Self-assembled nanotubes and helical tapes from diacetylene nonionic amphiphiles. Structural studies before and after polymerization

Article abstract of DOI:10.1021/la202162q

We synthesized new amphiphiles comprised of a single diacetylenic chain and an oligoethylenoxide polar chain linked by an amide bond. In aqueous medium, they are not soluble at room temperature but form weak gels. Electron microscopy studies have shown that they self-assemble into helical tapes or nanotubes with lengths of several micrometers, and inner and outer diameters of 50 ± 1 and 59 ± 1 nm, respectively. The wall has a thickness of 10 ± 1 nm for both kinds of objects and has an amphiphile bilayer structure. The hydrophobic chains are ordered, and the amide groups are linked with each other by H-bonds. The dissociation of the tubes is a first-order transition with an enthalpy of ca. 40 kJ mol^-1. The nanotubes were photopolymerized to yield purple solutions consisting of helical tapes and almost flat ribbons. The polymers exhibit irreversible thermochromism upon heating.

Full text of DOI:10.1021/la202162q

ARTICLE
pubs.acs.org/Langmuir
Self-Assembled Nanotubes and Helical Tapes from Diacetylene
Nonionic Amphiphiles. Structural Studies before and after
Polymerization
†
‡
‡
,‡
†
Aur ꢀe lia Perino, Marc Schmutz, St ꢀe phane Meunier, Philippe J. M ꢀe sini,* and Alain Wagner
†
Laboratory of Functional ChemoSystems, UMR 7199, Facult ꢀe de Pharmacie, Universit ꢀe de Strasbourg, 74 route du Rhin, 67401 Illkirch,
France
‡
Institut Charles Sadron, 3 rue du Loess, BP 84047, 67034 Strasbourg Cedex, France
ABSTRACT: We synthesized new amphiphiles comprised of a
single diacetylenic chain and an oligoethylenoxide polar chain
linked by an amide bond. In aqueous medium, they are not
soluble at room temperature but form weak gels. Electron
microscopy studies have shown that they self-assemble into
helical tapes or nanotubes with lengths of several micrometers,
and inner and outer diameters of 50 ( 1 and 59 ( 1 nm,
respectively. The wall has a thickness of 10 ( 1 nm for both
kinds of objects and has an amphiphile bilayer structure. The
hydrophobic chains are ordered, and the amide groups are
À1
linked with each other by H-bonds. The dissociation of the tubes is a first-order transition with an enthalpy of ca. 40 kJ mol . The
nanotubes were photopolymerized to yield purple solutions consisting of helical tapes and almost flat ribbons. The polymers exhibit
irreversible thermochromism upon heating.
iacetylenic lipids were the first compounds discovered to
form helical tapes or tubes. Since then, more molecules
atomic resolution is known only in a few cases where the tubes
1
18,19
D
have been successfully ordered.
Yet generally, their limited
able to form self-assembled nanotubes have been discovered with
size gives diffuse low-intensity signals on selected area diffraction,
and scattering experiments on solutions give powder patterns.
The helical tapes and nanotubes have triggered a lot of theoretical
research by physicists to understand the underlying of their self-
2
various chemical structures. The assemblies are driven by
noncovalent bonds such as H-bonds, van der Waals, or solvo-
phobic interactions. The resulting objects can find applications
due to their unique shape and their high aspect ratio. They have
20À27
assembly.
The proposed models derive the helical and
3
been used for instance as scaffolds for cell growth, as supports
tubular shape from the chirality of the constituent molecules
that induces the intrinsic bending of the self-assembled layers.
They cannot explain the few examples of nonchiral molecules
4
À6
for 2-D crystallization of proteins,
or as templates to form
7
silica nanotubes or mesopores in organic matrices. Because their
self-assembly is based on noncovalent bonds, the formation of
these nanotubes is thermoreversible and therefore depends on
temperature or concentration. Yet for some applications it is
necessary to rely on structures independent from the concentra-
tion. For this reason, there has been a growing interest to develop
17,28
forming nanotubes.
These models also rely on a phenom-
enological description of the elastic energy but do not take into
account the local packing or interactions between the molecules
and cannot identify the molecular parameters that govern the
formation of the tubes. In this context, the discovery of new tube
forming compounds, especially nonchiral ones, is of interest and
may help to identify the chemical interactions involved in the
self-assemblies.
Herein, we report a new nonchiral amphiphile (1, Scheme 1)
able to form nanotubes in aqueous medium. This amphiphile has
a single diacetylenic chain as the hydrophobic part and an
oligoethylenoxide chain as the hydrophilic part. The structure
and properties of the self-assemblies have been investigated by
electron microscopy, DSC, and FTIR to establish which inter-
actions are involved in the tube formation. The nanotubes were
8
À11
self-assembled tubes that can be locked by polymerization.
The diacetylenic amphiphiles are appropriate for this purpose
because their topochemical polymerization proceeds without
drastic reorganization of the self-assembly as it has been proved
1
12
very early. However, only a few thermal studies have been
performed on those polymerized tubes. The first examples of
molecules forming tubes were phosphatidylcholines bearing two
1
,13
diacetylenic chains. Efforts have been directed toward simpler
structures, more easily synthesized, especially with a single
1
4À16
acetylenic chain as the hydrophobic part and sugar
or
17
ammonium as the polar part. However, there are no rules to de-
sign such molecules, and, most often, a slight chemical modi-
fication of the parent molecule annihilates the formation of tubes.
If predictability is lacking it is mainly because the structure at the
Received: June 8, 2011
Revised:
July 31, 2011
Published: September 08, 2011
r 2011 American Chemical Society
12149
dx.doi.org/10.1021/la202162q
|Langmuir 2011, 27, 12149–12155

Langmuir
ARTICLE
Scheme 1. Chemical Structure of the Amphiphiles 1À4
CH
2
ÀCtC), 2.12 (t, J = 7.6 Hz, 2H, CH
2
COOH), 1.58À1.20 (m,
1
3
36H), 0.83 (t, J = 6.4 Hz, 3H, CH ). C NMR (100 MHz, CDCl ): δ =
3
3
1
73.5 (CONH), 77.7 (C12
H
25ÀCtCÀCtC), 77.6 (C12
H
25ÀCt
CÀCtC), 72.7 70.9, 70.8, 70.7, 70.7, 70.6, 70.6, 70.5, 70.4, 70.3, 70.1
(
(
CH O), 65.4 (CtCÀCtC), 61.8 (CH OH), 39.1 (HNÀCH ), 36.8
2
2
2
CH COOH), 32.1, 29.8, 29.8, 29.7, 29.6, 29.6, 29.6, 29.5, 29.5, 29.2,
2
2
9.0, 28.5, 25.9, 22.8, 19.3 (CH
2
ÀCtC), 19.3 (CH
2
ÀCtC), 14.3
ppm (CH ). FTIR (solid): ν = 3301 (ν NH), 2916 (νas CH
3
2
), 2847 (ν
s
CH ), 1738 (amide I), 1555 (amide II), 1463 (ν CH ), 1348, 1244,
2
2
À1
+
1
110 cm . HRMS m/z 666.5255 (M + H , C39
H
7
71NO requires
666.5309).
FT-IR. Spectra were recorded on a Bruker Vertex 70 spectrophot-
ometer equipped with a thermostatic cell holder and a temperature-
controlling unit (Specac West 6100+). The gels or solutions were placed
Scheme 2. Synthesis of Amphiphile 1
in a CaF cell with an optical path of 0.1 mm. The temperature was
2
increased of 1 °C before each spectrum (rate 1 °C/min), and the
measurement was performed after the temperature was stabilized (about
one-half a minute). The samples were visually checked after each heating
phase to ensure that no loss of solvent had occurred. The spectra were
2
recorded at different temperatures, were corrected from CO and water
vapor, and compensated for D O. They were processed and fitted with
2
Igor (Wavemetrics, Inc.). The maxima of the peaks were found with the
built-in algorithm using second derivatives. For the curves with higher
signal/noise ratio, the maxima were found by fitting the peaks with a
then photopolymerized, and the resulting self-assemblies were
studied.
Gaussian curve. Both calculation techniques gave the same results within
À1
an error of 0.1 cm
.
UV. Spectra were recorded on a Cary/Varian 500 spectrometer
equipped with a Linkam stage heating cell. The solutions to be measured
were put in quartz cells. The cells were heated by increments of 1 °C
(rate of 1 °C/min), the temperature was allowed to stabilize 30 s before
each measurement, and the corresponding spectrum was recorded
during each temperature plateau.
DSC. Thermograms were recorded with a microcalorimeter DSC III
from Setaram. The heating and cooling rates were set at 0.1 °C/min.
Electron Microscopy À Negative Staining. A droplet of the
solution to be studied was cast on a 400 mesh copper grid with a carbon
film. After a few seconds of adsorption, the excess solvent was blotted
with a small piece of Whatmann paper (No. 4 or 5). A droplet of a
solution of uranyl acetate (2%) was laid on the grid and soaked again.
The grid was observed with a Philips CM 120 electron microscope
operating at 100 kV.
Electron Microscopy À Cryogenic Preparation (cryo-TEM).
A 400 mesh lacey carbon film copper grid was rendered hydrophilic by a
mild glow discharge. A 5 μL aliquot of the studied solution was deposited
and left to adsorb for 2 min. Excess of sample solution was removed with a
piece of filter paper (Whatmann 2 or 5) to leave a thin film. The grid was
rapidly dipped into liquid nitrogen-cooled ethane and stored in liquid
nitrogen until observation. The grid was transferred in a Gatan 626 cryo
holder and observed with a FEI Tecnai G2 at 200 kV equipped with a FEI
eagle 2k ssCCD camera. Distances were measured with Analysis software
(SIS-Olympus, M €u nster, Germany). More than 200 measurements were
done and averaged. The uncertainty was taken equal to the standard
deviation.
’
EXPERIMENTAL SECTION
Syntheses. The synthesis of the amphiphiles 2, 3, and 4 will
be described elsewhere. Amphiphile 1 was synthesized according to
Scheme 2. Compound 5 was synthesized according to a published
procedure as well as hexaethyleneglycol monoamine.
Heptacosa-12,14-diynoic Acid (6). A solution of acid 5 (240 mg;
29
30,31
0.54 mmol; 1 equiv) inTHF (4.3 mL) was mixed with a solution ofHCl in
acetic acid (0.1% of concentrated acid; 4.3 mL). The mixture was stirred
under microwave for 21 min at 150 °C. Water (5 mL) was added, and the
aqueous layer was extracted with EtOAc (3 Â 5 mL). Combined organic
2 4
layers were dried (Na SO ) and concentrated under vacuum. The residue
was chromatographed (SiO
2 3
, CHCl eluent) to give 6 as a white solid in a
1
quantitative yield (215 mg): mp 68.5 °C. H NMR (300 MHz, CDCl ):
3
δ = 2.32 (t, J = 7.5 Hz, 2H, CH
2
COOH), 2.22 (t, J = 6.9 Hz, 4H,
ÀCtC), 1.63À1.57 (m, 2H, CH CH COOH), 1.52À1.24 (m,
4H), 0.86 ppm (t, J = 6.8 Hz, 3H, CH ). C NMR (75 MHz, CDCl ): δ
CH
3
2
2
2
1
3
3
3
=
179.9 (COOH), 77.8 (C H ÀCtCÀCtC), 77.7 (C H ÀCt
12
25
12 25
CÀCtC), 65.5 (CtCÀCtC), 34.2 (CH
2
COOH), 32.1, 30.0, 29.9,
29.8, 29.7, 29.6, 29.6, 29.4, 29.3, 29.3, 29.1, 29.0, 28.6, 24.9, 22.9, 19.4
(
(
(
CH ÀCtC), 14.3 ppm (CH ). FTIR (solid): ν = 2918 (ν CH ), 2848
2
3
as
2
À1
ν
s
CH
2
), 1721 (ν CO), 1467 (δ CH
2
), 727 cm . HRMS: m/z 401.3432
+
M À H , C27
46 2
H O requires 401.3498).
N-(17-Hydroxy-3,6,9,12,15-pentaoxaheptadec-1-yl)hep-
tacosa-12,14-diynamide (1). In a solution of acid 6 (215 mg; 0.53
mmol; 1 equiv) in anhydrous DMF (4 mL) were added EDC (153.5 mg;
0
.80 mmol; 1.5 equiv), HOBt (108.5 mg; 0.80 mmol; 1.5 equiv), a
Polymerization. Solutions of 1 in water (1 mg/mL) were irradiated
in a thermostatic homemade cabinet with a 1000 W UV Hg lamp, under
Ar at 10 °C during 15 min. The rate conversion was measured as
follows: A solution of the polymer prepared as described above was
precipitated in THF (10 times the volume of the solution), and the
resulting suspension was stirred 12 h under argon and in the dark. The
suspension was centrifuged, and the supernatant was evaporated and
diluted in a known volume of THF to reach a concentration of about
1.5 mg/mL. 200 μL aliquots of this solution were injected on
chromatography setup composed of a pump (Shimadzu LC20AD)
operating at a flow of 1 mL/min, 4 columns PLgel (granulometry 5 μ),
solution of hexaethyleneglycol monamine (181 mg; 0.64 mmol; 1.2
equiv), and DIPEA (140 μL; 0.80 mmol; 1.5 equiv) in anhydrous DMF
(
1.5 mL). The solution was stirred 18 h at room temperature, diluted
with a saturated solution of NH
Cl, and extracted with DCM (3 Â
0 mL). The combined organic layers were washed with a saturated
solution of NH Cl (20 mL) and dried over Na SO . Chromatography of
4
1
4
2
4
the residue (SiO , MeOHÀCHCl : 3/97 (v/v) eluent) yielded pure 1 as
2
3
1
a white solid (248 mg, 70%): mp 62.3 °C. H NMR (400 MHz, CDCl
δ = 6.27 (br s, 1H, NH), 3.49À3.68 (m, 22H, CH O), 3.38À3.40
m, 2H, CH NH), 3.06 (br s, 1H, OH), 2.18 (t, J = 6.8 Hz, 4H,
3
):
2
(
2
1
2150
dx.doi.org/10.1021/la202162q |Langmuir 2011, 27, 12149–12155

Langmuir
ARTICLE
Figure 1. Thermogram of a gel of 1 in water (1 wt %). Scan rate 0.1
°
C/min.
5
0, 100, 500, 1000 Å (separation area: 100À20 000 g/mol), a dif-
ferential refractometer from Shimadzu (RID10A), and a UV detector
from Shimadzu (SPD 10 Avp). The pure starting materials, which is
soluble in THF, was injected separately to identify the elution times
and to calculate the molar extinction at 239 nm (ε) and refraction
increment (dn/dc). The supernatant showed one main peak cor-
Figure 2. Top: VT-FTIR spectra of a gel of 1 in water (1 wt %); amide I
and II bands 25À38 °C. Bottom: Frequency and intensity of the amide I
band versus temperature.
responding to the monomer and traces of oligomers (DP = 10) not
n
detectable by RI but only by UV. The results obtained with a PEO
calibration were in good agreement for both detections. The amounts
of the monomer in the supernatant were measured from the areas
from both the RI and the UV traces, and a conversion rate of 71 ( 5%
was found.
’
RESULTS AND DISCUSSION
Associating Properties of 1. The structure of amphiphile 1
studied in this Article is shown in Scheme 1. It bears a diacetylene
chain comprised of 25 carbon atoms, and hexaethylenglycol as
the hydrophilic part. This compound was synthesized readily at
the gram scale by a malonic synthesis as described in the
Experimental Section. It is not soluble at 25 °C in water, for
concentrations above 1 mg/mL. Suspensions at this concentra-
tion dissolve above 36 °C and slowly precipitate when cooled
back to room temperature. By comparison, amphiphiles 2 and 3
solubilize and form spherical micelles under the same conditions,
and 4 is unsoluble in water even under heating. The solubilities
can be explained by the length of the hydrophilic chains, which
modifies the hydrophobic/hydrophilic balance. However, com-
pound 3 has the same structure as 1 except that the hydrophobic
chain comprises 23 carbon atoms instead of 25, and this slight
difference leads to surprisingly different associating behavior.
Indeed, a suspension of 1 at concentrations higher than 1 wt %
dissolves upon heating and forms an opaque gel when it is cooled
back at 25 °C. When the gel is heated, it melts again at 32À34 °C,
giving a clear solution. This thermoreversibility encountered for
Figure 3. Top: VT-FTIR spectra of a gel of 1 in water (1 wt %) in the
CH stretching from 25 to 38 °C. Bottom: Frequencies of the νas and ν
CH bands versus temperature.
s
2
À1
25 to 38 °C. The spectrum at 25 °C shows a band at 1634 cm
identified as the amide I band. Its frequency is characteristic of amide
groups H-bonded with each other. As the temperature increases, the
intensity of this band slightly decreases. The spectrum also shows a
À1
broad peak between 1400 and 1500 cm , which comprises the
CH bending and the amide II bands. The frequency of the former
2
3
2,33
all gelators
is the signature at the macroscopic level of
also shows that the amides are H-bonded. As the temperature
increases, the intensity of the amide I decreases abruptly between 32
association through noncovalent bonds, and the melting of the
gel corresponds to the dissociation of these bonds.
The gel-to-sol transition was studied by DSC with 1% wt gels.
The thermograms (Figure 1) show a sharp endothermic peak at
À1
and 34 °C, and the maximum of the band shifts to 1628 cm
(Figure 2, bottom). This change coincides with gel-to-sol transition
and attests a transformation of the H-bonds involving the amides.
The amide I band above 34 °C is deconvoluted into two peaks; the
3
3.0 °C (onset 32.2 °C), indicating a first-order transition, the
À1
À1
gel-to-sol transition, with an enthalpy of 42 kJ mol . Cooling the
same melted gel yields an exothermic transition, with a slight
hysteresis (onset 29.5 °C; max 29.0 °C) and with the same heat
absolute value, within the experimental uncertainty.
first one is the above-mentioned maximum at 1628 cm , and the
À1
second one is at 1647 cm . The lower frequency is consistent with
a stronger H-bond than in the gel state and is attributed to the amide
37
group H-bonded with water. The second peak has been observed
The nature of the interactions involved in gel formation was
explored by variable temperature FTIR, a technique that has proven
for other amide bearing compounds and can be interpreted as an
amide forming H-bonds with two water molecules.
In the CH stretching area (Figure 3), the symmetric and
38
34À36
to be powerful to investigate gels of self-assembled amides.
Spectra of a 1% gel in D O were measured at temperatures from
antisymmetric CH stretching bands (ν CH and ν CH ) are
2
2
s
2
as
2
1
2151
dx.doi.org/10.1021/la202162q |Langmuir 2011, 27, 12149–12155

Langmuir
ARTICLE
Scheme 3. Array of the Amphiphiles in the Wall of the Tubes
Figure 4. Top: Negative staining TEM of 1 (0.1% aqueous solution);
arrow, helical defects. Bottom: Cryo-TEM of 1 (0.05% aqueous
solution); arrow, self-assembled nanotubes; arrowhead, helical tapes.
Inset: Magnification of the tube showing the triple sheet structure of
the walls.
À1
found at 2848 and 2918 cm . These values are close to those
obtained for crystalline paraffins and indicate that the chains are
well ordered and in an all-trans conformation. It has been shown
by the same FTIR measurements that similar diacetylenic
molecules have no ordered alkyl chains in other nanostructures
3
9
40
such as micelles or thin films. As the temperature increases,
Figure 5. DSC of solution of polymerized 1 (0.5% in water, scan rate
0.1 °C/min). (A) First heating and cooling cycle. (B) Second cycle.
the frequencies of these bands exhibit a sharp increase from 2848
À1
À1
to 2854 cm and from 2915 to 2924 cm between 32 and
3 °C. These shifts characterize the presence of many gauche
3
defects in the alkyl chains. Hence, the first-order transition
observed from the DSC experiments is concomitant with an
evolution of the alkyl chains to a disordered state. These studies
show that the self-assembly of tubes involves the packing
between the alkyl chains and that H-bonds are crucial for the
self-assembly into nanotubes.
tapes are 59 ( 1 nm and 50 ( 1 nm, respectively. Sections of the
tubes and tapes are clearly visible (Figure 4, inset) and exhibit a
triple sheet structure, with a bright layer between two darker
layers. The thickness of the inner layer is 4 ( 1 nm, and that of
the outer layer is 3 ( 1 nm. The staining pattern ascertains the
composition of the layers: the brighter inner sheet corresponds
to a lower scattering density and is made of the alkyl parts,
whereas the darker outer layers correspond to the oligoethylen-
oxide chains. Therefore, the self-assembly is a bilayer with the
alkyl-diyne chains in the center (Scheme 3). The length of the
hydrophobic chain is 3.2 nm, and that of the hydrophilic chain is
2.3 nm. To account for the thickness of the inner sheet, the
hydrophobic part has to be tilted of 40° to match the measured
thickness of the inner layer. This hypothesis is reasonable
because, as shown below, the self-assembled amphiphile is
polymerizable and it has been proved that this polymerization
The structure of the aggregates was investigated by electron
microscopy. The self-assemblies in 0.1% wt solutions were first
studied by negative staining after their adsorption on carbon-
coated grids. The micrographs (Figure 4, top) show mixtures of
nanotubes and helical tapes that are several micrometers long.
Some tubes display helical defects (arrow) that show their
relationship with helical tapes. The apparent diameter of the
tubes is about 70 nm but is overestimated due to the adsorption
of the objects on the substrate and their subsequent deformation.
We also studied the self-assemblies by cryo-TEM. This technique
has two advantages: the observed contrast is due solely to the
chemical nature of the objects; and it preserves the objects in
their native environment by a deep freeze of the water to yield
amorphous ice. The micrographs (Figure 4, bottom) show the
same helical tapes and tubes as in the previous preparation, which
confirms that the objects observed by the previous technique are
not due to concentration variations (during adsorption and
blotting steps) or reorganization of the objects because of the
substrate. The external diameters of the tubes and the helical
41,42
requires a tilt angle of about 45°.
Moreover, many structures,
determined by crystallography, have shown similar tilt, due to the
42
interaction between the diyne groups.
Study of the Polymerized Tubes. When suspensions of 1 are
subjected to UV-irradiation, they turn into purple solutions,
which shows that polymerization takes place. The conversion is
not complete and stops at ca. 70%, as determined by SEC, which
shows that the self-assemblies may not offer ideal polymerization
conditions. Such limited conversion rates have been observed for
1
2152
dx.doi.org/10.1021/la202162q |Langmuir 2011, 27, 12149–12155

Langmuir
ARTICLE
Figure 6. VT-FTIR spectra of solutions of the polymer in D
from 27 to 80 °C: amide I and II region.
2
O (0.5%)
39
the polymerization of diacetylenic amphiphiles in micelles. We
also have noticed that X-rays induce polymerization of 1.
Solutions of freshly photopolymerized species at 5 mg/mL were
heated to 75 °C and cooled backed to room temperature while
their thermal behavior was measured by DSC.
Figure 7. Cryo-TEM of polymerized 1 (0.05% aqueous solution).
White arrow: Nanotubes. Large black arrow: Helical tapes. Arrowhead:
Ribbons.
During the heating phase, a broad endothermic peak is
observed with an onset at 27 °C and with a maximum at
À1
3
4.9 °C (Figure 5). The corresponding enthalpy, 23.0 kJ mol
,
is less than the one measured during the gel-to-sol transition of
the monomer solutions, which shows that less interactions are
involved in this transition than in the melting of the monomer
gels. Upon the first cooling step, one observes an exothermic
À1
peak at 26.1 °C, with the same area (22.6 kJ mol ) as the
endothermic one, but sharper. The second heating step yields a
different thermogram with an exothermic peak at 31.5 °C,
À1
corresponding to an enthalpy of 28.6 kJ mol , much sharper
than in the first cycle (onset: 29.4 °C). This difference between
the first and second cycles shows an irreversible transformation
of the solution upon heating. The solutions contain about 30% of
remaining monomers, and the observed peaks involve the
dissociation and reassociation of the monomer, but can account
only for part of the observed enthalpy. After the first heating
ramp, the thermograms show only single peaks. It proves that the
monomer and the polymer do not segregate and suggests that
part of the monomer associate with the polymer in the same phase
during the cooling step. In our studies, the polymerization of the
nanotubes does not provide any gain in thermal stability, which
can be due to the limited conversion rate. It has been shown for
other nanotubes self-assembled from diacetylenic amphiphiles
that the polymerization was incomplete in solution and did not
increase the thermal stability, whereas a second irradiation in dry
Figure 8. VT-UV spectra of solutions of the polymer (0.1% aqueous
solution) from 25 to 90 °C.
amides rather than with water. The transition can also corre-
spond to a reorganization of the aggregates without complete
dissolution of the polymer chain and to rearrangement of the
H-bonds between the amides.
The structure of the polymers was also investigated by electron
microscopy using a cryogenic preparation. The micrographs
(Figure 7) show tubes and helical tapes with the same dimen-
sions as the objects formed by the monomers before polymer-
ization. However, their proportion is much lower than before
polymerization. The samples exhibit a large proportion of
ribbons that are most often flat and are slightly twisted (Figure 7,
white arrowhead). The polymer remains self-assembled, but
tends to rearrange into flat ribbons.
12
state yielded tubes that do not show any thermal transition.
The IR spectra of the polymerized tube solutions were
measured during the first heating step; they show strong
The electronic spectrum of the purple polymer at room
temperature (Figure 8) exhibits three peaks at 502, 542, and
631 nm. The first two bands are the major ones, whereas the peak
at 631 nm is less intense. Polydiaecetylenes obtained in other
systems show a blue color corresponding to a much higher
similarity with those of the monomer. The stretching CH bands
2
show the same frequencies as for the monomer and the same
increase between 31 and 37 °C, which shows that the first-order
transition observed in DSC coincides with the apparition of
disorder in the alkyl chains. The amide I and II bands (Figure 6)
43,44
intensity of the peak at higher wavelength.
This blue phase
À1
are found at frequencies of 1632 and 1450 cm , respectively,
corresponds to higher effective conjugation length in the eneÀ
and show a steep decrease between 27 and 33 °C. Yet the amide I
transforms into a band that can be deconvoluted into two peaks
that differ from those of the pure monomer: a major one at
yne backbone with few nonplanar defects and to rod conforma-
45À48
tion in solution.
The purple color obtained for polymerized
1 suggests that the backbone of the self-assembled polymer is
subjected to constraints shortening the conjugation length lead-
ing to more defects. Alternatively, the purple color may reveal
that this constraint limits the degree of polymerization and leads
À1
À1
1
641 cm and a shoulder at 1626 cm . These values suggest
that most of the amide groups are H-bonded with each other and
a few of them with the solvent. The transition can correspond to a
dissociation of the self-assemblies, mainly into polymer chains
where the amide groups are close. Therefore, even after disas-
sembly, amide groups tend to have H-bonds with neighboring
48
to shorter degrees of polymerization.
When they are heated above 37 °C, the purple solutions of
polymer become red. The change is irreversible: the solution
1
2153
dx.doi.org/10.1021/la202162q |Langmuir 2011, 27, 12149–12155

Langmuir
ARTICLE
the dissociation or rearrangement of the chains. This reorganiza-
tion favors the relaxation of the polymer and its evolution to a
conformation with more nonplanar defects. Other systems where
interacting H-bonding side groups govern the electronic transi-
44,46
tions of polydiacetylenes have been described.
Upon cooling, the peak at 630 nm does not reappear. The
shifts and amplitudes of the peaks display an irreversible evolu-
tion (Figure 10). IR spectra show that the H-bonds between the
amide groups re-form (not shown) like in the freshly polymer-
ized solutions. However, because TEM shows that mainly flat
ribbons form after cooling, it is probable that in these ribbons, the
polymer is in a conformation that is close to its conformation in
the hot solution, above the transition.
Figure 9. Intensities (A) and wavelengths (B) of the three UV
absorption bands as a function of the temperature.
’
CONCLUSION
In the present study, we have prepared a new amphiphile
consisting of a single diacetylenic chain and an oligoethylen-
oxide hydrophilic linked by an amide group. We have shown
that it self-assembles in water to form nanotubes or helical
tapes. It represents a unique example of nonionic and non-
chiral diacetylenic able to form nanotubes. The walls of the
tubes are bilayers of the amphiphiles, where the alkyl chains
are ordered and the amide bonds are connected with each other
through H-bonds, as shown by FTIR. The self-assemblies can be
photopolymerized to yield purple solutions. A minor part of the
polymerized aggregates retain the tubular or helical shape, but
many form loosely twisted ribbons. The polymerized assemblies
undergo a thermochromic transition that can be detected by DSC
and corresponds to a reorganization of H-bond between the
amide groups.
Figure 10. Wavelength (b and O) and intensity (2 and 4) of the
second band. Plain symbols, heating phase; open symbols, cool-
ing phase.
remains red when it is cooled back. We measured the UV
spectra of a solution of freshly prepared polymer (1 mg/mL)
while it was heated from 25 to 90 °C and cooled back to room
temperature. When the temperature increases from 27 to 40 °C,
the first two intensities of the bands at 502 and 542 nm increase,
while the last band decreases and totally disappears above 45 °C
The comparison of a few analogues shows that the formation
of tubes depends on a subtle hydrophobicÀhydrophilic balance.
The use of an oligoethylenoxide allows one to tune precisely the
length of the amphiphile part, and new analogues will be further
studied.
(
3
Figure 9A). The inflection point for the curves in this range is at
4 °C, which is close both to the peak observed by DSC and to
the orderÀdisorder transition observed by FTIR. The frequen-
cies of the first two bands vary slightly in this temperature range,
while the shifts of the last band decrease from 631 to 617 nm. For
temperatures above 40 °C, the intensities bands at 502 and 542
nm further evolve, especially with a parallel and sharp increase
above 76 °C. The maxima of the peaks at 502 and 542
’ AUTHOR INFORMATION
Corresponding Author
*
Phone: +33 388 414 070. Fax: +33 388 414 099. E-mail:
mesini@ics-cnrs.unistra.fr.
(Figure 9B) show a very slight evolution. The peak at 631 nm
shifts toward the blue. The variation parallels that of the
intensity and is also centered around 34 °C. Solutions of the
polymer have been heated at 45 and 85 °C for 20 min, allowed
to cool at room temperature, and observed by electron
microscopy. The micrographs of the first solution showed
mainly the flat ribbons observed in Figure 7, and very rarely
helical tapes, which can be due to the reassembly of the
unpolymerized monomers. However, the drastic decrease of
the proportion of nanotubes confirms the irreversibility
observed in DSC and in UV. When heated at 85 °C and
cooled at room temperature, the solutions showed also
mainly flat ribbons but with smaller width.
’
ACKNOWLEDGMENT
The International Center for Frontier Research in Chemistry
(
Strasbourg) is acknowledged for financial support. A.P. thanks
ANR for financial support. We thank M. Seemann (Institut de
Chimie, Strasbourg) for the VT-UV spectra. We thank O. Gavat,
C. Foussat, and A. Rameau for the SEC experiments.
’
REFERENCES
(1) Yager, P.; Schoen, P. E. Mol. Cryst. Liq. Cryst. 1984, 106, 371–81.
(2) Shimizu, T.; Masuda, M.; Minamikawa, H. Chem. Rev. 2005,
1
05, 1401–1443.
3) Vauthey, S.; Santoso, S.; Gong, H.; Watson, N.; Zhang, S. Proc.
Natl. Acad. Sci. U.S.A. 2002, 99, 5355–5360.
4) Wilson-Kubalek, E. M.; Brown, R. E.; Celia, H.; Milligan, R. A.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 8040–8045.
5) Ringler, P.; Muller, W.; Ringsdorf, H.; Brisson, A. Europ. Chem.
J. 1997, 3, 620–625.
6) Huetz, P.; van Neuren, S.; Ringler, P.; Kremer, F.; van Breemen,
(
The thermochromic transition of the polymer observed during
heating corresponds to a reduction of the conjugation length of
the eneÀyne backbone mainly because of torsions of the back-
(
45,47
bone leading to nonplanar defects.
When the polymer is
(
formed, its conformation is imposed by the geometry of the self-
assembly, which is locked by the H-bonds between the amides
and the packing of the alkyl chains. When the polymer is heated,
the H-bonds between the amides are destroyed, which results in
(
J. F. L.; Wagenaar, A.; Engberts, J. B. F. N.; Fraaije, J. G. E. M.; Brisson, A.
Chem. Phys. Lipids 1997, 89, 15–30.
1
2154
dx.doi.org/10.1021/la202162q |Langmuir 2011, 27, 12149–12155

Langmuir
ARTICLE
(
7) Simon, F. X.; Khelfallah, N. S.; Schmutz, M.; Diaz, N.; Mesini,
(42) Enkelmann, V. Polydiacetylene. In Advances in Polymer Science;
Cantow, H. J., Ed.; Springer-Verlag: Berlin, 1984; Vol. 63, pp 91À136.
(43) Britt, D. W.; Hofmann, U. G.; M €o bius, D.; Hell, S. W. Langmuir
2001, 17, 3757–3765.
P. J. J. Am. Chem. Soc. 2007, 129, 3788–3789.
(
(
8) Stewart, S.; Liu, G. Angew. Chem., Int. Ed. 2000, 39, 340–344.
9) Kim, Y.; Mayer, M. F.; Zimmerman, S. C. Angew. Chem. 2003,
1
15, 1153–1158.
(44) Dautel, O. J.; Robitzer, M.; L ꢁe re-Porte, J.-P.; Serein-Spirau, F.;
Moreau, J. J. E. J. Am. Chem. Soc. 2006, 128, 16213–16223.
(45) Orchard, B. J.; Tripathy, S. K. Macromolecules 1986, 19, 1844–
1850.
(46) Tanaka, H.; Gomez, M. A.; Tonelli, A. E.; Thakur, M. Macro-
molecules 1989, 22, 1208–1215.
(47) Lim, K. C.; Heeger, A. J. J. Chem. Phys. 1985, 82, 522–530.
(48) Patel, G. N.; Miller, G. G. J. Macromol. Sci., Part B: Phys. 1981,
20, 111–131.
(
(
10) Huang, K.; Rzayev, J. J. Am. Chem. Soc. 2009, 131, 6880–6885.
11) Huang, K.; Canterbury, D. P.; Rzayev, J. Macromolecules 2010,
4
3, 6632–6638.
12) Lee, S. B.; Koepsel, R. R.; Russell, A. J. Nano Lett. 2005,
, 2202–2206.
13) Georger, J. H.; Singh, A.; Price, R. R.; Schnur, J. M.; Yager, P.;
Schoen, P. E. J. Am. Chem. Soc. 1987, 109, 6169–75.
14) Frankel, D. A.; O’Brien, D. F. J. Am. Chem. Soc. 1991, 113,
436–7437.
15) Fuhrhop, J. H.; Blumtritt, P.; Lehmann, C.; Luger, P. J. Am.
Chem. Soc. 1991, 113, 7437–7439.
16) Frankel, D. A.; O’Brien, D. F. J. Am. Chem. Soc. 1994, 116,
0057–69.
17) Lee, S. B.; Koepsel, R.; Stolz, D. B.; Warriner, H. E.; Russell, A. J.
J. Am. Chem. Soc. 2004, 126, 13400–13405.
18) Oda, R.; Artzner, F.; Laguerre, M.; Huc, I. J. Am. Chem. Soc.
008, 130, 14705–14712.
19) Valery, C.; Paternostre, M.; Robert, B.; Gulik-Krzywicki, T.;
(
5
(
(
7
(
(
1
(
(
2
(
Narayanan, T.; Dedieu, J. C.; Keller, G.; Torres, M. L.; Cherif-Cheikh,
R.; Calvo, P.; Artzner, F. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
10258–62.
(
(
20) Helfrich, W. J. Chem. Phys. 1986, 85, 1085–1087.
21) Seifert, U.; Shillcock, J.; Nelson, P. Phys. Rev. Lett. 1996,
77, 5237.
(
22) Selinger, J. V.; Schnur, J. M. Phys. Rev. Lett. 1993, 71, 4091.
(23) Selinger, J. V.; Spector, M. S.; Schnur, J. M. J. Phys. Chem. B
2
001, 105, 7157–7169.
24) Selinger, J. V.; MacKintosh, F. C.; Schnur, J. M. Phys. Rev. E
996, 53, 3804–18.
(
1
(25) Chen, C. M. Phys. Rev. E 1999, 59, 6192–6195.
(26) Chung, D. S.; Benedek, G. B.; Konikoff, F. M.; Donovan, J. M.
Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 11341–5.
(
(
27) Helfrich, W.; Prost, J. Phys. Rev. A 1988, 38, 3065–3068.
28) Singh, A.; Schoen, P. E.; Schnur, J. M. J. Chem. Soc., Chem.
Commun. 1988, 1222–1223.
29) Contal, E.; Klymchenko, A.; M ꢀe ly, Y.; Meunier, S.; Wagner, A.
Soft Matter 2010, 7, 1648–1650.
30) Nagahori, N.; Abe, M.; Nishimura, S.-I. Biochemistry 2008,
8, 583–594.
31) Tamiaki, H.; Ogawa, K.; Enomoto, K.; Taki, K.; Hotta, A.;
Toma, K. Tetrahedron 2010, 66, 1661–1666.
(
(
4
(
(
(
32) Terech, P.; Weiss, R. G. Chem. Rev. 1997, 97, 3133–3159.
33) Weiss, R. G.; Terech, P. Molecular Gels. Materials with Self-
Assembled Fibrillar Network; Springer: Dordrecht, The Netherlands, 2006.
34) Bouas-Laurent, H.; Desvergne, J.-P. Optical Spectroscopic
(
Methods as Tools to Investigate Gel Structures. In Molecular Gels;
Weiss, R. G., Terech, P., Eds.; Springer: Netherlands: 2006; pp 363À
429.
(
35) Palui, G.; Simon, F.-X.; Schmutz, M.; M ꢀe sini, P. J.; Banerjee, A.
Tetrahedron 2008, 64, 175–185.
36) Simon, F.-X.; Nguyen, T.-T.-T.; Schmutz, M.; Decher, G.;
Nicoud, J.-F.; Mesini, P. J. New J. Chem. 2009, 33, 2028–2033.
37) Eaton, G.; Symons, M. C. R.; Rastogi, P. P. J. Chem. Soc.,
Faraday Trans. 1 1989, 85, 3257–3271.
38) Maeda, Y.; Sakamoto, J.; Wang, S.-y.; Mizuno, Y. J. Phys. Chem.
B 2009, 113, 12456–12461.
39) Perino, A.; Klymchenko, A.; Morere, A.; Contal, E.; Rameau,
A.; Guenet, J.-M.; M ꢀe ly, Y.; Wagner, A. Macromol. Chem. Phys. 2011,
12, 111–117.
40) Gatebe, E.; Herron, H.; Zakeri, R.; Ramiah Rajasekaran, P.;
Aouadi, S.; Kohli, P. Langmuir 2008, 24, 11947–11954.
41) Sarkar, A.; Okada, S.; Matsuzawa, H.; Matsuda, H.; Nakanishi,
H. J. Mater. Chem. 2000, 10, 819–828.
(
(
(
(
2
(
(
1
2155
dx.doi.org/10.1021/la202162q |Langmuir 2011, 27, 12149–12155