Flavylium-supported poly(N -isopropylacrylamide): A class of multistimuli responsive polymer

Article abstract of DOI:10.1021/ma401497q

A new multistimuli responsive polymer was prepared by means of the free-radical copolymerization of N-isopropylacrylamide and a vinyl flavylium derivative. The polymer shows response to temperature (conferred by the N-isopropylacrylamide units) and pH and light stimuli (conferred by the flavylium moieties). The polymer follows the typical pH dependent network of chemical reactions of flavylium compounds in water that was studied by means of UV-vis spectroscopy and stationary fluorescence emission. Regarding the temperature response, the presence of the flavylium cation in the polymeric chains reduces the LCST, and it is possible to take advantage of this effect to photo-collapse partially the polymer by irradiating the system in defined ranges of pH and temperature.

Full text of DOI:10.1021/ma401497q

Article
pubs.acs.org/Macromolecules
Flavylium-Supported Poly(N‑isopropylacrylamide): A Class of
Multistimuli Responsive Polymer
Noemi Jordao, Raquel Gavara,* and A. Jorge Parola*
́
̃
̂
REQUIMTE, Departamento de Química, Faculdade de Ciencias e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica,
Portugal
S
* Supporting Information
ABSTRACT: A new multistimuli responsive polymer was
prepared by means of the free-radical copolymerization of N-
isopropylacrylamide and a vinyl flavylium derivative. The
polymer shows response to temperature (conferred by the N-
isopropylacrylamide units) and pH and light stimuli (conferred
by the flavylium moieties). The polymer follows the typical pH
dependent network of chemical reactions of flavylium
compounds in water that was studied by means of UV−vis
spectroscopy and stationary fluorescence emission. Regarding
the temperature response, the presence of the flavylium cation
in the polymeric chains reduces the LCST, and it is possible to take advantage of this effect to photo-collapse partially the
polymer by irradiating the system in defined ranges of pH and temperature.
pH is increased, the nucleophilic attack of water at position 2
(hydration reaction) originates the formation of the hemiketal
B. This species subsequently reorganizes by means of a ring−
chain tautomerization reaction to give a cis-chalcone Cc.
Finally, the cis-to-trans isomerization of the double bond results
on the trans-chalcone species Ct. Moreover, if hydroxyl
substituents are present in the compound (usually at position
7 and/or 4′) the deprotonation reaction of the flavylium cation
gives rise to the formation of the quinoidal base A. This
reaction competes with hydration and is by far the fastest
process taking place in the network. Thus, when the pH is
increased from very acidic to higher pH values, the quinoidal
base is the kinetic product, but usually disappears with time
because the equilibrium network shifts to the formation of the
trans-chalcone, the most stable species at moderately acidic and
neutral pH values. This multiequilibria system can be described
as the global acid−base equilibrium shown in eq 1, where the
conjugate base CB represents the sum of all the other species in
the network (eq 2). The equilibrium constant of eq 1, K′a,
relates with those describing the other equilibria (see Scheme
1) according to eq 3.13
INTRODUCTION
■
The development of stimuli-responsive polymers has attracted
much attention in the last years1,2 due to their potential
applications in several fields such as biomedicine,3 separation
processes,4 and responsive surfaces.5 A usual way to prepare
multiresponsive polymers consists on the copolymerization of
different monomers, each responding to a specific stimulus.
Poly(N-isopropylacrylamide), pNIPAM, is a very well-known
thermoresponsive polymer that presents a phase transition
around 32 °C in water,6 but in fact the actual value depends on
several factors such as molecular weight7 and concentration.8
The temperature of phase change is known as the lower critical
solution temperature (LCST). Below this temperature the
polymer is solvated in water due to the formation of a hydrogen
bond network involving the amide groups of the polymeric
chains and the water molecules. Heating the polymer above the
LCST provokes the collapsing and precipitation of the polymer
due to the disruption of the hydrogen bonds. The proximity of
the LCST to the temperature of human body makes pNIPAM-
based polymers very attractive for the development of
biomedical applications, such as drug or gene delivery systems.
There are several examples in the literature which report on the
copolymerization of pNIPAM with chromophores which
undergo reversible photochemical processes. The selected
chromophores were traditionally spiropyran and azobenzene
derivatives.9 But additionally other photoactive molecules have
been explored such as coumarin10 and cinnamate derivatives.11
Flavylium compounds are synthetic analogues of natural
anthocyanins and present in aqueous solution a well-known
network of chemical reactions that responds to several stimuli,
usually pH and light (Scheme 1).12 At very acidic pH values the
flavylium cation AH+ is the most stable species but when the
AH+ + H2O ⇆ CB + H3O+
(1)
[CB] = [A] + [B] + [Cc] + [Ct]
(2)
K′a = Ka + Kh + KhKt + KhKtKi
(3)
One interesting feature of flavylium networks is the
photochemical reactivity exhibited by the trans-chalcone Ct
Received: July 16, 2013
Revised: October 18, 2013
Published: November 4, 2013
© 2013 American Chemical Society
9055
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
Scheme 1. Network of Chemical Reactions Originated by Flavylium Cations in Solution
Scheme 2. Synthesis of the Polymerizable Monomer 7-Hydroxy-4′-vinylflavylium Hydrogen Sulfate
Scheme 3. Polymerization Reaction To Obtain P1
that photoisomerizes to give the cis-chalcone Cc. At appropriate
pH values, Cc subsequently evolves to give the colored species
in the network namely AH+ and A, then the system reverts back
to the equilibrium with a rate which depends on the
substituents present on the flavylium compound. This behavior
can be employed to design photochromic systems as well as
models of optical memories based on flavylium com-
pounds.12,13
In this paper, we report on the synthesis of a new smart
copolymer where a 4′-vinylflavylium derivative was copoly-
merized with N-isopropylacrylamide in order to extend the
known temperature dependence of the pNIPAM polymers to
include as well pH and light responses conveyed by the
flavylium units. The results show that it is possible to collapse
the polymer by the action of light at the appropriate pH and
temperature values.
RESULTS AND DISCUSSION
■
Synthesis of P1. The synthesis of the multistimuli-
responsive polymer P1 was performed in two stages: (i) first
the polymerizable monomer 7-hydroxy-4′-vinylflavylium hydro-
gen sulfate (1) was synthesized by acidic condensation at room
temperature14 between 2,4-dihydroxybenzaldehyde and 4-
vinylacetophenone which was previously obtained by palla-
dium-catalyzed Hiyama reaction of vinyltrimethoxysilane and 4-
bromoacetophenone15 (Scheme 2); (ii) in a second step
polymer P1 was obtained by thermal free-radical polymer-
ization involving compound 1 (1.8% mol) and N-isopropyla-
crylamide (NIPAM, 98.2% mol) (Scheme 3). Short time and
acidic conditions were selected to carry out the polymerization,
trying to avoid the evolution of the flavylium cation to the cis-
and trans-chalcone species, which could introduce some cross-
9056
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
linking affecting different polymeric chains through the reaction
of the second double bond (see Scheme 1).16 For comparative
purposes, a control polymer (pNIPAM) was as well prepared
in the same conditions but in the absence of compound 1.
The polymerization was followed by UV−vis spectroscopy
revealing a blue shift of the absorption maximum of the
flavylium cation during the reaction which moved from 452 nm
for compound 1 to 437 nm after 1 h of polymerization (Figure
1). This shift of the band can be explained by a loss of
incorporated into the polymer was about 0.4% mol, by
comparison between the integration areas of the most
deshielded proton in the flavylium moiety (H4, δ = 9.35
ppm) and the proton corresponding to −CH− in the isopropyl
side group of N-isopropylacrylamide (δ = 3.97 ppm, see Figure
1
S2, Supporting Information). Careful analysis of the H NMR
shows some peaks in the aromatic region besides those of the
flavylium moieties in P1, suggesting that the flavylium partially
evolves to the other species of the network (hemiketal,
chalcones) in this solvent and/or that some chalcone of the
monomer forms during the polymerization16 and, reacting
through its middle double bond, leads to aromatic moieties.
Regarding the infrared analysis (Figure S3, Supporting
Information), the spectrum of P1 was practically identical to
that of pure pNIPAM, due to the low content of the flavylium
compound, and showed their characteristic bands: 3298 cm−1
(secondary amide N−H stretching), 2972 cm−1 (−CH3
asymmetric stretching), 1651 cm−1 (amide I bond), 1547
cm−1 (amide II bond).19
The molecular weight distribution was analyzed by means of
MALDI−TOF MS.20 For pNIPAM control, the values Mw =
6521 g/mol and Mn = 7842 g/mol were found, which give a
relatively narrow polydispersity index, PDI = 1.2 (Figure S4,
Supporting Information). It was not possible to obtain a
reliable signal for P1 (probably because the sample did not
desorb well from the MALDI matrix), but due to the similarity
of the reaction conditions for the synthesis of both polymers, it
is reasonable to assume a molecular weight distribution of P1
close to that found for the control.
Thermodynamics of the Flavylium Network. From the
spectral variations occurring immediately after a series of pH
jumps, from pH = 1 to higher pH values, it was possible to
determine the value of the acidity constant (Ka) associated with
the equilibrium between the flavylium cation AH+ and the
quinoidal base A (λmax ∼ 480 nm, Figure 2). The calculated
value, Ka = 10−3.9, compares with Ka = 10−4.1 for the model
compound 2.17
Figure 1. Blue shift of the absorption maximum on going from
monomer 1 to polymer P1 was used to monitor the polymerization
reaction (inset). Monitorization of the polymerization reaction by
UV−vis was carried out by taking 10 μL aliquots from the reaction
vessel and diluting to 3 mL with 0.3 M HCl; the solutions were
allowed to equilibrate before recording the spectra at room
temperature.
The pH-dependent absorption of the equilibrated solutions
was measured after 24 h from the pH jumps, when no more
changes in absorption were noticed over time. Figure 3A shows
that at acidic pH values the flavylium moiety is basically in the
form of the flavylium cation while at moderately acidic pH
values two species are dominant, namely the trans-chalcone
(λmax ∼ 360 nm) and the quinoidal base (λmax ∼ 480 nm).
Representation of the absorbance at 442 nm as a function of
pH reveals a data trend that cannot be fitted with a sigmoid
function (see inset in Figure 3A). In other words, it is not
possible to determine a single global acidity constant (see eqs
1−3), which suggests that several microenvironments are
present in the polymer that influence the equilibrium between
AH+ and the other species (CB) in the network. Indeed, the
fittings become increasingly better when two or more pK′a
values are used. The data trend was adequately fitted using a
power law model, eq 4, where A is the absorbance and a and b
are constants. Stretched exponential functions and scaling
functions such as the power law are often used to describe
polymer properties, e.g., dielectric relaxation and rheological
parameters, that show up with a distribution function behavior
due to microheterogeneities in the system.21−23 The fitting was
achieved with K′a = 10−2.1, a = 0.33, and b = 0.12. While a is
just a normalization constant for the absorbance, b is sometimes
interpreted as an indication of the heterogeneity degree of the
system. When b = 1, eq 4 reduces to the usual sigmoid function
electronic conjugation in the flavylium molecule by reaction of
the double bound, indicating that it polymerizes. Interestingly,
the maximum of the flavylium after the polymerization
compares with that shown by 7-hydroxy-4′-methylflavylium
(2, Scheme 4, 436 nm) bearing a methyl substituent in the 4′
Scheme 4. Chemical Structure of 7-Hydroxy-4′-
methylflavylium
position,17 which can be considered as a model in solution for
the study of the properties of flavylium 1 linked to the
polymeric backbone.
1H NMR analysis of the isolated polymer showed the typical
signals of poly(N-isopropylacrylamide)18 (see Figures S1 and
S2, Supporting Information). The amplification of the aromatic
region revealed that the amount of the flavylium molecule
9057
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
On the other hand, the fraction of quinoidal base at the
equilibrium at pH ∼ 6 is significantly larger (∼55%) than that
observed for the model compound 217 and in general for the
majority of flavylium compounds in water (usually less than
10%).12 This is compatible with the stabilization of this neutral
form relatively to the positively charged AH+ species in the
polymer backbone, when compared to water. Moreover
aggregation can occur more easily in the case of the quinoidal
base due to its planar and noncharged structure, stabilizing this
species against the other species in the flavylium network. This
phenomenon is well-known for anthocyanins, the natural
analogues of flavylium compounds, where the quinoidal base is
present at the equilibrium in large quantities (up to 50% or
more, depending on the concentration), stabilized by self-
aggregation.24
Direct pH jumps to pH ≥ 12 leads to the formation of the
species Ct−2 in a few minutes (due to the hydroxyl nucleophilic
attack to quinoidal base)25 and this species can be titrated back
to acidic pH values to give the species Ct‑ and Ct (Scheme 5).
The titration yielded the acidity constants pKCt‑ = 9.9 and pKCt
= 8.6 respectively (Figure 3B). These values are slightly higher
that those found for the model compound 2, namely pKCt‑
=
9.4 and pKCt = 7.9 respectively.17 Again, the explanation can be
that the polymeric environment stabilizes neutral/less charged
species against more charged species. It must be noted in
Figure 3B, that at the end of the titration a significant
percentage of quinoidal base is already in equilibrium with the
neutral trans-chalcone, which remarks the high stabilization of
the quinoidal base in the polymer.
Figure 2. Spectral variations taking place immediately after a series of
pH jumps from pH = 1 to higher pH values (1.7% MeOH, ionic
strength = 0.1 M (NaClO4 or HClO4), [P1] = 0.017 wt %, T = 21
°C). Inset: fittings of the absorbance values at the specified
wavelengths allowed the determination of pKa.
on pH; when 0 < b < 1, the smaller b is the more stretched is
the sigmoid. In that sense, a value closer to 1 would be
indicative of a homogeneous system; the calculated value b =
0.12 seems to indicate a relatively important number of
microenvironments affecting the global acid−base equilibrium
relating AH+ and the other species.
Because of the polymerization procedure employed to obtain
P1 (thermal free-radical polymerization and short times), the
polymeric chains probably present a significant heterogeneity in
the distribution of the flavylium moieties regarding to the local
concentration and number of flavylium units per chain, among
other factors. These heterogeneities can induce different
microenvironments that can affect the attainment of the
equilibrium. Although the acid−base processes regarding to
both the formation of the quinoidal base and the consecutive
b
[H+]
⎛
⎜
⎝
⎞
⎟
⎠
A = a
[H+] + K′
(4)
a
Figure 3. Spectral variations of equilibrated aqueous solutions (1.7% MeOH, ionic strength =0.1 M (NaClO4 or HClO4)) of P1 at different pH
values, c = 0.017 wt %, T = 21 °C. (A) pH range from 1 to 6; the absorbance variations at 442 nm were fitted using a power law model (inset). (B)
pH range from 6 to 12; data fitted with normal sigmoid curve (inset).
9058
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
Scheme 5. Acidity Constants of the trans-Chalcones Attached to the Polymer Backbone (Deprotonation Order Not Confirmed)
Figure 4. (A) Normalized absorption (a), emission (b, λexc = 442 nm) and excitation (c, λem = 615 nm) spectra at pH = 1.2. (B) Normalized
absorption (a), emission (b, λexc = 476 nm) and excitation (c, λem = 610 nm) spectra at pH = 8.0. Solution conditions: 1.7% MeOH, ionic strength =
0.1 M (NaClO4 or HClO4), [P1] = 0.007 wt %, and T = 21 °C.
Figure 5. Time evolution of UV−vis spectra upon a pH jump from a stock aqueous solution of P1 (c = 0.017 wt %, 1.7% MeOH, [NaClO4] = 0.1 M,
T = 21 °C) at pH = 1 to pH = 3.4 (A) and to pH = 6.2 (B). Insets: Fittings of the absorbance values at the specified wavelengths to a single or
double exponential.
deprotonation equilibria of the trans-chalcone can be well fitted
with individual constants, we cannot exclude the presence of
several equilibrium conditions corresponding to slightly
different chemical environments in the polymer. It can simply
happen that we do not reach the experimental resolution to
distinguish between these slight differences. In fact, it is difficult
to detect two pKa values that differ in less than one unit.26
Comparing with the global acid−base equilibrium, which
implies several reactions (acid−base, hydration, tautomerization
and isomerization), it can be more affected by small differences
in the microenvironment than a single protonation−deproto-
nation process and thus it is easier the establishment and
detection of different equilibrium conditions. The protona-
tion−deprotonation processes take place in the μs time scale
9059
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
Figure 6. (A) Spectral variations of a solution of P1 at pH = 1 upon increase of the temperature (inset: transmittance changes at 700 nm and
determination of the LCST value by the inflection point method). (B) pH dependence of the LCST of P1 and pNIPAM control (in all the cases the
ionic strength was kept constant at 0.1 M).
while the hydration, tautometization and isomerization
reactions occur in longer time scales (often in the ms, s and
h time scales, respectively). The overall equilibria leading to K′a
(Scheme 1) may then discriminate the several microenviron-
ments present in the polymer which is reflected in the large
nonsigmoidal mole fraction vs. pH distribution curve (inset
Figure 3A) comprising a series of close equilibrium constants.
The equilibrated solutions were studied as well by steady-
state fluorescence emission. At pH = 1.2 and pH = 8.0 (Figure
4) the emission spectra are very similar and they can thus be
assigned to the emission of the quinoidal base (λmax = 610 nm),
also on the basis of its usually vibrationally resolved emission
band.12 This result can be explained by an excited state proton
transfer process (ESPT),27 i.e. the acidity of the flavylium
cation is much higher in the excited state than in the ground
state and as a result the emission of the quinoidal base is
collected when exciting a solution of the flavylium cation. At
pH = 1.2, an additional contribution centered at ca. 500 nm is
observed that can be attributed to the emission of the flavylium
cation itself. At pH = 5.3 (see Figure S5 in Supporting
Information) together with the emission of the quinoidal base,
it is possible to observe a new emission band centered at 488
nm that is assigned to the trans-chalcone.
Kinetics of the Flavylium Network. The kinetics of the
formation of the equilibrium species after direct pH jumps from
pH = 1 to moderately acidic and neutral pH values, revealed the
lack of a high cis-to-trans isomerization barrier for the formation
of Ct, as reported previously for flavylium compounds bearing
an hydroxyl substituent in position 7 (Figure 5).28 This
behavior indicates that the polymeric chains are averagely
solvated, facilitating the reaction of the flavylium moieties. In
general, after a pH jump, and depending on pH, AH+ can be
partially or completely transformed into the quinoidal base A (a
process too fast to be followed in a conventional spectropho-
tometer) and then, the pair AH+/A evolves to the equilibrium
through a biexponential process which is also dependent on
pH.12 The faster rate can be attributed to the formation of a
pseudoequilibrium involving the pair AH+/A, the hemiketal B
and cis-chalcone Cc; the slower rate corresponds to the
formation of Ct from the previous pseudoequilibrium. As can
be seen in Figure 5, the spectral variations taking place in P1 at
the absorption wavelength of AH+ and/or A reveal a
biexponential process, whereas the trace at 366 nm (absorption
of Ct) can be fitted as a monoexponential growth, with a delay
during the first seconds. In the last case, the observed rate
constant is coincident with the slowest rate calculated for the
biexponential decay of AH+/A. This behavior is in fact
compatible with the attainment of the pseudoequilibrium
involving AH+/A, B, and Cc previously to the formation of Ct.
On the other hand, in flavylium cations possessing hydroxyl
groups in position 7, the hydration and tautomerization
reactions are often too fast to be observed by conventional
spectrophotometers for pH < 4, and monoexponential kinetic
laws are observed.28 Thus, the biexponential behavior in Figure
5A (pH = 3.4) seems to indicate that the polymer backbone
slows down the hydration process because the tautomerization
reaction is an intramolecular process, and therefore less prone
to be affected.
Temperature Dependence Behavior. The lower critical
solution temperature (LCST) of P1 and pNIPAM control was
determined for several equilibrated solutions at different pH
values by measuring the optical density of the solutions at 700
nm (see Figure 6A for P1 at pH = 1).
As can be seen in Figure 6B, the LCST for the polymer
control pNIPAM decreases with decreasing pH, going from
34.0 to 31.7 °C when the pH changes from 7 to 1. It is known
that the addition of salts generally lowers the phase transition
temperature of pNIPAM.29−31 This behavior is thought to be
related to interactions of the ions with the hydration water
molecules in the polymer and to the increase of the surface
tension on the cavity around the isopropyl side chains and the
polymeric backbone (hydrophobic effect).29,30 Similar effects
can take place in the presence of hydronium ions, especially the
first one, due to the singular and cooperative hydrogen-bond
network that solvates protons in bulk water, which should
compete and weaken the solvation of the polymeric chains.
In the case of P1, the LCST decreases as well in acidic media
but the effect is higher than for pNIPAM. At pH = 1, the value
9060
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
is 28.2 °C (about 3.5 °C lower than in the control) and
increases with increasing pH up to 32.2 °C at pH = 6.4, see
Figure 6B. Interestingly, for P1 the pH dependence of the
LCST can be fitted to a sigmoid curve with an inflection point
at pH = 3.6 which corresponds to a mole fraction of AH+ ∼
0.5.32 This behavior indicates that the presence of AH+ in the
polymer leads to a reduction in the value of the LCST. This
effect can be explained by the hydrophobic character of the
flavylium cation which could prevent to a certain extension the
hydrogen bonding network between the amide groups in the
polymer and the water molecules and as a consequence
facilitate the collapse of the polymer at lower temperatures.
Photochemical Behavior: Photocontrol of Polymer
Collapsing. Since the presence of AH+ in the polymer reduces
the value of the LCST, it should in principle be possible to
induce the photochemical collapse of P1 by irradiating a
solution at the appropriate pH and temperature values. As
commented above, in the flavylium system the photoactive
species is the trans-chalcone, Ct. When a solution of Ct is
irradiated, the photoproduct, cis-chalcone, Cc, is formed in
picoseconds and then two competitive thermal processes occur,
namely the back isomerization to Ct and the formation of AH+
through the hemiketal B. The dehydration of B to give AH+ is
pH-dependent becoming slower at higher pH values, and at a
certain point, when the pH is too high, the dehydration is too
slow and Ct is recovered from the mixture of Cc and B without
formation of AH+.12 Taking these factors into consideration we
can rationalize that a good pH to photoproduce AH+, should be
one high enough to permit the existence of a significant amount
of Ct at the equilibrium and low enough to allow B to
dehydrate and give AH+.
We selected for the irradiation an equilibrated solution at pH
= 3.9 because at this pH value a significant amount of trans-
chalcone is present at the equilibrium and it is acidic enough to
permit the formation of AH+ from B. On the other hand, the
temperature selected for the experiment was 29 °C since in
these conditions the formation of AH+ is expected to produce a
collapse of the polymer, as the comparison of the LCST values
for pH =1 (28.2 °C) and pH = 3.9 (31.0 °C) suggests (Figure
7).
The sample was kept at 29 °C during 10 min before the
irradiation, and then was irradiated with a Xe lamp (450W) at
370 nm. The spectra corresponding to different times of
irradiation are shown in Figure 8A. At the beginning of the
irradiation the baseline did not change, but after a certain time
the baseline started to rise, indicating that the polymer was
collapsing. Figure 8B shows the spectra with the baseline
corrected where it can be clearly seen that the irradiation
resulted on the consumption of trans-chalcone and appearance
of flavylium cation (and quinoidal base). Comparison of both
figures makes clear that after the formation of a certain
percentage of flavylium cation the polymer started to collapse.
After approximately 1 h of exposition to light the system started
to degrade and the irradiation was stopped. The subsequent
monitoring of the absorbance changes at 29 °C (see Figure S6
in Supporting Information) showed the slow recovery of Ct;
nevertheless the collapsed polymer remained precipitated
during much more time (several hours) until the polymeric
chains could be resolvated.
Figure 7. Comparison between the transmittance changes at 700 nm
as a function of temperature for P1 at pH = 1.0 and pH = 3.9.
based on the copolymerization of N-isopropylacrylamide and a
vinyl flavylium derivative, the former being responsible for the
temperature-induced changes and the latter for the response to
pH and light. The polymer follows the typical pH-dependent
network of reactions of flavylium compounds in water and
shows an unusual large stabilization of the quinoidal base. The
presence of flavylium cation reduces the LCST of the polymer,
due to the hydrophobic character of the species which may
prevent at a certain extension the formation of hydrogen bonds
between the polymer and water. It is possible to take profit of
this behavior to photocollapse partially the polymer, by
irradiating a solution at the appropriate pH and temperature.
Under suitable irradiation conditions, the trans-chalcone leads
to the formation of the flavylium cation and the polymer
precipitates. Because of the chemical versatility of flavylium
compounds we believe that their utilization on the develop-
ment of multistimuli responsive polymers opens a wide range
of attractive possibilities. Currently, additional efforts to prepare
multistimuli responsive polymers by means of the postfunction-
alization of preformed polymers with flavylium derivatives are
ongoing.
EXPERIMENTAL SECTION
All the reagents and solvents used were of analytical grade.
■
Synthesis. Synthesis of 4′-Vinylacetophenone. The compound
was prepared according to the Hiyama coupling reaction as described
previously in the literature, using sodium hydroxide as activating agent,
palladium(II) acetate and tetra-n-butylammonium bromide (TBAB) as
catalyst and phase transfer catalyst, respectively.15 In a high pressure
glass vessel with septum, p-bromoacetophenone (10.12 mmol; 2.01 g),
vinyltrimethoxysilane (20.37 mmol; 3.02 g), palladium acetate (0.06
mmol; 0.01 g), and TBAB (0.01 mmol; 4.06 g) were mixed, and then
water (20 mL) was added and the mixture deoxygenated with argon.
Next, 2 mL of a previously deoxygenated 50 wt % NaOH aqueous
solution was added dropwise to the mixture under argon atmosphere,
and the vessel was locked, while a white-off solid appeared. After that,
the reaction was carried out on a microwave reactor by performing 8
consecutive cycles of 5 min duration with 80 W power and ca. 1.4 bar
pressure. After every cycle, the mixture was cooled with an air stream
during 2 min. Afterward, the reaction mixture was let to cool at room
temperature. The product was extracted with diethyl ether (8 × 75
mL), and then washed with 2 M NaOH (2 × 60 mL), 2 M HCl (2 ×
CONCLUSIONS
A new multistimuli responsive polymer with response to
temperature, pH, and light was synthesized. The polymer is
■
9061
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
Figure 8. (A) Spectral variations upon irradiation of a solution of P1 at pH = 3.9 and T = 29 °C (λirrad = 370 nm, [P1] = 0.017 wt %, 1.7% MeOH,
[NaClO4] = 0.1 M). The arrow indicates the rise of the baseline due to the collapsing of the polymer. Irradiation times (s): 0, 20, 80, 140, 260, 560,
1160, 1760, 2360, 2960, and 3260. (B) Same as part A with the baseline corrected showing the consumption of Ct (arrow down) and the formation
of AH+/A (arrow up).
60 mL) and water (2 × 60 mL). The ether extract was then dried with
anhydrous sodium sulfate and the solvent evaporated. Finally, the
product was dried under vacuum yielding a yellow oil (1.18 g; 80%).
1H NMR (400.13 MHz, 298 K, CDCl3) δ = 7.92 (d, 2H, 3J = 7.2 Hz),
7.48 (d, 2H, 3J = 7.3 Hz), 6.76 (dd, 1H, 3J1 = 17.6, 3J2 = 10.8 Hz), 5.88
solvent was evaporated and the brown polymeric solid dried under
vacuum (70 mg; 26%). The mole fraction content of the flavylium
1
moiety in the polymer is ∼0.4% from comparison between the H
NMR integrations of proton H4 of the flavylium and the proton
corresponding to −CH− in the isopropyl group of NIPAM (see
Figure S2 in Supporting Information). 1H NMR (400.15 MHz, 298 K,
CD3OD/DCl, pD < 1, AH+ species for flavylium): δ = 9.35 (br, 1H,
H4), 8.61−8.22 (br, 4H, H3, H2′, H6′, H5), 7.71−7.34 (br, 4H, H8,
H3′, H5′, H6), 3.97 (br, 1H, (CH3)2CH from NIPAM), 2.13 (br, 1H,
CH in polymeric backbone), 1.62 (br, 2H, CH2 in polymeric
backbone), 1.17 (br, 6H, (CH3)2CH from NIPAM). IR (solid): ν =
3298 cm−1 (secondary amide N−H stretching), 2972 cm−1 (−CH3
asymmetric stretching), 1651 cm−1 (amide I bond), 1547 cm−1 (amide
II bond).
3
3
(d, 1H, J = 17.6 Hz,), 5.40 (d, 1H, J = 10.8 Hz), 2.60 (s, 3H).
Synthesis of 7-Hydroxy-4′-vinylflavylium Hydrogen Sulfate (1).
This compound was prepared by condensation of 4′-vinylacetophe-
none (3.54 mmol; 520 mg) with 2,4-dihydroxybenzaldehyde (3.54
mmol; 490 mg) in acidic medium. The reagents were dissolved in 7.1
mL of acetic acid and 1.8 mL of sulfuric acid and the reaction mixture
was stirred at room temperature overnight. Then, the product was
precipitated using ethyl acetate and the solid washed with this solvent
several times. Finally the compound was dried under vacuum yielding
1
a brownish powder (530 mg, 44%). H NMR (400.15 MHz, 298 K,
Synthesis of Poly(N-isopropylacrylamide) (pNIPAM). N-Isopropy-
lacrylamide ((2.23 mmol; 250 mg) was polymerized in the presence of
AIBN (2.5 mg; 1 wt % with respect to the monomers) using the same
reaction and purification conditions that in the case of P1. The
3
CD3OD/DCl, pD < 1, AH+ species): δ = 9.32 (d, 1H, J = 8.5 Hz,
H4), 8.54 (d, 1H, 3J = 8.6 Hz, H3), 8.49 (d, 2H, 3J = 8.4 Hz, H2′,H6′),
8.29 (d, 1H, 3J = 9.0 Hz, H5), 7.82 (d, 2H, 3J = 8.4 Hz, H3′,H5′), 7.67
(d, 1H, 4J = 1.9 Hz, H8), 7.52 (dd, 1H, 3J = 9.0, 4J = 2.1 Hz, H6), 6.91
1
polymer was isolated as a white powder (88 mg; 35%). H NMR
3
3
3
(dd, 1H, J1 = 17.6, J2 = 11.0 Hz, CH=CH2), 6.14 (d, 1H, J = 17.6
(400.13 MHz, 298 K, CD3OD): δ = 8.16−7.16 (br, 1H, NH), 3.97 (br,
1H, (CH3)2CH), 2.09 (br, 1H, CH in polymeric backbone), 1.80−1.27
(br, 2H, CH2 in polymeric backbone), 1.16 (br, 6H, (CH3)2CH). IR
(solid): ν = 3286 cm−1 (secondary amide N−H stretching), 2972
cm−1 (−CH3 asymmetric stretching), 1651 cm−1 (amide I bond), 1549
cm−1 (amide II bond). MALDI−TOF MS: Mw = 7842 g/mol, Mn =
6521 g/mol, PDI = Mw/Mn = 1.2.
Characterization. NMR. NMR spectra were run on a Bruker ARX
400 instrument operating at 400.13 MHz (1H) or a Bruker Avance III
400 operating at 400.15 MHz (1H) and 100.62 MHz (13C). Signals of
deuterated solvents were used as internal references.
IR Spectroscopy. Infrared analyses were carried out using a Nicolet
Nexus spectrophotometer coupled to a Continuum microscope (15×
objective) with a MCT-A detector cooled by liquid nitrogen. The
spectra were collected in transmission mode, in 50−100 mm areas;
resolution set to 4 cm−1 and 128 or 256 scans, using a Thermo
diamond anvil compression cell.
Mass Spectrometry. ESI−TOF HRMS mass spectra were
performed on a Bruker microTOF (Bruker Corporation). MALDI−
TOF MS analyses were performed on a Voyager-DETM PROBio-
spectrometry Workstation model (Applied Biosytems) and data were
analyzed with Voyager V5.1 software. Dithranol was employed as the
matrix and NaBF4 was added to improve the ionization. Samples and
Hz, CHCH(trans)), 5.57 (d, 1H, 3J = 10.9 Hz, CHCH(cis)). 13
C
NMR (400.15 MHz, 298 K, CD3OD/DCl, pD < 1, AH+ species): δ =
172.8, 171.5, 161.3, 156.2, 146.6, 136.7, 134.5, 131.0, 129.5, 128.9,
123.6, 121.5, 119.9, 114.3, 103.8. ESI−TOF HRMS: m/z (%) calcd for
+
C17H13O2 , 249.0910; found, 249.0909 [M+] (100).
Synthesis of Poly(N-isopropylacrylamide-co-7-hydroxy-4′-vinyl-
flavylium) (P1). The synthesis was carried out by free radical
polymerization in a solvent mixture DMF/H2O 90/10 v/v in the
presence of 0.3 M HCl and employing 2,2′-azobis(isobutyronitrile)
(AIBN) as the initiator. First, the solvent mixture was deoxygenated
during 15 min with argon. Then, N-isopropylacrylamide (NIPAM,
2.23 mmol; 250 mg), 7-hydroxy-4′-vinylflavylium hydrogen sulfate
(0.04 mmol; 13 mg) and AIBN (2.6 mg, 1 wt % with respect to the
monomers) were dissolved in 1.67 mL of the solvent mixture. The
flask was sealed with a rubber septum, the temperature was raised to
85 °C, and the mixture let to react during 1 h. The reaction mixture
was then cooled at room temperature and the product was precipitated
in ethyl acetate. The polymer was dissolved in 0.5 M HCl, and
reprecipitated by warming at 50 °C. The precipitate was isolated by
filtration and washed carefully with hot water. A last precipitation was
carried out by dissolving the polymer in methanol, precipitating with
hot water and washing with diethyl ether. The polymeric solid was
recovered from the sintered glass funnel by dissolving in methanol, the
9062
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063

Macromolecules
Article
matrix were dissolved in acetone and the spectra were recorded on the
positive linear mode.
(11) Szczubialka, K.; Moczek, L.; Blaszkiewicz, S.; Nowakowska, M. J.
Polym. Sci., Part A: Polym. Chem. 2004, 42, 3879−3886.
(12) Pina, F.; Melo, M. J.; Laia, C. A. T.; Parola, A. J.; Lima, J. C.
Chem. Soc. Rev. 2012, 41, 869−908.
Solution Measurements. Solutions were prepared in Millipore
water containing 1.7% of spectroscopic MeOH. Polymeric concen-
tration was set at 0.017 wt %. The pH of the solutions was adjusted by
the addition of HClO4, NaOH and the universal buffer of Theorell and
Stenhagen33, and it was measured on a Radiometer Copenhagen
PHM240 pH/ion meter. In all measurements, the ionic strength was
kept constant at 0.1 M by the addition of NaClO4. UV−vis absorption
spectra were recorded on a Varian Cary 100 Bio spectrophotometer.
Emission spectra were registered on a Jobin Yvon Spex, Fluorolog
FL3−22. Photoexcitation in continuous irradiation experiments was
carried out with a Xe lamp (450 W) using a monochromator to select
the 370 7 nm window and a thermostatized bath was employed to
keep the sample at the selected temperature. LCST values were
determined by measuring the transmittance at 700 nm on a Varian
Cary 100 Bio spectrophotometer and using the inflection point
method. The temperature of the cell was regulated to 0.1 °C by a
circulating water bath.
(13) Pina, F.; Petrov, V.; Laia, C. A. T. Dyes Pigment. 2012, 92, 877−
889.
(14) Melo, M. J.; Sousa, M.; Parola, A. J.; Seixas de Melo, J. S.;
Catarino, F.; Marcalo, J.; Pina, F. Chem.Eur. J. 2007, 13, 1417−1422.
̧
(15) Alacid, E.; Naj
́
era, C. J. Org. Chem. 2008, 73, 2315−2322.
(16) UV−vis analysis during the reaction confirmed partial evolution
of the flavylium cation to the trans-chalcone.
(17) Diniz, A. M.; Pinheiro, C.; Petrov, V.; Parola, A. J.; Pina, F.
Chem.Eur. J. 2011, 17, 6359−6368.
(18) Kotsuchibashi, Y.; Kuboshima, Y.; Yamamoto, K.; Aoyagi, T. J.
Polym. Sci., Polym. Chem. 2008, 46, 6142−6150.
(19) Li, D.; Zhang, X.; Yao, J.; Simon, G. P.; Wang, H. Chem.
Commun. 2011, 47, 1710−1712.
(20) Schulte, T.; Siegenthaler, K. O.; Luftmann, H.; Letzel, M.;
Studer, A. Macromolecules 2005, 38, 6833−6840.
(21) P. G. de Gennes, Scaling Concepts in Polymer Physics; Cornell
University Press: Ithaca, NY, 1979.
ASSOCIATED CONTENT
* Supporting Information
■
(22) Williams, G.; Watts, D. C. Trans. Faraday Soc. 1970, 66, 80−85.
(23) Kluver, E.; Meyer, M. J. Appl. Polym. Sci. 2013, 128, 4201−4211.
(24) Leydet, Y.; Gavara, R.; Petrov, V.; Diniz, A. M.; Parola, A. J.;
Lima, J. C.; Pina, F. Phytochemistry 2012, 83, 125−135.
(25) McClelland, R. A.; McGall, G. A. J. Org. Chem. 1982, 47, 3730−
3736.
S
NMR, FTIR, MALDI−TOF MS, and fluorescence data of the
polymers as well as the thermal recovery after irradiation at pH
3.9 and 29 °C. This material is available free of charge via the
Internet at http://pubs.acs.org/.
(26) Edsall, J. T.; Wyman, J., Biophysical Chemistry; Academic Press:
New York, 1958; Vol. 1.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: (R.G.) r.castell@fct.unl.pt.
*E-mail: (A.J.P.) ajp@fct.unl.pt.
Notes
■
(27) Pina, F.; Melo, M. J.; Santos, H.; Lima, J. C.; Abreu, I.;
Ballardini, R.; Maestri, M. New. J. Chem. 1998, 1093−1098.
(28) Roque, A.; Lima, J. C.; Parola, A. J.; Pina, F. Photochem.
Photobiol. Sci. 2007, 6, 381−385.
(29) Freitag, R.; Garret-Flaudy, F. Langmuir 2002, 18, 3434−3440.
(30) Zhang, Y.; Furyk, S.; Bergbreiter, D. E.; Cremer, P. S. J. Am.
Chem. Soc. 2005, 127, 14505−14510.
The authors declare no competing financial interest.
(31) Du, H.; Wickramasinghe, R.; Qian, X. J. Phys. Chem. B 2010,
114, 16594−16604.
ACKNOWLEDGMENTS
■
Fundaca
̧
o para a Cien
̂
cia e Tecnologia is acknowledged through
̃
(32) Calculated from Figure 3A.
the National Portuguese NMR Network, Grant PEst-C/EQB/
LA0006/2013, and Projects PTDC/QUI-QUI/104129/2008
and PTDC/QUI-QUI/119932/2010 as well as postdoctoral
grant SFRH/BPD/44639/2008 (R.G.). C. A. T. Laia and J. C.
Lima are acknowledged for helpful discussions regarding the
nonsigmoidal behavior of the equilibrated system with pH.
(33) Kuster, F. W., Thiel, A. , Eds. Tabelle per le Analisi Chimiche e
̈
Chimico-Fisiche, 12th ed., Hoepli: Milan, Italy, 1982; pp 157−160.
REFERENCES
■
(1) Liu, F.; Urban, M. W. Prog. Polym. Sci. 2010, 35, 3−23.
(2) Special Issue: Stimuli-Sensitive Polymers; Lendlein, A., Prasad
Shastri, V., Eds., Adv. Mater. 2010, 22, 3339−3494.
(3) de las Heras Alarcon
Rev. 2005, 34, 276−285.
(4) Tan, I.; Roohi, F.; Titirici, M.-M. Anal. Methods 2012, 4, 34−43.
(5) Luzinov, I.; Minko, S.; Tsukruk, V. V. Prog. Polym. Sci. 2004, 29,
635−698.
́
, C.; Pennadam, S.; Alexander, C. Chem. Soc.
(6) Schild, H. G. Prog. Polym. Sci. 1992, 17, 163−249.
(7) Schild, H. G.; Tirrell, D. A. J. Phys. Chem. 1990, 94, 4352−4356.
(8) Boutris, C.; Chatzi, E. G.; Kiparissides, C. Polymer 1997, 38,
2567−2570.
(9) (a) Guragain, S.; Bastakoti, B. P.; Ito, M.; Yusa, S.; Nakashima, K.
Soft Matter. 2012, 8, 9628−9634. (b) Li, L. L.; Xing, X. D.; Liu, Z. L. J.
Appl. Polym. Sci. 2012, 124, 1128−1136. (c) Boissiere, O.; Han, D.;
Tremblay, L.; Zhao, Y. Soft Matter. 2011, 7, 9410−9415. (d) Zhao, Y.;
Tremblay, L.; Zhao, Y. J. Polym. Sci., Polym. Chem. 2010, 48, 4055−
4066. (e) Garcia, A.; Marquez, M.; Cai, T.; Rosario, R.; Hu, Z. B.;
Gust, D.; Hayes, M.; Vail, S. A.; Park, C. D. Langmuir 2007, 23, 224−
229. (f) Sumaru, K.; Kameda, M.; Kanamori, T.; Shinbo, T.
Macromolecules 2004, 37, 4949−4955.
(10) Zhao, Y. Macromolecules 2012, 45, 3647−3657.
9063
dx.doi.org/10.1021/ma401497q | Macromolecules 2013, 46, 9055−9063