Smectic liquid crystals from supramolecular guanidinium alkylbenzenesulfonates

Article abstract of DOI:10.1002/1521-3765(20020517)8:10<2248::AID-CHEM2248>3.0.CO;2-B

A homologous series of guanidinium alkylbenzenesulfonates from ethyl to tetradecyl were synthesized and characterised. Their thermotropic polymorphism was investigated by polarizing optical microscopy, differential scanning calorimetry, and dilatometry. The structure of the smectic liquid crystal phases obtained at high temperature with the compounds from octyl to tetradecyl was analysed by X-ray diffraction. The supramolecular assembling of the ionic species inside the smectic layers was investigated by infrared spectroscopy.

Full text of DOI:10.1002/1521-3765(20020517)8:10<2248::AID-CHEM2248>3.0.CO;2-B

FULL PAPER
Smectic Liquid Crystals from Supramolecular Guanidinium
Alkylbenzenesulfonates
[a]
Fabrice Mathevet, Patrick Masson,*Jean-FranÁois Nicoud, and Antoine Skoulios
Abstract: A homologous series of guanidinium alkylbenzenesulfonates from ethyl to
tetradecyl were synthesized and characterised. Their thermotropic polymorphism
was investigated by polarizing optical microscopy, differential scanning calorimetry,
and dilatometry. The structure of the smectic liquid crystal phases obtained at high
temperature with the compounds from octyl to tetradecyl was analysed by X-ray
diffraction. The supramolecular assembling of the ionic species inside the smectic
layers was investigated by infrared spectroscopy.
Keywords: amphiphiles ¥ hydrogen
bonds liquid crystals
supramolecular chemistry
¥
¥
ion packing in the crystals (the cross-sectional area of linear
paraffins in the crystalline state is of only 18.5 ä2).[3] In such a
para alkylated compound, the benzene rings should of course
disentangle and the ionic sublayers should be drawn apart in
order to make enough room for the alkyl chains to fit into the
structure (Figure 1c). More interestingly, the packing area of
the benzene rings is large enough even to permit the disordering
of the alkyl chains (the cross-sectional area of disordered alkyl
chains in smectic liquid crystals generally lies between 21and
27 ä2).[3, 4] Taking this last remark as a basis, one may then
formulate the conjecture that para alkylated guanidinium
benzenesulfonate salts should be able to produce smectic liquid
crystals after melting of the alkyl chains at high temperature.
The objective of the present work is to check the soundness
of the above conjecture. It is also to test whether the strength
of the hydrogen bonding of the guanidinium and sulfonate
ions is sufficient to maintain the supramolecular architecture
of the layers despite the melting of the alkyl chains in the
smectic state. For this purpose, a series of guanidinium
alkylbenzenesulfonates from ethyl to tetradecyl (abbreviated
Introduction
The crystal structure of a variety of guanidinium alkane- and
arenesulfonates has only recently been investigated by X-ray
diffraction.[1] Strongly interacting through coulombic forces,
the positive and negative ions were, interestingly enough,
found to self-assemble into supramolecular two-dimensional
networks as a result of ™equal numbers of donor and acceptor
hydrogen-bonding sites and three-fold topology common to
both guanidinium and sulfonate ions∫. Separated by the
organic residues of the sulfonate ions, these networks were, in
turn, found to pile up periodically over one another to form
lamellar structures. With bulky residues, unable for apparent
reasons to sit all on the same side of the supramolecular
networks, the ionic sheets are covered by the residues on both
sides, and the lamellar structure is single-layered. Inversely,
with thin residues, the ionic sheets, covered only on one side
are associated in pairs by bracketing together their polar
surfaces, and the lamellar structure is double-layered. Figure 1
illustrates the double-layered structure of guanidinium ben-
zenesulfonate, showing the in-layer hydrogen-bonded supra-
molecular network of the guanidinium and sulfonate ions, the
double-layered arrangement of the ionic species, and the
interdigitated packing of the benzene rings.
to GABS-n, with n even ranging from 2 to 14) were prepared;
their thermotropic polymorphism was investigated by polar-
izing optical microscopy, differential scanning calorimetry,
and dilatometry; their structure at high temperature was
determined by X-ray diffraction; and their supramolecular
architecture was checked by infrared spectroscopy.
Careful inspection of the crystal structure of the benzene
derivative (Figure 1) gives one interesting observation. As
deduced from the lattice parameters,[1] the packing area of the
interdigitated benzene rings, ³4 acsinb ˆ 22.6 ä , is large
enough to permit the addition of long alkyl chains in the
para position of the benzene rings without disruption of the
2
1
[a] Dr. P. Masson, F. Mathevet, Prof. J.-F. Nicoud, Dr. A. Skoulios
Results and Discussion
¬
Groupe des Materiaux Organiques
¬
Institut de Physique et Chimie des Materiaux
23 rue du Loess, 67037 Strasbourg Cedex (France)
Fax : (‡33) 3 88 1 0 72 46
Synthesis: The GABS-n compounds were synthesized follow-
ing the general method described in the literature,[1] by
precipitation from solutions of equimolar amounts of guani-
E-mail: patrick.masson@ipcms.u-strasbg.fr
2248
¹ WILEY-VCH Verlag GmbH, 69451Weinheim, Germany, 2002
0947-6539/02/0810-2248 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 1 0

2248 2254
Figure 1. Crystal structure of guanidinium benzenesulfonate drawn by molecular modelling from experimental data of reference:[1] a) projection on b, c
lattice plane of two layers of interdigitated molecules, b) packing mode in a, c lattice plane of sulfonate and guanidinium ions sitting on upper surface of
upper molecular layer (benzene rings are omitted for clarity), c) splitting of the upper molecular layer, the molecules being drawn apart in order to leave
enough space for the alkyl chains to be attached to the benzene rings, d) projection on a, c lattice plane of the upper layer of the interdigitated molecules
(guanidinium ions are omitted for clarity), e) projection on a, c lattice plane of half the upper molecular layer (all guanidinium ions and molecules with
sulfonate groups pointing downwards are omitted). Carbon atoms in green, sulphur atoms in yellow, oxygen atoms in red, nitrogen atoms in blue, hydrogen
atoms in white. Cell parameters: a ˆ 7.50, b ˆ 23.287, c ˆ 12.060 ä, b ˆ 92.248.
dine hydrochloride and sodium 4-alkylbenzensulfonate. To
ensure purity, special care was taken to select the appropriate
solvent (pure water for n ꢀ8, or water/ethanol 1:1 v/v mixture
for n >8) and to control its rate and extent of evaporation
during the precipitation process. Not commercially available,
the sodium 4-alkylbenzensulfonates used were synthesized
beforehand by action of chlorosulfonic acid on alkylbenzene
and subsequent saponification, a method known to produce
para isomers exclusively.[5]
Thermal and optical studies: Studied by thermogravimetry,
the GABS-n compounds were all found to be thermally stable
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2249 $ 20.00+.50/0
2249

P. Masson et al.
FULL PAPER
up to temperatures of about 3208C. Examined by polarizing
optical microscopy in the temperature range from ambient up
to 3008C, they appeared to melt sharply at the temperatures
indicated in Figure 2, transforming either into an isotropic
liquid for chain lengths n ꢀ6, or into a strongly birefringent
structure requires the knowledge of the molecular volumes.
Three GABS-n compounds, namely GABS-8, GABS-10, and
GABS-12, were therefore analysed by dilatometry as a
8]
function of increasing, then of decreasing temperature.[6
Figure 4 shows the phase transition from the low-temperature
crystal to the high-temperature smectic phase in the particular
case of GABS-12 taken as an example. In agreement with the
results of optical microscopy and differential scanning calo-
rimetry, the molecular volume was found to grow with
temperature in the stability range of each one of the phases
observed and to jump abruptly at the crystal to smectic phase
transition by some 36 ä3. Upon cooling, the volume decreases
quite reversibly in the whole range explored, and the smectic
phase supercools, as usual, well below the melting temper-
ature.
Figure 2. Phase transition temperatures of the guanidinium alkylbenzene-
sulfonates measured by polarizing optical microscopy and confirmed by
differential scanning calorimetry (onset of hottest endotherm) as a function
of the number of carbon atoms in the alkyl chains (Cr: crystal, I: isotropic
liquid, Sm: smectic).
fluid stable up to the highest temperature explored for chain
lengths n ꢁ8. The focal-conic textures observed (Figure 3)
clearly show that the birefringent fluid phases are genuine
smectic liquid crystals. With the object of characterising more
precisely the crossing over from the isotropic to the smectic
state as a function of growing chain length, a set of binary
mixtures of GABS-6 and GABS-8 were carefully investigated
at high temperature by optical microscopy and X-ray
diffraction. Practically independent of temperature, the cross-
ing over was thus found to proceed through a very narrow
biphasic region at an average chain length of n 6.7.
Figure 4. Variation of the molar volume of guanidinium dodecylbenzene-
sulfonate as a function of increasing (open circles) and decreasing (solid
circles) temperature, showing the melting of the crystals and the super-
cooling of the smectic phase. Arrow indicates the transition temperature
detected by optical microscopy.
Figure 4 further shows, in agreement with previous obser-
vations of liquid crystals, that the molecular volume of the
smectic phase increases with temperature in a linear fashion,
according to the following Equations (1) (3) established by a
least-squares linear fit of the experimental data (R > 0.9999):
GABS-8: V [ä3] ˆ 480.43Æ0.02 ‡ 0.3196Æ0.0001 T [8C]
GABS-10: V [ä3] ˆ 534.21Æ0.03 ‡ 0.3623Æ0.0002 T [8C]
GABS-12: V [ä3] ˆ 587.91Æ0.03 ‡ 0.4111Æ0.0002 T [8C]
(1)
(2)
(3)
Figure 3. Focal-conic smectic texture of guanidinium octylbenzenesulfo-
nate at 1848C.
The same for the three compounds, the relative thermal
expansion coefficient deduced from the equations above, (qV/
qT)/V 7  10À4 KÀ1, is equal to that currently given in the
literature for organic materials in the molten state.[4] Finally,
the molecular volume at 1908C (where the smectic structure is
to be analysed in the section below) increases with the number
n of carbon atoms in the alkyl chains in a linear fashion
(Figure 5), according to the Equation (4) (R ˆ 0.99999):
The thermotropic transition from a crystal to a smectic
phase was fully confirmed by differential scanning calorim-
etry. Ranging from 60 to 120 JgÀ1, the enthalpies measured
could not be analysed in detail, particularly as a function of
increasing chain length, because of a complex polymorphism
in the solid state difficult to control.
Dilatometric investigation of the smectic phases: As stated in
the section below, a quantitative interpretation of the smectic
V [ä3] ˆ 291.2Æ1.6 ‡ 31.22Æ0.16
n
(4)
2250
¹ WILEY-VCH Verlag GmbH, 69451Weinheim, Germany, 2002
0947-6539/02/0810-2250 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 1 0

Smectic Liquid Crystals
2248 2254
Figure 5. Variation of the molar volume of guanidinium alkylbenzenesul-
fonates in the smectic state at 1908C as a function of chain length.
Figure 7. Schematic representation of the structure of the guanidinium
&
*
alkylbenzenesulfonates in the smectic state: benzene rings; sulfonate
*
anions; guanidinium cations; wavy lines: molten alkyl chains.
Structural investigation of the smectic phases: The smectic
nature of the GABS-n compounds suggested by optical
microscopy was fully confirmed by X-ray diffraction. Indeed,
the powder patterns recorded at high temperature contain
two sharp, equidistant reflections in the small-angle region,
related to the smectic layering of the molecules, and a unique
diffuse ring in the wide-angle region at about 4.7 ä, related to
the liquid-like conformation of the alkyl chains and to the lack
of long-range ordering of the ionic species within the smectic
layers (Figure 6). With the concept of microphase separation
The slope of the d versus n line represents the growth of the
thickness of the smectic layers per methylene group added to
the molecules. As for its extrapolation down to n ˆ 0, it gives
an estimate of the thickness of the guanidinium benzenesul-
fonate polar sublayers (d0 ˆ 16.87 ä), which turns out per-
fectly consistent with that calculated otherwise by molecular
modelling (ꢂ16.1 ä).
Figure 6. X-ray powder diffraction pattern of guanidinium dodecylbenze-
nesulfonate in the smectic phase at 1908C. For clarity, intensity at Bragg
angles 2q > 4.88 is multiplied by 15.
Figure 8. Smectic period of the guanidinium alkylbenzenesulfonates in the
smectic state at 1908C as a function of the (average) number of carbon
atoms in the alkyl chains. Solid circles correspond to pure compounds, open
circles to hexyl/octyl, octyl/decyl, and decyl/dodecyl binary mixtures.
of amphiphiles (with their constituent parts being incompat-
ible with one another) taken for granted,[9, 10] the smectic
structure consists of an alternate superposition of nonpolar
and polar sublayers, here made respectively of the alkyl chains
and the ionic headgroups of the molecules (Figure 7).
The smectic periods measured increase linearly with the
number n of carbon atoms in the alkyl chains (Figure 8),
according to the Equation (5) below, determined at 1908C by
a least-squares linear fit of the experimental data (R ˆ
0.9996).
Considering that a layer of N molecules, occupying a
volume N ¥ V and covering an area N ¥ S, has a thickness d ˆ
V/S, one may easily calculate S using the values found
experimentally for d and V (except for GABS-14, where the V
value was extrapolated from the V versus n equation). It is
worth emphasising that parameter S measures the contribu-
tion of one molecule to the lateral spreading of the layers and
hence represents the cross-sectional area of two molecules
coupled end on through their ionic headgroups. As shown in
Figure 9, this parameter is not constant but increases slowly
with the length of the alkyl chains, so proving beyond question
d [ä] ˆ 16.87Æ0.15 ‡ 1.228Æ0.015
Chem. Eur. J. 2002, 8, No. 10
n
(5)
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2251 $ 20.00+.50/0
2251

P. Masson et al.
FULL PAPER
(Figure 10), permit the detection of the phase transitions quite
clearly: One transition at about 808C between two crystal
forms and one other at about 1208C between a crystal and the
smectic phase. In the crystalline state at room temperature,
Figure 9. Variation of S with the chain length for guanidinium alkylben-
zenesulfonates in the smectic state at 1908C. Solid circles correspond to
volumes measured experimentally, and open circle corresponds to a
volume determined by extrapolation.
Figure 10. IR absorption spectra of guanidinium dodecylbenzenesulfonate
that the bulkiness of the chains does play an important part in
the packing of the ionic species.
in the 3000 3600 cmÀ1 N H stretching region, recorded as a function of
À
increasing temperature in the range from 308C (upper curve) to 2208C
(lower curve) with steps of 108C.
It is important to note that the value of S (ranging from
20.3 ä2 for GABS-8 to 21.4 ä2 for GABS-14) is actually
slightly smaller than that (acsinb/4 ˆ 22.6 ä2) measured
crystallographically for the guanidinium benzenesulfonate
crystals.[1] Surprising at first sight, this result is, however, easy
to understand–it simply originates in the puckering of the
ionic layers, that is, in the distortion of the hydrogen-bonded
sheets (by bending of the hydrogen bonds) from strict
planarity due to the packing requirements of the bulky
residues.[1] As shown in Figure 1a, the ionic species in the
puckered sheets are indeed alternately shifted up and down
along the c direction, and the interdigitated benzene rings
alternately tilted left and right with respect to the layer
normal to maximise van der Waals contacts and to accom-
modate closer packing. Moreover, as often mentioned in the
literature,[11] the benzene rings adopt an edge-to-face arrange-
ment in the nonpolar sublayers to satisfy p interactions
(Figure 1d). Now, in the presence of long alkyl chains
preventing the interdigitation of the benzene rings, the
molecules find themselves arranged in double rows oriented
along the a direction (Figure 1c and e). Strengthened by the
p-interactions and leaving empty spaces between them, these
rows have a possibility to draw nearer to one another and so to
allow a reduction of the molecular area S by accentuating the
puckering of the ionic sheets.
the spectra contain four absorption bands (3198, 3263, 3351
and 3430 cmÀ1, as determined by a least-squares multiple-
Lorentzian fit of the experimental data, see for instance
Figure 11) that are very close to those reported in the
literature for GABS-0 (guanidinium benzenesulfonate)
(3186, 3258, 3341 and 3341 3400 cmÀ1).[1] Evidently, the
Figure 11. Deconvolution of the IR absorption spectrum of guanidinium
dodecylbenzenesulfonate in the crystalline state at room temperature.
Infrared spectroscopy: Since the smectic phases described
belong to the class of the so-called disordered smectic phases,
in which the in-layer ordering extends over only a few
molecules, the question immediately arises of whether the
hydrogen bonding that holds the molecules together in the
crystal survives at high temperature in the liquid crystal. In a
search of specific information about this bonding, one of the
compounds, namely GABS-12, was carefully investigated by
infrared spectroscopy.
hydrogen-bonding pattern of the molecules in both GABS-12
and GABS-0 is essentially the same. In the smectic phase at
high temperature, on the other hand, the absorption bands,
which are lower but wider, have basically the same position
(3186, 3282, 3369 and 3442 cmÀ1) and the same area (84 cmÀ1
at 1508C, instead of 101 cmÀ1 at 308C) as in the crystal. The
number of hydrogen bonds is therefore little affected by the
melting process–even though disorderly assembled within
the smectic layers, the molecules continue to be extensively
hydrogen-bonded to form rather large two-dimensional
supermolecules. This result may appear surprising at first
The spectra recorded as a function of increasing temper-
ature, particularly in the 3000 3600 cmÀ1 N H stretching
À
region where the hydrogen bonding shows up most clearly
2252
¹ WILEY-VCH Verlag GmbH, 69451Weinheim, Germany, 2002
0947-6539/02/0810-2252 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 1 0

Smectic Liquid Crystals
2248 2254
elemental analysis calcd (%) for C20H34 (274.49): C 87.52, H 12.45; found C
87.60 H 12.45.
sight on account of the disordering that takes place inside the
layers at the crystal to smectic phase transition; it is, however,
easy to understand because the acquisition time of the spectra
(ꢂ10À3 s) is way larger than the characteristic time (ꢂ10À9 s)
of the local-order fluctuations in the smectic layers.
Synthesis of sodium 4-alkylbenzenesulfonates: The synthesis of the decyl
derivative, taken as an example, is as follows. Decylbenzene (from Aldrich)
(40 mmol) in dichloroethane (30 mL) was added dropwise at room
temperature with chlorosulfonic acid (40 mmol) in dichloroethane
(10 mL). After stirring overnight, 4-decylbenzenesulfonic acid was isolated
as a yellow viscous liquid (occasionally crystallising at room temperature)
by evaporation to dryness. It was carefully added under stirring with NaCl
(5 g) in water (85 mL). The sodium 4-decylbenzenesulphonate formed as a
white precipitate was filtered off and recrystallized in water (90% yield).
The whole set of sodium 4-alkylbenzenesulfonates synthesized (yields
growing from 55% for ethyl to 90% for higher than octyl) were
Conclusion
To conclude, it is appropriate to emphasise the main result of
the present work, namely, the production of supramolecular
smectic liquid crystals from guanidinium alkylbenzenesulfo-
nates. The idea that prepared the way for this venture–and
that proved perfectly valid–originated in the simple remark
that the packing area of the ionic species in the layers of
guanidinium benzenesulfonate crystals is large enough, not
only to permit the introduction of long alkyl chains in the para
position of the benzene rings without disruption of the
supramolecular hydrogen-bonded association of the ions,
but also to permit eventually their melting at high temper-
ature, leading to the formation of smectic liquid crystals. The
disordered nature of the smectic phases obtained, implying
that the ordering of the molecules inside the layers is not
extending over long distances, could at first sight be taken to
suggest that the supramolecular self-assembly of the ions
vanishes completely in the smectic state. In fact, this is
definitely not the case as proven by infrared spectroscopy. The
supramolecular organisation described is all the more satisfy-
ing as the smectic phases, rather fluid, reach their thermody-
namic equilibrium easily.
1
3
characterised by H NMR: d ˆ 7.48 (d, J(H,H) ˆ 8.0 Hz, 2H, Ar-H ortho
SO3), 7.10 (d, 3J(H,H) ˆ 8.0 Hz, 2H, Ar-H meta SO3), 2.53 (t, 3J(H,H) ˆ
7.3 Hz, 2H, Ar-CH2(CH2)nÀ2CH3), 1.52 (m, 2H, Ar-CH2CH2(CH2)nÀ3CH3),
1.22 (m, (2n À 6)H, Ar-CH2CH2(CH2)nÀ3CH3), 0.83 (t, 3J(H,H) ˆ 6.3 Hz,
3H, Ar-(CH2)nÀ1CH3).
Synthesis of GABS-n: Salts with n ꢀ8 were prepared in pure water. For
example, GABS-2 was synthesized by dissolving guanidine hydrochloride
(used as received from Aldrich) (3.1mmol) and sodium 4-ethylbenzene-
sulfonate (prepared as described above) (3.1mmol) in hot water (10 mL).
Precipitating as transparent crystals by simple cooling, the expected
product was filtered off, dried in vacuum, and controlled by 1H NMR (70%
yield). On the other hand, salts with n >8, were prepared in water/ethanol
mixtures. For example, GABS-10 was synthesized by dissolving guanidine
hydrochloride (3.1mmol) and sodium 4-decylbenzene-sulphonate
(3.1mmol) inhot water/ethanol 1:1 v/v mixture (20 mL). Precipitating by
cooling to room temperature and partial evaporation of the solvent, the
product was filtered off, dried, and controlled by 1H NMR (82% yield).
Following this procedure, the whole set of GABS-n compounds from n ˆ 2
to 14 were synthesized. To make up the shortage in guanidinium of the
long-chained compounds, due to co-crystallisation with small amounts of
unreacted sodium 4-decylbenzenesulphonate, the products obtained were
redissolved in hot water/ethanol 1:1 v/v mixture (20 mL), treated with
additional guanidine hydrochloride (0.3 mmol), and isolated by precipita-
tion and evaporation as previously. The final compounds were character-
ised by 1H NMR: d ˆ 7.48 (d, 3J(H,H) ˆ 8.0 Hz, 2H, Ar-H ortho SO3), 7.10
‡
(d, 3J(H,H) ˆ 8.0 Hz, 2H, Ar-H meta SO3), 6.95 (s, 6H, [C(NH2)3] ), 2.53 (t,
3J(H,H) ˆ 7.2 Hz, 2 H, Ar-CH2(CH2)nÀ2CH3), 1.52 (m, 2H, Ar-
CH2CH2(CH2)nÀ3CH3), 1.22 (m, (2n À 6)H, Ar-CH2CH2(CH2)nÀ3CH3),
Experimental Section
3
0.83 (t, J(H,H) ˆ 6.3 Hz, 3H, Ar-(CH2)nÀ1CH3)]; elemental analysis calcd
General: 1H NMR spectroscopy: Bruker spectrometer at 200 MHz in
[D6]DMSO, 258C. Elemental analysis: performed at Institut Charles
Sadron, Strasbourg. Thermogravimetry: Setaram TGA 92, argon flow,
heating rate of 108CminÀ1. Polarizing optical microscopy: Leitz-Orthoplan,
Mettler FP82 hot stage. Differential scanning calorimetry: Perkin Elmer
DSC7, heating-cooling rates of 108CminÀ1. Dilatometry: home-made
computer-driven dilatometer, 1g of carefully degassed sample immersed
in about 10 mL of mercury, heating and cooling steps of 0.18C every 2 min.
X-ray diffraction: Guinier focusing camera, CuKa1 radiation, powder
samples in Lindemann capillaries, INSTEC hot stage, INEL CPS-120
curved position-sensitive detector. Molecular modelling: MSI Insight II
software, Silicon Graphics. Infrared spectroscopy: ATI Mattson FTIR
spectrometer, resolution of 4 cmÀ1, samples in KBr pellets heated in a
home-made temperature-controlled hot stage.
(%) for GABS-2 (245.08): C 44.07, H 6.16, N 17.13; found C 43.88, H 6.21, N
17.13; elemental analysis calcd (%) for GABS-4 (273.11): C 48.33, H 7.01, N
15.37; found C 48.27, H 7.02, N 15.33; elemental analysis calcd (%) for
GABS-6 (301.15): C 51.81, H 7.69, N 13.94; found C 51.66, H 7.73, N 13.82;
elemental analysis calcd (%) for GABS-8 (329.18): C 54.69, H 8.26, N
12.75; found C 54.61, H 8.30, N 12.75; elemental analysis calcd (%) for
GABS-10 (357.21): C 57.11, H 8.75, N 11.76; found C 57.28, H 8.80, N 11.75;
elemental analysis calcd (%) for GABS-12 (385.24): C 59.19 H 9.15, N
10.90; found C 59.60, H 9.26, N 10.79; elemental analysis calcd (%) for
GABS-14 (413.27): C 60.98, H 9.50, N 10.16; found C 60.90, H 9.54, N 10.22.
Synthesis of tetradecylbenzene: Tetradecylbenzene, which is not commer-
Acknowledgement
cially available, was synthesized by Kumada×s cross-coupling reaction.[12]
A
Grignard reagent was prepared under argon by slow addition of 1-bromo-
tetradecane (0.2 mol) to magnesium (0.2 mol) in dry ether (100 mL) and
heating at reflux for 2 h. Chlorobenzene (0.16 mol) and NiCl2(dppp)
(0.4 mmol) used as a catalyst were dissolved in diethyl ether (60 mL) and
added with the ether solution of the Grignard reagent; the mixture, turning
from red to pale green, was then heated at reflux overnight. After cooling
to 08C, the mixture was hydrolysed with aqueous 2n HCl (100 mL). The
organic layer was washed with water and a saturated aqueous solution of
NaHCO3, the organic layer was dried over MgSO4 and evaporated to
dryness. The liquid tetradecylbenzene obtained was finally purified by
distillation at 164 1668C (7 mmHg) (71% yield). 1H NMR: d ˆ 7.18 (d,
3J(H,H) ˆ 8.0 Hz, 5H, Ar-H), d ˆ 2.59 (t, 3J(H,H) ˆ 7.6 Hz, 2H, Ar-
CH2CH2)12CH3), 1.57 (m, 2H, Ar-CH2CH2(CH2)11CH3), 1.26 (m, 6H, Ar-
CH2CH2(CH2)11CH3), 0.88 (t, 3J(H,H) ˆ 6.35 Hz, 3H, Ar-(CH2)13CH3)];
The authors wish to thank Dr. B. Heinrich for skilled assistance and helpful
discussions in the dilatometry experiments.
[1] V. A. Russell, M. C. Etter, M. D. Ward, J. Am. Chem. Soc. 1994, 116,
1941 1952.
[2] A. I. Kitaigorodskii, Organic Chemical Crystallography 1961, Con-
sultants Bureau, New York, p. 180.
[3] P. Seurin, D. Guillon, A. Skoulios, Molec. Cryst. Liq. Cryst. 1981, 71,
51 63.
[4] D. Guillon, A. Skoulios, J. J. Benattar, J. Phys. Fr. 1986, 47, 133 138.
[5] E. E. Gilbert, Synthesis 1969, 1, 3 5.
[6] D. Guillon, A. Skoulios, Mol. Cryst. Liq. Cryst. 1977, 39, 139 157.
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2253 $ 20.00+.50/0
2253

P. Masson et al.
FULL PAPER
[7] B. Heinrich, A. Halbwachs, A. Skoulios, D. Guillon, unpublished
results.
[11] F. Vˆgtle, C. Seel, Angew. Chem. 1992, 104, 542 563; Angew. Chem.
Int. Ed. Engl. 1992, 31, 528 549.
[8] D. Kardassi, D. Tsiourvas, C. M. Paleos, B. Heinrich, A. Skoulios, Mol.
Cryst. Liq. Cryst. 1999, 326, 49 54.
[12] M. Kumada, Pure Appl. Chem. 1980, 52, 669 679; M. Kumada, K.
Tamao, K. Sumitani, Org. Syn. 1988, Vol. VI, 407 411.
[9] A. Skoulios, Ann. Phys. 1978, 3, 421 450.
[10] D. Guillon, A. Skoulios, J. Phys. Fr. 1984, 45, 607 621.
Received: October 4, 2001[F3593]
2254
¹ WILEY-VCH Verlag GmbH, 69451Weinheim, Germany, 2002
0947-6539/02/0810-2254 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 1 0