Utilization of evaporation during the crystallization process: Self-templation of organic parallelogrammatic pipes

Article abstract of DOI:10.1002/chem.200801408

Analogues of 4-dodecyloxy-2-trifluoromethylbenzamide (12FH2) consisting of a hydrophobic alkyl chain, a trifluoromethylated aromatic ring, and a self-complementary hydrogenbonding amido group were synthesized, and the structural effect of each component o

Full text of DOI:10.1002/chem.200801408

DOI: 10.1002/chem.200801408
Utilization of Evaporation during the Crystallization Process:
Self-Templation of Organic Parallelogrammatic Pipes
Myungeun Seo,[a] Jung Hak Kim,[a] Gon Seo,[b] Chae-Ho Shin,[c] and Sang Youl Kim*[a]
Abstract: Analogues of 4-dodecyloxy-
2-trifluoromethylbenzamide (12FH2)
consisting of a hydrophobic alkyl chain,
appropriate molecular balance between
the benzamide and alkyl groups. Scan-
ning electron microscopy and powder
XRD analyses of solids obtained by a
drying-mediated assembly process re-
vealed that production of the L poly-
morph by polymorphic transition from
the S polymorph resulted in evolution
of a three-dimensional structure when
the alkyl group has more than 12
carbon atoms. Among the series of
compounds, 12FH2 and 4-tetradecy-
loxy-2-trifluoromethylbenzamide
atomic force microscopy study of
12FH2 suggested that a single pipe
may be composed of platelike crystalli-
tes of L polymorph. From a mercury-
intrusion porosimetry study, it was de-
termined that macroporous materials
with average pore diameters of about
40 mm and porosity of about 80% were
obtained. The previously proposed self-
templation mechanism by polymorphic
transition from S to L polymorph was
further discussed in view of poly-
morphism and the crystallization rate.
An appropriate molecular balance be-
tween the benzamide and alkyl groups
is necessary to induce a proper poly-
morphic transition for the development
of a three-dimensional hollow structure
in the evaporation process.
a
trifluoromethylated aromatic ring,
and a self-complementary hydrogen-
bonding amido group were synthesized,
and the structural effect of each com-
ponent on the formation of parallelog-
rammatic pipes was investigated. Dif-
ferential scanning calorimetry and
powder XRD analyses revealed that
all-trans L and gauche-rich S poly-
morphic forms appeared for the ana-
logues with more than eight carbon
atoms in the alkyl chain, that is, the
polymorphism originates in the confor-
mation of the alkyl groups and hydro-
gen-bonding patterns of the benzamide
group. Also, the trifluoromethyl sub-
stituent is crucial in that it provides an
(14FH2) formed parallelogrammatic
pipes with micrometer dimensions. An
Keywords: amides · crystal growth ·
hydrogen bonds · polymorphism ·
porous materials · self-assembly
Introduction
bilities in manufacturing functional nano- and microstruc-
tures on a large scale.[1–4] The structures of self-assembled
materials largely depend on their thermodynamic energy
levels at equilibrium states. When self-assembly proceeds
under nonequilibrium conditions, however, the process
Self-assembly is a powerful tool for fabrication of two- and
three-dimensional structures, and spontaneous formation of
complex supramolecular structures opens promising possi-
strongly influences the structure of self-assembled ACHTUNGTRENNUNGmater-
ials.[2a,5a] Development of different macroscopic orders and
shapes from a homogeneous solution is frequently observed
in nonequilibrium self-assembly owing to a different evolu-
tion process.[6,7]
One dramatic example of this can be found in a system in
which irreversible evaporation of the solvent leads the
system far from the equilibrium.[8b] A continuous change of
concentration and temperature brings about a subsequent
change of interactions between the components and gener-
ates nonequilibrium phenomena such as Rayleigh–Bꢀnard
convection flow and fingering inhomogeneity.[5] Physical
properties of solutes and solvents such as dewetting, surface
tension, and intermolecular interaction strongly affect the
[a] Dr. M. Seo, J. H. Kim, Prof. S. Y. Kim
Department of Chemistry and School of Molecular Science (BK21)
KAIST
Gwahangno 335 (373-1) Guseong-dong, Yuseong-gu
Daejeon 305-701 (Korea)
Fax : (+82)42-350-8177
E-mail: kimsy@kaist.ac.kr
[b] Prof. G. Seo
School of Applied Chemical Engineering and
The Center for Functional Nano Fine Chemicals
Chonnam National University, Gwangju 500-757 (Korea)
[c] Prof. C.-H. Shin
Department of Chemical Engineering
Chungbuk National University, Chungbuk 361-763 (Korea)
612
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 612 – 622

FULL PAPER
formation of architectures.[9b] Much effort has concentrated
on fabrication of mesoporous materials and arrays of nano-
particles/microspheres by drying-mediated self-assem-
bly,[8,10,11] but there have been few reports on the use of or-
ganic materials in the fabrication of hierarchically structured
self-assembled materials.[9,12]
Recently, we reported on the formation of parallelogram-
matic pipes with micrometer dimensions through drying-
Figure 1. Illustration of two-dimensional hydrogen-bonded network of
mediated self-assembly of
a small organic compound,
benzamide groups with R2(8) and C(4) synthons.
namely, 4-dodecyloxy-2-trifluoromethylbenzamide
(12FH2).[13] Pipes with a unique shape were readily prepared
by evaporation of a hot ethanolic solution of 12FH2 on a sil-
icon wafer within a minute. Observation of the existence of
two polymorphs and transition in their population during
the evolution process led us to propose self-templated
growth of a hollow structure, in which one polymorph
serves as a template and the other as a wall. This demon-
strates that the utility of drying-mediated assembly can be
expanded into the micrometer regime. In addition, it pro-
vides a novel route for the generation of unique shapes with
lower symmetries in self-assembled structures.
structural components: alkoxyl and trifluoromethyl groups.
We designed analogues of 12FH2 to examine the structural
effect of each component on self-assembly of the benzamide
moiety, especially on the ability to form parallelogrammatic
pipes. Synthetic routes to the compounds are shown in
Scheme 1.
The synthesis was based on a nitrile-activated nucleophilic
aromatic substitution (SNAr) reaction to build an ether link-
age between the benzamide and alkyl groups. The nitrile
group was transformed into a benzamide group by subse-
quent hydrolysis. The alkyl group was varied by treating the
starting compound 4-nitro-2-trifluoromethylbenzonitrile (1)
with the corresponding aliphatic alcohols. The trifluoro-
methyl group also played an important role in the reaction:
when potassium hydroxide was used in water/ethanol, hy-
drolysis of the nitrile group stopped at the amide stage in-
stead of yielding carboxylic acid when the carbon number of
the alkyl chain was less than 14 (Method I). Generation of
the carboxylic acid from 4-dodecyloxy-2-trifluoromethylben-
zonitrile required very severe hydrolytic conditions, reflect-
ing the stability of the corresponding benzamide. It appears
that the trifluoromethyl group stabilizes the intermediate
benzamide structure during hydrolysis, through an electronic
effect and weak hydrogen bonding between fluorine and hy-
drogen atoms.[21] Thus, a series of molecules with different
lengths of the alkyl groups including 4-octyloxy-2-trifluoro-
methylbenzamide (8FH2), 4-decyloxy-2-trifluoromethylben-
zamide (10FH2), and 12FH2 were synthesized. When the
Here we report on the syntheses of structural analogues
of 12FH2, their polymorphic nature, and self-assembly
ACHTUNGTRENNUNGbehavior.
Results and Discussion
Synthesis: The benzamide group is a well-known hydrogen-
bonding motif in crystal engineering.[14] Basically, it is a self-
complementary motif: in self-association of benzamides, two
kinds of supramolecular synthons can be defined and de-
scribed as R22(8) (face-to-face hydrogen bond involving syn
hydrogen atoms) and C(4) (side-to-side hydrogen bond in-
volving anti hydrogen atoms) according to Etterꢂs rule
(Figure 1).[15] Combination of two synthons yields the two-
dimensionally hydrogen bonded network observed in the
crystal structure of molecules containing benzamide groups,
such as terephthalamide.[16]
While most studies concerning benzamide compounds
have concentrated on cocrystal-
lization with other hydrogen-
bonding motifs and subsequent
control of hydrogen-bonding
patterns,[17] there have been a
surprisingly small number of re-
ports on utilizing them as a
structural motifs. A few papers
have illustrated the growth of
conducting single crystals,[18]
physically cross-linked net-
works,[19] and reinforcement of
the polymer interface.[20]
The novel benzamide com-
pound we reported, 4-dodecyl-
oxy-2-trifluoromethylbenza-
mide (12FH2), contains two Scheme 1. Synthetic routes for a series of compounds containing benzamide and alkyl groups.
Chem. Eur. J. 2009, 15, 612 – 622
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
613

S. Y. Kim et al.
carbon number was greater than 12, the use of potassium
hydroxide was insufficient to complete the hydrolysis. In
these cases, hydrogen peroxide in DMSO solution yielded
the desired products 4-tetradecyloxy-2-trifluromethylbenza-
mide (14FH2) and 4-hexadecyloxy-2-trifluoromethylbenza-
mide (16FH2) (Method II).[22]
Also, 4-dodecyloxybenzamide (12HH2) was synthesized
to examine the effect of the trifluoromethyl group on self-
assembly. 4-Dodecyloxybenzonitrile was obtained by the re-
action of 4-cyanophenol with 1-bromododecane, and the ni-
trile group was hydrolyzed with sodium percarbonate, which
is known to selectively generate a primary aromatic amide
when there is no substituent at the ortho position.[23]
Investigation of polymorphism: The benzamide group, dis-
covered in 1832 by Wçhler and Liebig,[24] was the first mo-
lecular polymorph to be identified. Recently, the crystal
structures of two different polymorphs of benzamide were
revealed, 175 years after the discovery of polymorphism.[25]
The different polymorphs show different p–p interaction
patterns (i.e., face-to-face versus face-to-edge interaction),
although the two-dimensional hydrogen-bonded network is
conserved. However, the possibility of face-to-edge interac-
tion was neglected here due to the existence of an alkyl
chain along the edge direction that blocks approach of the
other aromatic ring.
A previous study revealed that 12FH2 also has two poly-
morphic forms.[13] The L polymorph, which has a lamellar
structure with a longer layer spacing of d=2.39 nm, was ob-
served in a solid crystallized from a melt and from n-
hexane. X-ray crystallographic analysis revealed that the L
polymorph has an intercalated head-to-head bilayer struc-
ture with all-trans conformation of the alkyl groups, induced
by a hydrogen-bonded network of benzamide groups with
strong side-to-side hydrogen bonds along the layer and face-
to-face hydrogen bonds between the layers. The S poly-
morph, another lamellar structure with a shorter layer spac-
ing of d=1.48 nm, was observed in a solid crystallized from
polar solvents such as ethanol. Based on FTIR and powder
X-ray diffraction (XRD) analyses, the structure was inter-
preted as a different head-to-head bilayer structure with
strong face-to-face hydrogen bonds and significantly disor-
dered alkyl groups having a large fraction of gauche confor-
mation. The polymorphic transition from S to L polymorph
during the irreversible evaporation of ethanol was suggested
to be a key step in the formation of unique macroporous
parallelogrammatic pipes. Thus, further investigation of the
polymorphism for synthesized analogues of 12FH2 was con-
ducted to obtain more information about the formation of
parallelogrammatic pipes.
Figure 2. Powder XRD patterns of 8FH2–16FH2 and 12HH2. a) Solids
crystallized from the melt. b) Solids crystallized from ethanol. Patterns
were normalized by the intensity of the largest peaks.
peaks corresponding to (001) diffraction than the patterns in
Figure 2b, with the exception of 8FH2, which shows identi-
cal patterns in both. Also, from 10FH2 to 16FH2, the d
spacing of the (001) diffraction increased with increasing
length of the alkyl groups. More importantly, the patterns of
analogues 10FH2, 14FH2, and 16FH2 were consistent with
that of 12FH2, that is, the L and S polymorphic forms are
preserved throughout the series. It appears that analogues
10FH2–16FH2 have the same polymorphism as 12FH2: S
polymorphs are obtained by crystallization from ethanol,
and L polymorphs by crystallization from the melt.
In Figure 2a, the peaks commonly appearing at about 208
were assigned to crystalline oligomethylene.[26] Growing in-
tensity of the peaks from 8FH2 to 16FH2 indicates that crys-
talline packing of the alkyl groups is favored for higher
numbers of carbon atoms. Also, slight shifts of the peaks for
8FH2 and 10FH2 to wider angles compared to 12FH2–
16FH2 suggest that the crystalline structure of the alkyl
groups is slightly different. In Figure 2b only 10FH2 showed
a peak with strong intensity at about 208, that is, crystalliza-
Figure 2 shows the powder XRD patterns of the ana-
logues, which were obtained from solids crystallized from
the melt (Figure 2a) and from ethanol solution (Figure 2b).
Both patterns clearly show peaks corresponding to (00n) dif-
fractions with remarkably high intensities, that is, all of the
analogues have a lamellar structure. The patterns illustrated
in Figure 2a showed much longer d spacing for the first
614
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 612 – 622

Self-Templation of Macroporous Pipes
FULL PAPER
tion from ethanol produces poorly crystallized alkyl groups.
In the case of 10FH2, the length of the alkyl group may be
appropriate to form highly crystalline packing in the prede-
fined geometry of the S polymorphic form, which is domi-
nated by the hydrogen bonds of benzamide in ethanol.
However, the spacing would not be suitable for the shorter
alkyl group in 8FH2 or the longer alkyl groups in 12FH2–
16FH2, and would result in poor crystalline packing.
By taking into consideration previously reported FTIR
and 13C solid-state CP-MAS NMR studies,[13] molecular
packings of the L and S polymorphic forms were suggested
(Figure 3). Crystallization from a melt would favor produc-
tion of the L polymorph with a stable packing of alkyl
groups and aromatic rings. At the cost of a large fraction of
gauche conformation in the alkyl group, the S polymorph
appears to have a more optimal geometry for face-to-face
hydrogen bonding between benzamide groups. This is an ex-
ample of solvent-mediated polymorphism, in which the sol-
vent molecules interact with the solute and preferentially
produce a certain polymorph.[27] In the case of 8FH2, it ap-
pears that face-to-face hydrogen bonds are not preferred,
because the alkyl group cannot afford a gauche-rich confor-
mation.
Compound 12HH2 also follows the aforementioned trend
in showing two polymorphs; however, the solid obtained
from ethanol consisted of both polymorphs, whereby that
with shorter d spacing was proven to be the same as the
polymorph obtained from the melt. The extremely poor sol-
ubility of 12HH2 resulted in rapid crystallization, even from
ethanol. However, the occurrence of the other polymorph
with longer d spacing in ethanol suggests that the lamellar
structure of 12HH2 is substantially different from that of
12FH2. Also, 12HH2 did not show peaks corresponding to
crystalline packing of oligomethylene chains in either poly-
morphic form: instead, a strong peak appeared at about 238
in both polymorphic forms, and it was assigned to p–p stack-
ing of the aromatic rings. This difference indicates that the
trifluoromethyl group at the ortho position of the benzamide
moiety plays an important role in the crystal structure and
for the solubility. The role of the trifluoromethyl group may
be disruption of p–p stacking of the aromatic ring to pro-
vide good solubility in common organic solvents. Also, it
may stabilize the face-to-face hydrogen bonds of benzamide
moieties via F···H interaction.[21] This leaves the Hanti atom
to be solubilized by ethanol molecules and leads to forma-
tion of the S polymorph.
The fact that 8FH2 does not exhibit polymorphism pro-
vides strong evidence for the origin of the polymorphism:
an alkyl group consisting of more than 10 carbon atoms in-
duces polymorphism based on difference in conformation.
The octyl group may have only one conformation, because
it cannot provide sufficient space for the benzamide groups
to form stronger face-to-face hydrogen bonding. Based on
the powder XRD analysis, the crystal structure of 8FH2 was
assigned as the L polymorph, because the d spacing was too
large for the S polymorph.
To gain further insight into the polymorphism and the rel-
evant thermal transitions, a differential scanning calorimetry
(DSC) study was conducted. Figure 4 shows DSC thermo-
grams of the solids crystallized from ethanol. In the first
heating cycle, the series 8FH2–16FH2 commonly showed a
large endothermic peak around 1058C (Figure 4a). This
peak was assigned as the melting transition (Tm) of the com-
pounds, in which the hydrogen bonds between benzamide
groups are cleaved. When the carbon number of the alkyl
group was greater than ten, an endothermic premelting tran-
sition (Tpm) appeared in prior to the melting transition, and
Figure 3. a) Illustration of crystal structures of L and S polymorphic forms appearing in 8FH2–16FH2. Red cross: benzamide group. Blue rectangles:
alkyl group. Green rectangle: ethereal oxygen atom. b) Suggested hydrogen-bonded molecular packing of L (top) and S (bottom) polymorphic forms of
12FH2. White H, gray C, red O, blue N, yellow F.
Chem. Eur. J. 2009, 15, 612 – 622
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
615

S. Y. Kim et al.
Figure 4. DSC thermograms of the compounds under N2 at a heating rate of 58CminÀ1 and a cooling rate of 28CminÀ1. Peak intensities were normalized.
a) First heating scan of 8FH2–16FH2. b) First cooling scan of 8FH2–16FH2. c) Second heating scan of 8FH2–16FH2. d) First heating, first cooling, and
second heating scan of 12HH2.
the transition temperature increased with increasing carbon
number. However, no morphological change was observed
around that temperature under the polarized optical micro-
scope. Thus, it was assigned to the phase transition corre-
sponding to a change in the conformation of the alkyl chain
from an ordered to disordered structure.[28]
Figure 4b shows the thermograms recorded during the
first cooling cycle. Large exothermic peaks with a tilted
shape were assigned to abrupt crystallization of the com-
pounds from the melt. Compounds 12FH2, 14FH2, and
16FH2 showed an additional exothermic peak, and the peak
temperature increased with increasing carbon number. This
exothermic peak was assigned to rearrangement of alkyl
chains. The above DSC results indicate that hydrogen bond-
ing and rearrangement of the alkyl chains occur at different
temperatures.
to the L polymorph. Indeed, the thermograms of the second
heating cycle were identical to those of the L polymorph,
and the transition temperature corresponding to the confor-
mational change of the alkyl group decreased in all cases.
This suggests that the conformation of the alkyl groups is
different between the L and S polymorphs. The temperature
was maintained during the third and fourth scans (data not
shown), that is, the L polymorphic form is preserved.
The enthalpies corresponding to the melting of the hydro-
gen bonds of 30–40 kJmolÀ1 reflect the strength of two-di-
mensional hydrogen bonds between benzamide moieties.
The enthalpies corresponding to the transition of alkyl
chains were much smaller, in the range of 10–20 kJmolÀ1
(17.7 for 12FH2, 18.9 for 14FH2, and 16.9 kJmolÀ1 for
16FH2 in the first heating cycle). When divided by the
number of carbon atoms in the alkyl chains, the enthalpies
for the transitions during the first heating cycle were about
1 kJmolÀ1. These values are similar to the reported premelt-
ing transition enthalpies of n-alkanes.[26b]
Because crystallization from the melt yields only L poly-
morph, the initial S polymorph during the first heating cycle
should be transformed into the L polymorph. Therefore, the
second cycle should show thermal transitions corresponding
616
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 612 – 622

Self-Templation of Macroporous Pipes
FULL PAPER
Also, the effect of the trifluoromethyl group at the ortho
position of the benzamide moiety was examined for 12FH2
and 12HH2. Similar to 12FH2, 12HH2 showed two transi-
tions corresponding to the conformational change of the
alkyl group and the cleavage of the hydrogen bonds be-
tween benzamide groups. However, the much higher melting
temperature (ca. 408C) indicated that the hydrogen-bonded
network of 12HH2 is stronger due to the absence of the
sterically bulky trifluoromethyl group.
Drying-mediated assembly: To utilize the difference be-
tween the two polymorphs in assembly, we evaporated etha-
nolic solutions of 8FH2–16FH2 and 12HH2 on a silicon
wafer. Evaporation of a solution on a substrate is in princi-
ple a nonequilibrium process: it is complex enough to form
dissipative structures, and irreversible because it freezes the
final structure.[5a,8b] Although complete understanding of the
structural evolution during evaporation is very difficult, the
coffee-stain mechanism and the pinhole mechanism have
been applied to explain the formation of unique struc-
tures.[29,30] The coffee-stain mechanism explains the forma-
tion of ringlike stains from solution droplets by precipitation
of solute at the perimeter of the droplet and subsequent out-
ward flow of liquid from the interior to the edge of the
drop. The pinhole mechanism has been applied when the
formation of holes in the thinning liquid films accounts for
ring formation. The evaporation speed is higher at the edge
of the pinhole than in the bulk of the solution film, and
inward flow of solute concentrates the ring structure.[9b]
However, to the best of our knowledge, drying-mediated
assembly has not been applied to crystalline compounds. We
speculated that rapid evaporation at high temperature
would provide an interesting environment in which both
crystallization in the solution and crystallization from the
melt can occur, yielding unique crystal morphologies com-
posed of two polymorphs. In a typical experiment, 20% (w/
v) ethanolic solutions of the compounds at around 1008C
were simply dropped onto a silicon wafer at room tempera-
ture and allowed to evaporate. Figure 5 shows selected SEM
images of the resulting solids after complete evaporation.
In the cases of 8FH2 (Figure 5a) and 10FH2 (Figure 5b),
two-dimensional sheetlike morphologies with cracks and
holes formed by evaporation of the solvent were observed.
Formation of the structure can be explained by the pinhole
mechanism, including thinning of the concentrated solution
on the substrate. However, we observed that rapid crystalli-
zation during evaporation strongly affected the drying-medi-
ated assembly process by defining the growth direction of
the structure, especially at high concentration. In the case of
8FH2, linear holes developed rather than the usual spherical
holes, due to crystalline growth around the boundaries of
the pinholes. Compound 10FH2, which is more crystalline
due to the longer alkyl group, developed cracks with lengths
of several tenths of a micrometer along the domain bounda-
ries. It appears that the crystallization of 10FH2 was initiat-
ed mainly by the decrease in temperature and rapidly
propagated in two dimensions. These results also imply that
Figure 5. SEM images of solids obtained from ethanolic solution (20%
w/v) by drying-mediated assembly process. a) 8FH2. b) 10FH2. c,
d) 12FH2. e, f) 14FH2. g) 16FH2. h) 12HH2. Samples were prepared by
dropping hot ethanolic solutions of the compounds onto a silicon wafer
at room temperature, followed by evaporation of ethanol.
the number of carbon atoms in the alkyl group determines
the crystallization rate of the compounds, and this number
should exceed ten in order to grow three-dimensional crys-
tals during the evaporation process.
In the cases of 12FH2 (Figure 5c and d) and 14FH2 (Fig-
ure 5e and f), evaporation resulted in an unusual crystal
growth. The SEM images clearly show that hemispherical
bundles of rectangular pipes with micrometer size formed
over a wide region. The observation of hemispherical bun-
dles spread over the substrate indicates that Bꢀnard convec-
tion flow was locally produced due to the temperature dif-
ference between the surface of the solution drop and the
substrate. The flow originates from the region near the sub-
strate, where it is relatively hot. Due to the temperature gra-
dient from the surface to the substrate, which is generated
by subsequent cooling, the flow rises up with the aid of
buoyant force. When the flow reaches the surface of the
drop and contacts cold air, the temperature abruptly de-
creases and drives the flow downward. At that point, crystal-
lization appears to be initiated from a locally supersaturated
solution and propagates along the direction of the sinking
Chem. Eur. J. 2009, 15, 612 – 622
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
617

S. Y. Kim et al.
flow, forming hemispherical bundles. Because of the appear-
ance of platelike crystals at the boundary of the droplet, the
evaporation process is thought to occur by a coffee-stain
mechanism.
The cross section of the pipes obtained from 12FH2 was
about 20ꢃ15 mm with an interior angle of À808 and an exte-
rior angle of À1008, similar to those of 14FH2. The highest
end of the pipe was about 1 mm high. The thickness of the
wall was about 150 nm for 12FH2 and about 300 nm for
14FH2. Although most pipes had a parallelogrammatic
shape, some were more complex in shape. Platelike crystals
were observed in regions around the bundle.
Although a small number of studies have reported the for-
mation of organic self-assembled tubes of micrometer
size,[31,32] macroporous nature of the tubes, which is impor-
tant for practical applications, has yet to be fully realized.
We used mercury-intrusion porosimetry to evaluate the
average pore diameter of 12FH2 and 14FH2. The pore size
distribution of 12FH2 and 14FH2 is depicted in Figure 6,
which shows that macropores with a median pore diameter
of 47.0 mm for 12FH2 and 34.4 mm for 14FH2 were devel-
oped. The total pore area was calculated as 77.2 m2 gÀ1 for
12FH2 and 50.2 m2 gÀ1 for 14FH2. Also, the multimodal
shape of the distribution suggests that the size of the initial
S polymorph, which acts as a template, is broadly distributed
because the growth process occurs in heterogeneous envi-
ronments. The porosity of the samples was 73.0 and 80.6%
for 12FH2 and for 14FH2, respectively.
The next task was to identify the supramolecular structure
of the parallelogrammatic pipes, which should have two dif-
ferent faces and junctions between them. Using an atomic
force microscope (AFM), we scanned two different faces of
the pipe of 12FH2 and obtained the height image of each
face. The face depicted in Figure 7a showed a considerably
flat surface consisting of several
Figure 6. Pore size distribution of parallelogrammatic pipes obtained
from ethanolic solution (20% w/v) by drying-mediated assembly.
a) 12FH2. b) 14FH2.
terraces and steps. The height
of the steps was about 1.6 nm
on average. This value did not
match any repeat distance ob-
served in the powder XRD pat-
tern of 12FH2, but was smaller
than the d spacing of 2.39 nm.
Thus, it was thought to be a
stack of lamellar layers, and we
envisioned the whole face as
the (100) surface of a horizon-
tally stacked lamellar structure
of the L polymorph. The other
face, depicted in Figure 7b, re-
vealed a highly rough surface
on which the ledges of platelike
structures all protruded verti-
cally. Following the assignment
of the (100) surface, the face
should be the (010) surface of
the L polymorph. The distance
Figure 7. AFM images of pipes of 12FH2. a) Image of horizontal face. b) Image of vertical face. c) Proposed
between the plates was about structure based on the results.
618
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 612 – 622

Self-Templation of Macroporous Pipes
FULL PAPER
1.8 nm on average, similar to the height of the steps ob-
served in Figure 7a.
rare that crystals can be exploited in the formation of
unique hollow structures which are macroscopic objects with
controlled symmetries.[32] A previous study for 12FH2 indi-
cated that the appearance of an L polymorph consisting of
pipes is actually the result of the polymorphic transition
from the initial S polymorph, and this is crucial for the for-
mation of a parallelogrammatic cross section of micrometer
size.[13] Compared to other methodologies, utilization of
polymorphism may be advantageous as it does not require
additional materials.
Based on the morphology shown in Figure 7 and consider-
ation of the anisotropic unit-cell structure of the L poly-
morph, we could draw the three-dimensional structure of
the pipe (Figure 7c). It is reasonable to regard the pipe as a
hollow crystal of the L polymorph. The stacked lamellar
structures of similar thicknesses observed at both faces,
however, suggest that the whole pipe is composed of small
crystallites, similar to a sectioned column. This picture also
fits the drying-mediated assembly process, because rapid nu-
cleation due to evaporation is likely to generate small crys-
tallites and promote their aggregation.
Powder XRD measurements were carried out to investi-
gate the composition of solids obtained by evaporation.
Figure 8 depicts XRD patterns of the compounds after com-
The features described above strongly support the tenta-
tive mechanism we previously suggested for the origin of
the formation of the unusual parallelogrammatic cross sec-
tion of micrometer-size, that is, polymorphic transition of S
into L polymorph during evaporation.[13] An XRD study of
quenched samples of 12FH2 and 14FH2 verified that crys-
tallization of the S polymorph occurs in the drop of solution
as it becomes saturated during evaporation. Thus, the initial
solvent-rich crystal composed of the S polymorph should be
transformed into the L polymorph on contact with air. The
existence of the hollow space inside the pipes suggests that
the transformation is initiated from the surface of the crystal
to form a wall, and a subsequent decrease in the amount of
solute leaves the inner space solvent-filled. Because of the
anisotropic shape of the molecules, growth of crystals of the
L polymorph would favor a certain direction along the [001]
axis and yield large platelike crystals on the (100) surface
and small crystallites on the (010) surface. Eventually, mac-
roscopic tubes would be generated by packing of crystals
after removal of the solvent.
In the case of 16FH2, the SEM images showed three-di-
mensionally grown crystals arranged in a thin platelike mor-
phology. The width of the crystals of about 20 mm is similar
to those of 12FH2 and 14FH2, and several platelike crystals
were attached such that they appeared to form half of a
pipe. However, no hollow structures were observed, that is,
a very sophisticated balance in the molecular structure is re-
quired. Compound 12HH2 also revealed a similar but thin-
ner morphology compared to 16FH2, consisting of fibers
with diameters of several hundred nanometers adhering to-
gether to form a petal-like structure.
Figure 8. Powder XRD patterns of the solids obtained from ethanol solu-
tion (20% w/v) by drying-mediated assembly. dL and dS denote peaks
corresponding to the (001) diffraction of L and S polymorphs of 10FH2–
16FH2. d1 and d2 denote peaks corresponding to the (001) diffraction of
each polymorph of 12HH2.
plete evaporation. The patterns clearly show that both L
and S polymorphs were simultaneously developed in the
drying process, except for the case of 8FH2.
The intensity ratio between the first diffraction peaks of
the two polymorphs (L/S ratio) varied depending on the
length of the alkyl group. The increasing L/S ratio from
10FH2 to 16FH2 indicates that compounds with longer alkyl
groups favor the L polymorph during evaporation. This sug-
gests that rapid crystallization induced by evaporation pro-
motes the formation of the L polymorph because of in-
creased intermolecular interaction between alkyl groups.
Also, the crystallization rate became faster as the L/S ratio
increased.
Investigation of structural analogues of 12FH2 reveals
that the formation of uniquely shaped pipes occurs under
very precise conditions, and also requires a sophisticated
molecular structural balance between hydrogen-bonding in-
teractions of benzamide groups and van der Waals interac-
tions of alkyl groups. Given that the benzamide moiety is es-
sential to afford the desired two-dimensional hydrogen
Molecular/mesomorphic structure relationship and criteria
for the formation of pipe structures: One of the prerequi-
sites for the formation of parallelogrammatic pipes is the
ability to develop three-dimensional structures during evap-
oration. The SEM study indicates that the molecule should
have an alkyl chain length of more than 12 carbon atoms,
thus ruling out 8FH2 and 10FH2. Also, overly crystalline
molecules such as 16FH2 and 12HH2 were inappropriate
for forming parallelogrammatic pipes, although they formed
three-dimensional structures.
The other factor is thought to be a proper polymorphic
transition during the evaporation process. It is surprisingly
Chem. Eur. J. 2009, 15, 612 – 622
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
619

S. Y. Kim et al.
bonds and the subsequent crystallinity, the length of the
alkyl group is critical for generating the hollowness based
on the alkoxybenzamide backbone. Specifically, an alkyl
group with more than ten carbon atoms is necessary for
polymorphism. The compounds consisting of an alkyl group
with 12 or 14 carbon atoms generate a three-dimensional
structure through an appropriate balance between S and L
polymorphs, where one serves as a template and the other
as a negative replica. If the process conditions are inclined
toward one polymorph, such as in the case of 8FH2 or
16FH2, a three-dimensional hollow structure cannot form.
The relationship between molecular and mesomorphic struc-
ture and the criteria for formation of pipe structures are
summarized in Figure 9
polymorphic transition from S to L polymorph. Only 12FH2
and 14FH2 met the criteria for polymorphic transition and
produced parallelogrammatic pipes.
Experimental Section
General: 4-Nitro-2-trifluoromethylbenzonitrile was synthesized according
to procedures reported elsewhere.[13] Other starting materials were ob-
tained from commercial suppliers and used without further purification.
MgSO4 was used to dry all organic solutions during workup procedures.
Analytical TLC was performed on a Kieselgel 60 F-254 precoated TLC
plates. Silica for flash chromatography was silica 60 (230–400 mesh) from
1
Merck. The H and 13C NMR spectra were recorded on a Bruker Fourier
Transform AVANCE 400 spectrometer. Chemical shifts are expressed in
parts per million relative to residual
solvent protons. Scanning electron mi-
croscopy (SEM) studies were per-
formed on a Philips XL30S field-emis-
sion SEM (FE-SEM) and Leo
1455 VP environment SEM (ESEM).
Samples prepared on a silicon wafer
were sputtered with gold and investi-
gated by SEM. X-ray diffraction pat-
terns were obtained on a Rigaku D/
MAX III diffractometer with a scan
speed of 28minÀ1 and
a sampling
width of 0.018 by using CuKa radiation.
Mercury-porosimetry isotherms were
obtained on an AutoPore III (Micro-
meritics). Prior to intrusion of mercury
the samples were evacuated under
vacuum
maintained
below
3ꢃ
10À2 Torr at room temperature. The
Kelvin equation was used to deter-
mine the pore size distribution of the
samples. The contact angle between
the sample and mercury was chosen as
1408. Differential scanning calorimetry
(DSC) was conducted on a TA Q100
calorimeter. Samples were measured
in aluminum hermetic cells under N2
atmosphere at
a
a
heating rate of
cooling rate of
58CminÀ1 and
28CminÀ1
.
Atomic force microscopy
(AFM) was conducted on Veeco
a
Multimode Scanning Probe Micro-
scope in tapping mode.
General method for the synthesis of 4-
alkoxy-2-trifluoromethylbenzonitriles:
Aliphatic alcohol (43.74 mmol) was
Figure 9. Molecular/mesomorphic structure relationship and criteria for formation of pipe structures.
added to
sodium hydride (60% (w/w) oil disper-
a stirred suspension of
Conclusion
sion, 1.75 g, 43.85 mmol) in THF (60 mL) at 08C under N2. The mixture
was heated to reflux for 30 min and cooled to room temperature. 4-
Nitro-2-trifluoromethylbenzonitrile (10 g, 46.27 mmol) was dissolved in
THF (60 mL) and dropped into the flask. The reaction mixture was re-
fluxed for 15 h under N2 and cooled to room temperature. A saturated
aqueous solution of NaCl (120 mL) was added to the reaction mixture
and an organic layer was collected. The aqueous layer was extracted with
ethyl acetate. The combined organic layers were dried and concentrated
to give a reddish liquid. The desired product was purified by flash chro-
matography on silica gel with ethyl acetate/hexane as eluent.
A series of structural analogues of 12FH2 were synthesized
to evaluate their molecular structures in relation to the for-
mation of parallelogrammatic pipes. The polymorphic
nature of the analogues, characterized by powder XRD and
DSC analyses, originated from the conformation of alkyl
groups and was preserved through the series when the
number of carbon atoms in the alkyl group was greater than
eight. Investigation of the solids obtained by a drying-medi-
ated assembly process revealed that the formation of paral-
lelogrammatic pipes required sophisticated evolution condi-
tions with a moderate crystallization rate that would allow a
4-Octyloxy-2-trifluoromethylbenzonitrile (2): Yield 58%; m.p. <RT;
1H NMR (400 MHz, CDCl3, 258C): d=7.71 (d, 1H, ArH), 7.23 (d, 1H,
ArH), 7.07 (q, 1H, ArH), 4.03 (t, 2H, OCH2), 1.77 (p, 2H, OCH2CH2),
1.27 (m, 10H), 0.85 ppm (t, 3H, CH2CH3); 13C NMR (100 MHz, CDCl3,
620
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 612 – 622

Self-Templation of Macroporous Pipes
FULL PAPER
258C): d=162.3, 136.5, 134.5 (q, 2J
G
G
345.1908 (calcd: 345.1916); elemental analysis calcd (%) for
C18H26F3NO2: C 62.59, H 7.59, F 16.50, N 4.06, O 9.26; found: C 63.08, H
8.04, N 4.58.
272.3 Hz), 117.1, 115.9, 113.8 (q, 3J
AHCTUNGTRENNUNG
29.1, 28.8, 25.8, 22.6, 14.0 ppm; MS: m/z: 299.1518 (calcd: 299.1497); ele-
mental analysis calcd (%) for C16H20F3NO: C 64.20, H 6.73, F 19.04, N
4.68, O 5.35; found: C 63.92, H 7.38, N 5.10.
4-Dodecyloxy-2-trifluoromethylbenzamide (12FH2): Yield 85%; m.p.
1038C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=7.79 (s, 1H, NH of
amide), 7.46 (d, 1H, ArH), 7.43 (s, 1H, other NH of amide), 7.21 (q, 1H,
ArH), 7.18 (d, 1H, ArH), 4.04 (t, 2H, OCH2), 1.69 (p, 2H, OCH2CH2),
1.23 (m, 18H), 0.84 ppm (t, 3H, CH2CH3); 13C NMR (100 MHz,
4-Decyloxy-2-trifluoromethylbenzonitrile (3): Yield 40%; m.p. <RT;
1H NMR (400 MHz, CDCl3, 258C): d=7.71 (d, 1H, ArH), 7.23 (d, 1H,
ArH), 7.07 (q, 1H, ArH), 4.03 (t, 2H, OCH2), 1.77 (p, 2H, OCH2CH2),
1.27 (m, 14H), 0.85 ppm (t, 3H, CH2CH3); 13C NMR (100 MHz, CDCl3,
[D6]DMSO, 258C): d=168.8, 159.0, 130.3, 128.9, 128.0 (q, 2J
19.7 Hz), 123.5 (q, 1J(C,F)=272.2 Hz), 117.0, 112.5 (q, 3J
(C,F)=4.7 Hz),
ACHTUNGTRENNUNG(C,F)=
258C): d=162.2, 136.5, 134.5 (q, 2J
277.5 Hz), 117.2, 115.9, 113.8 (q, 3J
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
68.1, 31.4, 29.2–28.5 (m), 25.5, 22.2, 13.8 ppm; MS: m/z: 373.2229 (calcd:
373.2229); elemental analysis calcd (%) for C20H30F3NO2: C 64.32, H
8.10, F 15.26, N 3.75, O 8.57; found: C 63.22, H 8.04, N 3.77.
29.2, 28.8, 25.8, 22.6, 14.1 ppm; MS: m/z: 327.1813 (calcd: 327.1810); ele-
mental analysis calcd (%) for C18H24F3NO: C 66.04, H 7.39, F 17.41, N
4.28, O 4.89; found: C 66.10, H 8.08, N 4.85.
Method II for the synthesis of 4-alkoxy-2-trifluoromethylbenzamides: 4-
Alkoxy-2-trifluoromethylbenzonitrile (2.670 mmol) was dissolved in
DMSO (10 mL). Potassium carbonate (0.6 g, 4.341 mmol) was added to
the flask and aqueous H2O2 solution (35%, 2 mL) was dropped into the
mixture. The reaction mixture was stirred at 608C for 10 h and cooled to
room temperature. Water (30 mL) was added to the mixture. The brown
solid was collected by filtration and purified by flash chromatography on
silica gel with ethyl acetate/hexane as eluent.
4-Dodecyloxy-2-trifluoromethylbenzonitrile (4): Yield 55%; m.p. 33–
1
358C; H NMR (400 MHz, CDCl3, 258C): d=7.71 (d, 1H, ArH), 7.22 (d,
1H, ArH), 7.07 (q, 1H, ArH), 4.03 (t, 2H, OCH2), 1.77 (p, 2H,
OCH2CH2), 1.27 (m, 18H), 0.85 ppm (t, 3H, CH2CH3); 13C NMR
(100 MHz, CDCl3, 258C): d=162. 2, 126.5, 134.1 (q, 2J
ACHTUNGTRENNUNG
122.2 (q, 1J
A
100.8, 69.0, 31.2, 29.7–28.8 (m), 25.8, 22.6, 14.0 ppm; MS: m/z: 355.2125
(calcd: 355.2123); elemental analysis calcd (%) for C20H28F3NO: C 67.58,
H 7.94, F 16.04, N 3.94, O 4.50; found: C 65.74, H 7.66, N 3.74.
4-Tetradecyloxy-2-trifluoromethylbenzamide (14FH2): Yield 97%; m.p.
1048C; 1H NMR (400 MHz, CDCl3, 258C): d=7.54 (d, 1H, ArH), 7.17
(d, 1H, ArH), 7.03 (q, 1H, ArH), 5.88 (s, 1H, NH of amide), 5.79 (s, 1H,
other NH of amide), 3.98 (t, 2H, OCH2), 1.78 (p, 2H, OCH2CH2), 1.42
(p, 2H, OCH2CH2CH2), 1.30 (m, 20H), 0.86 ppm (t, 3H, CH2CH3);
13C NMR (400 MHz, CDCl3, 258C): d=169.6, 160.3, 130.7, 128.7 (q, 2J-
4-Tetradecyloxy-2-trifluoromethylbenzonitrile (5): Yield 47%; m.p. 43–
1
448C; H NMR (400 MHz, CDCl3, 258C): d=7.71 (d, 1H, ArH), 7.23 (d,
1H, ArH), 7.07 (q, 1H, ArH), 4.03 (t, 2H, OCH2), 1.77 (p, 2H,
OCH2CH2), 1.27 (m, 10H), 0.85 ppm (t, 3H, CH2CH3); 13C NMR
(100 MHz, CDCl3, 258C): d=162.3, 136.5, 134.5 (q, 2J
ACHTUNGTRENNUNG
A
(C,F)=272.0 Hz), 117.1, 112.9 (q, 3J-
ACHTUNGTRENNUNG
122.0 (q, 1J
A
AHCTUNGTRENNUNG
100.9, 69.1, 31.9, 29.7–29.2 (m), 28.8, 25.8, 22.7, 14.1 ppm; MS: m/z:
383.2401 (calcd: 383.2436); elemental analysis calcd (%) for C22H32F3NO:
C 68.90, H 8.41, F 14.86, N 3.65, O 4.17; found: C 69.43, H 9.32, N 4.10.
401.2506 (calcd: 401.2542); elemental analysis calcd (%) for
C22H34F3NO2: C 65.81, H 8.54, F 14.20, N 3.49, O 7.97; found: C 65.89, H
9.04, N 3.93.
4-Hexadecyloxy-2-trifluoromethylbenzonitrile (6): Yield 58%; m.p. 51–
528C; H NMR (400 MHz, CDCl3, 258C): d=7.71 (d, 1H, ArH), 7.23 (d,
4-Hexadecyloxy-2-trifluoromethylbenzamide (16FH2): Yield 81%; m.p.
1058C; 1H NMR (400 MHz, CDCl3, 258C): d=7.52 (d, 1H, ArH), 7.16
(d, 1H, ArH), 7.02 (q, 1H, ArH), 6.16 (s, 1H, NH of amide), 5.84 (s, 1H,
other NH of amide), 3.98 (t, 2H, OCH2), 1.78 (p, 2H, OCH2CH2), 1.43
(p, 2H, OCH2CH2CH2), 1.30 (m, 24H), 0.86 ppm (t, 3H, CH2CH3);
13C NMR (400 MHz, CDCl3, 258C): d=169.8, 160.2, 130.6, 128.8 (q, 2J-
1
1H, ArH), 7.07 (q, 1H, ArH), 4.03 (t, 2H, OCH2), 1.80 (p, 2H,
OCH2CH2), 1.44 (p, 2H, OCH2CH2CH2), 1.27 (m, 24H), 0.85 ppm (t,
3H, CH2CH3); 13C NMR (100 MHz, CDCl3, 258C): d=162.3, 136.5,
134.5 (q, 2J
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
(C,F)=31.5 Hz), 122.0 (q, 1J
ACHTUNGTRENNUNG
113.8 (q, 3J
G
(C,F)=273.6 Hz), 117.0, 113.0 (q, 3J-
ACHTUNGTRENNUNG
14.1 ppm; MS: m/z: 411.2701 (calcd: 411.2749); elemental analysis calcd
(%) for C24H36F3NO: C 70.04, H 8.82, F 13.85, N 3.40, O 3.89; found: C
70.82, H 9.79, N 3.87.
AHCTUNGTRENNUNG
429.2917 (calcd: 429.2855); elemental analysis calcd (%) for
C24H38F3NO2: C 67.11, H 8.92, F 13.27, N 3.26, O 7.45; found: C 66.72, H
9.74, N 3.74.
Method I for the synthesis of 4-alkoxy-2-trilfluoromethylbenzamides: 4-
Alkoxy-2-trifluoromethylbenzonitrile (0.5 g, 1.407 mmol) was dissolved in
ethanol (5 mL). Potassium hydroxide (1.34 g, 23.91 mmol) was dissolved
in water (5 mL) and the solution was added to the flask. The reaction
mixture was heated to reflux for 20 h and cooled to room temperature.
Water (3 mL) was added to the reaction mixture. The mixture was fil-
tered to give a white solid, which was dried under vacuum.
4-Dodecyloxybenzonitrile (7): A procedure reported in the literature[33]
was adjusted as follows. 4-cyanophenol (0.361 g, 3.029 mmol) and 1-bro-
mododecane (0.906 g, 3.635 mmol) were dissolved in acetone (20 mL).
Potassium carbonate (2.344 g, 16.96 mmol) was added and the mixture
was heated to reflux for 2 days with vigorous stirring. After cooling to
room temperature, the mixture was filtered and the residue washed with
acetone. The filtrate was concentrated in a rotary evaporator and redis-
solved in ethyl acetate. The solution was washed with water (2ꢃ20 mL)
and brine (20 mL). Finally, the solution was dried on MgSO4 and evapo-
rated to yield the product (0.720 g, 83%). M.p. 408C; 1H NMR
(400 MHz, CDCl3, 258C): d=7.55 (d, 2H, ArH), 6.90 (d, 2H, ArH), 3.97
(t, 2H, OCH2), 1.76 (p, 2H, OCH2CH2), 1.44 (p, 2H, OCH2CH2CH2),
1.24 (m, 18H), 0.86 ppm (t, 3H, CH2CH3); 13C NMR (400 MHz, CDCl3,
258C): d=162.5, 133.9, 121.9, 115.2, 103.6, 68.4, 31.9, 29.7–29.0 (m), 25.9,
22.7, 14.1 ppm; MS: m/z: 287.2241 (calcd:287.2249); elemental analysis
calcd (%) for C19H29NO: C 79.39, H 10.17, N 4.87; O, 5.57. Found: C,
75.80; H, 9.97; N, 4.33.
4-Octyloxy-2-trifluoromethylbenzamide (8FH2): Yield 88%; m.p. 1058C;
1H NMR (400 MHz, CDCl3, 258C): d=7.52 (d, 1H, ArH), 7.16 (d, 1H,
ArH), 7.01 (q, 1H, ArH), 6.27 (s, 1H, NH of amide), 5.89 (s, 1H, other
NH of amide), 3.98 (t, 2H, OCH2), 1.78 (p, 2H, OCH2CH2), 1.44 (p, 2H,
OCH2CH2CH2), 1.30 (m, 8H), 0.86 ppm (t, 3H, CH2CH3); 13C NMR
(100 MHz, CDCl3, 258C): d=169.9, 160.2, 130.6, 128.7 (q, 2J
ACHTUNGTRENNUNG
33.1 Hz), 126.6, 123.3 (q, 1J(C,F)=272.3 Hz), 117.0, 113.0 (q, 3J
G
ACHTUNGTRENNUNG
5.0 Hz), 68.6, 31.7, 29.2, 29.1, 29.0, 25.9, 22.6, 14.1 ppm; MS: m/z:
317.1628 (calcd: 317.1603); elemental analysis calcd (%) for
C16H22F3NO2: C 60.56, H 6.99, F 17.96, N 4.41, O 10.08; found: C 60.67,
H 7.32, N 4.94.
4-Dodecyloxybenzamide (12HH2): A mixture of acetone (90 mL) and
water (60 mL) was added to 4-dodecyloxybenzonitrile (0.915 g,
3.183 mmol) and sodium percarbonate (11.07 g, 70.50 mmol). The mix-
ture was heated for 9 h at 508C. The mixture was cooled to room temper-
ature and acetone was removed on a rotary evaporator. Water (100 mL)
was added and the solid was collected by filtration to yield crude prod-
uct. Reprecipitation from ethanol/water gave the product (0.901 g, 93%).
M.p. 1438C (Lit. 96–988C[34]); 1H NMR (400 MHz, [D6]DMSO, 258C):
4-Decyloxy-2-trifluoromethylbenzamide (10FH2): Yield 79%; m.p.
1058C; H NMR (400 MHz, CDCl3, 258C): d=7.49 (d, 1H, ArH), 7.14 (s,
1
1H, ArH), 6.99 (d, 1H, ArH), 6.52 (s, 1H, NH of amide), 5.98 (s, 1H,
other NH of amide), 3.97 (t, 2H, OCH2), 1.77 (p, 2H, OCH2CH2), 1.43
(p, 2H, OCH2CH2CH2), 1.30 (m, 12H), 0.85 ppm (t, 3H, CH2CH3);
13C NMR (400 MHz, CDCl3, 258C): d=170.1, 160.2, 130.5, 128.6 (q, 2J-
ACHTUNGTRENNUNG ACHTUNGTRENNUGN
(C,F)=32.3 Hz), 126.7, 123.3 (q, 1J(C,F)=272.0 Hz), 116.9, 112.9 (q, 3J-
ACHTUNGTRENNUNG(C,F)=6.0 Hz), 68.5, 31.8, 29.5–29.0 (m), 25.9, 22.6, 14.0 ppm; MS: m/z:
Chem. Eur. J. 2009, 15, 612 – 622
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
621

S. Y. Kim et al.
d=7.78 (dd, 2H, ArH), 7.76 (s, 1H, NH of amide), 7.10 (s, 1H, other
NH of amide), 6.90 (dd, 2H, ArH), 3.96 (t, 2H, OCH2), 1.67 (p, 2H,
OCH2CH2), 1.36 (p, 2H, OCH2CH2CH2), 1.20 (m, 18H), 0.81 ppm (t,
3H, CH2CH3); 13C NMR (100 MHz, [D6]DMSO, 258C): d=167.4, 161.0,
129.3, 126.3, 113.7, 67.6, 31.2, 30.0–28.5 (m), 25.4, 22.0, 13.9 ppm; MS:
m/z: 305.2357 (calcd: 305.2355); elemental analysis calcd (%) for
C19H31NO2: C 74.71, H 10.23, N 4.59, O 10.48; found: C 71.79, H 9.46, N
4.43.
2004, 108, 8; c) K. Chen, A. Taflove, Y. L. Kim, V. Backman, Appl.
Phys. Lett. 2005, 86, 033101.
[12] a) M. Srinivasarao, D. Collings, A. Philips, S. Patel, Science 2001,
292, 79; b) S. A. Jenekhe, X. L. Chen, Science 1999, 283, 372; c) T.
Hayakawa, S. Horiuchi, Angew. Chem. 2003, 115, 2387; Angew.
Chem. Int. Ed. 2003, 42, 2285; d) H. Li, Q. Liu, L. Qin, M. Xu, X.
Lin, S. Yin, L. Wu, Z. Su, J. Shen, J. Colloid Interface Sci. 2005, 289,
488.
[13] M. Seo, G. Seo, S. Y. Kim, Angew. Chem. 2006, 118, 6454; Angew.
Chem. Int. Ed. 2006, 45, 6306; .
[14] A. Nangia, G. Desiraju, Top. Curr. Chem. 1998, 198, 57.
[15] a) M. C. Etter, Acc. Chem. Res. 1990, 23, 120; b) M. C. Etter, J.
Phys. Chem. 1991, 95, 4601.
[16] R. E. Cobbledick, R. W. H. Small, Acta Cryst. 1972, B28, 2893.
[17] a) P. Vishweshwar, A. Nangia, V. M. Lynch, Cryst. Growth Des.
2003, 3, 783; b) V. R. Pedireddi, J. PrakashaReddy, K. K. Arora, Tet-
rahedron Lett. 2003, 44, 4857; c) C. M. Reddy, L. S. Reddy, S. Aitipa-
mula, A. Nangia, C.-K. Lam, T. C. W. Mak, CrystEngComm 2005, 7,
44; d) K. K. Arora, J. PrakashaReddy, V. R. Pedireddi, Tetrahedron
2005, 61, 10793.
Preparation of pipes: The compound (0.2 g) and ethanol (1 mL) were put
into a vial. The vial was closed and heated in a silicone oil bath at 968C
(for 12FH2) or 1018C (for 14FH2) until the whole compound dissolved.
The vial was then removed from the bath and opened, and one drop of
the hot solution was dropped onto a silicon wafer. Evaporation of etha-
nol in ambient atmosphere produced parallelogrammatic pipes on the
wafer.
Acknowledgement
[18] a) K. Heuzꢀ, M. Fourmiguꢀ, P. Batail, J. Mater. Chem. 1999, 9, 2373;
b) S. A. Baudron, N. Avarvari, P. Batail, C. Coulon, R. Clꢀrac, E.
Canadell, P. Auban-Senzier, J. Am. Chem. Soc. 2003, 125, 11583;
c) M. Fourmiguꢀ, P. Batail, Chem. Rev. 2004, 104, 5379.
[19] U. Beginn, S. Sheiko, M. Mçller, Macromol. Chem. Phys. 2000, 201,
1008.
Financial support by the BK 21 program and by grant Nos. R01-2006-
000-11258-0 and R11-2007-050-04001-0f of the KOSEF is acknowledged.
We also would like to thank the Korea Basic Science Institutue in Dae-
jeon for the use of ESEM.
[20] B. D. Edgecombe, J. M. J. Frꢀchet, Chem. Mater. 1998, 10, 9944.
[21] a) P. Metrangolo, H. Neukirch, T. Pilati, G. Resnati, Acc. Chem. Res.
2005, 38, 386; b) K. Reichenbꢄcher, H. I. Suess, J. Hulliger, Chem.
Soc. Rev. 2005, 34, 22.
[22] A. R. Katritzky, B. Pilarski, L. Urogdi, Synthesis 1989, 949.
[23] G. W. Kabalka, S. M. Deshpande, P. P. Wadgaonkar, N. Chatla,
Synth. Commun. 1990, 20, 1445.
[24] J. Liebig, F. Wçhler, Ann. Pharm. 1832, 3, 249.
[25] a) W. I. F. David, K. Shankland, C. R. Pulham, N. Blagden, R. J.
Davey, M. Song, Angew. Chem. 2005, 117, 7194; Angew. Chem. Int.
Ed. 2005, 44, 7032; b) J. Thun, L. Seyfarth, J. Senker, R. E. Dinnebi-
er, J. Breu, Angew. Chem. 2007, 119, 6851; Angew. Chem. Int. Ed.
2007, 46, 6729.
[26] a) T. Shimizu, R. Iwaura, M. Masuda, T. Hanada, K. Yase, J. Am.
Chem. Soc. 2001, 123, 5947; b) O. S. Tyagi, H. S. Bisht, A. K. Chat-
terjee, J. Phys. Chem. B 2004, 108, 3010.
[27] a) M. R. Caira, Top. Curr. Chem. 1998, 198, 164; b) R. J. Davey, N.
Blagden, S. Righini, H. Alison, E. S. Ferrari, J. Phys. Chem. B 2002,
106, 1954.
[28] R. Azumi, G. Gçtz, P. Bꢄuerle, Synth. Met. 1999, 101, 544.
[29] a) R. D. Deegan, O. Bakajin, T. F. Dupont, G. Huber, S. R. Nagel,
T. A. Witten, Nature 1997, 389, 827; b) R. D. Deegan, Phys. Rev. E
2000, 61, 475.
[30] P. C. Ohara, W. M. Gelbart, Langmuir 1998, 14, 3418.
[31] a) B. H. Hong, J. Y. Lee, C.-W. Lee, J. C. Kim, S. C. Bae, K. S. Kim,
J. Am. Chem. Soc. 2001, 123, 10748; b) S. Leclair, P. Baillargeon, R.
Skouta, D. Gauthier, Y. Zhao, Y. L. Dory, Angew. Chem. 2004, 116,
353–357; Angew. Chem. Int. Ed. 2004, 43, 349–353; c) J.-S. Hu, Y.-
G. Guo, H.-P. Liang, L.-J. Wan, L. Jiang, J. Am. Chem. Soc. 2005,
127, 17090; d) H. Y. Lee, S. R. Nam, J.-I. Hong, J. Am. Chem. Soc.
2007, 129, 1040.
[32] a) H. Nakamura, Y. Matsui, J. Am. Chem. Soc. 1995, 117, 2651; F.
Miyaji, S. A. Davis, J. P. H. Charmant, S. Mann, Chem. Mater. 1999,
11, 3021; b) F. Mallet, S. Petit, S. Lafont, P. Billot, D. Lemarchand,
G. Coquerel, Cryst. Growth Des. 2004, 4, 965.
[1] a) J.-M. Lehn, Science 2002, 295, 2400; b) J.-M. Lehn, Proc. Natl.
Acad. Sci. USA 2002, 99, 4763.
[2] a) G. M. Whitesides, B. Grzybowski, Science 2002, 295, 2418;
b) G. M. Whitesides, M. Boncheva, Proc. Natl. Acad. Sci. USA 2002,
99, 4769.
[3] I. W. Hamley, Angew. Chem. 2003, 115, 1730; Angew. Chem. Int. Ed.
2003, 42, 1692.
[4] a) S. Fçrster, T. Plantenberg, Angew. Chem. 2002, 114, 712; Angew.
Chem. Int. Ed. 2002, 41, 688; b) T. P. Lodge, Macromol. Chem. Phys.
2003, 204, 265; c) O. Ikkala, G. ten Brinke, Science 2002, 295, 2407;
d) H.-A. Klok, S. Lecommandox, Adv. Mater. 2001, 13, 1217; e) M.
Lee, B.-K. Cho, W.-C. Zin, Chem. Rev. 2001, 101, 3869.
[5] a) M. Shimomura, T. Sawadaishi, Curr. Opin. Colloid Interface Sci.
2001, 6, 11; b) N. Maruyama, O. Karthaus, K. Ijiro, M. Shimomura,
T. Koito, S. Nishimura, T. Sawadaishi, N. Nishi, S. Tokura, Supramol.
Sci. 1998, 5, 331; c) M. Gonuguntla, A. Sharma, Langmuir 2004, 20,
3456.
[6] N. Glade, J. Demongeot, J. Tabony, C. R. Biol. 2002, 325, 283.
[7] a) B. A. Grzybowski, H. A. Stone, G. M. Whitesides, Nature 2000,
405, 1033; b) B. A. Grzybowski, G. M. Whitesides, J. Phys. Chem. B
2001, 105, 8770.
[8] a) J. Tang, G. Ge, L. E. Brus, J. Phys. Chem. B 2002, 106, 5653; b) E.
Rabani, D. R. Reichman, P. L. Geissler, L. E. Brus, Nature 2003,
426, 271; c) L. V. Govor, G. Reiter, J. Parisi, G. H. Bauer, Phys. Rev.
E 2004, 69, 061609; d) L. V. Govor, G. Reiter, G. H. Bauer, J. Parisi,
Appl. Phys. Lett. 2004, 84, 4774.
[9] a) A. P. H. J. Schenning, F. B. G. Benneker, H. P. M. Geurts, X. Y.
Liu, R. J. M. Nolte, J. Am. Chem. Soc. 1996, 118, 8549; b) M. C.
Lensen, K. Takazawa, J. A. A. W. Elemens, C. R. L. P. N. Jeukens,
P. C. M. Christianen, J. C. Maan, A. E. Rowan, R. J. M. Nolte,
Chem. Eur. J. 2004, 10, 831; c) C. R. L. P. N. Jeukens, M. C. Lensen,
F. J. P. Wijnen, J. A. A. W. Elemens, P. C. M. Christianen, A. E.
Rowan, J. W. Gerritsen, R. J. M. Nolte, J. C. Maan, Nano Lett. 2004,
4, 1401.
[10] a) C. J. Brinker, Y. Lu, A. Sellinger, H. Fan, Adv. Mater. 1999, 11,
579; b) D. Grosso, F. Cagnol, G. J. de A. A. Soler-Illia, E. L. Crepal-
di, H. Amenitsch, A. Brunet-Bruneau, A. Bourgeois, C. Sanchez,
Adv. Funct. Mater. 2004, 14, 309.
[33] Y. S. Chong, K. D. Shimizu, Synthesis 2002, 1239.
[34] C. Tan, L. Su, R. Lu, P. Xue, C. Bao, X. Liu, Y. Zhao, J. Mol. Liq.
2006, 124, 32.
Received: July 11, 2008
[11] a) E. Lauga, M. P. Brenner, Phys. Rev. Lett. 2004, 93, 238301;
b) A. P. Sommer, M. Ben-Moshe, S. Magdassi, J. Phys. Chem. B
Revised: October 5, 2008
Published online: December 9, 2008
622
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 612 – 622