Alcantar, C.; Yatsimirsky, A. K.; Lehn, J.-M. J. Phys. Org.
Chem. 2005, 18, 979−985.
(31) The departure from statistical distribution, with marked
imbalance between the different constituents observed here already
in apparently homogeneous solution, may be considered as an
indication for preferential solvation of the dynamic solutes. Such
behavior of constitutional dynamic systems could well be a signature
for microheterogeneity of the medium, in connection with the
physicochemical data on AN/W mixtures16 as used here. One may
consider the operation of solute-enhanced microheterogeneity, a
feature which would be of particular interest within the present
framework, if the solute(s) is (are) of dynamic type. It may have a
broad bearing on the complexity of liquid structure, in particular in
mixed solvents, by translating/expressing it into constitutional
adaptation to the microstructure.
(32) (a) Newman, M. E. J. Phys. Rev. E 2004, 70, 056131. (b) Barrat,
́
A.; Barthelemy, M.; Pastor-Satorras, R.; Vespignani, A. Proc. Natl.
Acad. Sci. U.S.A. 2004, 101, 3747−3752. (c) Weighted Network
Analysis; Horvath, S., Ed.; Springer: Heidelberg, Germany, 2011.
H
dx.doi.org/10.1021/ja305379c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX