Self-assembly of π-conjugated gelators into emissive chiral nanotubes: Emission enhancement and chiral detection

Article abstract of DOI:10.1002/asia.201301518

A series of new π-conjugated gelators that contain various aromatic rings (phenyl, naphthyl, 9-anthryl) and amphiphilic L-glutamide was designed, and their gel formation in organic solvents and self-assembled nanostructures was investigated. The gelators

Full text of DOI:10.1002/asia.201301518

FULL PAPER
DOI: 10.1002/asia.201301518
Self-Assembly of p-Conjugated Gelators into Emissive Chiral Nanotubes:
Emission Enhancement and Chiral Detection
Xiufeng Wang, Pengfei Duan, and Minghua Liu*[a]
Chem. Asian J. 2014, 9, 770 – 778
770
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
Abstract: A series of new p-conjugated
gelators that contain various aromatic
rings (phenyl, naphthyl, 9-anthryl) and
amphiphilic l-glutamide was designed,
and their gel formation in organic sol-
vents and self-assembled nanostruc-
tures was investigated. The gelators
showed good gelation ability in various
organic solvents that ranged from polar
to nonpolar. Those gelator molecules
with small rings such as phenyl and
naphthyl self-assembled into nanotube
structures in most organic solvents and
showed strong blue emission. However,
the 9-anthryl derivative formed only
a nanofiber structure in any organic
solvent, probably owing to the larger
steric hindrance. All of these gels
showed enhanced fluorescence in orga-
nogels. Furthermore, during the gel for-
mation, the chirality at the l-glutamide
moiety was transferred to the nano-
structures, thus leading to the forma-
tion of chiral nanotubes. One of the
nanotubes showed chiral recognition
toward the chiral amines.
Keywords: amines · chirality · gels ·
nanotubes · self-assembly
Introduction
of the building motifs have been carried out, it still remains
a great challenge to realize both structural and functional
control.[8] The self-assembly of p-conjugated molecules pro-
vided one of the most convenient ways to meet such a chal-
lenge. Various p-conjugated molecules,[9] such as hexabenzo-
coronene (HBC), porphyrins, and p-conjugated oligomers,
have been designed and widely used as building blocks in
self-assembled architectures to prepare nanotubes, nano-
wires, and so forth. Despite these excellent examples of ar-
chitectures, p-conjugated nanotubes[10] are still less investi-
gated relative to nanofiber structures.[11] Furthermore, con-
sidering the structure and functions of the nanotube, chirali-
ty is also an important issue. Although many functional
chiral nanotubes have been fabricated, their unique func-
tions have not been extensively investigated. In this paper,
we have developed a simple way to fabricate p-conjugated
chiral nanotubes through gel formation and we have found
a novel chiral recognition toward the chiral amine.
Relative to general amine detection, chiral amine detec-
tion has been deemed more difficult and challenging.
Amine vapor is teratogenic, carcinogenic, and can be ab-
sorbed by the human body through the skin as well as the
gastrointestinal and respiratory tract. Therefore the detec-
tion of various amines is very important. A unique enantio-
selective gelation in response to the chirality of a chiral
amine was reported in the system of poly(phenylacetylene)-
bearing cyclodextrin pendants.[12] Molecularly imprinted
polymers can be used as fluorescent sensors for chiral
amines.[13] In addition, a fluorescent sensor based on chiral
trifluoromethyl ketone was developed to detect chiral dia-
mine with high sensitivity and enantioselectivity by Puꢁs
group.[14] In spite of these studies, however, no report has
appeared on using functional p-conjugated nanotubes for
enantioselective fluorescent recognition of chiral amines or
chiral amine vapor.
One-dimensional nanotube structures have been receiving
a great deal of attention in the fields of both materials
chemistry and biological systems. As materials, the one-di-
mensional hollow nanostructures provide a special and con-
fined space, and some unique properties or enhanced per-
formances can be expected,[1] which could provide many in-
teresting applications. In biological systems, bundled tubular
structures on the nano- to microscale, such as axons of the
sciatic nerve in the human body, function in the mainte-
nance of cell structures or the transport of proteins or or-
ganelles in the cell,[2] which has encouraged many efforts to
mimic these systems.
Since the 1990s, several kinds of nanotubes have been the
subject of much attention. One is the carbon nanotube,
which was initially observed by Iijima in 1991 and has at-
tracted long and continuous interest owing to its unique
electronic properties as well as its stability and various ap-
plications.[3] The second kind is self-assembled nanotubes
based on peptides or various amphiphiles. The peptide
nanotube (PNT), which was first reported by Ghadiriꢁs
group and self-assembled based on the rational design of
oligopeptide,[4] has been extensively investigated, and appli-
cations of PNTs in drug delivery, biosensors, filters, and mi-
croelectronics have been proposed.[5] Organic or lipid nano-
tubes (ONT, LNT), which were generally self-assembled
from various kinds of amphiphilic building blocks[6] such as
phospholipids, bola-amphiphiles, glucolipids, amphiphilic
metal complexes, and so on, provided an important type of
soft nanomaterials in which diverse functions can be tail-
ored. Since these two kinds of nanotube are generally self-
assembled through noncovalent bonds, they provide the op-
portunity to tune their properties as well as their functions
through molecular design.[7] Although a deeper understand-
ing of the noncovalent bond and great efforts in the design
Low-molecular-weight organogels provide numerous op-
portunities not only for gelling various organic solvents but
also for the formation of nanostructures. Through the gel
formation, relatively uniform nanostructured materials
could be fabricated in large quantities. Although many of
the nanostructures in the organogels were revealed to be fi-
brillar structures, other forms such as nanorods and nano-
tapes are frequently encountered.[15] Nanotubes are also one
of the important structures that organogels can provide. Pre-
viously, we have realized the formation of chiral nanotubes
[a] X. Wang, Dr. P. Duan, Prof. Dr. M. Liu
Beijing National Laboratory for Molecular Science
CAS Key Laboratory of Colloid, Interface and
Chemical Thermodynamics, Institute of Chemistry
Chinese Academy of Sciences, Beijing, 100190 (China)
Fax : (+86)10-6256-9564
E-mail: liumh@iccas.ac.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201301518.
Chem. Asian J. 2014, 9, 770 – 778
771
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
and tuned their chirality by mixing different enantiomers of
N,N’-bis(octadecyl)-l-glutamic diamide (LGAm) and N,N’-
bis(octadecyl)-d-glutamic acid diamide (DGAm).[16a] The
good gelation properties of l-glutamide made us wonder
whether we could attach p-conjugated groups onto one end
of this moiety and realize p-conjugated nanotubes. Our
group has further designed several gelators that contain
LGAm with different functional p-conjugated sections.[16]
Although these gelators did not form a nanotube structure,
they showed a multiresponsive chiroptical switch, universal
nanotwists, chiral recognition of l-/d-tartaric acid, and so
forth. In this experiment, we linked several simple aromatic
rings of different sizes from phenyl, naphthyl, and anthryl to
LGAm through amide bonds. Figure 1 shows the molecular
to the boiling point of the solvent. Then a transparent solu-
tion could be obtained. Upon cooling to room temperature,
the solution became solidified. The gel formation was con-
firmed by the inverted test-tube method. Various organic
solvents from polar to nonpolar ones were tested, as shown
in Table 1. Compounds BLG, 1NLG, and 2NLG were able
Table 1. Gelation properties of BLG, 1NLG, 2NLG, and ALG in various
organic solvents.
Solvent
BLG
1NLG
2NLG
ALG
phase[a]
phase[a]
phase[a]
phase[a]
toluene
DMSO
DMF
TG (1.3)
G (1.3)
G (2.5)
TG (1.5)
G (1.4)
G (2.3)
TG (6.0)
G (3.0)
G (2.5)
G (5.0)
TG (1.1)
G (1.8)
G (2.4)
TG (6.0)
G (3.0)
G (2.4)
G (5.0)
TG (2.5)
TG (2.5)
TG (6.5)
G (12)
G (4.8)
S
PG
P
tetrahydrofuran TG (6.0)
acetonitrile
ethyl acetate
ethanol
n-hexane
cyclohexane
G (3.0)
G (2.6)
G (2.0)
TG (2.4)
TG (2.6)
P
TG (2.5)
TG (2.5)
TG (3.2)
TG (3.3)
[a] G=gel; TG=transparent gel; PG=partial gel; P=precipitate; S=so-
lution. For gels, the minimum gelation concentrations, CGC [mgmLÀ1],
at room temperature are shown in parentheses.
to gel the various organic solvents tested. Except for the
transparent and colorless gel that formed in toluene and cy-
clohexane, translucent or white gels were essentially ob-
served in polar solvents. Depending on the solvent, the criti-
cal gelation concentration (CGC) is somewhat different, and
the lowest CGC was obtained in toluene. Whereas com-
pounds BLG, 1NLG, and 2NLG were able to gel various or-
ganic solvents, ALG could only gel a few kinds of solvents.
In addition, its CGC increased significantly. Similar to the
other three compounds, transparent yellow gels were ob-
tained in toluene, n-hexane, and cyclohexane, and translu-
cent yellow gels only formed in DMSO and DMF.
The organogels in various organic solvents were fabricat-
ed into xerogels by evaporating the solvents, and their mor-
phologies were investigated by scanning and transmission
electron microscopy (SEM and TEM). The polarity of the
solvents affected the assembly of gelators. In nonpolar sol-
vents, such as toluene, n-hexane, and cyclohexane, the gela-
tors of BLG, 1NLG, and 2NLG can all assemble into
random nanofibers, which became entangled with each
other to immobilize the solvents (see the Supporting Infor-
mation). In strong polar solvents such as acetonitrile, DMF,
and DMSO, BLG assembled into nanotube structures,
whereas in other polar solvents, nanofibers were generally
obtained. For 1NLG and 2NLG, nanotubes could be ob-
tained in most polar solvents such as ethanol, THF, acetone,
ethyl acetate, acetonitrile, DMF, and DMSO. When the
head group was changed to a large aromatic anthryl, no
nanotube was observed from any ALG organogel (see the
Supporting Information). Instead, only nanofibers or other
random structures could be obtained. SEM images of nano-
tubes in various polar solvents are shown in the Supporting
Information. The smooth surface and the straight and open-
Figure 1. Molecular structures designed with different aromatic rings:
phenyl, 1-naphthyl, 2-naphthyl, and 9-anthryl are called (from left to
right) BLG, 1NLG, 2NLG, and ALG, respectively.
structures we designed. We have found that these molecules
could gel many organic solvents that ranged from nonpolar
to polar. In addition to the derivative with larger anthracene
derivatives, those of the derivative with phenyl and naphthyl
could form nanotube structures in many organic solvents.
Relative to those p-conjugated nanotubes,[9–11] our deriva-
tives are easily synthesized and with simpler p-conjugated
aromatic rings. Upon gel formation, the fluorescence of the
gel was significantly enhanced. However, the xerogels of
these compounds are in a nanotube form, and were found to
be sensitive to amine and could serve as an amine detector.
Furthermore, since the gelator molecules have chiral cen-
ters, when they form organogels, their molecular chirality
can be transferred to the supramolecular nanotube. Interest-
ingly, when enantiomeric amine vapor was introduced to the
nanotube, the nanotube showed excellent discrimination
toward the derivatives, thus working as a chiral sensing
system, although such recognition could not happen at a mo-
lecular level. Thus, through a simple molecular design, we
have realized uniform nanotubes from 1NLG that can be
used to detect chiral amine vapor.
Organogelation: Solvent-Regulated Nanofibers and
Nanotubes
The gel was fabricated in a general way. The solid was dis-
persed into a certain amount of organic solvent and heated
Chem. Asian J. 2014, 9, 770 – 778
772
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
Figure 3. The fluorescence spectra of a) a solution and gel of 1NLG and
b) 2NLG in DMSO. The fluorescence intensity of BLG (dashed line) is
twenty times the original data.
gel formation, the fluorescence emission was significantly
enhanced; it appeared at 344 and 356 nm upon excitation at
290 nm. The emission intensity of 1NLG gel increased by
more than fifty times than that of the solution state, which
could be ascribed to the aggregation-induced enhanced
emission.[17] Such an unusual phenomenon has been fre-
quently observed in the organogel as well.[18] In the present
gel, the aggregation-induced enhanced emission (AIEE) was
attributed to inhibition of intramolecular rotation in the gel
state and the intermolecular stacking. The emission of the
gel is observed at longer wavelengths than in solution,
which suggests that 1NLG piled up tightly in a J-aggregation
fashion. In J aggregation, the lower excited-state energy
level formed by dipole–dipole interactions is an allowed
transition to the ground state, which favors aggregate emis-
sion.[18e] As a result, the fluorescence emission of the gel is
much stronger than that of the monomer. And the result of
2NLG is similar to that of 1NLG (Figure 3b).
Figure 2. SEM images of a DMSO gel of a) BLG, b) 1NLG, c) 2NLG;
and acetonitrile gel of d) BLG, e) 1NLG, f) 2NLG. TEM images of g) a
BLG gel in DMSO, h) a 1NLG gel in DMSO, and i) 2NLG in acetoni-
trile, separately.
ended features revealed a tubular structure. Then we chose
DMSO and acetonitrile as representative solvents to study
their structures in detail. Figure 2 shows the SEM and TEM
images for some of the organized structures.
The SEM revealed the nanotube structures for the three
gelators. The high-quality SEM image of the assemblies in-
dicated the formation of nanotubes with a high aspect ratio
and uniform diameters for 1NLG and 2NLG. Although the
tubular structure is not clear for BLG, the open mouth of
the nanotube can be clearly seen in the SEM image. It is fur-
ther evident that the nanotubes are rolled from the nano-
belt, as a distinct helix turn can be seen in Figure 2b,c,e,f
for 1NLG and 2NLG. The tubular structures were solidly
confirmed by TEM, in which two dark parallel ribbons sepa-
rated by a light center (Figure 2g–i) can be seen. The diame-
ter of the BLG nanotube is approximately (63Æ2) nm,
1NLG is (50Æ2) nm, and 2NLG is (50Æ2) nm.
It has been further observed that the organic solvents also
have a strong effect on the fluorescence through either their
emission band or the intensity. Figure 4 shows the fluores-
cence spectra of various organogels. As shown in Figure 4a,
the emission spectra of BLG gels vary with the solvents,
which range from 287 (toluene) to 327 nm (DMSO). The
Gelation-Induced Enhancement of Fluorescence Emission
Upon gel formation, these compounds showed strong blue
emission upon UV irradiation. The luminescence spectra of
the gels in various solvents were investigated. The gel 2NLG
showed the strongest emission, and BLG was the worst at
the same concentration in DMSO solvent. Figure 3 shows
the fluorescence spectra of the compounds in solution and
the gel state. For BLG gel, a weak blue emission was ob-
served with an excitation of 270 nm, although there is no
fluorescence in the solution state.
For compound 1NLG, the fluid solution in DMSO
showed a very weak fluorescence under illumination at
254 nm UV light. The concentration-dependent fluorescence
spectra of 1NLG in solution were also measured and are
shown in the Supporting Information. Before the gel forma-
tion, the emission peak at 339 nm redshifted to 344 nm, and
the intensity increased as the concentration increased. Upon
Figure 4. The fluorescence spectra of a) BLG, b) 1NLG, c) 2NLG, and
d) ALG in various solvents.
Chem. Asian J. 2014, 9, 770 – 778
773
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
fluorescence of 1NLG gels in various organic solvents indi-
cated that there are two clear peaks at around 345 and
355 nm in most solvents. The 2NLG gels showed different
intensities in relation to the polarity of the solvents. The
peaks of nonpolar solvents such as toluene, n-hexane, and
cyclohexane are at 351 and 362 nm, which become clearly
redshifted relative to the polar solvents at 344 and 354 nm
or so. In addition, the gels in nonpolar solvents are much
stronger than those in polar solvents. For ALG, the fluores-
cence spectra are bound up with the solvents used, and the
peak range is from 415 to 444 nm (shown in Figure 4d).
packed in an all-trans zigzag conformation.[19] The FTIR
spectral features indicate that multiple hydrogen bonds be-
tween head groups together with the closely packed tails are
the driving force for the formation of multi-bilayers and
supramolecular nanotubes. In addition, the amide I and ami-
de II bands of 1NLG/2NLG were a little different from
BLG; they split clearer than BLG and appeared as three
kinds of amide groups, thereby revealing that 1NLG/2NLG
molecules arranged in a more orderly fashion . For 1NLG
and 2NLG, nas
ACHTUNGTRENNUNG
wavenumber, and nAHCTUNGTRENNUNG
a lower one, which indicated the presence of stronger hydro-
gen bonds than BLG.[20]
Analysis of Organogel Driving Forces
X-ray diffraction measurements were performed to gain
further insights into the different structural features. Then
we chose DMSO and acetonitrile as representative polar
solvents to further disclose the structures of the nanotubes,
whereas toluene represents the nonpolar solvents for nano-
fiber structures (shown in Figure 5b and the Supporting In-
formation). Well-defined diffraction patterns were observed
in the xerogels from different gelling solvents. The two q
values appeared at 2.53 and 5.418 for the BLG xerogel, at
2.54 and 5.278 for 1NLG, and 2.42 and 5.188 for 2NLG, re-
spectively. The layer distance is estimated and listed in Fig-
ure 5b according to Braggꢁs equation. The d-spacing ratio is
about 1:0.5, which was consistent with the lamellar structure.
The value of the layer distance is less than twice the molecu-
lar dimension but larger than a single extended one, which
suggests that BLG or NLG formed bilayer structures with
interdigitation of the alkyl chain.[21] On the basis of the
XRD pattern, the d-spacing value of 2NLG is larger than
that of 1NLG, and the value for BLG is the smallest one.
This is concordant with the single extended molecule length,
which results from the different steric hindrance of the aro-
matic head group. In nonpolar solvents, the bilayer struc-
tures then stack to form nanofibers. In polar solvents, such
as DMSO and acetonitrile, the bilayer structures further
rolled into helices and then formed nanotubes,[22] as illustrat-
ed in Figure 7. By taking BLG gel in DMSO as an example,
3.15 nm corresponds to the d-spacing value of BLG in a bi-
layer structure. The wall thickness of the nanotubes is
around 16 nm, as revealed by the TEM image in Figure 2g,
therefore it can be inferred that the nanotubes were rolled
up from 5 bilayers or orderly accumulated multiple bilayers.
A similar lamellar structure and FTIR spectrum for ALG
xerogel from DMSO are shown in the Supporting Informa-
tion.
To further characterize the gel formation and their driving
forces, FTIR and XRD were measured on the xerogels. Fig-
ure 5a,b show the FTIR spectra and XRD patterns of the
xerogels. We chose the typical assembly of nanotubes and
focused on the assembly of BLG, 1NLG, and 2NLG in
DMSO. In the FTIR spectra, characteristic vibration bands
Figure 5. a) FTIR spectra and b) the XRD patterns of the BLG, 1NLG,
and 2NLG xerogels and c) CD spectra of the corresponding gels in
DMSO.
À
are observed clearly. The appearance of an N H stretching
vibration band around 3300 cmÀ1 indicated the formation of
the hydrogen bond. The two splits are assigned to two kinds
À
of N H bond; one kind is adjacent to the aromatic ring, and
the other one links to the long alkyl chain. All the amide I
and amide II bands appeared at around 1659–1630 and
À1
À
1557–1532 cm , which indicates that both C=O and N H
Chirality of the Nanotube and Chiral Sensing
are in the hydrogen-bonded form. At the same time, there
are different degrees of splits in the amide I band at 1658
and 1644 cmÀ1 with a shoulder at 1636 cmÀ1 for BLG; 1658,
1644, and 1633 cmÀ1 for 1NLG; and 1658, 1644, and
1628 cmÀ1 for 2NLG, which suggest the existence of three
kinds of amine bond. The asymmetric and symmetric
stretching vibrations of CH2 always appear at 2916 and
2849 cmÀ1, respectively, which suggests the alkyl chains
Since the gelator molecules have chiral centers, we mea-
sured the circular dichroism (CD) spectra of the organogels.
As seen in Figure 5c, the CD spectrum of the BLG organo-
gel in DMSO exhibits a negative band at 291 nm and a posi-
tive band at 254 nm, with a crossover at around 274 nm. For
1NLG and 2NLG, there are clear positive signals, which is
in accordance with its adsorption band and indicates that
Chem. Asian J. 2014, 9, 770 – 778
774
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
gradually as the volume of ani-
line vapor increases. The
volume of the injected saturat-
ed aniline vapor was 20, 60,
100, 200, 300, 400, 500, 600, 800,
and 1000 mL in turn. Because
the volume of the sealed quartz
cuvette is 4.6 mL, the aniline
converted into vapor pressure is
3.8, 7.6, 19.1, 38.3, 57.4, 76.5,
95.7, 114.8, 153, and 191 ppm.
The film of 2NLG showed
a similar fluorescence quench-
ing phenomenon (Supporting
Information). We have also in-
vestigated the effect of the dif-
ferent nanostructures on the de-
tection of aniline. The films fab-
ricated from toluene and aceto-
nitrile gave nanofibers and
nanotubes, respectively. The
fluorescence results of the dif-
ferent nanostructures on expo-
sure to aniline are shown in the
Supporting Information. For
the 1NLG film, it was observed
Figure 6. The fluorescence spectra of the 1NLG film in acetonitrile: A) exposure to the vapor of aniline; the
intensity decreases as the volume of aniline vapor increases. Exposure to the vapor of chiral amine: B) (R)- that the nanotube showed
TEA and C) (S)-TEA versus time. D) Fluorescence quenching efficiency of the 1NLG film by adding chiral
amine versus time (from 0 to 75 min).
a rapid decrease in fluorescence
in response to the aniline
vapor, but it was nearly the
the structure was chiral. As has been noted before, 1NLG
and 2NLG assemble into helical nanotubes with left-hand
chirality in Figure 2b,c,e,f. On the basis of these CD spectra
and the above SEM observation, it is clear that the chirality
that is localized in the l-glutamide transferred to the whole
assemblies during the gel formation.
same for 2NLG, which might be owing to the larger steric
hindrance in 1NLG. However, in any case, the nanofiber
structure exhibited a larger decrease of fluorescence at
a higher vapor pressure.
In our study, chiral organic amines (R)-(+)-1-(p-tolyl)e-
thylamine and (S)-(À)-1-(p-tolyl)ethylamine ((R)-TEA and
(S)-TEA) were tested too. The saturated vapor pressure of
1-(p-tolyl)ethylamine (TEA) is 279 ppm, which is lower
than that of aniline. The fluorescence spectra were obtained
in the following way. First, the cast film was sealed in
a quartz cuvette. Then saturated (R)- or (S)-TEA vapor
(1 mL) was injected into the cuvette. Figure 6B,C shows the
change in fluorescence as a function of time. The fluores-
cence of the films was quenched by the chiral TEA and
became saturated at 75 min. We compared the degree of the
fluorescence quenching by the chiral amines and the I/I0
values in the equilibrium. No clear differences appeared in
the cast films of 1NLG-toluene, 2NLG-toluene, 2NLG-ace-
tonitrile, ALG-toluene, and ALG-acetonitrile, as shown in
the Supporting Information. Interestingly, the result of
1NLG-acetonitrile, which is a nanotube structure, showed
interesting results. When (R)-TEA was detected, I/I0 is
about 0.60; this value changed to 0.43 when the S enantio-
mer was used. This indicated that during the interaction of
the TEA enantiomer with the 1NLG nanotube, (S)-TEA is
more favorable, which might be due to their chirality match.
Since these nanotubes showed strong emission, they were
further used to test if these nanotubes can be used to detect
amines. Aniline, owing to its low saturated vapor (880 ppm)
among other organic amines, is a commonly examined or-
ganic amine.[23] Since aniline is aromatic, it is expected that
the aromatic ring together with the electron-donor amino
group could extinguish the fluorescence from the naphthyl
ring in both 1NLG and 2NLG. Therefore, the nanotube
films of both the compounds were prepared by casting a hot
solution of 1NLG or 2NLG in acetonitrile onto a clean
quartz plate. When the solvent volatilized completely, a thin
film composed of the nanotube structures of either 1NLG
or 2NLG was fabricated. These films were subsequently
used to detect aniline vapor. The fluorescence emission
spectra at different diluted vapor pressures were measured
upon exposure to the vapor of aniline. Different vapor pres-
sures were acquired by the injection of different volumes of
saturated aniline vapor into the 10 mm sealed quartz cuvette
with a cushion and screw cap.
In Figure 6A it can clearly be seen that the fluorescence is
quenched. The intensity of the 1NLG fluorescence decreases
Chem. Asian J. 2014, 9, 770 – 778
775
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
parison, the sensing performan-
ces of other kinds of film
toward chiral amines were also
investigated. For the same
1NLG gelator, when the struc-
ture is a nanofiber, the fluores-
cence quenching with different
enantiomers is nearly the same.
The enantiomers were not dis-
tinguished from each other
even with a different gelator
with the same nanostructure.
This further indicated that, in
addition to the steric hindrance
and the supramolecular chirali-
ty in the nanostructures, the
curvature of the nanotube
might also play an important
role, which would make (S)-
TEA more likely to access the
nanotube, as illustrated in Fig-
ure 7C,D.
Figure 7. Illustration of A) nanotube formation. By taking 1NLG nanotubes as an example, the detection of
aniline using B) 1NLG nanotubes, and 1NLG nanotube detection of chiral amine: C) R enantiomer and D) S
enantiomer. The 1NLG nanotube favors the S enantiomer and disfavors the R enantiomer.
Conclusion
The self-assembly of gelators
with a p-conjugated aromatic
By combining the above data, the formation of nanotubes
and nanofibers can be illustrated as in Figure 7. In strong
polar solvents, BLG, 1NLG, and 2NLG molecules first
formed a bilayer structure with interdigitated aliphatic tails.
Then multiple bilayer structures were stacked into nano-
belts. To reduce the large surface energy, the nanobelts were
further rolled into helical structures owing to the chiral
nature of the layered structure. Finally, these gelator mole-
cules assembled into a nanotube structure. It is noteworthy
that, to roll into nanotubes, the head group should not be
too big. When ALG was used, the larger steric hindrance of
the head group prevented its rolling and could only yield
a fibrous structure.
Both nanofiber and nanotube structures of 1NLG and
2NLG can be used to detect aniline. Therefore, we propose
a mechanism for the fluorescence sensing of gaseous amines,
which is shown in Figure 7B. By taking 1NLG nanotubes ob-
tained from acetonitrile as an example, aniline molecules
can adsorb on the surface of the nanostructure by means of
p–p interactions between aniline and naphthyl. The other
aromatic ring exhibits similar properties. This indicates that
the gelators as electron-accepting materials are suitable for
sensing electron-rich compounds such as organic amines.
Similar to aniline detection, chiral organic amines that con-
tain aromatic rings were also tested. On the basis of the re-
sults shown in Figure 6D, a 1NLG-acetonitrile cast film
showed a distinct decrease in the fluorescence between (R)-
TEA and (S)-TEA. This is directly linked to the nanostruc-
ture and the steric hindrance effect of the gelator. As a com-
ring was investigated in various organic solvents. The com-
pounds with small rings such as phenyl and naphthyl can
self-assemble into uniform nanotubes in many polar sol-
vents, whereas they formed nanofiber structures in some
nonpolar solvents. With its large steric hindrance, the 9-an-
thryl derivative could only form a nanofiber structure. Upon
gel formation, the organogels exhibited clear fluorescence
enhancement with blue emission. During the gel formation,
the molecular chirality was transferred to the chromophores
and the subsequent nanostructures. Strong CD signals were
detected in the organogels. Furthermore, the helical nano-
tubes of 1NLG and 2NLG show the same left-handed chir-
ality. Combined with enhanced fluorescence and supra-
molecular chirality, the film composed of the nanostructures
by means of gelation was used to sense organic amines. Re-
markably, the nanotube film of 1NLG showed enantioselec-
tivity toward enantiomeric (R)- and (S)-1-(p-tolyl)ethyla-
mine, in which the latter was favored.
Experimental Section
Materials
Compounds (R)- and (S)-1-(p-tolyl)ethylamine were purchased from TCI
and used as received. The remaining chemicals and solvents were pur-
chased from Beijing Chemical Reagent Industry. The p-conjugated gela-
tors (BLG, 1NLG, 2NLG, and ALG) were synthesized as follows.
Chem. Asian J. 2014, 9, 770 – 778
776
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
Synthesis of the Aromatic l-Glutamic Lipid (BLG, 1NLG, 2NLG, and
ALG)
4H; CH2), 4.85–4.90 (q, 1H; CH), 6.09 (s, 1H), 7.38–7.30 (d, 1H), 7.45–
7.51 (m, 4H), 8.00–9.02 (m, 4H), 8.48 ppm (s, 1H); MALDI-TOF MS:
m/z calcd for C56H91N3O3: 853.7; found: 876.9 [M+Na]+; elemental analy-
sis calcd (%) for C56H91N3O3: C 78.73, H 10.74, N 4.92; found: C 78.07, H
10.68, N 5.15.
The synthesis of N,N’-bis(octadecyl)-l-glutamic diamide (LGAm) was re-
ported previously.[16a]
Gelator BLG
General Methods
LGAm (0.85 g, 1.3 mmol) was dispersed in dichloromethane (40 mL) and
stirred for 30 min. Benzoyl chloride (0.5 mL, 4.3 mmol) was dissolved in
dichloromethane (20 mL) in a dropping funnel. Then the chloride was
dropped into the above mixture and stirred at 08C overnight. After that,
the solvent was removed by rotary evaporation and a light yellow solid
was obtained. The crude product was dissolved in THF (10 mL) and
poured into an aqueous saturated solution of NaHCO3 (300 mL). After
filtration, the product was purified by recrystallization in ethanol to give
a white solid (0.84 g, 85.7%). 1H NMR (CDCl3, 400 Hz): d=0.86–0.89 (t,
6H; CH3), 1.25 (m, 60H; CH2), 1.50–1.53 (m, 4H; CH2), 1.85 (s, 1H;
NH), 2.19–2.28 (q, 2H; CH2), 2.51–2.63 (m, 2H; CH2), 3.24–3.30 (m, 4H;
CH2), 3.43–3.47 (t, 1H; NH), 4.58 (s, 1H; CH), 7.26–7.47 (m, 2H), 7.51–
7.54 (m, 1H), 7.91–7.93 (d, 2H), 8.15 ppm (s, 1H; NH); MALDI-TOF
MS: m/z calcd for C48H87N3O3: 753.7 [M]; found: 754.9 [M+Na]+, 776.9
[M+K]+, 792.9; elemental analysis calcd (%) for C48H87N3O3: C 76.44, H
11.63, N 6.36; found: C 76.16, H 11.80, N 5.61.
UV/Vis spectra were measured with a Hitachi U-3900 spectrophotome-
ter. Fourier transform-infrared (FTIR) studies were performed with
a JASCO FTIR-660 spectrometer. 1H NMR spectra were recorded with
a Bruker ARX400 (400 MHz) with Me4Si used as the internal standard
and CDCl3 for solvent. MALDI-TOF MS spectra were recorded with
a
BIFLEX III instrument. Elemental analyses were recorded with
a Carlo–Erba-1106 instrument.
Circular Dichroism Spectroscopy
CD spectra were recorded with a JASCO J-810 CD spectrophotometer
under a nitrogen atmosphere. Experiments were performed at room tem-
perature in a quartz cell with a 0.1 mm path length over a range of 200–
650 nm.
Fluorescence Spectroscopy Measurements
Fluorescence spectra were recorded with a Hitachi F-4600 spectrometer.
The organogel was heated to a solution state, then transferred to a quartz
cell with a 2 mm or 1 cm path length and heated to keep it transparent;
the fluorescence was measured rapidly. Different solvents were used in
the same way. After waiting for fifteen minutes, gelation took place com-
pletely, and then the fluorescence of the gel was measured. Fluorescence
emission spectra of BLG were measured with excitation at 270 nm,
1NLG and 2NLG were measured with excitation at 290 nm, and ALG at
380 nm.
Gelator 1NLG
LGAm (1.26 g, 1.94 mmol) was dispersed in dichloromethane (40 mL)
and stirred for 30 min. 1-Naphthoyl chloride (0.7 g, 3.7 mmol) was dis-
solved in dichloromethane (20 mL) in a dropping funnel. The following
procedure was the same as that for BLG. After filtration, the product
was purified by recrystallization in ethanol to give a white solid (1.38 g,
85.9%). 1H NMR (CDCl3, 400 Hz): d=0.86–0.89 (t, 6H; CH3), 1.24–1.30
(m, 60H; CH2), 1.40–1.52 (m, 4H; CH2), 2.17–2.27 (m, 2H; CH2), 2.50–
2.63 (m, 2H; CH2), 3.28 (m, 4H; CH2), 4.70 (s, 1H; CH), 7.10 (s, 1H;
NH), 7.26–7.47 (m, 1H), 7.49–7.57 (m, 3H), 7.69–7.71 (d, 1H), 7.86–7.88
(d, 1H), 7.93–7.95 (d, 1H), 8.34–8.36 ppm (d, 1H); MALDI-TOF MS: m/
z calcd for C52H89N3O3: 803.7; found: 827.0 [M+Na]+; elemental analysis
calcd (%) for C52H89N3O3: C 77.65, H 11.15, N 5.22; found: C 77.58, H
11.00, N 5.27.
X-ray Diffraction Measurements
The organogel was spin-coated onto the glass substrate and dried under
vacuum before measuring. XRD of the xerogels was performed with
a Holland PANalytical X’Pert PRO MPD operating at 40 kV and 40 mA
using CuKa radiation (l=1.5418 ꢃ). The scan range (2q) was from 1 to
68.
Gelator 2NLG
SEM and TEM Measurements
LGA (1.3 g, 2 mmol) was dispersed in dichloromethane (40 mL) and
stirred for 30 min. 2-Naphthoyl chloride (0.8 g, 4.2 mmol) was dissolved
in dichloromethane (20 mL) in a dropping funnel. The following proce-
dure was the same as that for BLG. After filtration, the product was puri-
fied by recrystallization in ethanol to give a white solid (1.45 g, 87.2%).
1H NMR (CDCl3, 400 Hz): d=0.86–0.89 (t, 6H; CH3), 1.20–1.30 (m,
60H; CH2), 1.50–1.64 (m, 4H; CH2), 2.26–2.55 (m, 2H; CH2), 2.86 (m,
2H; CH2), 3.22–3.40 (m, 4H; CH2), 4.79 (s, 1H; CH), 7.26–7.60 (m, 2H),
7.86–7.92 (q, 2H), 7.97–7.99 (d, 1H), 8.03–8.05 (d, 1H), 8.64 ppm (s,
1H); MALDI-TOF MS: m/z calcd for C52H89N3O3: 803.7; found: 804.9
[M+Na]+, 827.0 [M+K]+, 843.0; elemental analysis calcd (%) for
C52H89N3O3: C 77.65, H 11.15, N 5.22; found: C 78.28, H 10.29, N 4.76.
The fully aged gel was cast onto single-crystal silica plates (Pt-coated)
and carbon-coated Cu grids (unstained), and the trapped solvent in the
gel was first evaporated under ambient conditions, then vacuum-dried for
12 h. After that, the performances were determined with a Hitachi S-
4800 FESEM and JEOL TEM-2011 instrument operating at accelerating
voltages of 10 and 200 kV, respectively.
Preparation of Organic Gels for Analysis
Different solvents (1 mL) were added to BLG, 1NLG, or 2NLG (3 mg)
in a 5 mL vial, then the mixture was heated until the entire solid dis-
solved into a transparent solution. After standing at room temperature,
the self-supporting organogel was usually complete within 10 to 30 min.
Samples for SEM, TEM, UV/Vis, fluorescence, and CD studies were pre-
pared without any dilution.
Gelator ALG
Anthracene-9-carboxylic acid (0.50 g, 2.25 mmol) was suspended in dry
benzene (50 mL), and then SOCl2 (3 mL) was added. The mixture was
heated at reflux under strong stirring for 4 h. The solvent and the surplus
SOCl2 were then removed by rotary evaporation, and a dark brown solid,
anthracene-9-carbonyl chloride, was obtained and used without further
purification. LGAm (1 g, 1.5 mmol) was dispersed in dichloromethane
(40 mL) and stirred for 30 min. The obtained anthracene-9-carbonyl chlo-
ride was dissolved in dry CH2Cl2 (30 mL) and was then added dropwise
to the LGAm solution within 30 min. The mixture was stirred for another
3 h at 08C. After the reaction, the solution was washed with 0.01m HCl
(3ꢂ30 mL) and pure water (3ꢂ30 mL) and dried over magnesium sulfate
under vacuum. The product was purified by recrystallization in ethanol
to give a brown solid (0.77 g, 58.8%). 1H NMR (CDCl3, 400 Hz): d=
0.86–0.89 (t, 6H; CH3), 1.20–1.30 (m, 60H; CH2), 1.45–1.59 (m, 4H;
CH2), 2.25–2.30 (m, 2H; CH2), 2.42–2.77 (m, 2H; CH2), 3.16–3.32 (m,
Fabrication of Cast Film
By taking the cast film of NLG nanotubes as an example, the samples of
1NLG and 2NLG (3.0 mg) were weighed separately in a 5 mL vial. Ace-
tonitrile (1 mL) was added to the vial. The white solid sample was heated
until it dissolved, and a transparent solution was obtained. The hot solu-
tion was cast into a clean quartz plate, and after the solvent evaporation
was complete, a nanotube film was obtained. The structure of the film
was determined by SEM, and many nanotubes were observed. Other cast
films were prepared by following the same method.
Chem. Asian J. 2014, 9, 770 – 778
777
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Minghua Liu et al.
Fluorescence Measurement by Aniline Vapor
jayakumar, R. Varghese, S. J. George, Angew. Chem. 2006, 118, 470–
474; Angew. Chem. Int. Ed. 2006, 45, 456–460.
The fluorescence of aniline vapor with different concentrations was
monitored following a method reported by Zang et al.[23a] The saturated
vapor of aniline was achieved in a sealed serum bottle (25 mL). Before
use the bottle was sealed overnight to attain a saturated vapor inside it.
The cast film was measured inside a 10 mm sealed quartz cuvette with
cushion and screw cap. The actual measured volume of the sealed quartz
cuvette was 4.6 mL. Fluorescence quenching with different vapor pres-
sures of aniline was performed in the cuvette. A small volume of the sa-
turated vapor of aniline was immediately injected into the sealed cuvette
by using a microsyringe to achieve the diluted vapor. After stabilizing for
120 s, the fluorescence of the film was measured. Throughout the whole
measurement process, the cast film was kept immobile to guarantee that
the fluorescence was measured at the same point and to ensure the accu-
racy and comparability of the monitoring data.
[10] a) H. Shao, J. Seifert, N. C. Romano, M. Gao, J. J. Helmus, C. P. Jar-
oniec, D. A. Modarelli, J. R. Parquette, Angew. Chem. 2010, 122,
7854–7857; Angew. Chem. Int. Ed. 2010, 49, 7688–7691; b) Y. Chen,
B. Zhu, F. Zhang, Y. Han, Z. Bo, Angew. Chem. 2008, 120, 6104–
6107; Angew. Chem. Int. Ed. 2008, 47, 6015–6018; c) Z. Huang,
S. K. Kang, M. Banno, T. Yamaguchi, D. Lee, C. Seok, E. Yashima,
M. Lee, Science 2012, 337, 1521–1526.
[11] a) S. S. Babu, S. Mahesh, K. K. Kartha, A. Ajayaghosh, Chem. Asian
J. 2009, 4, 824–829; b) D. Dasgupta, S. Srinivasan, C. Rochas, A.
Ajayaghosh, J.-M. Guenet, Soft Matter 2011, 7, 9311–9315.
[12] K. Maeda, H. Mochizuki, K. Osato, E. Yashima, Macromolecules
2011, 44, 3217–3226.
[13] T. H. Nguyen, R. J. Ansell, Org. Biomol. Chem. 2009, 7, 1211–1220.
[14] S. Yu, W. Plunkett, M. Kim, L. Pu, J. Am. Chem. Soc. 2012, 134,
20282–20285.
[15] a) J. Liu, Y. Feng, Z. X. Liu, Z. C. Yan, Y. M. He, C. Y. Liu, Q. H.
Fan, Chem. Asian J. 2013, 8, 572–581; b) Y. W. Huang, B. G. Quan,
Z. X. Wei, G. T. Liu, L. F. Sun, J. Phys. Chem. C 2009, 113, 3929–
3933; c) G. T. Wang, J. B. Lin, X. K. Jiang, Z. T. Li, Langmuir 2009,
25, 8414–8418; d) X. D. Yu, Q. A. Liu, J. C. Wu, M. M. Zhang,
X. H. Cao, S. Zhang, Q. Wang, L. M. Chen, T. Yi, Chem. Eur. J.
2010, 16, 9099–9106; e) P. Sahoo, D. K. Kumar, S. R. Raghavan, P.
Dastidar, Chem. Asian J. 2011, 6, 1038–1047.
Acknowledgements
We gratefully acknowledge funding of this research by the Basic Re-
search Development Program (2013CB834504) and the National Natural
Science Foundation of China (grant nos. 91027042, 21021003, and
21227802).
[16] a) X. F. Zhu, Y. G. Li, P. F. Duan, M. H. Liu, Chem. Eur. J. 2010, 16,
8034–8040; b) P. F. Duan, Y. G. Li, L. C. Li, J. G. Deng, M. H. Liu,
J. Phys. Chem. B 2011, 115, 3322–3329; c) P. F. Duan, X. F. Zhu,
M. H. Liu, Chem. Commun. 2011, 47, 5569–5571; d) Q. X. Jin, L.
Zhang, X. F. Zhu, P. F. Duan, M. H. Liu, Chem. Eur. J. 2012, 18,
4916–4922; e) X. F. Wang, P. F. Duan, M. H. Liu, Chem. Commun.
2012, 48, 7501–7503; f) Q. Jin, L. Zhang, M. Liu, Chem. Eur. J.
2013, 19, 9234–9241.
[1] E. Gazit, Chem. Soc. Rev. 2007, 36, 1263–1269.
[2] C. Conde, A. Caceres, Nat. Rev. Neurosci. 2009, 10, 319–332.
[3] J. N. Coleman, U. Khan, W. J. Blau, Y. K. Gun’ko, Carbon 2006, 44,
1624–1652.
[4] M. R. Ghadiri, J. R. Granja, R. A. Milligan, D. E. McRee, N. Khaza-
novich, Nature 1993, 366, 324–327.
[5] a) S. Scanlon, A. Aggeli, Nano Today 2008, 3, 22–30; b) R. J. Brea,
C. Reiriz, J. R. Granja, Chem. Soc. Rev. 2010, 39, 1448–1456;
c) H. X. Xu, A. K. Das, M. Horie, M. S. Shaik, A. M. Smith, Y. Luo,
X. F. Lu, R. Collins, S. Y. Liem, A. M. Song, P. L. A. Popelier, M. L.
Turner, P. Xiao, I. A. Kinloch, R. V. Ulijn, Nanoscale 2010, 2, 960–
966.
[6] a) S. Svenson, P. B. Messersmith, Langmuir 1999, 15, 4464–4471;
b) H. Shao, M. Gao, S. H. Kim, C. P. Jaroniec, J. R. Parquette,
Chem. Eur. J. 2011, 17, 12882–12885; c) J. H. Jung, Y. Do, Y. A.
Lee, T. Shimizu, Chem. Eur. J. 2005, 11, 5538–5544; d) Q. Jin, L.
Zhang, H. Cao, T. Wang, X. Zhu, J. Jiang, M. Liu, Langmuir 2011,
27, 13847–13853; e) Y. Tang, L. Zhou, J. Li, Q. Luo, X. Huang, P.
Wu, Y. Wang, J. Xu, J. Shen, J. Liu, Angew. Chem. 2010, 122, 4012–
4016; Angew. Chem. Int. Ed. 2010, 49, 3920–3924.
[17] J. D. Luo, Z. L. Xie, J. W. Y. Lam, L. Cheng, H. Y. Chen, C. F. Qiu,
H. S. Kwok, X. W. Zhan, Y. Q. Liu, D. B. Zhu, B. Z. Tang, Chem.
Commun. 2001, 1740–1741.
[18] a) Z. Zhao, J. W. Y. Lam, B. Z. Tang, Soft Matter 2013, 9, 4564–
4579; b) Y. L. Chen, Y. X. Lv, Y. Han, B. Zhu, F. Zhang, Z. S. Bo,
C. Y. Liu, Langmuir 2009, 25, 8548–8555; c) P. Xue, R. Lu, G. Chen,
Y. Zhang, H. Nomoto, M. Takafuji, H. Ihara, Chem. Eur. J. 2007, 13,
8231–8239; d) T. H. Kim, M. S. Choi, B. H. Sohn, S. Y. Park, W. S.
Lyoo, T. S. Lee, Chem. Commun. 2008, 2364–2366; e) C. Y. Bao, R.
Lu, M. Jin, P. C. Xue, C. H. Tan, G. F. Liu, Y. Y. Zhao, Org. Biomol.
Chem. 2005, 3, 2508–2512; f) S. Y. Li, L. M. He, F. Xiong, Y. Li,
G. Q. Yang, J. Phys. Chem. B 2004, 108, 10887–10892.
[19] D. L. Allara, R. G. Nuzzo, Langmuir 1985, 1, 45–52.
[20] M. Jackson, H. H. Mantsch, Can. J. Chem. 1991, 69, 1639–1642.
[21] H. Cao, Q. Z. Yuan, X. F. Zhu, Y. P. Zhao, M. H. Liu, Langmuir
2012, 28, 15410–15417.
[7] R. S. Devan, R. A. Patil, J. H. Lin, Y. R. Ma, Adv. Funct. Mater.
2012, 22, 3326–3370.
[8] a) T. Shimizu, M. Masuda, H. Minamikawa, Chem. Rev. 2005, 105,
1401–1443; b) D. T. Bong, T. D. Clark, J. R. Granja, M. R. Ghadiri,
Angew. Chem. 2001, 113, 1016–1041; Angew. Chem. Int. Ed. 2001,
40, 988–1011; c) R. Chapman, M. Danial, M. L. Koh, K. A. Jolliffe,
S. Perrier, Chem. Soc. Rev. 2012, 41, 6023–6041; d) N. Kameta, H.
Minamikawa, M. Masuda, Soft Matter 2011, 7, 4539.
[9] a) W. Zhang, W. Jin, T. Fukushima, N. Ishii, T. Aida, J. Am. Chem.
Soc. 2013, 135, 114–117; b) Z. Wang, C. J. Medforth, J. A. Shelnutt,
J. Am. Chem. Soc. 2004, 126, 16720–16721; c) H. B. Liu, J. L. Xu,
Y. J. Li, Y. L. Li, Acc. Chem. Res. 2010, 43, 1496–1508; d) V. K.
Praveen, S. S. Babu, C. Vijayakumar, R. Varghese, A. Ajayaghosh,
Bull. Chem. Soc. Jpn. 2008, 81, 1196–1211; e) A. Ajayaghosh, C. Vi-
[22] P. F. Duan, L. Qin, X. F. Zhu, M. H. Liu, Chem. Eur. J. 2011, 17,
6389–6395.
[23] a) Y. Che, X. Yang, S. Loser, L. Zang, Nano Lett. 2008, 8, 2219–
2223; b) B.-P. Jiang, D.-S. Guo, Y. Liu, J. Org. Chem. 2010, 75, 7258–
7264; c) H. Peng, L. Ding, T. Liu, X. Chen, L. Li, S. Yin, Y. Fang,
Chem. Asian J. 2012, 7, 1576–1582; d) P. C. Xu, H. T. Yu, X. X. Li,
Anal. Chem. 2011, 83, 3448–3454; e) B.-P. Jiang, D.-S. Guo, Y. Liu,
J. Org. Chem. 2011, 76, 6101–6107.
Received: November 12, 2013
Published online: January 21, 2014
Chem. Asian J. 2014, 9, 770 – 778
778
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim