Pyrene derived functionalized low molecular weight organic gelators and gels

Article abstract of DOI:10.1039/b801708e

Pyrene derived binary functionalized low molecular weight organic gelators (FLMOGs) and gels thereof in selected organic solvents were synthesized and characterized. The functionality refers to a functional group that does not take part in formation of the supramolecular gel network, but remains free and available for other purposes, such as to bind nanoparticles or other molecules into the gel structure. Functional groups were observed to disturb gel formation strongly, if they interact with each other within the same supramolecule due to the formation of competitive structures. Preventing such interactions restored the original gel properties. A gel with weaker supramolecular bonding than the binding between the functional groups was successfully made by separating the functional groups by distance. The π-π-interaction was found to be of negligible significance to the supramolecular binding energy, but probably essential to align the molecules to a one-dimensional chain and bring them into the range of van der Waals forces mainly responsible for the binding in this system. Solvent was observed to increase the binding energy of the supramolecule. All molecules were characterized by spectroscopic techniques and elemental analysis. Selected gels were characterized with rheometry, scanning electron microscopy, UV- and fluorescence spectroscopy. Gelation kinetics and hysteresis were measured by UV-spectroscopy and a fast gelation process was observed for all the gelators studied. The melting enthalpies were measured by DSC and calculated theoretically by PM3 level of theory. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/b801708e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Pyrene derived functionalized low molecular weight organic gelators and
gelsw
a
bc
a
a
a
Arno Hahma,* Shreedhar Bhat, Kimmo Leivo, Juha Linnanto, Manu Lahtinen
b
and Kari Rissanen
Received (in Montpellier, France) 30th January 2008, Accepted 13th March 2008
First published as an Advance Article on the web 23rd April 2008
DOI: 10.1039/b801708e
Pyrene derived binary functionalized low molecular weight organic gelators (FLMOGs) and gels
thereof in selected organic solvents were synthesized and characterized. The functionality refers to
a functional group that does not take part in formation of the supramolecular gel network, but
remains free and available for other purposes, such as to bind nanoparticles or other molecules
into the gel structure.
Functional groups were observed to disturb gel formation strongly, if they interact with each other
within the same supramolecule due to the formation of competitive structures. Preventing such
interactions restored the original gel properties. A gel with weaker supramolecular bonding than the
binding between the functional groups was successfully made by separating the functional groups by
distance. The p–p-interaction was found to be of negligible significance to the supramolecular
binding energy, but probably essential to align the molecules to a one-dimensional chain and bring
them into the range of van der Waals forces mainly responsible for the binding in this system. Solvent
was observed to increase the binding energy of the supramolecule.
All molecules were characterized by spectroscopic techniques and elemental analysis. Selected gels
were characterized with rheometry, scanning electron microscopy, UV- and fluorescence
spectroscopy. Gelation kinetics and hysteresis were measured by UV-spectroscopy and a fast
gelation process was observed for all the gelators studied. The melting enthalpies were measured by
DSC and calculated theoretically by PM3 level of theory.
research. Many of the reported materials are derived from
4
complicated molecules that are laborious to synthesize and
Introduction
A low molecular weight organogel is an organic solvent with a
considerable increase in viscosity by means of formation of a
expensive. In addition, many compounds are effective gelators,
but do not have any functional groups to be used as binding
sites to other materials, such as nanoparticles. In this context,
pyrene based LMOGs and functionalized LMOGs are simple
compounds, which have proved extremely effective in gelling
3
D-network of entangled supramolecular fibers, which are
formed by the self-arrangement of small gelator molecules
instead of a network of polymer chains. Aggregation of low
molecular weight organic gelators (LMOGs) is usually driven
by specific noncovalent intermolecular forces, such as p–p
interaction, hydrogen-bond formation, metal coordination,
solvo/hydrophobic, dipole–dipole or van der Waals interac-
tions. All such interactions lead to a great diversity of possible
5
many organic solvents in less than 1% w/w concentrations as
such or as mixtures. However, pyrene derived gelators inves-
tigated so far possess no functionality except van der Waals
interactions to bind materials into the gel network or to
control the gel structure.
1
organic gelators. A large number of different organic gelators
In this paper we present a new class of very simple functio-
nalized low molecular weight gelators (FLMOGs) derived
from pyrene, which have the ability to effectively bind to polar
surfaces and incorporate for instance metallic particles as an
integral part of the gel structure yet forming a gel. The
functional groups of choice are polar groups that readily bind
to metal and metal oxide surfaces or react with metals to form
covalent or ionic compounds. Functional groups such as
carboxylic acid, alcohol, hydrazide and amine are especially
suitable for this task in diminishing order of acidity. In
addition, these functional groups can readily be used to bind
other molecules to the gelator structure.
have been discovered serendipitously, but recently there
has been a tendency to actively design new organic gelator
2
molecules.
The ease of formation and the reversibility of gelation by
LMOGs are attractive strategies for synthesizing novel mate-
3
rials. Material design is a highly competitive area of current
a
Department of Chemistry, University of Jyva¨skyla¨, P.O. Box 35,
FIN-40014 JYU, Finland. E-mail: arno@jyu.fi;
Fax: +358 50 8428 5243; Tel: +358 50 428 5243
Nanoscience Center, Department of Chemistry, University of
b
Jyva¨skyla¨, P.O. Box 35, FIN-40014 JYU, Finland
Department of Organic Chemistry, Indian Institute of Science,
c
The application for the gelators in this study is to create gels
that can bind particulate material, such as metal powders and
metallic nanoparticles, into the gel network resulting in a
stable gel, where the particles cannot sediment even under
Bangalore, 560012, India
w Electronic supplementary information (ESI) available: Experimental
and characterization data, and photographs of the gels. See DOI:
10.1039/b801708e
1
438 | New J. Chem., 2008, 32, 1438–1448
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
severe acceleration. Unfortunately, all gelators based on any
kind of strong polar interactions contain functional groups,
which form the intermolecular bonds to form a gel. Those
functional groups are usually reserved for forming the gel
network and hence are not available to bind to other materials
into the gel. Mixing adsorbent materials into such gels im-
mediately destroys the gel if the surface of the extra material
binds to the same functional groups that would otherwise
form a gel. In order to use the gelator to bind other materials
than the solvent it is essential to design a system that forms the
gel by such interactions that do not bind readily to adsorbent
surfaces, which are typically very polar compared to the
organic solvent to be gelled. On the other hand, the functional
groups attached to the gelator molecules have to bind strongly
onto the surfaces in order to displace any gelator adsorbed
from a wrong location or block the surfaces before the gelating
site is adsorbed on them. This requirement excludes single
component LMOGs, since there is not much control available
over which functional group of the molecule binds to the
the carboxylic based functional group was transformed after
the first attachment of the side chain to the corresponding
alcohol, hydrazide and amine derivatives. All the new com-
pounds were identified by spectroscopic techniques and ele-
mental analysis.zy The synthesis is simple in principle, but the
purification of the compounds containing functional groups
may be difficult and requires chromatographic techniques in
many cases, while non-functional gelators can be readily
recrystallized from hexane after filtration through a short
column of silica.
Gels and gelation properties
All gel samples were prepared by adding equimolar amounts
of gelator and 2,4,7-trinitrofluorenone (TNF, 12, Scheme 1) to
a solvent and heating, if necessary, up to the boiling point of
the solvent, until at least the gelator, and preferably both
compounds, were fully dissolved. The solution was allowed to
cool down to room temperature, whereby the gel formed, if
TNF was soluble enough (Fig. 1).
5
adsorbent surface first. The classical pyrene based gelators
Gelation tests indicated that compounds 1–11 do not form
gels alone in any of the organic solvents tested so far. How-
ever, all of these compounds do gel many organic solvents in
the presence of TNF, if the gelator and TNF are sufficiently
soluble. The results from the gelation experiments at 15 mmol
were found to be more tolerant to adsorbent materials, showed
the best gelation properties and gels thereof were found to
have excellent rheological properties and stability for applica-
tion as fuel gelators. These materials were therefore selected as
the base system to be extended to a group of functionalized
gelators.
ꢁ1
ꢁ1
kg concentration (+TNF 15 mmol kg ) have been listed in
Table 1 and shown in the ESI.w The gels are optically
transparent but dark red or brown at low gelator and charge
transfer (CT) agent concentrations (at and below 2% w/w). At
higher concentration (above 3% w/w) the gels are translucent
and contain dispersed CT agent particles inside the gel. When
heated up to and above the melting point, the colour fades
gradually and the solution becomes clear and pale yellow or
colourless showing the breakup of the CT complex. The dark
colour returns gradually upon cooling.
Results and discussion
Gelator syntheses
Compounds 1–11 (Scheme 1) were synthesized starting from
5a
pyrene by following a modified known route through Friedel–
Crafts acylation using TiCl catalyst, which furnished an o-
4
keto acid. The keto group was subsequently reduced by the
Wolff–Kishner method into the corresponding methylene
group and the acid group was then transformed into the
appropriate functional groups. The reactants were chosen to
have the desired functionality in advance (carboxylic, ester) or
Compounds 1–11 formed the strongest gels in higher alco-
hols. For obtaining strong gels alcohols above 1-butanol have
been found to be optimal. Dissolution in lower chain alcohols
(
oC ) resulted in precipitation of the red CT complex in many
5
cases. Compounds 1–3 gel even alkane solvents while com-
pounds 4–11 are not soluble enough. This is partly due to the
poor solubility of the CT agent TNF (12) in aliphatic hydro-
carbons. Interestingly, compounds 1 and 2 were found to
produce stable gels in petroleum ether in the presence of only
a trace of TNF.y These gels are stable, but lightly coloured and
weaker than the 1 : 1 gelator TNF gels. This supports the
observation that the CT mechanism is not the main source of
binding energy, but nevertheless, without any CT agent no gel
is formed and the CT mechanism is needed at least as an
initiator.
The gelation properties of compounds 1–11 in 1-decanol
have been studied in more detail. In 1-decanol, these com-
pounds form optically transparent CT gels that are deep red in
colour. The visually observed gel transparencyw against a
white illuminated background follows an order 9 4 6 4 7
5
a
8b
8a
z Known compounds: 2, but literature data incomplete: 4, 5–6.
y CT-gels can also be obtained at lower ratios of CT-agent and pyrene
analogs. We have found that as small as 5% doping is sufficient to gel
petroleum ether (boiling range 80–100 1C) using compounds 1–3.
Scheme 1 Functional, pyrene based organic gelators with TNF.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1438–1448 | 1439

View Article Online
Fig. 1 Compound 5 in 1-octanol as gel at 25 1C (above) and as sol at
ꢁ1
ꢁ1
gelator + 30 mmol kg TNF
80 1C (below) at 30 mmol kg
concentration. The sol is lightly colored, the gel is dark red. The black
shadow at the bottom of in the sol is a stirrer magnet.
4 5 4 3 4 2 4 1, which agrees well with the gelator polarity.
However, Tgel followed the reverse order, transparent gels
melting at lower temperatures than the opaque gels.
Fig. 2 DSC of gels made of compounds 3, 5 and 7 in 1-decanol, 15
ꢁ1
ꢁ1
ꢁ1
The gels proved extremely stable to the effects of accelera-
tion. Less than 1% of free solvent was released by centrifuging
mmol kg gelator, 15 mmol kg TNF, heating rate: 20 1C min
.
ꢁ
1
a 5 ml gel sample at 15 mmol kg concentration in 1-octanol
ꢁ1
very low: +13 kJ mol for compound 7 with 1 : 1 TNF. This
under 1000 g acceleration for one hour.
is much less than the melting enthalpy of the pure gelator.
Spectroscopic properties and gelation kinetics
Thermoanalysis
Gelator molecules 1–11 are highly UV and fluorescence active.
All these compounds show characteristic UV/visible absorp-
tion bands of pyrene. A very strong absorption band starting
at 560 nm and extending to 360 nm was observed (Fig. 3) when
the charge transfer agent TNF was added to sufficiently
concentrated gelator solutions of 1–11 in 1-decanol. The
absorption is very broad and all wavelengths at and below
DSC experiments were performed to study the melting beha-
viour of the CT complexes of 3, 5 and 7 in 1-decanol at 15
1
ꢁ
mmol kg concentration to determine the melting enthalpies
Fig. 2). The DSC profiles show several exothermic transitions
(
along with the endothermic melting transition, showing that
many structural changes take place in the material during
heating. All these CT complexes melt in a relatively narrow
temperature range (ranging from 65 to 75 1C) depending on
the functional groups present on the pyrene system. At this
stage we do not clearly understand the reason for the exother-
mic phase transitions observed by DSC and the melting
enthalpies of complexes of compounds 3 and 4 could not be
determined due to unexplained endothermic transitions, while
compound 7 showed a clear endothermic peak. It should be
noted, however, that these very small enthalpies are at the
limits of the instrument, even though the measured curves
were repeatable. The curves shown are an average of three
runs. The enthalpy changes measured for these transitions are
5
50 nm are extremely strongly absorbed in the gel while the sol
is at least partially transparent over the entire range. The
broad absorption band is responsible for the deep red colour
of the gels resulting from the CT complex. Note that the gel
UV/Vis spectrum exceeds the device range due to the high
ꢁ1
concentration of the material in the gel (8 mmol kg ) from
the point of view of the UV/Vis measurement. The samples
a
Table 1 Observations from gelation tests with TNF
Gelator Hexane CHCl
3
6 6
C H Toluene Octanol Decanol Dodecanol
1
2
3
4
5
5
7
8
9
1
1
g
g
g
i
i
i
i
i
i
i
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
sg
g
g
sg
sg
sg
wg
sg
sg
sg
sg
wg
sg
sg
sg
g
wg
wg
wg
wg
wg
wg
wg
wg
sg
wg
wg
g
wg
wg
sg
g
wg
0
1
i
Fig. 3 UV/Vis spectrum of gel at 25 1C and sol at 80 1C made from
compound 5 in 1-decanol at 8 mmol kg gelator and 8 mmol kg
a
ꢁ1
ꢁ1
1
5 mmol kg gelator, 15 mmol kg TNF; sg = strong gel, g = gel,
ꢁ1
ꢁ1
wg = weak gel, s = sol, i = solubility too small, otherwise would
form a gel.
TNF concentration. The gel spectrum is over scale, but the gel could not
be diluted more because it would break down at lower concentrations.
1
440 | New J. Chem., 2008, 32, 1438–1448
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
Fig. 5 Gelation hysteresis of compound 6 in 1-decanol by UV/Vis.
ꢁ1
Fig. 4 Gelation kinetics of compounds 3, 5 and 7 by UV/Vis in 1-decanol
ꢁ1
Heating and cooling rates were 10 1C min
.
ꢁ1
at 15 mmol kg gelator and 15 mmol kg TNF concentration.
4
70–500 nm with excitation wavelength 336 nm. As the
could not be diluted more, since no gel forms at concentrations
concentration increases the fluorescence properties of the
molecule change and processes such as the internal filter effect
and scattering may affect the fluorescence.
ꢁ1
below 8 mmol kg and no change in the spectrum is observed
at lower concentrations than that. The kinetic measurements
were carried out at 550 nm, i.e. at the edge of the absorption
band to avoid exceeding the spectrometer range even in a gel
state. Absorption and fluorescence maxima for the gel were at
Although the UV/Vis and fluorescence experiments at high-
er concentrations are not recommended in general, they can
give a general idea about the material properties. Therefore,
the emissive properties of both sols and gels of compounds
1–11 have been studied, which showed a strong fluorescence
(lmax at 480 nm) resulting from the excited state complex with
TNF (Fig. 7). The fluorescence intensity of the gels is also
temperature dependent. The gel melts completely before 75 1C
if determined by fluorescence. This is in accordance with visual
Tgel values of the gels and with the results from the rheometry
and UV/Vis experiments.
4
80 nm, but this wavelength could not be used in UV/Vis or
kinetic measurements due to excessive absorption.
The gelation kinetics for gels of 3, 5–9 (Fig. 4) were studied by
monitoring the charge transfer band. During the cooling pro-
cess of a hot sol the initial change in the absorbance at 550 nm
is very low for few seconds but is enhanced rapidly with
cooling and reaches a maximum steady value at the gelling
point, since the CT band is very sensitive to temperature. A
kinetic plot indicated that the compounds 3, 6 and 7 form CT
gels twice as fast (2 min to steady absorption) as 5 and 9 (5
min) containing hydroxyl groups at the chain end. The effect
of the finite cooling rate is negligible on this time scale, since
cooling will only take a few seconds in the setup used. Never-
theless, all these compounds form gels very fast and the
process is fully reversible unlike with many polymeric gelators,
where re-equilibration can take days or weeks.
Rheological properties
Rheologically, the gels behave almost ideally as a function of
temperature. First the storage modulus decays only slowly, but
We carried out both melting and cooling studies of the gels
of 6 with the help of the CT band. The hysteresis curve showed
that the deaggregation process commences already at 35 1C
upon heating while the aggregation starts at 55 1C during the
cooling process (Fig. 5). A rather small hysteresis effect was
observed despite the relatively fast heating and cooling rate (10
ꢁ
1
1
hysteresis results are in accordance with rheology measure-
C min ) also showing the gels form and equilibrate fast. The
ments, which show the onset of the reduction of the storage
0
modulus G at around 40 1C. An example is presented for the
compounds 1 and 5 in Fig. 6.
The pyrene ring fluoresces strongly and shows five emission
bands at low concentrations. The third and first bands are
highly sensitive to the dielectric constant of the solvent med-
ium. At higher concentrations pyrene is expected to show an
excimer band at 480 nm due to the formation of an excited
state dimer. A broad fluorescence band was observed at
Fig. 6 Rheological properties of gels of compounds 1 and 5 in
0
1
-octanol as a function of temperature. G is the storage modulus,
0
0
G is the loss modulus and d is the phase angle of the stress vs. strain at
1 Hz frequency.
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1438–1448 | 1441

View Article Online
Fig. 8 Xerogel of compound 7 from 1-decanol has both fibrillar and
Fig. 7 Fluorescence spectra of sol at 80 1C and gel at 25 1C of
compound 5 in 1-decanol at 15 mmol kg gelator and 15 mmol kg
ꢁ1
ꢁ1
granular structure.
TNF. The excitation wavelength was 336 nm.
drops sharply close to the melting point of the gel. The melting
much better solubility of TNB. Complexing with 2,4,5,7-
tetranitrofluorenone (TeNF) formed dark green, almost black
gels,w which precipitated on standing within a few days
showing the complex was too strong. Aromatic hydrocarbons
could not be gelled with any combination, but they dissolved
the compounds into a sol and prevented gelation by complex-
ing to the CT agent themselves or by dissolving the resulting
CT complex too well. Pyrene itself also forms a CT complex in
organic solvents, but the gelation process requires a long alkyl
point is defined as the point where the phase angle reaches 451
(
Fig. 6) and the material becomes more viscous than elastic.
The functionalized gelator 5 is considerably weaker than the
non-functionalized analog 1, however, as a mixture the gel
shows only a slight reduction of the storage modulus, prob-
ably due to different polarity and length of the side chains and
not due to functional group interactions. At this concentra-
tion, every sixth molecule in the gelator chain has a functional
group on average spacing the functional groups so far from
each other that they no longer interact with adjacent groups
within the same supramolecular chain preventing distortion of
the supramolecular structure.
2
chain (47 CH ) on the pyrene moiety. This alkyl chain
provides rigidity to the 1D-aggregates and binds solvent
molecules to the backbone of the gelator chain and increases
solubility in aliphatic solvents.
Xerogel structure
Supramolecular structure and properties
The xerogels were imaged with high resolution scanning
electron microscopy (HR-SEM) to study the supramolecular
structure of the gel. As an example, an HR-SEM image of 7 is
presented in Fig. 8. The image showed an interconnected fiber
network with evenly sized spherical clusters. The fibers were
uniformly 50 nm and the clusters 1.5 mm in diameter. All gels
showed an at least fibrous structure, but the diameter of the
fibres varied from 20 nm to about 300 nm with different
molecules: the longer the side chain in the pyrene nucleus,
the thicker the fibres in the xerogel. A dual structure was
observed in all gels from gelators with functional groups at the
end of the side chain, e.g. in gels of 5–7 and 9.
The charge transfer complex enthalpy of formation of 7 was
6
calculated with PM3 level of theory using Gaussian 03 in the
gas phase at 0 K. Two possible structures with almost exactly
the same energy of formation were found: a linear chain (DH
f
ꢁ1
=
ꢁ7.4 kJ mol ) and a spherical aggregate (DH
f
= ꢁ7.6 kJ
ꢁ1
mol ), in which the p–p-interactions are broken, but hydro-
2
gen bonding between –CQO and –NHNH is established
twisting the chain structure into a spheroid or helix (Fig. 9).
Calculations on larger than dimeric aggregates have been
unsuccessful so far due to the very flat potential energy surface
and due to the very large number of local minima in these
systems. A similar dual structure is observed at a much larger
scale in the SEM image (Fig. 8). Unfortunately, crystal
structures of the CT complexes could not be determined so
far and more rigorous evidence of the theoretically proposed
dual structure could not be obtained.
Gelation process
The exact mechanism for the gelation of pyrene analogs in the
presence of TNF is not completely understood.y TNF is used
as a coupling reagent in the gelation and is presumed to help
the one dimensional growth of the substituted pyrene mole-
cules by forming a charge transfer complex with the pyrene
nucleus. 1,3,5-Trinitrobenzene (TNB) also proved effective
when used in a 2 : 1 molar ratio with the gelator. TNB-gels
were opaque and bright orange in colour. With TNB, the
aliphatic hydrocarbons can be more readily gelled due to the
The measured melting enthalpy of the CT complex of 7 by
ꢁ
1
DSC is +13 kJ mol . The difference from the calculated
values at 0 K in the gas phase is surprisingly small, indicating
that the gelator system has a low entropy change at its melting
point and retains some of its ordered structure even in a
molten state, or that adding solvent to the CT complex
increases its bonding energy or both of the above.
1
442 | New J. Chem., 2008, 32, 1438–1448
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
Charge transfer
The sum of atom charges on TNF by MKS method was
found to be ꢁ0.0055 electron charges in the linear chain and
zero in the twisted structure showing the CT mechanism is
weak and present only for the linear chain structure.
7
Similar results have also been observed previously. Less than
5
% of the bonding energy in the linear chain structure is due
to CT interaction, the rest being due to van der Waals
bonding.
Consequently, the CT mechanism alone cannot be
responsible for gelator chain formation nor the melting en-
thalpy observed. It may, however, be necessary to first align
the molecules by p–p-interactions into the correct positions,
where short range multipolar field coulombic interactions
between individual atoms in the gelator and TNF can take
over. Individual atoms in TNF have charges as much as ꢂ0.5
and in the gelator molecule up to ꢂ0.3 electron charges.
Such charges create much larger bonding energies
than 0.0055 total electron charge for the entire molecule
despite the large surface area. In order for ideal atom-to-atom
van der Waals bonding to occur, the atoms have to be
aligned in such a way that a positively charged atom always
matches a negatively charged counterpart in the adjacent
molecule. Even a weakly electronegative substituent on
the pyrene nucleus inhibits gelation, e.g. supramolecule
formation, since it swaps the order of negative and positive
atoms in the pyrene nucleus and the match with TNF at the
optimal CT complex geometry is lost in addition to reduction
of the p-electron density.
Fig. 9 Two possible minimum energy structures of compound 7 with
TNF: linear p–p-configuration (above) and hydrogen bonded helical
or spherical aggregate (below).
Solvent effects
The solvent was observed to increase the bonding energy of
the complex as expected. The calculations were carried out
without solvent molecules present while DSC was carried out
with solvent producing approximately twice larger enthalpy
than the calculations.
Conclusions
We have successfully implemented a group of pyrene based
functionalized low molecular weight organic gelators
(
FLMOGs) and shown that it is possible to make a gel with
Hydrogen bonding
free, unhindered functional groups available as binding sites to
other molecules and materials. We have implemented a gel
that forms with a weaker bonding mechanism than the gelator
functional groups can bond with. We have also shown that
such gelator systems are not allowed to contain structurally
competing intermolecular binding sites within a single supra-
molecule or the resulting gels become weak and behave
unpredictably (Fig. 6). In addition, the first atom of the side
chain attached to the pyrene nucleus may not have substitu-
ents other than hydrogen or the formation of gel will be
inhibited due to steric effects, even though such substituents
were favourable to the formation of the CT complex (electron
donating groups) (Table 1). If the pyrene substituent is
electron withdrawing it causes unfavourable charge distribu-
tion in the pyrene nucleus at optimal CT complex geometry
preventing gelation.
The alternative twisted structure bonding energy originates
from hydrogen bonding only. As this configuration has
only one binding site, it cannot form a gel network, but
works to break such ordering, if present in the gel chain.
The side chains in the supramolecule are located so far
from each other (Fig. 9) that hydrogen bonding at the other
end of the side chain twists the pyrenes apart despite the
flexibility available in the side chain. Thus, hydrogen bonding
between the side chains should be prevented by keeping the
functional groups apart. This is necessary also, since the
functional groups are not available for binding other materi-
als, if they coordinate to each other. The solution to this is to
add enough non-functionalized pyrene based gelator or even
neat pyrene to the mixture to separate the hydrogen bonding
groups to at least two pyrene–TNF-units apart preventing
mutual interactions. This approach proved very effective
as seen in Fig. 6 and a LMOG based gel with large amounts
of free and sterically unhindered functional groups could
be made.
The adverse effects of the gelator functionality can be
controlled by mixing non-functionalized and functional gela-
tors or to a limited amount even pyrene to effect a supramo-
lecular copolymer (Fig. 6) and also by crosslinking the gel at
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1438–1448 | 1443

View Article Online
supramolecular level with compound 11 in analogy to polymer
chemistry separating functional groups from each other within
an individual gelator supramolecule. We have also shown that
charge transfer mechanism plays only a minor role as far as
the total agglomeration energy of the gelator is concerned. It
may, however, be essential to have p–p- and CT complexation
to align the molecules correctly for the van der Waals forces to
take over the main binding role. The solvent increases the
supramolecular binding energy compared to the correspond-
ing vacuum structure, as expected.
uncertainty for measured temperatures was less than 0.8 1C for
all measurements.
Rheology
Rheological properties were measured with Haake Rheos-
tress1 rheometer using a cone–plate configuration. The sample
was spread over the measurement disc thermostated at 10 1C.
The measurement head (35 mm diameter 21 titanium cone,
type C35/2 Ti) was lowered on to the sample slowly by the
control software squeezing excess gel out of the gap. Oscilla-
tory mode measurement was started immediately the measure-
ment head reached its position, since lowering the head took
about one minute allowing the gel to cool down in time.
Constant frequency of 1 Hz and constant stress of 1 Pa
were used while sweeping the temperature from 10 1C up to
80 1C within one hour, e.g. at a rate of 1.17 1C per minute. The
stress level was chosen according to prior stress sweep
measurements to a linear range of the modulus vs. stress
dependency within the entire temperature range of interest.
A PTFE cover was placed over the measurement head to
prevent condensation of moisture at low temperatures and
evaporation of the solvent at the high temperature end. Three
parallel measurements were carried out and the average of the
three was used as the final result. Stray points were removed
from the data prior to averaging, if the said points were clearly
erratic.
Experimental
Spectral characterizations
Compounds. NMR spectra were recorded with Bruker
Avance DPX 250 (250 MHz) or Bruker Avance DRX 500
1
3
(
500 MHz). The solvent peak in C-NMR was adjusted to
7.23 ppm for CDCl and to 39.51 ppm for DMSO-d at 30
C. J values are given in Hz. Mass spectra were measured with
Micromass LCT (electrospray MS) or with VG AutoSpec
electron impact MS). The reference material used for high
7
3
6
1
(
resolution mass spectroscopy (HRMS) was leucine enkephalin
acetate hydrate for ESI-MS and perfluorokerosene for EI-MS.
Gels
The gel transparency and turbidity were estimated visually
against a background lighted table. A photograph of this
experiment can be found in the ESI.w
The storage and loss modulus and the phase angle were
automatically calculated by the Haake Rheowin v3.21 con-
trolling software according to calibrations of the instrument.
The temperature was swept upwards in order to have the gel as
intact as possible at the beginning of the measurement and
avoid melting it prematurely. Additionally, the samples were
stored overnight at room temperature before measuring them
to ensure the gel had fully cured and an equilibrium state had
been reached.
UV/Vis and fluorescence spectra were recorded in a 10 ꢃ 10
mm thermostated quartz cuvette at 25 1C using Shimadzu UV-
2
100 UV/Vis spectrophotometer. The hysteresis studies were
carried out with the same instrument with a heating and
cooling rate of 10 1C per minute.
The kinetic experiments by UV-spectroscopy were carried
out by transferring a hot sol (at 80 1C) consisting of a gelator
and TNF in a 2 : 1 ratio into a UV cuvette with path length 1
mm maintained at 25 1C. Less TNF than a 1 : 1 ratio was used
to prevent precipitation of any substance and to reduce the
already very high absorbance level. Absorbance at 550 nm was
monitored against time. Fluorescence spectra were measured
using Perkin-Elmer LS-50B luminescence spectrometer with a
Microscopy
The gels were imaged with RAITH E-Line high resolution
field emission scanning electron microscope equipped with
Schottky tungsten field effect emitter. The samples were pre-
pared on pieces of silicon wafers by melting 1 to 3 mg of gel on
the wafer, cooling and drying at room temperature in an air
stream for 3 days. The samples were then gold coated with
sputtering with Jeol Fine Coat Ion Sputter JFC-1100. A
sputtering time of 30 sec up to 1 min was used with 1 kV
voltage, 10 mA of current and 5 mbar air as the gas
medium producing a gold coating less than 1 nm thick on
average. The samples were imaged using an aperture size of
10 mm path length quartz cuvette at 25 1C using 336 nm
excitation wavelength.
Thermoanalysis
Thermal transitions for the gels were determined on power
compensation type Perkin-Elmer Pyris Diamond DSC. The
measurements were carried out under nitrogen atmosphere
ꢁ
1
(
flow rate 50 ml min ) using 50 ml sealed aluminium sample
1
0 or 30 mm, collector current 100 mA, spot size 20 nm,
acceleration voltage 6 kV, working distance 6 mm and using
pans. The gel samples were melted in a test tube by heating it
with a hot air gun to 100–130 1C and a droplet of the molten
gel was then quickly placed on the sample pan, which was
allowed to cool to room temperature before sealing the pan.
The temperature calibration of DSC was made using two
standard materials (n-decane, indium) and energy calibration
ꢁ7
ultra high vacuum (o4 ꢃ 10
Torr) with an Everhart
Thornley in-lens detector.
Chemicals
ꢁ1
by an indium standard (28.45 J g ). Each sample was heated
ꢁ
All chemicals and solvents were used as received from sealed
containers. The list of chemicals and their sources are given in
the ESI.w
1
from 25 to 100 1C with a heating rate of 20 1C min . Sample
weights of 10–20 mg were used on the measurements. The
1
444 | New J. Chem., 2008, 32, 1438–1448
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
ꢁ
1
Syntheses
C
28
H
34: C, 90.75; H, 9.25%); nmax(thin film)/cm 1465 and
40; d /ppm (300 MHz; CDCl ) 0.88 (3 H, t, J 6.8), 1.26–1.37
16 H, m), 1.37 (2 H, q, J 7.5 Hz), 1.49 (2 H, m), 3.33 (2 H, t, J
.7), 7.87 (1 H, d, J 7.8), 8.02 (2 H, m), 8.17 (5 H, m) and 8.29
1 H, d, J 9.3 Hz); d /ppm (75 MHz; CDCl ) 22.07, 29.36,
8
H
3
1
-(Pyren-1-yl)decan-1-one (1a). Decanoic acid (86 g, 0.5
(
mol) was first treated with an excess of thionyl chloride (300
7
2
ml, 4.2 mol) and refluxed for 3–4 h. Excess of SOCl was
(
C
3
removed by distillation under reduced pressure. The decanoyl
chloride (95 g, 0.5 mol) was mixed with pyrene (101 g, 0.5 mol)
and 1,2-dichloroethane (DCE, 1200 ml). To this mixture was
2
1
1
9.61, 29.65, 29.68, 29.85, 31.93, 31.98, 33.64, 123.55, 124.59,
24.76, 124.77, 125.08, 125.74, 126.46, 127.07, 127.25, 127.54,
1
29.68, 130.96, 131.47 and 137.39; LRMS: m/z 384 (M ).
added TiCl (150 g, 0.75 mol) dropwise with stirring, in an ice
4
bath at below 5 1C, during the addition. The brown complex
formed was stirred for a further 16 h at RT. The brown
complex was cleaved by the addition of water and DCE was
distilled off. The product was extracted into chloroform in a
Soxhlet (1500 mL) and washed with water, saturated sodium
bicarbonate (2 ꢃ 100 mL) and finally with water. The chloro-
1
-(Pyren-1-yl)octadecan-1-one (3a). Stearic acid (4 g, 14.12
mmol) was first treated with excess of thionyl chloride (2.5 g,
1.2 mmol) and refluxed for 3–4 h. Excess of SOCl was
2
2
removed by distillation. The stearoyl chloride (3.6 g, 11.9
mmol) was mixed with pyrene (2 g, 9.9 mmol) and DCE (10
ml). To this mixture was added TiCl4 (2.25 g, 11.9 mmol)
dropwise with stirring over an ice bath. The temperature was
maintained below 5 1C during the addition. The brown
complex formed was stirred for further 12 h. The brown
complex was cleaved by the addition of water and dilute
HCl. The product was extracted into DCM (20 mL) and
washed with saturated sodium bicarbonate (2 ꢃ 85 mL), dil.
HCl and finally with water. The DCM layer was dried over
4
form layer was dried over anhydrous MgSO and concen-
trated under reduced pressure. The crude material was purified
by recrystallization from hexane–chloroform 99–1. The pure
product 1a (156 g, 84%) was isolated as yellow solid; mp.
5
8
1
9–61 1C (Found: C, 87.96; H, 7.86. Calc. for C H O: C,
7.60; H, 7.92%); nmax(KBr)/cm 3045, 2920, 1594, 1506,
466, 1413, 1382, 1254, 1218, 1182, 1063, 958, 842 and 714; dH/
26 38
ꢁ1
ppm (500 MHz; CDCl
2 H, m), 1.87 (2 H, q, 7.5), 3.20 (2 H, t, 7.5), 7.97–8.35 (8 H,
m) and 8.86 (1 H, d, 9.5); d /ppm (126 MHz; CDCl ) 14.06,
2.64, 25.03, 29.26, 29.43, 29.46, 29.49, 31.86, 42.74, 123.99,
3
) 0.88 (3 H, t, 7.0), 1.29 (10 H, m), 1.46
2 4
anhydrous Na SO and concentrated under reduced pressure.
(
The crude material was purified further by chromatography
over Merck 60 silica gel using DCM–hexane (1 : 9) as eluent.
The pure product 3a (3.25 g, 70%) was isolated as yellow solid;
mp 74–76 1C (Found: C, 87.09; H, 9.47. Calc. for C H O: C,
7.12; H, 9.46%); nmax(KBr)/cm 2953–2849, 1672, 1471, 843
and 716; d /ppm (250 MHz; CDCl ) 0.88 (3 H, t, J 6.75 Hz),
.25 (30 H, br s), 1.86 (2 H, m), 3.21 (2 H, t, J 7.5), 8.00–8.33 (8
H, m) and 8.87 (1 H, d, J 9.3 Hz); d /ppm (63 MHz; CDCl
4.11, 22.69, 25.03, 29.36, 29.44, 29.50, 29.62, 29.66, 29.70,
C
3
2
1
1
1
24.40, 124.84, 124.92, 125.05, 125.91, 125.92, 126.13, 126.34,
27.08, 127.37, 129.23, 129.33, 129.41, 130.61, 131.13, 133.08,
33.54 and 205.44; HRMS: m/z (TOF-ESþ) 357.2237 (Calc.
3
4
44
ꢁ
1
8
H
3
for: C H O þ H 357.2216).
2
6
28
1
C
3
)
1
-Decylpyrene (1). 1-(Pyren-1-yl)decan-1-one 1a (148 g, 0.4
1
3
1
1
mol) and KOH (90 g, 1.6 mol) were mixed with digol (800 ml)
and warmed to 180 1C. The solution turned brownish in
colour. To this solution hydrazine monohydrate (40 ml, 0.8
mol) was added slowly keeping the temperature at 190–200 1C
under a nitrogen atmosphere for 12 h. Finally, the temperature
was slowly raised to 220 1C. Vapours were flushed out with
nitrogen flow. The mixture was cooled to 50 1C. The product
was extracted into 3 ꢃ 1000 ml hexane, which were filtered
1.93, 42.73, 123.99, 124.40, 124.84, 125.04, 125.92, 126.13,
26.34, 127.08, 129.24, 129.33, 129.42, 130.60, 131.12, 133.05,
33.54 and 205.43; HRMS: m/z (EI 70 eV) 469.3383 (Calc. for
C H O: 469.3426).
34 44
1
-Octadecylpyrene (3). 1-(Pyren-1-yl)octadecan-1-one 3a
(
0.5 g, 1.07 mmol) was mixed with KOH (0.36 g, 6.4 mmol)
and digol (2 mL) and warmed to 110 1C. The hydrazine
hydrate was then added dropwise to the above stirring mix-
ture. This was stirred at 120 1C for 2–3 h. The temperature of
the reaction was raised to 200 1C. After 1 h the reaction
mixture was cooled to room temperature. The mixture was
acidified with dil. HCl till neutral to litmus. The product was
extracted in to chloroform and washed with saturated sodium
bicarbonate (2 ꢃ 85 mL), dil. HCl and finally with water. The
DCM layer was dried over anhydrous Na SO and concen-
4
through 50 g of silica and dried over anhydrous MgSO . The
combined extracts were evaporated to 1000 ml volume and left
to crystallize over 3 days at RT. Additional crop was obtained
by evaporating the hexane down to 400 ml and crystallizing
overnight with seed crystals from the first batch. The pure
product 1 (125 g, 88%) was obtained as bright yellow solid;
mp 69–71 1C (Found: C, 91.09; H, 8.92. Calc. for C H : C,
2
6
30
ꢁ
1
9
1.17; H, 8.83%); nmax(KBr)/cm 3039, 2954, 2917, 2850,
463, 961, 841, 824, 760, 722, 707, 680 and 625; d /ppm (500
) 0.88 (3 H, t), 1.2-1.35 (10 H, m), 1.38 (2 H, m),
.50 (2 H, q), 1.85 (2 H, q), 3.33 (2 H, t), 7.87 (1 H, d, J 8.0),
.97–8.20 (7 H, m) and 8.29 (1 H, d, J 9.0 Hz); d /ppm (126
) 14.10, 22.68, 29.34, 29.60, 29.636, 29.644, 29.84,
1
H
2
4
trated under reduced pressure. The crude material was purified
further by chromatography over Merck silica 60 using hexane
as eluent. The pure product 3 (0.34 g, 70%) was isolated as
white solid; mp 83–84 1C (Found: C, 89.57; H, 10.27. Calc. for
MHz; CDCl
3
1
7
C
MHz; CDCl
3
ꢁ1
C
34
H
46
:
C, 89.80, H, 10.19%);
nmax(KBr)/cm
3039,
3
1
1
1.91, 31.93, 33.60, 124.57, 124.73, 124.75, 125.107, 125.114,
25.70, 126.44, 127.04, 127.19, 127.52, 128.62, 129.70, 130.97,
31.48 and 137.35; HRMS: m/z (EI 70 eV) 342.2355 (Calc. for
2
965–2849, 1463 and 840; d /ppm (250 MHz; CDCl
H
3
) 0.89
(
3 H, t, J 6.8), 1.26 (30 H, br s), 1.85 (2 H, m), 3.34 (2 H, t, J 7.5
Hz), 7.87 (1 H, d, J 7.8), 7.95–8.18 (7 H, m) and 8.29 (1 H, d, J
.3); d /ppm (63 MHz; CDCl ) 14.11, 22.70, 29.37, 29.61,
C H : 342.2348).
6
2
30
9
C
3
1-Dodecylpyrene (2). Synthesis and characterization from
29.71, 29.85, 31.94, 33.62, 123.54, 124.59, 124.75, 125.12,
ref. 5a; mp 72–73 1C (Found: C, 90.95; H, 9.33. Calc. for
125.73, 126.46, 127.06, 127.23, 127.54, 128.63, 129.70,
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1438–1448 | 1445

View Article Online
1
30.98, 131.49 and 137.38; HRMS: m/z (EI 70 eV) 454.3595
(20 mL) was cooled in an ice bath for 15 min. To this stirring
(
Calc. for C34 46: 454.3599).
H
4
solution TiCl (5.7 g, 52 mmol) was added dropwise for 20
min. The dark brown complex formed was allowed to warm to
room temperature. The stirring was continued until the start-
ing material disappeared by TLC. The complex was cleaved by
the addition of water and the product was extracted into
4
-Oxo-4-(pyren-1-yl)butanoic acid (4a). Succinic acid (0.4 g)
was refluxed with excess of acetic anhydride (5 mL) for 10 h.
After the reaction acetic anhydride and acetic acid were
removed by distillation. A mixture containing pyrene (0.5 g,
CHCl
3
. Washed with water (2 ꢃ 5 mL) and dried over
SO . The solvent was removed under reduced
2
.5 mmol) and succinic anhydride (about 0.39 g, 3 mmol) were
dissolved in DCE (3 mL) and kept stirring on an ice bath. To
this stirring solution TiCl (0.56 g, 3 mmol) was added
anhydrous Na
2
4
pressure. The product was further purified by chromatography
over Merck silica 60 using acetone–DCM (1 : 9) as eluent. The
pure product 5a (5.0 g, 25%) was obtained as pale yellow
4
dropwise for 5 min and the stirring was continued for 24 h.
The complex was hydrolyzed by the addition of water and dil.
HCl. The product formed was extracted into EtOAc (15 mL)
and washed with water (2 ꢃ 5 mL). The organic layer was
solid; mp 116–117 1C (Found: C, 50.27; H, 8.20. Calc. for
ꢁ1
C
28
H
30
O
3
þ 14H
2
O: C, 50.43; H, 8.76%); nmax(KBr)/cm
3
425, 3043, 2924, 2889, 2848, 1716, 1666 and 843; d /ppm (250
H
2 4
dried over anhydrous Na SO and concentrated under re-
MHz; DMSO-d6) 1.18 (10 H, br s), 1.26–1.45 (4 H, m), 1.71 (2
duced pressure. The final product was further purified by
chromatography over Merck silica 60 using acetone–DCM
H, m), 2.14 (2 H, t, J 7.5 Hz), 3.22 (2 H, t, J 7.0 Hz), 8.09–8.39
(
7 H, m), 8.50 (1 H, d, J 8.0 Hz) and 8.70 (1 H, d, J 9.3Hz); dC/
ppm (63 MHz; DMSO-d6) 24.38, 24.48, 28.52, 28.61, 28.68,
8.79, 28.83, 33.67, 42.00, 123.56, 124.06, 124.36, 124.47,
125.99, 126.43, 126.45, 126.77, 127.20, 128.15, 129.22,
29.27, 129.99, 130.68, 132.88, 132.96, 174.51 and 205.09;
HRMS: m/z (TOF-ESꢁ) 413.2171 (Calc. for C28 ꢁ H:
13.2117).
(
1 : 9) as eluent. The pure product 4a (0.51 g, 68%) was
obtained as pale yellow crystalline solid; mp 184–185 1C
2
1
4
(
Found: C, 78.80; H, 4.52. Calc. for C H O þ H O: C,
2
0
13
3
2
ꢁ1
7
8.55; H, 4.28%); nmax(KBr)/cm 3453, 1695, 1665 and 842;
/ppm (250 MHz; DMSO-d6) 2.76 (2 H, t, J 6.0 Hz), 3.49 (2
1
d
H
H, t, J 6.0 Hz), 8.12–8.43 (7 H, m), 8.57 (1 H, d, J 8.3 Hz) and
30 3
H O
4
8
3
1
1
C
.77 (1 H, d, J 9.5 Hz); d /ppm(63 MHz; DMSO-d6) 28.55,
6.92, 123.47, 123.96, 124.44, 125.97, 126.39, 126.44, 126.72,
27.15, 128.12, 129.14, 129.29, 129.94, 130.63, 132.48, 133.01,
73.89 and 203.18; HRMS: m/z (TOF-ESꢁ) 301.0365 (Calc.
1
2-(Pyren-1-yl)dodecanoic acid (5). 12-Oxo-12-(pyren-1-yl)-
dodecanoic acid 5a (0.3 g, 0.7 mmol) and KOH (0.3 g, 5.5
mmol) were mixed with digol (1 ml) and warmed to 110 1C the
solution turned brownish in colour. To this solution hydrazine
monohydrate was added slowly and stirred at 120 1C for 2–3 h.
Then the temperature was slowly raised to 200 1C. All the
vapours were allowed to escape out under nitrogen flow. The
mixture was cooled to room temperature and acidified with
dil. HCl till neutral to litmus. The product was extracted into
EtOAc, washed with water (2 ꢃ 5 ml) and dried over anhy-
drous Na SO . The crude product was concentrated to dryness
for C20
14
H O
3
ꢁ H: 301.0865).
4
-(Pyren-1-yl)butanoic acid (4). To a reaction mixture con-
taining 4-oxo-4-(pyren-1-yl)butanoic acid 4a (0.5 g, 1.65
mmol), digol (2.0 mL) and KOH (0.74 g, 13.2 mmol) at 110
1
C was added hydrazine monohydrate (0.83 g, 16.5 mmol)
slowly and stirred for 2 h at 120 1C. The temperature of the
reaction was increased to 180 1C with continued stirring for at
least 50 min. The reaction mixture was poured into an ice bath
and acidified until neutral to litmus and extracted with ethyl
acetate (15 mL, 3 ꢃ 5 mL). The organic layer washed with
2
4
under reduced pressure. The product was further purified by
chromatography over Merck silica 60 using acetone–DCM (5
:
95) as eluent. The pure product 5 (0.19 g, 67%) was obtained
as pale yellow solid; mp 110–111 1C (Found: C, 80.52; H, 7.72.
Batch 1: calc. for C28 O: C, 80.35; H, 8.18. Found:
þ H
C, 82.58; H, 7.96. Batch 2: calc. for C28 O: C,
2.20; H, 8.12%); nmax(KBr)/cm 3435, 2921, 2846, 1705 and
844; d /ppm (500 MHz; CDCl ) 1.27 (10 H, br s), 1.38 (2 H, t,
.5), 1.49 (2 H, q, 7.5), 1.63 (2 H, q, 7.5), 1.86 (2 H, q, 8.0), 2.34
2 H, t, J 7.5 Hz), 3.34 (2 H, t, 8.0), 7.87 (1 H, d, J 8.0 Hz),
.96–8.17 (7 H, m) and 8.29 (1 H, d, J 9.5 Hz); d /ppm (126
MHz; CDCl ) 24.67, 29.03, 29.19, 29.38, 29.54, 29.55, 29.56,
water (2 ꢃ 5 mL) and dried over anhydrous Na
2 4
SO . The
crude product was further purified by column chromatogra-
phy over Merck silica 60 using acetone–DCM (5 : 95) as
eluent. The final product 4 (0.31 g, 65%) was obtained as pale
yellow solid; mp 182–183 1C (Found: C, 80.41; H, 5.42. Calc.
H
32
O
2
2
þ 12
32 2 2
H O H
ꢁ
1
8
1
2
ꢁ1
H
3
for C H O þ H O: C, 80.84; H, 5.76%); nmax(KBr)/cm
2
0
16
2
2
7
3
2
433, 2920, 2845, 1701 and 843; d /ppm (250 MHz; CDCl )
H
3
(
.22 (2 H, m), 2.52 (2 H, t, J 7.0 Hz), 3.43 (2 H, m), 7.87 (1 H,
7
C
d, J 7.8 Hz), 7.95–8.20 (7 H, m) and 8.30 (1 H, d, J 9.3 Hz); d
ppm (63 MHz; CDCl 26.53, 32.68, 33.27, 123.25,
C
/
3
3
)
2
1
1
9.79, 31.92, 33.60, 33.88, 123.53, 124.59, 124.75, 124.76,
25.10, 125.11, 125.73, 126.45, 127.06, 127.23, 127.53,
28.62, 129.70, 130.97, 131.48, 137.35 and 179.21; HRMS:
1
1
24.83, 124.86, 124.97, 125.03, 125.17, 125.88, 127.37, 127.50,
28.79, 130.09, 130.94, 131.46, 135.49 and 177.69; HRMS:
m/z (EI 70 eV) 288.1145 (Calc. for C20
16 2
H O : 288.1150).
m/z (TOF-ESꢁ) 399.2137 (Calc. for C28
H
32
O
2
ꢁ H: 399.2324).
12-Oxo-12-(pyren-1-yl)dodecanoic acid (5a). Dodecanedioic
acid (23 g, 0.1 mol) was first treated with excess of thionyl
chloride (50 ml, 0.7 mol) and refluxed for 3–4 h. Excess of
Methyl-12-(pyren-1-yl)dodecanoate (6). 12-(Pyren-1-yl)do-
decanoic acid 5 (0.3 g, 0.75 mmol) was added to a solution
SOCl
2
was removed by distillation under reduced pressure.
2 4
containing 2 ml MeOH and 0.2 ml H SO and refluxed for 4–5
The dodecanedioyl dichloride was used without further
purification.
h. After the completion, the reaction the mixture was neutra-
lized with saturated NaHCO and extracted into CHCl . The
3
3
To a reaction mixture containing pyrene (10 g, 49.5 mmol)
and dodecanedioyl dichloride (14 g, 52.4 mmol) in DCE
chloroform layer was washed with water (2 ꢃ 5 ml), dried over
anhydrous Na SO . The solvent was removed under reduced
2
4
1
446 | New J. Chem., 2008, 32, 1438–1448
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008

View Article Online
pressure and purified further by column chromatography over
Merck silica 60 using CHCl –hexane (2 : 8) as eluent. The
and 138.68; HRMS: m/z (TOF-ESþ) 493.2900 (Calc. for
3
34
C H46ONa: 493.3446).
pure product 6 (0.28 g, 90%) was obtained as pale yellow
solid; mp 57.5–58.0 1C (Found: C, 84.02; H, 8.31. Calc. for
1
2-(Pyren-1-yl)dodecan-1-ol (9). Methyl-12-(pyren-1-yl)do-
1
ꢁ1
decanoate 6 (0.1 g, 0.24 mmol) was dissolved in 1 mL of dry
THF. This solution was added dropwise to a stirring suspen-
sion of LAH (0.092 g, 2.4 mmol) in 1 mL of dry THF. The
resulting mixture was stirred for 3–4 h at RT. After the
reaction, the excess of LAH was neutralized by addition
EtOAc and moist EtOH. The mixture was filtered. Filtrate
was concentrated and dried. The crude product was further
purified by column chromatography over Merck silica 60
column using DCM–EtOAc (98 : 2) as the eluent. The pure
product 9 (0.061 g, 66%) was isolated as a white powder; mp
C H O
9
þ
H O: C, 84.01; H, 8.27%); nmax(KBr)/cm
2
34
2
2
2
3
455, 3040, 2928, 2852, 1741, 1155 and 841; d /ppm (500
H
MHz; CDCl
.5), 1.63 (2 H, m), 1.87 (2 H, q, J 7.5 Hz), 2.31 (2 H, t, J 7.5
Hz), 3.34 (2 H, t, J 8.0 Hz), 3.68 (3 H, s), 7.87 (1 H, d, J 7.8),
.97–8.17 (7 H, m) and 8.29 (1 H, d, J 9.0 Hz); d /ppm (126
MHz; CDCl ) 24.94, 29.12, 29.21, 29.40, 29.54, 29.55, 29.56,
3
) 1.29 (10 H, br s), 1.39 (2 H, m), 1.50 (2 H, q,
7
7
C
3
2
1
1
9.79, 31.90, 33.57, 34.09, 51.36, 123.50, 124.56, 124.72,
24.74, 125.08, 125.09, 125.70, 126.43, 127.04, 127.19,
27.51, 128.60, 129.68, 130.95, 131.46, 137.31 and 174.26;
7
8
1
2–75 1C (Found: C, 86.54; H, 8.43. Calc. for C28
H
34O: C,
6.99; H, 8.86%); nmax(KBr)/cm 3411, 3038, 2920, 2849,
465, 1057 and 843; d /ppm (250 MHz; CDCl ) 1.18 (1 H, br
HRMS: m/z (EI 70 eV) 414.2551 (Calc. for C H O :
2
9
34
2
ꢁ
1
4
14.2558).
H
3
s), 1.28 (17 H, br s), 1.53 (2 H, m), 1.86 (2 H, q, J 8.0 Hz), 3.34
2 H, t, J 7.8 Hz), 3.63 (2 H, t, J 6.8 Hz), 7.87 (1 H, d, J 7.8 Hz),
.93–8.20 (7 H, m) and 8.29 (1 H, d, J 9.3 Hz); d (63 MHz;
1
2-(Pyren-1-yl)dodecanehydrazide (7). To a solution con-
(
taining methyl-12-(pyren-1-yl)dodecanoate 6 (0.25 g, 0.6
mmol) in 3 mL methanol was added hydrazine hydrate (0.21
g, 4.2 mmol) and refluxed for 4 h. After the reaction the
mixture was extracted into chloroform. The chloroform layer
was washed with water (2 ꢃ 5 ml) and dried over anhydrous
Na SO . Chloroform was removed under reduced pressure.
7
C
CDCl
3
) 25.96, 29.64, 29.81, 30.05, 32.17, 33.05, 33.84, 63.33,
23.77, 124.82, 124.99, 125.33, 125.97, 126.69, 127.29, 127.47,
27.77, 128.85, 129.93, 131.20, 131.71 and 137.60; HRMS: m/z
34ONa: 409.2507).
1
1
(
TOF-ESþ) 409.1206 (Calc. for C28
H
2
4
The crude material was purified further by column chromato-
graphy over Merck silica 60 using acetone–DCM (1 : 9) as the
eluent. The pure product 7 (0.21 g, 80%) was obtained as
1
-(Pyren-1-yl)dodecane-1,12-diol (10). 12-Oxo-12-(pyren-1-
yl)dodecanoic acid 5a (0.1 g, 0.24 mmol) was dissolved in 1
mL of dry THF. This solution was added dropwise to a
stirring suspension of LAH (0.09 g, 2.4 mmol) in 1 mL of
dry THF. The resulting mixture was stirred for 5–6 h at RT.
After the reaction, excess of LAH was neutralized by addition
EtOAc and moist EtOH. The mixture was filtered. Filtrate was
concentrated and dried. The crude product was further pur-
ified by column chromatography over Merck silica 60 column
using EtOAc–DCM (5 : 95) as eluent. The pure product 10
white solid; mp 122–123 1C (Found: C, 79.97; H, 8.113; N,
1
2
6
.39. Calc. for C28
H
34
ꢁ
N
2
O þ H
2
O: C, 79.50; H, 8.33; N,
1
6
.61%); nmax(KBr)/cm 3430, 3300, 2918, 2850, 1631, 1530
H 3
and 842; d /ppm (500 MHz; CDCl ) 1.27 (10 H, br s), 1.38 (2
H, m), 1.48 (2 H, m), 1.60 (2 H, m), 1.86 (2 H, q, 7.5), 2.12 (2
H, t, J 8.0 Hz), 3.34 (2 H, t, J 8.0 Hz), 3.88 (2 H, br s), 6.60 (1
H, br s), 7.87 (1 H, d, J 8.0 Hz), 7.96–8.17 (7 H, m) and 8.28 (1
H, d, J 9.5 Hz); d /ppm (126 MHz; CDCl ) 25.47, 29.25,
C
3
(
0.068 g, 70%) was isolated as a pale yellow powder; mp
2
1
1
9.41, 29.54, 29.55, 29.79, 31.93, 33.60, 34.59, 123.54, 124.60,
24.77, 125.11, 125.13, 125.75, 126.47, 127.07, 127.24, 127.54,
28.64, 129.71, 130.98, 131.49 and 137.36; HRMS: m/z (TOF-
1
11–112 1C (Found: C, 76.29; H, 8.93. Calc. for C28
H
34
O
2
þ 2
ꢁ1
H
2
O: C, 76.67; H, 8.73%); nmax(KBr)/cm 3415, 2920, 2850,
1
H 3
465, 1057 and 843; d /ppm (250 MHz; CDCl ) 1.25 (14 H, br
ESþ) 415.2757 (Calc. for C28
H
34
N
2
O þ H: 415.2749).
s), 1.54 (2 H, m), 2.05 (2 H, m), 3.62 (2 H, t, J 6.5 Hz), 5.80 (1
H, t, J 6.3 Hz), 7.97–8.21 (8 H, m) and 8.37 (1 H, d, J 9.5 Hz);
1
-(Pyren-1-yl)octadecan-1-ol (8). 1-(Pyren-1-yl)octadecan-1-
d
2
1
1
C 3
/ppm (63 MHz; CDCl ) 25.94, 26.47, 29.61, 29.73, 29.76,
one 3a (0.1 g, 0.21 mmol) was dissolved in 1 mL of dry THF.
This solution was added dropwise to a stirring suspension of
LAH (0.08 g, 2.1 mmol) in 1 mL of dry THF. The resulting
mixture was stirred for 3–4 h at RT. After the reaction, excess
of LAH was neutralized by addition EtOAc and moist EtOH.
The mixture was filtered and filtrate was concentrated and
dried. The crude product was further purified by column
chromatography over Merck silica 60 column using
EtOAc–DCM (5 : 95) as eluent. The pure product 8 (0.072
9.81, 33.04, 39.42, 63.33, 71.68, 122.81, 123.57, 125.16,
25.22, 125.29, 125.46, 126.16, 127.42, 127.73, 127.79,
27.83, 130.86, 130.94, 131.67 and 138.69; HRMS: m/z
(
TOF-ESþ) 425.2405 (Calc. for C H O Na: 425.2457).
2
8
34
2
0
1,1 -(Dodecane-1,12-diyl)dipyrene (11). To a reaction mix-
ture containing pyrene (10.1 g, 0.05 mol) and dodecanedioyl
dichloride (7 g, 0.026 mol; see 5a) in dichloroethane (150 ml)
was cooled in an ice bath for 15 min. To this stirred solution
g, 72%) was isolated as a pale yellow powder; mp 85–87 1C
4
TiCl (40 g, 0.2 mol) was added dropwise during 4 h at 5 1C or
1
(
Found: C, 85.89; H, 9.90. Calc. for C34
H
46O þ H
2
O: C,
lower temperature. The dark brown complex formed was
allowed to warm to room temperature. The stirring was
continued until the pyrene disappeared by TLC. The complex
was cleaved by the addition of water and the product was
4
ꢁ1
8
1
1
7
5.93; H, 9.81%); nmax(KBr)/cm 3440, 2920, 2850, 1466,
055 and 842; d
.24–1.61 (30 H, m), 2.00–2.13 (2 H, m), 5.79 (1 H, t, 6.5),
.95–8.20 (8 H, m) and 8.36 (1 H, d, J 9.3 Hz); d /pppm (63
H 3
/ppm (250 MHz; CDCl ) 0.88 (3 H, t, 6.8),
C
3
extracted into CHCl in Soxhlet apparatus and washed with
MHz; CDCl ) 14.33, 22.91, 26.49, 29.58, 29.92, 32.15, 39.42,
3
water (2 ꢃ 50 ml) and dried over anhydrous MgSO . The
4
7
1
1.67, 122.79, 123.55, 125.14, 125.20, 125.21, 125.27, 125.43,
26.13, 127.40, 127.72, 127.77, 127.81, 130.84, 130.93, 131.66
solvent was removed under reduced pressure yielding a mix-
ture of impure 1,12-di(pyren-1-yl)dodecane-1,12-dione and 12-
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008 New J. Chem., 2008, 32, 1438–1448 | 1447

View Article Online
oxo-12-(pyren-1-yl)dodecanoic acid 5a, which were used as
Chem., Int. Ed., 2001, 40, 3164–3166; (c) S. Tamaru, M. Nakamura,
M. Takeuchi and S. Shinkai, Org. Lett., 2001, 3, 3631–3634.
. (a) K. J. C. van Bommel, A. Friggeri and S. Shinkai, Angew. Chem.,
Int. Ed., 2003, 42, 980; (b) C. L. Chan, J. B. Wang, J. Yuan,
H. Gong, Y. H. Liu and M. H. Liu, Langmuir, 2003, 19, 9440;
(c) C. S. Love, V. Chechik, D. K. Smith, K. Wilson, I. Ashworth
and C. Brennan, Chem. Commun., 2005, 1971; (d) K. Sugiyasu,
S. Tamura, M. Takeuchi, D. Berthier, I. Huc, R. Oda and
S. Shinkai, Chem. Commun., 2002, 1212; (e) S. Kobayashi,
N. Hamasaki, M. Suzuki, M. Kimura, H. Shirai and
K. Hanabusa, J. Am. Chem. Soc., 2002, 124, 6550;
such in the next reaction.
3
2
00 ml of digol was added with KOH (25 g, 0.45 mol) and
the mixture was heated to 180 1C under a nitrogen purge.
Hydrazine hydrate (10 ml, 0.2 mol) was added dropwise
keeping the temperature above 180 1C under 20 h. After all
the hydrazine was added, the temperature was raised to 220 1C
for 4 h under nitrogen flush and the mixture was cooled to 50
1
C. Hexane (5 ꢃ 500 ml) was added and refluxed for 20
(
f) E. D. Sone, E. R. Zubarev and S. I. Stupp, Angew. Chem.
Int., Ed., 2002, 41, 1705; (g) Z. Hu and X. Xia, Adv. Mater., 2004,
6, 305; (h) M. Kimura, S. Kobayashi, T. Kuroda, K. Hanabusa
minutes under vigorous stirring for each batch. The hexane
phase was separated and mixed with 10 g of silica, filtered and
evaporated under reduced pressure to 20 ml and then let
evaporate to dryness slowly at normal pressure in a flask in
a hood. Colourless crystalline powder of 11 (2.2 g, 15%) was
obtained.
1
and H. Shirai, Adv. Mater., 2004, 16, 335; (i) M. Asai, K. Sugiyasu,
N. Fujita and S. Shinkai, Chem. Lett., 2004, 33, 120;
(
j) B. Simmons, S. Li, V. T. John, G. L. McPherson, C. Taylor,
D. K. Schwartz and K. Maskos, Nano Lett., 2002, 2, 1037.
4. (a) A. Bogershausen, S. J. Pas, A. J. Hill and H. Koller, Chem.
¨
The digol phase was neutralized with 10% HCl and ex-
tracted with chloroform (3 ꢃ 250 ml). The chloroform was
Mater., 2006, 18, 664–672; (b) S. J. Langford, M. J. Latter,
V.-L. Lau, L. L. Martin and A. McChler, Org. Lett., 2006, 7,
1
371–1373; (c) H. Jung, S. J. Lee, J. A. Rim, H. Lee, T.-S. Bae,
washed with 2 ꢃ 100 ml of water and dried over MgSO
4
. 75 g
S. S. Lee and S. Shinkai, Chem. Mater., 2005, 17, 459–462;
(d) S. Milijanic, L. Frkanec, Z. Meic and M. Zinic, Langmuir,
of silica was added to the chloroform and the solvent was
evaporated to dryness under reduced pressure. The silica was
eluted with THF–hexane–methanol (19 : 80 : 1) over 1.5 kg
of Merck silica 60 in a preparative MPLC. The pure product 5
2
005, 21, 2754–2760; (e) S. J. Lee, S. S. Lee, J. S. Kim, J. Y. Lee and
J. H. Jung, Chem. Mater., 2005, 17, 6517–6520; (f) A. Kishimura,
T. Yamashita and T. Aida, J. Am. Chem. Soc., 2005, 127, 179–183;
(g) S. H. Seo and J. Y. Chang, Chem. Mater., 2005, 17, 3249–3254;
(h) Y. Watanabe, T. Miyasou and M. Hayashi, Org. Lett., 2004, 6,
1547–1550; (i) M. George, S. L. Snyder, P. Terech, C. J. Glinka and
R. G. Weiss, J. Am. Chem. Soc., 2003, 125, 10275–10283;
(j) A. Friggeri, O. Gronwald, K. J. C. van Bommel, S. Shinkai
and D. N. Reinhoudt, J. Am. Chem. Soc., 2002, 124, 10754–10758;
(
3.5 g, 18%) was recovered.
0
1
,1 -(dodecane-1,12-diyl)dipyrene 11 (2.2 g, 15%); mp
1
9
1
29–132 1C (Found: C, 92.57; H, 7.49. Calc. for C H : C,
2.58; H, 7.42%); nmax(KBr)/cm 3038, 2918, 2851, 1604,
44 42
ꢁ1
(
1
k) M. George and R. G. Weiss, J. Am. Chem. Soc., 2001, 123,
0393–10394; (l) A. Ajayaghosh and S. J. George, J. Am. Chem.
466, 1418, 1180, 963, 844, 758, 723, 706, 681 and 620; d (500
H
MHz; CDCl ) 1.28 (8 H, m), 1.37 (4 H, m), 1.48 (4 H, q), 1.82
3
Soc., 2001, 123, 5148–5149; (m) D. J. Abdallah, L. Lu and
R. G. Weiss, Chem. Mater., 1999, 11, 2907–2911;
(n) K. Hanabusa, K. Hiratsuka, M. Kimura and H. Shirai, Chem.
Mater., 1999, 11, 649–655; (o) P. Terech and R. G. Weiss, Chem.
Rev., 1997, 97, 3133–3160; (p) S. Ray, A. K. Das and A. Banerjee,
Chem. Commun., 2006, 2816–2818; (q) X.-Q. Li, V. Stepanenko,
(
4 H, q), 3.33 (4 H, t), 7.86 (2 H, d, J 8.0 Hz), 7.96–8.03 (6 H,
m), 8.08–8.16 (8 H, m) and 8.28 (2 H, d, J 9.5 Hz); d (126
MHz; CDCl ) 29.556, 29.580, 29.607, 29.802, 31.926, 33.602,
C
3
1
1
1
23.536, 124.583, 124.746, 124.758, 125.107, 125.120, 125.726,
26.455, 127.058, 127.226, 127.531, 128.631, 129.701, 130.974,
31.486 and 137.374; HRMS: m/z (EI 70 eV) 570.3280 (Calc.
¨
Z. Chen, P. Prins, L. D. A. Sibbeles and F. Wurthner, Chem.
Commun., 2006, 3871–3873.
5
6
. (a) P. Babu, N. M. Sangeetha, P. Vijaykumar, U. Maitra,
K. Rissanen and A. R. Raju, Chem.–Eur. J., 2003, 9, 1922–1932;
(b) U. Maitra, P. Vijaykumar, N. M. Sangeetha, P. Babu and
A. R. Raju, Tetrahedron: Asymmetry, 2001, 12, 477–480.
for C44H42: 570.3287).
Acknowledgements
. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz,
Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson,
W. Chen, M. W. Wong, C. Gonzalez, J. A. Pople, GAUSSIAN 03
(Revision C.02), 2004, Gaussian, Inc., Wallingford, CT, 2001.
This work was supported by the Academy of Finland grants
1
06998 and 107014. Thanks are due to Paavo Niutanen for his
help in SEM sample preparation, to Reijo Kauppinen for
NMR and to Mirja Lahtipera for MS. We also thank Dr
¨
Roland Pein at DLR, Lampoldshausen, Germany for provid-
ing facilities for rheometry and Prof. Uday Maitra IISc,
Bangalore, India for equipment for UV/Vis and fluoresence
spectroscopy.
References
1
. (a) N. M. Sangeetha and U. Maitra, Chem. Soc. Rev., 2005, 34,
21–836; (b) ‘Low Molecular Mass Gelators’, in Topics in Current
Chemistry, ed. F Fages, Springer, Berlin, Heidelberg, 2005, vol.
56; (c) Molecular Gels: Materials with Self-Assembled Fibrillar
Networks, ed. R. G. Weiss and P Terech, Springer, Dordrecht,
8
2
7. C. A. Hunter, K. R. Lawson, J. Perkins and C. J. Urch, J. Chem.
Soc., Perkin Trans. 2, 2001, 651–669.
2006; (d) P. Terech and R. G. Weiss, Chem. Rev., 1997, 97,
3133–3159; (e) L. A. Estroff and A. D. Hamilton, Chem. Rev.,
2004, 104, 1201–1217.
8. (a) Cook and Hewett, J. Chem. Soc., 1933, 398–401; Cook and
Hewett, Chem. Ind. (London), 1936, 843; Winterstein, Vetter and
Schoen, Chem. Ber., 1935, 68, 1079–1082; Bachmann, Carmack and
Safir, J. Am. Chem. Soc., 1941, 63, 1682–1684; (b) S. Katusin-
Razem, Croat. Chem. Acta, 1978, 51, 163–164.
2
. (a) J. H. van Esch and B. L. Feringa, Angew. Chem., Int. Ed., 2000,
9, 2263–2266; (b) G. Mieden-Gundert, L. Klein, M. Fischer,
F. Vogtle, K. Heuze, J. L. Pozzo, M. Vallier and F. Fages, Angew.
3
¨
´
1
448 | New J. Chem., 2008, 32, 1438–1448
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2008