Design of Stimuli-Responsive Dynamic Covalent Delivery Systems for Volatile Compounds (Part 1): Controlled Hydrolysis of Micellar Amphiphilic Imines in Water

Article abstract of DOI:10.1002/chem.202102049

Despite their intrinsic hydrolysable character, imine bonds can become remarkably stable in water when self-assembled in amphiphilic micellar structures. In this work, we systematically studied some of these structures and the influence of various parameters that can be used to take control of their hydrolysis, including pH, concentration, the position of the imine function in the amphiphilic structure, relative lengths of the linked hydrophilic and hydrophobic moieties. Thermodynamic and kinetic data led us to the rational design of stable imines in water, partly based on the location of the imine function within the hydrophobic part of the amphiphile and on a predictable quantitative term that we define as the total hydrophilic–lipophilic balance (HLB). In addition, we show that such stable systems are also stimuli-responsive and therefore, of potential interest in trapping and releasing micellar components on demand.

Full text of DOI:10.1002/chem.202102049

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
Design of Stimuli-Responsive Dynamic Covalent Delivery
Systems for Volatile Compounds (Part 1): Controlled
Hydrolysis of Micellar Amphiphilic Imines in Water
Eric Lutz,[a] Emilie Moulin,[a] Vera Tchakalova,[b] Daniel Benczédi,[b] Andreas Herrmann,*[b] and
Nicolas Giuseppone*[a]
Dedicated to the memory of Prof. François Diederich
Abstract: Despite their intrinsic hydrolysable character, imine
bonds can become remarkably stable in water when self-
assembled in amphiphilic micellar structures. In this work, we
systematically studied some of these structures and the
influence of various parameters that can be used to take
control of their hydrolysis, including pH, concentration, the
position of the imine function in the amphiphilic structure,
relative lengths of the linked hydrophilic and hydrophobic
moieties. Thermodynamic and kinetic data led us to the
rational design of stable imines in water, partly based on the
location of the imine function within the hydrophobic part of
the amphiphile and on a predictable quantitative term that
we define as the total hydrophilic–lipophilic balance (HLB). In
addition, we show that such stable systems are also stimuli-
responsive and therefore, of potential interest in trapping and
releasing micellar components on demand.
Introduction
thioesters;[33,34] on Diels–Alder[35] and nitroaldol[36] reactions; on
dynamic native chemical ligation;[37] and on acetal,[38]
aminal,[39,40] boronic esters,[41,42] and imine (Schiff base)
condensations.[19,43,44] In the present article, we take a closer look
at the possibility of increasing imine bond stability in water by
taking advantage of surfactant micellization. Indeed, the hydro-
lysable character of imines is by essence detrimental to their
condensation in water according to Le Chatelier’s principle.
Although parameters such as concentration, stoichiometry, pH
and temperature can contribute to shifting the equilibrium
towards condensation, full hydrolysis is often observed in
various experimental contexts of interest requiring diluted
aqueous solutions.[24,43,45] Recently, we and others observed the
possibility of influencing the equilibrium and kinetic constants
of imine formation when imines are partially protected from
water in amphiphilic micellar structures.[46–49] In particular, it has
been shown that the use of imine-based surfactants can lead to
i) their autocatalytic condensation in a process related to
autopoiesis[48–50] and to ii) their modular and controllable
recombination in various types of dynamic micellar and
vesicular structures.[51–58]
Dynamic covalent chemistry extends the general principle of
“reversible binding” that is at the foundation of supramolecular
chemistry. It makes use of reversible covalent bonds to build
associative–dissociative molecular systems in which the equili-
brium position can be precisely shifted, depending on various
internal and external parameters.[1–7] Over the past 20 years, this
approach has been successfully implemented in the design of
dynamic
receptors,[8,9]
substrates
and
inhibitors,[10,11]
catalysts,[12,13] and stimuli-responsive materials.[14–28] Further-
more, the set-up of so-called dynamic mixtures by reversible
covalent bond formation of multicomponent systems has been
used to control the sensing, transport and release of volatile
compounds.[29] For practical implementation, dynamic covalent
bonds usually need to combine fast equilibration times
together with sufficient thermodynamic stabilities.[30,31] Among
the efficient systems described in the literature, one can find
examples based on the exchange of disulfides[32] and
[a] Dr. E. Lutz, Dr. E. Moulin, Prof. Dr. N. Giuseppone
SAMS Research Group, Institut Charles Sadron, CNRS
University of Strasbourg
Among a number of industrially relevant applications,
amphiphilic structures are commonly used to solubilize hydro-
phobic volatile compounds, such as fragrances, in aqueous
formulations of almost all perfumed body care or household
products. Cleavable surfactants with a labile chemical bond,
typically located between the hydrophilic and hydrophobic
section of the amphiphile, have been proposed in order to
increase the biodegradability of these products.[59–61] The
cleavable bond needs to be sufficiently stable during storage of
the formulation to ensure the performance of the surfactant in
application, but labile enough to decompose after use. From
23 rue du Loess, BP 84047, 67034 Strasbourg Cedex 2 (France)
E-mail: giuseppone@unistra.fr
[b] Dr. V. Tchakalova, Dr. D. Benczédi, Dr. A. Herrmann
Firmenich SA, Corporate R&D Division
Rue de la Bergère 7, 1242 Satigny (Switzerland)
E-mail: andreas.herrmann@firmenich.com
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/chem.202102049
This article belongs to a Joint Special Collection dedicated to François
Diederich.
Chem. Eur. J. 2021, 27, 1–12
1
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
our previous experience with micellar imines, we wondered
whether a profragrance[62–64] could be generated by reversible
condensation of a hydrophobic fragrance aldehyde with a
hydrophilic amine compound. Imine-based cleavable surfac-
tants would also be potentially useful to solubilize other volatile
perfumery compounds in an aqueous environment and would
hydrolyse as a response to dilution and/or a change of pH to
release the fragrance aldehyde by disintegration of the
amphiphile.
As a first approach, we decided to study the parameters
that stabilize – or destabilize – non-ionic amphiphilic imines in
water, the aim being to design stimuli-responsive surfactants
that would be able to encapsulate and release hydrophobic
molecules on demand and over a given period of time. The
coupled equilibria involved (molecular and supramolecular) are
schematically represented in Figure 1.
In this paper, we discuss the thermodynamic and kinetic
influence of various parameters on such systems, including pH,
concentration, chemical nature of the imine and position of the
imine in the amphiphilic structure, as well as the relative
lengths of the linked hydrophilic and hydrophobic moieties. As
a main result, we demonstrate that the location of the imine
function within the hydrophobic part of the amphiphile and its
total hydrophilic–lipophilic balance (HLB) are key parameters
for the performance of these systems as surfactants. In a second
study described in a separate article, we implement this
fundamental knowledge and demonstrate the controlled
release of volatiles (fragrances) in a real application context.
This aspect requires cheap starting materials, simple chemical
derivations, and products with low toxicity. It also invites us to
potentially integrate hydrophobic fragrances directly within the
amphiphilic imine structure (e.g., in the form of hydrophobic
volatile aldehydes) to form a profragrance. A series of imine-
based profragrances that form liquid crystals, polymers or gels
has been reported in the literature,[20,65–69] one study also
investigating some properties of micellar polymer
aggregates.[58]
From the present set of constraints and opportunities, we
decided to include in our study a series of six aldehyde
derivatives (A–F). Two of them (A and B) allowed us to
investigate fundamental aspects of the stability of the micelles;
the other four (C–F) were typical fragrance aldehydes, namely
hexylcinnamic aldehyde (C), citral (D) and (Z)-4-dodecenal (E) as
aliphatic and hydrophobic aldehydes, and vanillin (F) as an
aromatic and more hydrophilic aldehyde (Figure 2a).
As they are hydrophilic amines, we selected a series of poly
(ethylene glycol) (PEG)-derived structures without a linker (1) or
with variable lengths of a linker (X) between the amine and the
PEG units (2–7) to generate a more or less hydrophobic
environment in proximity to the imine bond to be formed
(Figure 2a). The different amines were derived from aniline (2),
glycinamide (3), carbamate urea (4), or tyramine (5). In addition,
we tested two commercially available derivatives of the Jeff-
amine family, which are diblock copolymers composed of a
poly(propylene oxide) and poly(ethylene oxide) unit of variable
lengths, namely M1000 (6) and M2070 (7).
release of fragrances in the context of
a
realistic
application.[65–69]
When condensed to imines, the generic surfactants that are
produced present three domains that play an important role in
imine stabilization (Figure 2b). In addition to the hydrophobic
and hydrophilic moieties, we will highlight the strong influence
of the relative hydrophobicity of the (small) group of atoms
immediately linking the primary amine with the hydrophilic tail,
denoted as linker (X) in Figure 2a and b. Finally, as an additional
example, we investigated the combination of a hydrophilic
PEG-derived aldehyde (G) with hydrophilic octylamine (8) to
form an imine with “inversed” polarity compared with that in
A1–F7 (Figure 2c).
Results and Discussion
Rational design and chemical structures of amphiphilic imine
derivatives
The various chemical structures used in this study reflect two
complementary objectives. The first objective is to rationalize
the stability of imine-based surfactants in water, depending on
structural parameters such as i) the type of imine bond involved
(e.g., aliphatic, aromatic), ii) the respective lengths of the
hydrophobic and hydrophilic parts within the surfactant, and
iii) the resulting overall HLB of the amphiphile to form a
cleavable surfactant. The second objective is to use such
potentially stimuli-responsive amphiphiles for the controlled
The synthetic protocols and characterizations of all products
presented in Figure 2 are detailed in the Supporting Informa-
tion.
General stability rules of amphiphilic imines in water
The kinetics of hydrolysis and thermodynamic stability of the
imine derivatives presented in Figure 2 were determined as
follows. Fully condensed imines, initially prepared in organic
solvents (see the Supporting Information), were dissolved at a
typical concentration (c=10 mM) in deuterium oxide (�1% w/
w), and the pD was immediately measured (pD=pHmeasured
+
Figure 1. Generic representation of the condensation of hydrophobic
aldehydes (blue) with hydrophilic amines (red) in water, yielding amphiphilic
imines. Depending on various parameters, amphiphilic imines can sub-
sequently self-assemble into micellar structures. Such supramolecular
structures can affect the stability of the imine bond by protecting it from
water when it is integrated in the hydrophobic micellar core.[49]
0.4).[24,70] The hydrolysis was then followed by integrating the
imine (8 ppm) and the aldehyde (9.5–10.0 ppm) resonance
signals (Iimine(t) and Ialdehyde(t), respectively) in the corresponding
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
2
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
Figure 3. a) Plot of the hydrolysis profile of imine A2 as a function of time
for different initial concentrations C0. b) Influence of C0 on the concentration
at thermodynamic equilibrium Ceq. c) Influence of pD on the half-time of
hydrolysis (h1/2) starting from a solution of imine A2 at C0 =10 mM.
Figure 2. a) Reversible condensation of aldehydes (A–F) with hydrophilic
amines (1–7) to form imine-based cleavable surfactants (A1–F7). b) General
structure of the condensed imines showing three domains within the
surfactant: a hydrophobic domain (blue), a linker (X, green) placing the imine
bond in a hydrophobic environment, and a hydrophilic domain (red).
c) Structures of hydrophilic aldehyde G and hydrophobic amine 8 to form
imine G8 with inversed polarity compared with that in A1–F7.
can also notice a linear decrease of imine A2 during the first
6 h, which cannot be fitted by a second-order kinetic plot and
which highlights the complex kinetics of hydrolysis associated
with the formation of spherical micelles of 7 nm diameter for
this particular surfactant.[46] Hence, in such “nanostructured
reservoirs”, the imine bond is, to a certain extent, protected
from water molecules within the micellar hydrophobic core.
Obviously, depending on the critical micellar concentration
(CMC), various amounts of unimers (i.e., single surfactant
imines) are also solubilized in water and can be easily hydro-
lysed. In addition, when considering the tendency of the
released hydrophobic aldehyde to in turn be solubilized in the
micellar core, the overall kinetics becomes even more complex.
1H NMR spectra and by plotting the imine concentration as a
function of time following Equation (1):
IimineðtÞ
CimineðtÞ ¼ C0 �
(1)
IaldehydeðtÞ þ IimineðtÞ
A typical concentration profile (illustrated here with imine
A2) is shown in Figure 3a. From this plot, one can determine
the amount of condensed product at equilibrium (Ceq =2.5 mM)
and the initial rate of hydrolysis (V0 =9×10À 3 mMminÀ 1). One
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
3
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
One can, however, determine a pseudo-first-order constant of
hydrolysis (kh =V0/C0 =0.9×10À 3 minÀ 1), reflecting the kinetics
that the system would have if it was not stabilized by the
micellar structure.
We then assessed the influence of the total concentration
on the thermodynamic and kinetic parameters for the same
reference imine (A2). We measured the hydrolysis profile for
three concentrations of starting material (10, 25 and 50 mM),
with an initial pD varying between 7.4 and 8.3, and we
determined the corresponding V0, kh and Ceq values (Figure 3a
and Table 1). Not surprisingly, reducing the initial imine
concentration slowed the hydrolysis and decreased the amount
of imine present at thermodynamic equilibrium (Figure 3b). This
is of interest for the intended use as delivery systems where,
during the final application, relatively small concentrations of
the system are typically encountered, for example, after
dilution.
Finally, we measured the influence of the pD on the rate of
hydrolysis – which is well known to be acid catalysed[71] – by
modifying at the initial time the acidity of the solution with a
controlled addition of NaOD or CF3CO2D. The corresponding
half-life times of hydrolysis (h1/2; which are necessary to reach a
concentration of 5 mM in imine) are shown as a function of pD
values (between 5 and 13) in Figure 3c. Interestingly, a very
large range of characteristic half-life times was observed, with
values being between 2 min (at pD 5.0) and several months (at
pD 13.0). Figure 3c illustrates the impact of the pH to generate
stimuli-responsive micellar systems that can be either highly
stable or rapidly dissociate on demand. This is also relevant for
practical uses of the system because structural modifications
might allow adjustment of the rates of hydrolysis to a specific
time window.
We then measured the size of the micelles during the
course of hydrolysis of imine A2. The hydrodynamic radii (Rh) of
the micelles measured by dynamic light scattering (DLS) are
illustrated in Figure 4 as a function of hydrolysis time until the
size of the micelles was found to be stable (after ca. 400 min).
The large polydispersity observed by DLS (Figure 4a),
ranging from half to twice the average micellar size, is here
partly due to the polydispersity of the PEG11OH chain used in
the synthesis (between PEG6OH and PEG16OH as determined by
ESI-MS). Figure 4b shows a linear increase of Rh as a function of
hydrolysis time, moving from 6.5 to 13.8 nm after 6 h, which
represents a speed of 0.02 nmminÀ 1. At this moment, the
concentration of imine is 5 mM, and the increase in size
corresponds to the entrapment of the released hydrophobic
aldehyde in the remaining micellar imine structures, as
described in our previous studies.[46,48,49] In addition, after 3 h,
another population of larger objects (Rh >100 nm) appeared
Figure 4. a) DLS measurements of the hydrodynamic radius (Rh) of micelles
of imine A2 during hydrolysis (C0 =10 mM; pD0 8.0). Hydrolysis time:
10 (black), 80 (red), 143 (green), 247 (blue), 317 (purple), and 373 min
(orange). b) Evolution of the micellar Rh as a function of hydrolysis time for
imine A2. c) Evolution of the hydrodynamic radius (black) and volume
percentage (blue) of an emulsion of aldehyde A as a function of time
monitored by DLS (A2 in D2O, C0 =10 mM).
with a rapidly increasing size of 0.65 nmminÀ 1, and this micro-
emulsion continued to grow after the disappearance of the
small micelle population, leading to a turbid solution and
eventually to a macroscopic phase separation (Figure 4c).
From these preliminary results using our reference imine
A2, we tried to explore the possibility of stabilizing the micellar
structure by introducing hydrogen bonding units into the imine
surfactants. We therefore investigated the hydrolytic behaviour
of imine B2, which contains a urea function in the hydrophobic
Table 1. Thermodynamic and kinetic values determined for the hydrolysis
of imine A2 in deuterated water.
C0 imine [mM] V0 [10À 3 mMmnÀ 1
]
kh [10À 3 minÀ 1] (h1/2 [h])
Ceq [mM]
10
25
50
9.0
40.9
126.3
0.9 (12.5)
1.64 (14)
2.53 (16)
2.5
6.0
20.0
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
part of its structure (Figure S1 in the Supporting Information).
Surprisingly, at C0 =10 mM and pD 8.9, we measured an initial
rate of hydrolysis of V0 =17 mMminÀ 1, h1/2 =5 h, and complete
hydrolysis after 8 h. This increased propensity to hydrolyse
might be attributed to the polarity of the urea group, which
attracts water molecules in their vicinity, but also to the
stronger tendency of aldehyde
B to phase separate as
compared with octyl benzaldehyde (A). Indeed, aldehyde B is a
solid, which was hardly soluble in water, and a white precipitate
was observed in the course of the hydrolysis of imine B2.
We then studied the influence of the chemical structure of
the hydrophilic amine on the hydrolysis rate (Figure S2). We first
compared the behaviour of the reference aromatic amine 2
leading to A2 with that of its corresponding aliphatic analogue
1 leading to A1. At C0 =10 mM and pD 10.2, the characteristic
hydrolysis time h1/2 was 3.5 h (compared with 120 h for imine
A2 at a similar pD), and full hydrolysis was reached after 24 h.
Despite the higher nucleophilicity of aliphatic amines compared
with that of aniline, we attributed the observed higher
sensitivity of A1 towards hydrolysis with respect to A2 to the
direct linkage of the primary amine to the hydrophilic PEG
chain in the case of A1, which increased the effective
concentration of water at the immediate vicinity of the imine
bond within the micellar structure.
Attempts to stabilize the micellar structure by adding
hydrogen bonding units in the form of amide functions into the
hydrophilic part of the surfactant (as in imine A3) were
unsuccessful (Figure S3). With h1/2 =1 h at pD 8.3 (12 times
faster than reference A2), and complete hydrolysis after 4 h, the
hydrogen bonds again appeared to destabilize the system. By
following the size of the micelles by DLS, we observed two
coexisting populations from the beginning of the hydrolysis, a
smaller one (Rh1 =3.8 nm) and a larger one (Rh2 =100 nm). Both
increased in size during the course of the reaction (up to Rh1 =
50 nm and Rh2 =400 nm) until micellar aggregates were no
longer observed due to macroscopic phase separation (Fig-
ure S4). A similar behaviour of enhanced hydrolysis was found
for imine A4, containing both a carbamate unit and a urea
group attached to the primary amine (Figure S5). Again, these
examples illustrated (as observed when we attempted to
stabilize the micelles by adding hydrogen bonding units in the
hydrophobic part) that macroscopic phase separation is a
strong driving force that pushes the equilibrium towards
hydrolysis by removing one product from solution and should
carefully be taken into account when designing these types of
dynamic amphiphiles.
Figure 5. a) Evolution of the concentration of imine G8 as a function of time
in D2O at different pDs (C0 =10 mM, unmodified pD0 10.2). b) Evolution of
the concentration of imine G8 as a function of time in D2O for different
initial concentrations and at pD0 8.3 (C0 =5 (blue), 10 (green) and 25 (red)
mM).
Table 2. Comparison of hydrolysis characteristics (h1/2 and Ceq) for imines
A1 and G8 in deuterated water starting from an initial concentration of
C0 =10 mM and depending on pD.
Imine [mM]
pD
h
1/2 [h]
Ceq [mM]
A1 (10 mM)
A1 (10 mM)
A1 (10 mM)
G8 (10 mM)
G8 (25 mM)
G8 (10 mM)
G8 (10 mM)
7.9
8.0
9.9
8.2
8.2
8.8
10.2
5.8
2.5
2.6
2.8
0
21.8
4.5
6.5
12.5
117
0.5
151
78
242
We then turned to another strategy to stabilize the imine
surfactants in water by introducing a hydrophobic linker
between the amine and the hydrophilic tail, as in imine G8.
Compared with A1, the aromatic aldehyde was switched from a
hydrophobic to a hydrophilic moiety and, conversely, the
aliphatic amine was switched from a hydrophilic to a hydro-
phobic moiety. The hydrolysis of G8 was performed at various
pD values and concentrations (Figure 5). Interestingly, imine G8
was found to be particularly stable against hydrolysis, although
it appeared to be particularly pH sensitive as well. Table 2
illustrates these variable characteristics when compared with
those of imine A1. Figure 5 and Table 2 together provide useful
information on several important factors that can be used to
stabilize amphiphilic imines in water. Imine G8 showed a high
hydrolysis dependency on pD, with four different domains:
a) for pD<7.4, complete hydrolysis was observed in less than
1 h; b) for 7.4<pD<9.9, a very fast initial hydrolysis was
followed by a very slow hydrolysis until near completion; c) for
9.9<pD<11.4, a very slow hydrolysis was observed from the
beginning until near complete hydrolysis; d) for pH>11.4,
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
5
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
hydrolysis proceeded at a moderate rate until completion due
to the cleavage of the ester bond. It was interesting to observe
that by changing the initial pD value by 2 units (from 8.2 to
10.2), the half-life time of hydrolysis extended by a factor of 500
(i.e., from 30 min to 10 days).
Table 3. Kinetic and thermodynamic values determined for the hydrolysis
of imine C1, C5 and C6 in deuterated water (C0 =10 mM).
Imine
pD
V0 [10À 3 mMmnÀ 1
]
h1/2 [h]
C1
C5
C6
10.5
10.1
10.5
14.8
1.3
6.8
4.5
150
12
This obviously correlated well with the pKa of the aliphatic
amine (ca. 10.6) and agrees with the higher stability of imine A1
at low pH, corresponding to a lower pKa of the aniline structure
(ca. 4.5). This also shows that imines involving aliphatic amines
can be very stable if not placed in a medium below their pKa.
Finally, the effect of the ester hydrolysis at pD>11.4 showed
that fast hydrolysis is reached for a similar function and pD if
the imine loses its amphiphilic character, thus further confirm-
ing the importance of the micellization in imine condensation.
However, as shown by the experiments performed with G8 at
pD 8.2 but for C0 =25 mM, Ceq showed a high level of
condensation, which was not necessarily expected from what
was observed at C0 =10 mM, and at least demonstrated the
nonlinearity of the condensation process. However, this result
was not too surprising with respect to the interplay between
the coupled molecular and supramolecular equilibria (Figure 1).
In such systems, the variation of the free Gibbs energy for the
imine condensation is superimposed with a phase diagram that
expresses various ratios between free imine and micellar imines
depending on the total concentration. In other words, when
crossing the phase diagram from unimers to micelles, there is a
concomitant nonlinear shift of the imine bond towards
condensation. These strong effects on Ceq can also be found
when changing the concentration of octylamine by hydrolysing
imine G8 (10 mM) in the presence of 10 mM of octylamine
(Figure S6). We observed that at a pD of 9.2, G8 rapidly
stabilized at Ceq =9.1 mM instead of 4.5 mM without extra
amine. This change can be evaluated by an apparent stability
constant, with Kapp =14 MÀ 1 at CamineT =10 mM and a much
higher Kapp =92 MÀ 1 at CamineT =20 mM.
Unsurprisingly, imine C1 showed the fastest hydrolysis of
this series, with a h1/2 of 4.5 h, similar to its previously studied
analogue A1. With a moderately hydrophobic linker, as in C6,
an approximately three times longer h1/2 was measured, while
the more hydrophobic tyramine linker in C5 led to an h1/2 of
150 h (33 times longer with respect to C1).
We also evaluated the kinetic effect of the concentration on
three amphiphilic hexylcinnamic aldehyde derivatives, C1, C6
and C7 (Figure 6). This effect was not studied in detail for C5
because of its already high stability at 10 mM and its h1/2 that
reached several weeks at higher concentrations.
Interestingly, h1/2 was very fast for C1 at all concentrations,
which is in agreement with poor stabilization of the imine. The
more stable micellar imines C6 and C7 showed a similar and
strong dependency on concentration. This result has two
implications of practical interest: i) it appears to be possible to
slow the release of volatiles by increasing the concentration of
imines, and ii) it appears to be possible to strongly accelerate
the release of the aldehyde from the imine by dilution of a
concentrated stock solution.
Furthermore, from these findings, one can assume that
imines that are not stabilized in water will exhibit pseudo-first-
order kinetics, whereas those that are stabilized within a
hydrophobic environment will exhibit pseudo-zero-order ki-
netics. We thus tried to fit the hydrolysis plots of C1, C5 and C6
as a function of time, using Equations (2) and (3) as follows:
Design of a micellar imine-based profragrance and its
controlled release
We then investigated the possibility of controlling the con-
densation and hydrolysis of hydrophobic fragrance aldehydes
(e.g., hexylcinnamic aldehyde (C)) with a series of hydrophilic
amines in water. For making the profragrances, we selected
aliphatic amines that are less problematic for the targeted
applications – from a toxicity point of view – than aromatic
amines. From what we learned earlier, we selected primary
amines with a hydrophobic linker between the amine group
and the hydrophilic PEG tail. For our tests, we selected PEG11
tyramine 5, as well as poly(propylene oxide) and poly(ethylene
oxide) diblock copolymer amines 6 and 7, which are commer-
cially available as Jeffamines M1000 (6) and M2070 (7). As a
reference, we also used amine 1 with no hydrophobic linker
group between the amine and the hydrophilic PEG group.
Starting with hexylcinnamic aldehyde (C), we compared the
rates of hydrolysis of imines C1, C5 and C6 at C0 =10 mM and
10.1<pD<10.5 (Table 3 and Figure S7).
Figure 6. Evolution of the half-time of hydrolysis for compounds C1, C6 and
C7 as a function of initial concentration at unmodified pD (10.5<pD<11.3).
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
6
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
CfitðtÞ ¼ C0th orderðtÞ þ C1st orderðtÞ
(2)
Thus:
CfitðtÞ ¼ CeqÀ k0t þ AeÀ k t
1
(3)
It can be seen that all the kinetic data can be fitted by this
simple model (Figure S8), showing that i) C1 is mainly hydro-
lysed with a pseudo-first-order rate, associated with a large
fraction of imine monomers in solution; that ii) C5 is mostly
hydrolysed by a pseudo-zero-order rate, associated with a large
amount of imine being stabilized in the hydrophobic micellar
core and with a constant low concentration in solution related
to the CMC; and that iii) C6 is hydrolysed equally by the two
mechanisms.
Not surprisingly, hydrolysis of a non-amphiphilic imine
profragrance, such as F6 with Jeffamine M1000 (6) as the amine
and vanillin (F) as the aldehyde, resulted in first-order kinetics
with an h1/2 of 1 h, even at an initially relatively high
concentration of 500 mM. Comparing the h1/2 of F6 with that of
C6 showed that the hydrolysis of C6 is slowed by about
120 times. This important difference occurs because F6 does
not aggregate into micelles and thus does not act as a
surfactant, whereas C6 does. This experiment further underlines
the high impact of self-aggregation on the stability of the
imine.
In addition to hexylcinnamic aldehyde (C), we also inves-
tigated the controlled release of citral (D) when condensed with
hydrophilic amines PEG11 amine 1, and Jeffamines M1000 (6)
and M2070 (7). Starting from an initial concentration of 10 mM,
the kinetics of hydrolysis of these citral-based imines is very
peculiar and unexpected. It displayed a very fast reaction rate
within the first 20 min, immediately followed by a very stable
state in which the imine no longer hydrolysed (Figure 7a). As
for hexylcinnamic aldehyde, the hydrolysis profiles of D6 and
D7 are very similar, here reaching a final concentration of Ceq =
2.8 mM, while D1 remains in its imine form at a noticeably
lower concentration (Ceq =1.5 mM). Diffusion ordered spectro-
scopy NMR (DOSY NMR) experiments showed that D6 is present
in water solution in its micellar form (Hr=4.7 nm) and as
unimers (Hr=1.2 nm), whereas citral D is present only within
the micellar structures.
The rapid hydrolysis of the citral-based imine at the very
beginning of the kinetic profile seems to reflect an intrinsic
sensitivity of that particular imine bond to water, as expected
from an aliphatic aldehyde. In addition, the partial solubility of
citral in water (0.6 mg/mL) induces a higher CMC and probably
a slower kinetic formation of the micelles when the imine is
transferred to water. Thus, at early time, a large quantity of
monomer is present and tends to hydrolyse. Upon micellar
formation, the high stabilization of the imine becomes suddenly
apparent and leads to a strong change in the slope of the
hydrolysis kinetics.
Figure 7. a) Evolution of the concentration of imines D1, D6 and D7 as a
function of time in D2O for an initial concentration of C0 =10 mM.
b) Evolution of the concentration of imines D1, D6 and D7 as a function of
time, in D2O for various initial concentrations (10 mM<C0 <100 mM).
c) Evolution of the concentration of amphiphilic imines C6 and D6, and non-
amphiphilic imine F6 as a function of time in D2O for an initial concentration
of C0 =500 mM.
an initial concentration of 100 mM for the three imines D1, D6
and D7. In another experiment at an initial concentration of
500 mM in D6, Ceq was reached within 10 min at 416 mM (83%).
This is similar behaviour to what is observed with amphiphilic
C6, but in striking difference when compared with the non-
amphiphilic vanillin derivative F6 (Figure 7c). This is an
The percentage of condensed imine at equilibrium is also
strongly dependent on the initial concentration in imine (Fig-
ure 7b), reaching about 50 to 60% of imine at equilibrium for
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
7
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
important observation from an application perspective because
when incorporated in sealed packaging, the imine concentra-
tion should be stable over time. However, when used upon
dilution, or simply by slow evaporation from the concentrated
solution in open conditions, the release of the volatile will shift
the imine equilibrium to full hydrolysis.
It is also important to note here that citral is known to be
poorly stable in aqueous solutions due to its aliphatic aldehyde
function and that its storage in stable form is a problem of
industrial relevance. This aspect will be discussed further in a
future paper.
exchange of amphiphilic imines between the micelle and the
solution.[75] In this case, the concentration of unimers available
in solution will not be constant and no longer equal to the
CMC, as limited by the exchange kinetics. This latter explanation
about the influence of viscosity on the rate of hydrolysis was
also supported by measuring an increase of h1/2 by a factor of
2.5 for C6 in the presence of Texapon (4% w/w) and of NaCl
(5% w/w), as compared with Texapon alone (Figure S10).
Presumably, the observed changes in the rates of hydrolysis are
due to the combination of the two phenomena (lower CMC and
change in viscosity). However, quantifying the contribution of
each of these effects is difficult and currently a subject of
debate in the literature. The use of co-surfactants in real
applications will be discussed further in the following paper.
Finally, we turned to the possibility to further stabilize
amphiphilic profragrance C6 by making use of co-surfactants.
The potential beneficial effects here were twofold: i) lowering of
the CMC value of the mixed system compared with that of the
imine alone and ii) investigating more realistic conditions of use
as encountered in industrial formulations containing fragrances
(such as body wash products and detergents). We studied the
effect of additional anionic (sodium dodecyl sulfate (SDS) and
Texapon), cationic (cetrimonium bromide) and nonionic (PEG25
monostearate, Triton X-100 and Tween 80) surfactants (see
Supporting Information for detailed chemical structures (Ta-
ble S1)). Their effects on the h1/2 when added at a concentration
of 4% w/w are reported in Table 4.
From all the surfactants tested, when compared with the
reference with no additive (h1/2 =12.2 h), only SDS and PEG25
monostearate were shown to decrease h1/2 by factors of 2.5 and
24.5, respectively. However, the SDS was shown to induce
partial precipitation of the system, and hydrolysis in the
presence of PEG25 monostearate was complex, with a very fast
initial period followed by a strong stabilization of the system at
Ceq =5 mM (Figure S9). In this case, the slow kinetics of
formation of stable micelles upon mixing might be the rate-
limiting step, leading to two different regimes along the course
of the reaction. Conversely, one can notice that two surfactants
considerably slow down the hydrolysis (by a factor of �3.5),
namely cationic cetrimonium bromide and nonionic Triton X-
100. The corresponding plots show pseudo-zero-order kinetics
associated with the hydrolysis of unimers in solution and
stabilized imines in the micelles. The explanation of a lower
CMC for the mixed systems imine/co-surfactant compared with
that for the imine alone is in agreement with the literature.[72–74]
A second explanation would be the possible viscosity increase
of the micellar core that would slow down the dynamic
Discussion on the total hydrophilic-lipophilic balance (HLB(T))
The surface activity of an amphiphilic molecule can be easily
determined by measuring the surface tension at the air-water
interface as a function of the concentration. The CMC, at which
the surface tension reaches a constant value, indicates self-
aggregation of the surfactant molecules in the bulk. To prove
the surfactant properties of the new imine compounds of the
present study, we measured surface tensions at the air-aqueous
solution interface by using the pendant drop method.
Figure 8 presents the surface tension as a function of the
concentration of imine compounds E6 and F6. The curve
corresponding to (Z)-4-dodecenal derivative E6 clearly demon-
strates the existence of a CMC above which the surface tension
kept a constant value. The further increase of the imine
compound concentration leads to micelle formation in the bulk
solution. In contrast, the curve corresponding to vanillin
derivative F6 does not follow the same trend, confirming the
absence of micelle formation of this compound in the studied
concentration range. The surface activity of the amphiphilic
molecules strongly depends on the equilibrium between the
Table 4. Kinetic values determined for the hydrolysis of imine C6 in
deuterated water at a concentration of 10 mM and unmodified pD (�10.5)
in the presence of various co-surfactants (4% w/w).
Surfactant type
Surfactant
h1/2 [h]
nonionic
cationic
anionic
anionic
nonionic amine
nonionic amine
no additive
anionic
Triton X-100
43.4
43.1
42.0
17.0
16.9
16.6
12.2
4.9
cetrimonium bromide
Texapon (+5% w/w NaCl)
Texapon
Jeffamine M2070
Jeffamine M1000
no additive
SDS
Figure 8. Surface tension at the air–water interface as a function of imine
compound concentration; black squares: E6 and blue circles: F6.
nonionic
PEG25 monostearate
0.5
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
8
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
hydrophobic and the hydrophilic part of the molecule. Imine E6
has a hydrophobic chain length of 12 C-atoms and a polar part
consisting of 19 PEG groups. This molecule is well balanced to
form spherical micelles. F6 has the same hydrophilic part of
19 PEG groups but a significantly smaller hydrophobic part with
a phenolic structure. This molecule is an example of an
amphiphilic molecule that cannot be considered as a surfactant
because of the significant difference between the hydrophilic
and hydrophobic molecular structures.
The hydrophilic-hydrophobic balance could be expressed
by the so-called HLB number, which could be assigned to each
amphiphilic molecule in order to scale its solubility in polar and
nonpolar media. It is an empirical parameter that can be
obtained experimentally or be estimated from the molecular
structures of the surfactant molecules by using the Davies’
group method.[76] In the present study, we calculated an HLB
value for each new imine by using the so-called effective chain
length (ECL) model, which is a modification of Davies’ method,
also considering group properties of the molecules.[77] The
values are reported in Table 5.
the HLB a sum of three components: HLB (Q) of the hydrophilic
part, HLB (X) of the linker and HLB (A) of the hydrophobic part.
The HLB (T) of a compound is calculated by using Equation (4).
The definition and calculation of the HLB values for the different
components is described in the Supporting Information.
HLB ¼ 7 þ HLBðQÞ þ HLBðXÞ þ HLBðAÞ
(4)
Correlation of measured surface activity to the calculated HLB
values: In order to know which calculated HLB values indicate
real surfactant properties, we obtained a correlation between
the experimental surface tension (measured at a concentration
of 1% w/w) of typical widely used and recognized nonionic
surfactants and their HLB values (Figure 9, Table S2). The
correlation curve obtained has a sigmoidal shape with three
different domains: i) a plateau at low surface tension and low
HLB values; ii) surface tension increasing with HLB value; and
iii) a plateau at high surface tension and high HLB value. The
correlation between the surface tension and the HLB for the
nonionic commercial surfactants (black squares) is very good.
The two limits correspond to highly water soluble and highly oil
soluble molecules. The surfactant molecules, which are efficient
solubilizers, are positioned in the range of HLB between 10 and
18. Molecules with structures corresponding to HLB values >18
are highly hydrophilic and not efficient to reduce the surface
tension and thus to solubilize nonpolar compounds.
The data for imines C6, D6, D7, E6 and F6, were added to
Figure 9 (blue circles, see also Table S2). The surface tension
values of imine surfactants confirm the strong surface activity of
these molecules, leading to a good solubilization capacity.
Vanillin derivative F6 has a high surface tension value, which is
positioned on the surface tension plateau on Figure 9, and an
HLB value exceeding the efficient surfactant limits fixed in the
present study. The good correlation between the surface
tension and HLB values in Figure 9 proves that the ECL method
for calculation of HLB values of non-ionic surfactants is
sufficiently precise to be used for the design and selection of
new non-ionic surfactant-like molecules. The additive and
modified method of Davies (ECL method) allows one to vary
the chemical structure of the different parts, hydrophilic,
lipophilic and linker, maintaining the surfactant properties of
the new molecules.
To estimate the role of each part of the compounds
according to the schematic structure in Figure 1, we considered
Table 5. Calculated HLB values (HLB(T)) as the sum of the corresponding
values of the hydrophilic part (Q), the linker (X) and the hydrophobic part
(A) of imines according to the ECL model based on the Davies’ group
method.
Imine
HLB (Q)
HLB (X)
HLB (A)
HLBtotal
C1
C5
C6
C7
D1
D6
D7
E6
E7
F6
10.01
10
0
3.31
13.69
11.26
13.41
12.79
14.77
14.49
13.87
13.64
13.02
18.83
2.43
2.28
4.44
0
À 2.2
4.44
2.28
4.44
2.28
À 3.31
À 3.31
À 3.31
2.23
À 2.23
À 2.23
À 3.08
À 3.08
2.1
12.01
13.55
10.01
12.01
13.55
12.01
13.55
12.01
As a further example, we explored the influence of the size
of the hydrophobic chain on the aldehyde part of imine A6 by
comparing its half-life time of hydrolysis with some analogues
involving alkylbenzaldehyde and alkoxybenzaldehydes that
have a smaller number of carbons in their hydrophobic side
chains (Figure 10).
One can observe that for a number of carbons between
zero and six, h1/2 remains constant around 0.5 h. A noticeable
slowdown of imine hydrolysis is observed only when there are
more than six carbons, with a large effect for more than eight
carbons, that is, with h1/2 becoming larger than 10 h. Interest-
ingly, this fits very well with HLB values, with a large gain in h1/2
for values below 15, which correspond to the inflexion point of
the surface tension curve plotted in Figure 9 (see detailed
calculations for HLB values in Figure S11 and Table S3).
Figure 9. Correlation between the experimentally determined surface ten-
sion and calculated HLB values for a) commercial surfactants (Table S2 and
black squares) and b) cleavable imines (Table S2 and blue circles).
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
9
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
Overall, both the stabilization and destabilization effects
highlighted in this paper offer particular insight into the
implementation of our approach to the controlled release of
fragrances in a realistic application context, and this is what we
shall describe in the following paper.[79]
Experimental Section
Sample preparation: The syntheses of the amines and aldehydes
are described in our previous work[25] and in the Supporting
Information. Imine condensation between amines and aldehydes
were performed in solutions of deuterated solvents at room
temperature. The molecular compositions of the systems were
determined by 1H NMR (Bruker Advance 400 spectrometer at
400 MHz), integrating the imine resonance signals comparatively to
those of the aldehydes. The spectra were internally referenced to
the residual proton solvent signal of D2O (4.79 ppm).
Figure 10. Influence of the hydrophobic chain length at the para position of
alkyl- and alkyloxybenzaldehydes on the half-time of hydrolysis of A6
(C0 =10 mM, unmodified pD 10.9).
Diffusion ordered spectroscopy NMR (DOSY NMR): 1H NMR
spectra were recorded on a Bruker Advance 500 spectrometer at
500 MHz from the NMR service of Institut de Chimie de Strasbourg.
The spectra were internally referenced to the residual proton
solvent signal. Residual solvent peaks were taken as reference (D2O:
4.79 ppm). The NMR probe is an Inverse Proton X equipped with a
gradient Z bobbin, able to generate pulse field gradients of
70 GausscmÀ 1 with a power amplifier of 10 A.
Conclusions
In this article, we have provided an overview of the hydrolytic
behaviour of imine-based dynamic covalent surfactants in water
with a twofold objective: first, to stabilize volatile aldehydes
into micellar cores in order to avoid their fast release by
hydrolysis and evaporation, and second, to trigger hydrolysis on
demand by changing some external parameters, such as
concentration or pH.[78]
During these investigations, we determined that both
aromatic and aliphatic imine bonds can indeed be strongly
stabilized in pure water, provided the imine surfactants form
Surface tension measurements: Surface tension measurements
were performed with a Kruss DSA 10 MK2 Drop Shape Analysis
System and pendant drop geometry. For this purpose, a liquid drop
was formed on the tip of a vertical needle suspended in the air
phase. The shape of the drop reflects the equilibrium between the
surface tension and the gravity. Surface tension was calculated
from the image of the pendant drop using the drop shape analysis
software.
micelles. This phenomenon can be predicted by calculating HLB Acknowledgements
values, which provide a first guideline for designing such
amphiphilic systems. Furthermore, the presence of a hydro-
phobic linker between the imine bonds and the hydrophilic
tails of the unimers reinforce their stability against hydrolysis
(here up to 30 times). Beyond these structural aspects, we Conflict of Interest
found that concentration effects can also be extremely efficient
E. L. thanks Firmenich SA for a doctoral fellowship.
in stabilizing or destabilizing imine bonds when crossing the
phase diagram from unimers to micelles and vice versa,
revealing strong nonlinear effects. Therefore, dilution effects
constitute an efficient trigger to boost the release of the
entrapped volatiles. The responsiveness of such systems can
also be supported by pH variation, with the half-time of
hydrolysis varying by more than five orders of magnitude
between pH 5 and 13.
Interestingly, the hydrolysis rates switch from purely first-
order kinetics when imines are not stabilized in micelles, to
zero-order kinetics when self-assembled in micelles because
these self-assemblies play the role of a reservoir in which the
imine bonds are fully protected from water. This reservoir effect
can also be improved by the use of nonhydrolysable co-
surfactants, which can decrease the CMC of the mixed system,
and therefore decrease the concentration of free imine unimers
in aqueous solution.
The authors declare no conflict of interest.
Keywords: amphiphiles · dynamic covalent chemistry · imines ·
micelles · self-assembly
[1] N. Giuseppone, Acc. Chem. Res. 2012, 45, 2178–2188.
[2] J.-M. Lehn, Chem. Eur. J. 1999, 5, 2455–2463.
[3] P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor, J. K. M. Sanders, S.
Otto, Chem. Rev. 2006, 106, 3652–3711.
[4] M. C. Misuraca, E. Moulin, Y. Ruff, N. Giuseppone, New J. Chem. 2014, 38,
3336–3349.
[5] F. Schleef, U. Lüning, Eur. J. Org. Chem. 2011, 2062–2065.
[6] O. Storm, U. Lüning, Chem. Eur. J. 2002, 8, 793–798.
[7] A. Herrmann, Chem. Soc. Rev. 2014, 43, 1899–1933.
[8] A. M. Escalante, R. L. E. Furlan, in Dynamic Combinatorial Chemistry (Eds:
J. N. H. Reek, S. Otto), Wiley-VCH, Weinheim, 2010, pp. 49–90.
[9] L. I. James, J. E. Beaver, N. W. Rice, M. L. Waters, J. Am. Chem. Soc. 2013,
135, 6450–6455.
[10] P. Frei, R. Hevey, B. Ernst, Chem. Eur. J. 2019, 25, 60–73.
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
10
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Paper
doi.org/10.1002/chem.202102049
Chemistry—A European Journal
[11] M. Mondal, A. K. H. Hirsch, Chem. Soc. Rev. 2015, 44, 2455–2488.
[12] P. Dydio, P. R. Breuil, J. N. H. Reek, Isr. J. Chem. 2013, 53, 61–74.
[13] F. Schaufelberger, O. Ramström, Chem. Eur. J. 2015, 21, 12735–12740.
[14] E. Moulin, G. Cormos, N. Giuseppone, Chem. Soc. Rev. 2012, 41, 1031–
1049.
[15] M. Barboiu, Chem. Commun. 2010, 46, 7466–7476.
[16] Y. Zhang, Y.-M. Legrand, A. van der Lee, M. Barboiu, Eur. J. Org. Chem.
2016, 1825–1828.
[47] K. Kurihara, M. Tamura, K. Shohda, T. Toyota, K. Suzuki, T. Sugawara,
Nat. Chem. 2011, 3, 775–781.
[48] R. Nguyen, L. Allouche, E. Buhler, N. Giuseppone, Angew. Chem. Int. Ed.
2009, 48, 1093–1096; Angew. Chem. 2009, 121, 1113–1116.
[49] R. Nguyen, N. Jouault, S. Zanirati, M. Rawiso, L. Allouche, E. Buhler, N.
Giuseppone, Chem. Eur. J. 2018, 24, 17125–17137.
[50] P. L. Luisi, Naturwissenschaften 2003, 90, 49–59.
[51] Y. Kang, K. Liu, X. Zhang, Langmuir 2014, 30, 5989–6001.
[52] Y. Kang, X. Tang, Z. Cai, X. Zhang, Adv. Funct. Mater. 2016, 26, 8920–
8931.
[17] N. Giuseppone, J.-M. Lehn, Chem. Eur. J. 2006, 12, 1715–1722.
[18] J. M. A. Carnall, C. A. Waudby, A. M. Belenguer, M. C. A. Stuart, J. J.-P.
Peyralans, S. Otto, Science 2010, 327, 1502–1506.
[53] C. B. Minkenberg, L. Florusse, R. Eelkema, G. J. M. Koper, J. H. van Esch, J.
[19] N. Giuseppone, J.-M. Lehn, Angew. Chem. Int. Ed. 2006, 45, 4619–4624;
Angew. Chem. 2006, 118, 4735–4740.
[20] A. Herrmann, N. Giuseppone, J.-M. Lehn, Chem. Eur. J. 2009, 15, 117–
124.
[21] Y. Zhang, M. Barboiu, ACS Omega 2018, 3, 329–333.
[22] L. Tauk, A. P. Schröder, G. Decher, N. Giuseppone, Nat. Chem. 2009, 1,
649–656.
Am. Chem. Soc. 2009, 131, 11274–11275.
[54] R. Nguyen, E. Buhler, N. Giuseppone, Macromolecules 2009, 42, 5913–
5915.
[55] R. Nguyen, N. Jouault, S. Zanirati, M. Rawiso, L. Allouche, G. Fuks, E.
Buhler, N. Giuseppone, Soft Matter 2014, 10, 3926–3937.
[56] C. Wang, G. Wang, Z. Wang, X. Zhang, Chem. Eur. J. 2011, 17, 3322–
3325.
[23] N. Giuseppone, G. Fuks, J.-M. Lehn, Chem. Eur. J. 2006, 12, 1723–1735.
[24] C. Godoy-Alcántar, A. K. Yatsimirsky, J.-M. Lehn, J. Phys. Org. Chem.
2005, 18, 979–985.
[25] A. M. Valdivielso, F. Puig-Castellví, J. Atcher, J. Solà, R. Tauler, I. Alfonso,
Chem. Eur. J. 2017, 23, 10789–10799.
[57] G. Wang, C. Wang, Z. Wang, X. Zhang, Langmuir 2011, 27, 12375–12380.
[58] Y. Liu, K. Liu, M. Zhao, S. Wang, Z. Zhou, Y. Shen, L. Jiang, React. Funct.
Polym. 2018, 132, 138–144.
[59] A. Tehrani-Bagha, K. Holmberg, Curr. Opin. Colloid Interface Sci. 2007, 12,
81–91.
[26] N. Giuseppone, J.-M. Lehn, J. Am. Chem. Soc. 2004, 126, 11448–11449.
[27] N. Giuseppone, J.-L. Schmitt, J.-M. Lehn, J. Am. Chem. Soc. 2006, 128,
16748–16763.
[28] N. Giuseppone, J.-L. Schmitt, L. Allouche, J.-M. Lehn, Angew. Chem. Int.
Ed. 2008, 47, 2235–2239; Angew. Chem. 2008, 120, 2267–2271.
[29] A. Herrmann, Chem. Eur. J. 2012, 18, 8568–8577.
[30] Y. Jin, C. Yu, R. J. Denman, W. Zhang, Chem. Soc. Rev. 2013, 42, 6634–
6654.
[60] D. Shukla, V. K. Tyagi, Tenside Surfactants Deterg. 2010, 47, 7–12.
[61] D. Lundberg, K. Holmberg, M. Stjerndahl, Encycl. Surf. Colloid Sci. 2016,
2, 1231–1249.
[62] A. Herrmann, Angew. Chem. Int. Ed. 2007, 46, 5836–5863; Angew. Chem.
2007, 119, 5938–5967.
[63] A. Herrmann, Photochem. Photobiol. Sci. 2012, 11, 446–459.
[64] A. Herrmann, Chimia 2017, 71, 414–419.
[65] H. Kamogawa, H. Mukai, Y. Nakajima, M. Nanasawa, J. Polym. Sci. Polym.
Chem. Ed. 1982, 20, 3121–3129.
[31] P. Chakma, D. Konkolewicz, Angew. Chem. Int. Ed. 2019, 58, 9682–9695;
Angew. Chem. 2019, 131, 9784–9797.
[32] S. P. Black, J. K. M. Sanders, A. R. Stefankiewicz, Chem. Soc. Rev. 2014, 43,
1861–1872.
[66] T. Tree-udom, S. P. Wanichwecharungruang, J. Seemork, S. Arayachu-
keat, Carbohydr. Polym. 2011, 86, 1602–1609.
[67] D. Hua, J. Jiang, L. Kuang, J. Jiang, W. Zheng, H. Liang, Macromolecules
2011, 44, 1298–1302.
[33] R. Larsson, Z. Pei, O. Ramström, Angew. Chem. Int. Ed. 2004, 43, 3716–
3718; Angew. Chem. 2004, 116, 3802–3804.
[68] L. Marin, B. Simionescu, M. Barboiu, Chem. Commun. 2012, 48, 8778–
8780.
[69] L. Marin, S. Moraru, M.-C. Popescu, A. Nicolescu, C. Zgardan, B. C.
Simionescu, M. Barboiu, Chem. Eur. J. 2014, 20, 4814–4821.
[70] A. K. Covington, M. Paabo, R. A. Robinson, R. G. Bates, Anal. Chem. 1968,
40, 700–706.
[34] B. T. Worrell, S. Mavila, C. Wang, T. M. Kontour, C.-H. Lim, M. K. McBride,
C. B. Musgrave, R. Shoemaker, C. N. Bowman, Polym. Chem. 2018, 9,
4523–4534.
[35] P. Reutenauer, E. Buhler, P. J. Boul, S. J. Candau, J.-M. Lehn, Chem. Eur. J.
2009, 15, 1893–1900.
[36] A. Karalius, Y. Zhang, O. Kravchenko, U. Elofsson, Z. Szabó, M. Yan, O.
Ramström, Angew. Chem. Int. Ed. 2020, 59, 3434–3438; Angew. Chem.
2020, 132, 3462–3466.
[71] A. C. Dash, B. Dash, S. Praharaj, J. Chem. Soc. Dalton Trans. 1981, 2063–
2069.
[72] X. Y. Hua, M. J. Rosen, J. Colloid Interface Sci. 1982, 87, 469–477.
[73] M. J. Rosen, ACS Symp. Ser. 1986, 311, 144–162.
[74] M. J. Rosen, Langmuir 1991, 7, 885–888.
[37] Y. Ruff, V. Garavini, N. Giuseppone, J. Am. Chem. Soc. 2014, 136, 6333–
6339.
[38] R.-C. Brachvogel, M. von Delius, Eur. J. Org. Chem. 2016, 3662–3670.
[39] G. Godin, B. Levrand, A. Trachsel, J.-M. Lehn, A. Herrmann, Chem.
Commun. 2010, 46, 3125–3127.
[40] B. Buchs, G. Godin, A. Trachsel, J.-Y. de Saint Laumer, J.-M. Lehn, A.
Herrmann, Eur. J. Org. Chem. 2011, 681–695.
[41] M. Demetriades, I. K. H. Leung, R. Chowdhury, M. C. Chan, M. A.
McDonough, K. K. Yeoh, Y.-M. Tian, T. D. W. Claridge, P. J. Ratcliffe,
E. C. Y. Woon, C. J. Schofield, Angew. Chem. Int. Ed. 2012, 51, 6672–6675;
Angew. Chem. 2012, 124, 6776–6779.
[42] M. Röttger, T. Domenech, R. van der Weegen, A. Breuillac, R. Nicolaÿ, L.
Leibler, Science 2017, 356, 62–65.
[75] G. S. Kwon, M. L. Forrest (Wisconsin Alumni Research Foundation), WO
2006/110862, 2006.
[76] J. T. Davies in Proceedings of the 2nd International Congress on Surface
Activity, Butterworths, London, 1957, p. 426.
[77] X. Guo, Z. Rong, X. Ying, J. Colloid Interface Sci. 2006, 298, 441–450.
[78] Parts of this publication are the subject of a patent application: E. Lutz,
N. Giuseppone, A. Herrmann, V. Tchakalova, D. Benczédi (Firmenich SA
& CNRS), WO 2018/134410, 2018.
[79] V. Tchakalova, E. Lutz, S. Lamboley, E. Moulin, D. Benczédi, N.
Giuseppone, A. Herrmann, Chem. Eur. J. 2021, 27, DOI: 10.1002/
chem.202102051 (following publication).
[43] N. Giuseppone, J.-L. Schmitt, E. Schwartz, J.-M. Lehn, J. Am. Chem. Soc.
2005, 127, 5528–5539.
[44] M. Ciaccia, S. Di Stefano, Org. Biomol. Chem. 2015, 13, 646–654.
[45] K. Meguellati, A. Fallah-Araghi, J.-C. Baret, A. El Harrak, T. Mangeat, C. M.
Marques, A. D. Griffiths, S. Ladame, Chem. Commun. 2013, 49, 11332–
11334.
Manuscript received: June 10, 2021
Accepted manuscript online: July 16, 2021
Version of record online: ■■■, ■■■■
[46] N. Jouault, R. Nguyen, M. Rawiso, N. Giuseppone, E. Buhler, Soft Matter
2011, 7, 4787–4800.
Chem. Eur. J. 2021, 27, 1–12
www.chemeurj.org
11
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPER
Dr. E. Lutz, Dr. E. Moulin, Dr. V. Tchaka-
lova, Dr. D. Benczédi, Dr. A. Herrmann*,
Prof. Dr. N. Giuseppone*
1 – 12
Design of Stimuli-Responsive
Dynamic Covalent Delivery Systems
for Volatile Compounds (Part 1):
Controlled Hydrolysis of Micellar
Amphiphilic Imines in Water
Imines are hydrolytically unstable,
but can be stabilized in aqueous
media by self-assembly in micellar
structures. Parameters such as pH,
concentration, position of the imine
function in the amphiphilic structure
and relative lengths of the linked hy-
drophilic and hydrophobic moieties
affect the stability of the imine bond
and can be adapted to form stimuli-
responsive micelles suitable for con-
trolling the release of bioactive
volatiles.