Synthesis and gelation behaviors of five new dimeric cholesteryl derivatives

Article abstract of DOI:10.1007/s11426-010-4208-4

Five new diacid amides of di-cholesteryl l-glycinates were designed and prepared. The compounds with linkers containing 0, 1, 2, 3, or 4 methylene units are denoted as 1, 2, 3, 4, and 5, respectively. Their gelation behaviors in 25 solvents were tested as novel low-molecular-mass organic gelators (LMOGs). It was shown that the length of the linker connecting the two-cholesteryl residues in a gelator plays a crucial role in the gelation behavior of the compound. 1 gels 11 of the 25 solvents tested at a concentration lower than 1.0%, while 2 gels 17 of the solvents tested. 4 and 5, however, gel only 2 and 4 of them, respectively. SEM observation reveals that the lengths of the linkers and the identity of the solvents are the main factors affecting the structures of the aggregates in the gels. Experimentally, a clear linker effect on the microstructures of the gels was observed. As example, the aggregates of 1, 2 and 3 in benzene or 1-heptanol adopt structures of thin fibers, rods or lamellas, respectively. Furthermore, it was found that the gelation and aggregation behaviors of 2, 3, 4, and 5 in DMSO showed an even-odd effect.

Full text of DOI:10.1007/s11426-010-4208-4

SCIENCE CHINA
Chemistry
March 2011 Vol.54 No.3: 475–482
doi: 10.1007/s11426-010-4208-4
• ARTICLES •
Synthesis and gelation behaviors of five new dimeric
cholesteryl derivatives
LIU KaiQiang, PENG JunXia, XUE Min, YAN Ni, LIU Jing & FANG Yu*
Key Laboratory of Applied Surface and Colloid Chemistry (Shaanxi Normal University), Ministry of Education; School of Chemistry and
Materials Science, Shaanxi Normal University, Xi’an 710062, China
Received March 21, 2010; accepted May 4, 2010
Five new diacid amides of di-cholesteryl L-glycinates were designed and prepared. The compounds with linkers containing 0, 1,
2, 3, or 4 methylene units are denoted as 1, 2, 3, 4, and 5, respectively. Their gelation behaviors in 25 solvents were tested as
novel low-molecular-mass organic gelators (LMOGs). It was shown that the length of the linker connecting the two-cholesteryl
residues in a gelator plays a crucial role in the gelation behavior of the compound. 1 gels 11 of the 25 solvents tested at a con-
centration lower than 1.0%, while 2 gels 17 of the solvents tested. 4 and 5, however, gel only 2 and 4 of them, respectively.
SEM observation reveals that the lengths of the linkers and the identity of the solvents are the main factors affecting the struc-
tures of the aggregates in the gels. Experimentally, a clear linker effect on the microstructures of the gels was observed. As
example, the aggregates of 1, 2 and 3 in benzene or 1-heptanol adopt structures of thin fibers, rods or lamellas, respectively.
Furthermore, it was found that the gelation and aggregation behaviors of 2, 3, 4, and 5 in DMSO showed an even-odd effect.
organogelators, cholesterol, linker effect, solvent effect
nature of the 3-D networks within the supramolecular gels
1 Introduction
promises accessibility for designing and constructing sen-
sors, actuators, and other molecular devices [13–15]. Most
of the gels based on LMOGs have been obtained by a heat-
ing-cooling method.
The phenomenon that some organic compounds with low
molecular mass can solidify organic solvents at low con-
centrations was discovered more than 50 years ago. These
compounds have been known as low-molecular-mass or-
ganic gelators (LMOGs). In recent years, physical gelation
of organic solvents by LMOGs has become one of the hot
areas in soft matter research because of its abundant poten-
tial applications [1–10]. The gels based on LMOGs are usu-
ally considered as supramolecular gels, in which the gelator
molecules self-assemble into three-dimensional (3-D) net-
works via various non-covalent interactions, such as hydro-
gen bonding, - stacking, van der Waals interaction, dipole-
dipole interaction, coordination, solvophobic interaction,
and host-guest interaction [1–9, 11, 12]. The non-covalent
The control of self-assembling mode at the molecular
level and the balance between precipitation and dissolution
in a given solvent are critical to the design of new LMOGs
[1–9]. This is because the gelation of a solvent by a LMOG
is the result of gelator-gelator and gelator-solvent interac-
tions, which involve specific and nonspecific intermolecular
forces. Therefore, the aggregation behavior of a gelator in
solvents can be adjusted by varying its structure, for exam-
ple, introducing directional self-assembling units, such as
amides, and urea [1–9]. As a versatile unit, the covalently
bound cholesteryl group with four rings fused together, of
which three rings are six-member rings and one is five-
member ring, has been widely chosen for designing new
LMOGs because of its unique directional self-association
*Corresponding author (email: yfang@snnu.edu.cn)
© Science China Press and Springer-Verlag Berlin Heidelberg 2011
chem.scichina.com
www.springerlink.com

476
Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
through van der Waals interactions in the aggregates of the
gelators comprising it [3]. Historically, molecules compris-
ing an aromatic (A) moiety and a steroidal (S) group, of
which the two units are connected by a linker (L), are called
ALS type LMOGs. These compounds can gel a range of
solvents including polar and apolar solvents, mainly de-
pending on the nature of A, L, and S components [1, 3, 16].
Another class of LMOGs is dimeric cholesterol-based de-
rivatives denoted as A(LS)2 or LS2 gelators. Via the com-
ponents of A and L, various functional units such as chro-
mophores, ligands, and polymerizable units have been suc-
cessfully incorporated into gelator structures, laying a
foundation for the applications of the gels [3, 17–22]. As an
example, Shinkai prepared a series of dimeric cholesterol
derivatives bearing various functional linkers as versatile
gelators, and some organogels comprising them were used
successfully as templates for the preparation of inorganic
materials possessing unique structures.
scanning electron microscopy spectrometer (Philips-FEI).
The xerogel was prepared by freezing the gel in liquid ni-
trogen, and then evaporated by a vacuum pump for 12–24 h.
The dried sample thus obtained was attached to a mica or
glass slice by conductive adhesive tape, and shielded by
gold. The accelerating voltage was 15 kV, and the emission
was 10 mA.
2.3 FTIR measurement
The solution and gel samples were measured by a Bruker
EQUINX55 spectrometer in an attenuated total reflection
(ATR) mode. The xerogel samples for measurements were
coated on a glass or mica slice as a gel film and frozen in
liquid nitrogen, and then evaporated by a vacuum pump for
12–24 h.
2.4 1H NMR measurement
In recent years, we designed and prepared some ALS, LS
and LS2 type gelators and studied their gelation behaviors in
various solvents [23–25]. It was found that the gelators of
LS2 type structures are generally much more efficient than
gelators of other structures. Furthermore, some organogels
comprising LS2 type gelators possess thixotropic properties
[26]. On the bases of these works and the structural infor-
mation cited above, we have designed and synthesized a
new class of LS2 type LMOGs, and regulated the gelation
and aggregation behaviors of these compounds by changing
the number of the methylene unit in the linker of them.
The gel sample containing one gelator and d6-benzene was
prepared in a NMR tube, and the chemical shifts of the
sample were detected by a Fourier Digital NMR spec-
trometer (AVANCF300MHZ, 300 MHz) at a given tem-
perature between 318 K and 338 K.
2.5 X-ray diffraction (XRD) analysis
The measurement was conducted using a Rigaku/MSC
D/Max-2550/PC X-ray diffraction system. Fresh gel sample
was directly loaded onto a rectangular glass sample holder
as smooth film, and then scanned immediately. The XRD
pattern was obtained using CuK radiation with an incident
wavelength of 0.1542 nm under a voltage of 40 kV and a
current of 50 mA. The scan rate was 0.5 °/min.
2 Experimental
2.1 Gelation test
A known weight of the potential gelator and a measured
aliquot of liquid were placed into a sealed glass tube (d =10
mm, V = 3 mL) and the system was heated in an oil or water
bath until all solid materials were dissolved completely.
Then, the solution was cooled to room temperature in air
and the test tube was inversed to see if a gel was formed.
When the gelator formed a transparent, translucent or white
gel by immobilizing the solvent at this stage, it was denoted
as “TG”, “TLG” or “G”. For the systems in which only so-
lution remained until the end of the tests, they were referred
to as solution (S). When the gelator formed into precipitate
in some solvent, it was denoted as a “precipitate (P)”. The
system in which the potential gelator could not be dissolved
even at the boiling point of the solvent was designated as an
insoluble system (I). In a few cases, turbid solution (TUS)
or viscous solution (VS) was also observed at 3.0% of the
gelator. Critical gelation concentration (CGC) refers to the
minimum concentration of the gelator for gel formation.
3 Synthesis of compounds
3.1 Cholesteryl glycinate primary amine
Step (a): 0.175 g (1 mmol) Boc-glycin and 0.387 g (1 mmol)
cholesterol were dissolved in 40 mL dichloromethane. The
solution was maintained at 0 °C using an ice bath. 0.206 g
(1 mmol) dicyclohexylcarbodiimide (DCC) and 0.012 g (0.1
mmol) N,N-dimethylamino-pyridine (DMAP) were added,
and the reaction mixture was being stirred for 4 h at 0 °C.
After the reaction, the mixture was filtered and the filtrate
was washed with 0.001 mol L1 hydrochloric acid (30 mL
3), 0.001 mol L1 sodium hydroxide solutions (30 mL  3)
and pure water (30 mL  3). The organic layer was evapo-
rated to dryness. The residue was purified by a silica gel
column eluting with THF/n-hexane (1:6, v/v) to give cho-
lesteryl Boc-glycinate in 64% yield as a white solid. Step
(b): 0.544 g (1 mmol) cholesteryl Boc-glycinate was dis-
solved in 100 mL dichloromethane and then bubbled dry
HCl gas for 30 min. The mixture was filtered and the resi-
2.2 SEM measurement
SEM pictures of the xerogel were taken on a Quanta 200

Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
477
due was washed with dichloromethane three times and dried
in vacuum to give the desired product in 90% yield as a
white crystal. For hydrochloric salts of cholesteryl glycinate
1H NMR (CDCl3/Me4Si, 300 MHz):  (ppm) 5.37 (1H, al-
kenyl), 4.66–4.69 (1H, m, oxycyclohexyl), 3.98 (2H, d,
CH2CO), 2.31–2.34 (2H, d, CH2), 0.67–1.85 (42H, m, cho-
lesteryl protons). Elemental analysis, calculated for
C29H50O2NCl (Mw = 479.35): C 72.54, H 10.50, N 2.92;
Found: C, 72.21, H, 10.40, N, 2.88. Step (c): 1.45 mL
triethylamine was added to the solution of 5.00 g hydro-
chloric salts of cholesteryl glycinate in 200 mL benzene.
The mixture was stirred and refluxed for 5 h, after which the
formed precipitate was filtered off, the resulting solution
was evaporated to dryness, and the solid was dried in vacuo
to give cholesteryl glycinate primary amine in 78% yield as
a white or yellowish solid.
3.4 Dimeric cholesterol-based amino acid derivative
3–5
Two mmol (0.887 g) cholesteryl glycinate primary amine
and 1 mmol dicarboxylic acid were dissolved in 40 mL THF.
The solution was maintained at 0 °C using an ice bath.
2 mmol DCC and 0.2 mmol DMAP were added, and the
reaction mixture was stirred for 4–7 h at 0 °C. After the
reaction, the mixture was filtered and the filtrate was
evaporated to dryness. The resulting solid was washed by
hot methanol three times to give in a dimeric choles-
terol-based amino acid derivative in 70%–90% yield as a
1
white solid. For 3: H NMR (CDCl3/Me4Si, 300 MHz): 
(ppm) 6.51 (2H, s, CONH), 5.38 (2H, s, alkenyl), 4.64–4.67
(2H, m, oxycyclohexyl), 3.99–4.01 (4H, d, CH2COO, J =
4.62 Hz), 2.31–2.34 (4H, d, CH2, J = 7.52), 1.85–2.00 (4H,
m, CH2CH2), 0.68–1.62 (82H, m, cholesteryl protons).
Elemental analysis, calculated for C62H100N2O6 (Mw =
969.47): C 76.81, H 10.40, N 2.88; Found: C 75.48, H 10.38,
N 2.79. For 4: 1H NMR (CDCl3/Me4Si, 300 MHz):  (ppm)
7.77 (2H, CONH), 5.39 (2H, s, alkenyl), 4.65–4.67 (2H, m,
oxycyclohexyl), 4.05–4.06 (4H, d, CH2COO, J = 4.2 Hz),
2.30–2.33 (4H, d, CH2, J = 8.99), 1.86–2.04 (6H, m,
CH2CH2CH2), 0.68–1.58 (82H, m, cholesteryl protons).
Elemental analysis, calculated for C63H102N2O6 (Mw =
983.49): C 76.94, H 10.45, N 2.85; Found: C 76.76, H 10.13,
N 2.847. For 5: 1H NMR (CDCl3/Me4Si, 300 MHz):  (ppm)
7.77 (2H, CONH), 5.39 (2H, alkenyl), 4.65–4.67 (2H, m,
oxycyclohexyl), 4.05–4.15 (4H, d, CH2COO, J = 4.19 Hz),
2.30–2.33 (4H, d, CH2, J = 2.99), 1.86–2.04 (8H, m,
CH2CH2CH2CH2), 0.68–1.60 (82H, m, cholesteryl protons).
Elemental analysis, calculated for C64H104N2O6 (Mw =
997.52): C 77.06, H 10.51, N 2.81; Found: C 76.84, H 10.38,
N 2.75.
3.2 Dimeric cholesterol-based amino acid derivative 1
Two mmol (0.887 g) cholesteryl glycinate primary amine
and 2 mmol (278.2 L) TEA were dissolved in 40 mL dried
benzene. The solution was maintained at 0 °C using an
ice-water bath. 40 mL benzene with 1 mmol (85.8 L) oxa-
lyl chloride was added dropwise to the above solution under
stirring. The reaction mixture was stirred for 4–5 h at 0 °C.
After the reaction, the resulting mixture was filtered and the
filtrate was evaporated to dryness. The solid was washed
with hot methanol three times to give a dimeric cholesterol-
based amino acid derivative 1 in 80% yield as a white solid.
1
For 1: H NMR (CDCl3/Me4Si, 300 MHz):  (ppm) 7.79
(2H, s, CONH), 5.39 (2H, s, alkenyl), 4.69–4.72 (2H, m,
oxycyclohexyl), 4.07–4.08 (4H, d, CH2COO, J = 5.25 Hz),
2.23–2.36 (4H, d, CH2, J = 7.29), 0.68-1.99 (82H, m,
cholesteryl protons). Elemental analysis, calculated for
C60H96N2O6 (Mw = 941.41): C 76.55, H 10.28, N 2.98;
Found: C 76.89, H 10.34, N 2.68.
4 Results and discussion
4.1 Design and synthesis of the gelators
3.3 Dimeric cholesterol-based amino acid derivative 2
It has been demonstrated in former research that organic
diacid monoamides of cholesteryl glycinate are poor gela-
tors for organic solvents [25]. However, neutralization of
the acids with ammonia enhanced their gelling ability sig-
nificantly [25]. Referring to the structures of these com-
pounds, another simple way to enhance their gelling ability
is to modify their structures by connecting two molecules in
order to change their structure from LS type to LS2 type. In
this way, novel dimeric cholesteryl derivatives should be
formed (c.f. Scheme 1). It was expected that the new com-
pounds are efficient LMOGs because (1) more cholesteryl
units in a molecule are favorable for the aggregation of the
compounds in solution, (2) similarly, increase in the number
of amide residue is also favorable for the aggregation due to
formation of hydrogen bonds and (3) change of the linker
lengths is another factor to adjust the balance between
Two mmol (0.887 g) cholesteryl glycinate primary amine
and 1 mmol (0.104 g) malonic acid were dissolved in 40 mL
dried THF. The solution was maintained at 40–50 °C using
a thermostated water bath. 2 mmol (0.412 g) DCC was
added, and the reaction mixture was stirred for 10–12 h at
40–50 °C. After the reaction, the mixture was filtered and
the filtrate was evaporated to dryness. The resulting solid
was washed with hot methanol three times to give a dimeric
cholesterol-based amino acid derivative 2 in 78% yield as a
1
white solid. For 2: H NMR (CDCl3/Me4Si, 300 MHz): 
(ppm) 7.19 (2H, CONH), 5.39 (2H, alkenyl), 4.67–4.79 (2H,
m, oxycyclohexyl), 4.03–4.04 (4H, d, CH2COO, J = 3.15
Hz), 2.32–2.35 (4H, d, CH2, J = 7.03), 2.03 (4H, s, CH2),
0.68–1.89 (82H, m, cholesteryl protons). Elemental analysis,
calculated for C61H98N2O6 (Mw=955.44): C 76.68, H 10.34,
N 2.93; Found: C 76.48, H 9.99, N 2.84.

478
Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
Scheme 1 Structure of amino acid derivatives of dimeric cholesterols. n = 0, 1, 2, 3, 4; named 1, 2, 3, 4, 5.
dissolution and precipitation. This also increases the chance
to find good LMOGs.
4.2 Gelation behaviors of the compounds
The gelation performances of the compounds, 1–5, in 25
solvents are listed in Table 1. Examination of the table re-
veals that 1, 2, and 3 are more versatile and efficient gela-
tors than 4 and 5. The data shown in Table 1 indicate that
the increase of the linker length between the two-cholesteryl
residues decreases the gelling abilities of the compounds.
Specifically, 1, 2 and 3 gel 11, 17 and 11 of the solvents
tested, respectively. 4 and 5, however, gel only 2 and 4 of
them, respectively. This result shows that a longer linker in
Synthesis of 1–5 has been accomplished by using two
different routes. One is that the primary amine was coupled
with acryl chloride in a suitable solvent. Another is the
condensation between the two carboxyl groups of organic
diacids and the primary amine group of the reactants with
dicyclohexylcarbodiimide (DCC) and N,N-dimethylamino-
pyridine (DMAP) as catalysts. For the system, in which
malonic acid was taken as a reactant, DMAP could not be
used due to its reactivity with the methylene unit of the diacid.
Table 1 Gelation behaviors of some amino acid derivatives of dimeric cholesterols
Solvents
Methanol
Ethanol
1
I
MGC
2
I
MGC
3
I
MGC
4
I
MGC
5
P
MGC
I
G
G
G
G
G
G
G
G
G
TLG
TLG
P
1.25
0.92
0.91
0.76
1.21
1.20
1.19
1.19
1.18
1.70
1.69
I
I
P
1-Propanol
1-Butanol
1-Pentanol
1-Hexanol
1-Heptanol
1-Octanol
1-Nonanol
1-Decanol
Toluene
P
P
P
P
P
P
P
P
P
P
P
P
P
I
P
G
0.26
0.33
0.20
0.28
0.40
0.89
0.36
P
P
Ga)
Ga)
Ga)
Ga)
Ga)
Ga)
VS
TLG
TUS
P
P
P
G
G
G
G
G
TG
TG
TG
P
3.40
3.31
1.79
1.78
1.77
4.15
3.47
4.34
P
P
P
P
P
P
Benzene
Xylene
3.10
P
G
G
P
4.40
3.23
Ethyl acetate
Butyl acetate
DMSO
G
G
G
G
TGa
P
1.10
1.12
2.70
1.25
0.17
G
0.38
0.16
G
P
3.36
TLG
G
P
I
3.37
2.57
TUS
Ga)
I
P
DMF
P
P
Cyclohexane
1-Heptane
TEA
G
TUS
G
P
0.43
1.02
TLG
I
1.5
P
I
P
P
P
P
I
P
Acetic acid
Diethyl ether
Acetone
G
0.94
G
I
0.81
G
I
0.38
I
I
P
P
P
I
I
THF
TUS
I
S
S
S
I
S
H2O
I
I
I
TLG, translucent gel; I, Insoluble; TG, transparent gel; TUS, turbid solution; S, transparent solution; P, precipitate. a) The color of gel is changing from
white to translucent with the decrease of gelator concentration.

Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
479
such gelators is not favorable for the gelation of most of the
organic solvents. Moreover, in most cases, for a given sol-
vent the minimum gelation concentrations (MGCs) of these
gelators increase with increasing the linker lengths. The
reasons for the linker effect on the gelation behaviors can be
the change of the spatial conformation of the gelators as
revealed by computer simulation (c.f. Figure S1 in Support-
ing Information), which may decrease the ability of the ge-
lator molecules to self-assemble into ordered structures, a
necessity for forming network structures.
gates shown in the SEM images may be rationalized by
considering a commonly accepted idea that highly direc-
tional intermolecular interactions, such as hydrogen bond-
ing or metal-ligand interactions, favor formation of fiber
structures. The increase in the number of methylene unit in
the linkers of the gelators provides the gelators more op-
portunities in adopting conformations of lower energies,
which must weaken the hydrogen-bonding interactions be-
tween gelator molecules, and result in loss of flexible fine
fibrous structure [27].
It is helpful to have a close inspection of the gelation be-
haviors of compounds 2, 3, 4 and 5 in DMSO. Table 1
shows that 3 and 5 do not gel DMSO even at a concentra-
tion of 5.0%, but 2 and 4 form stable gels in DMSO at con-
centrations below 3.0%. More interestingly, the structures
of the xerogels of 2 and 4 from DMSO are characterized by
lamellar structures (c.f. Figure 2(a), and (c)). By contrast,
the precipitates of 3 and 5 from DMSO take micro-particle
like structures (c.f. Figure 2(b) and (d)). Both the gelation
behaviors of the compounds and their aggregate structures
show an even-odd effect considering the number of the me-
thylene units in the linkers. This kind of phenomena con-
cerning the self-assembling morphologies has already been
reported by other groups [28–30]. The reasons for the effect
are not clear at this moment, but changes in the gelator-
gelator and gelator-solvent interactions are the origin of the
effect which may be interrogated by conducting FTIR
measurements as described below.
4.3 Morphology studies
To obtain a visual insight into the gel microstructures, the
linker effects, solvent effects and concentration effects on
the structures of the gel networks were studied by SEM
technique.
Figure 1 shows some typical SEM images of the gels. It
shows that the microstructures of the xerogels of 1, 2, and 3
in benzene are significantly different from each other, and
the morphologies of the aggregates change from the fiber,
rod to lamella with increasing the linker lengths of the ge-
lators. A similar linker effect was also observed in other
systems as shown in Figures 1(d)–(f), and Figure S2 (c.f.
Supporting Information). The morphologies of the aggre-
The polarity of a solvent may have a great effect on its
interaction with a given gelator. Accordingly, solvents of
different polarities (The relative polarities of acetic acid,
DMSO, benzene, and cyclohexane are 64.8, 44.4, 11.1, and
0.6, respectively.) have been chosen as sample solvents to
investigate the solvent effect on the aggregation and gelation
Figure 1 SEM images of the xerogels from the gels of 1/benzene (a),
2/benzene (b), and 3/benzene (c) and those from 1/1-heptanol (d),
2/1-heptanol (e), and 3/1-heptanol (f).
Figure 2 SEM images of the xerogels of 2 (a) and 4 (c) from DMSO and
those of the precipitates of 3 (b) and 5 (d) from the same solvent.

480
Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
behaviors of the compounds. The microstructures of the
xerogels of 2 in these solvents are shown in Figure S3 (c.f.
Supporting Information). It shows that 2 aggregated into
thick fibers and lamella structures in acetic acid and DMSO,
respectively. However, it aggregated into thin fabric bun-
dles in cyclohexane and benzene. According to the opinion
of Dordick [27], the different aggregation behavior of 2 in
different solvents can be explained by considering the dif-
ferent strengths of the gelator-solvent interactions. It is be-
lieved that in the presence of stronger gelator-solvent inter-
action, particularly in the presence of specific gelator-solvent
interaction, more solvent molecules will occupy the hydro-
gen bonding sites of the gelator, resulting in weakening of
the gelator-gelator interaction. As a result, gelator mole-
cules tend to form thicker fibers or lamellas via the media-
tion of solvent molecules.
To obtain information of the effect of gelator concentra-
tion on the molecular aggregation, the microstructures of
freezing-dried cyclohexane gels of 2 at concentrations from
0.5% to 3.0% were studied by conducting SEM measure-
ment. The results are shown in Figure 3. It shows that 2
aggregated into rigid and thick fibers at a high concentration
(3.0%) in cyclohexane, and flexible and thin fibers at lower
concentrations (2.0%, 1.0% and 0.5%). At the same time,
the fresh gels changed from white to transparent with de-
creasing the gelator concentrations. These results suggest
that the rigid and thick fiber may be clusters of many flexi-
ble fine fibers.
and 2 in benzene (gel state, 2.0%) were recorded and are
shown in Figure 4. It shows that the three typical absorption
bands of 2 corresponding to the stretching vibrations of
NH and C=O, and the bending vibration of NH appeared
at 3448.1 cm1, 1655.6 cm1, and 1537.6 cm1, respectively.
Upon gelation, the bands shifted to 3420.3 cm1, 1640.3
cm1 and 1558.4 cm1, respectively, suggesting direct par-
ticipation of the groups to intermolecular interactions. Con-
sidering the aprotic identity of the solvents involved, the
potential hydrogen bond formation properties of the groups
and the relatively extended conformation adopted by 2 (c.f.
Figure S1, Supporting Information), we believe changes in
the FTIR absorptions may be indications of intermolecular
hydrogen bond formation. This tentative conclusion was
further supported by the result from temperature-dependent
1H NMR studies. In the studies, the gel of 2/d6-benzene was
used as a sample, and the dissolution of the gel was moni-
1
tored by H NMR measurement. The results are shown in
Figure 5. Structurally, the solvent has no possibility to par-
ticipate in the hydrogen bond formation, and thereby the
hydrogen bond can solely be ascribed to the gelator-gelator
interaction provided it is present in the system. With refer-
4.4 Hydrogen bond formation
FTIR spectroscopy is a powerful technique to verify hydro-
gen bond formation between molecules [31, 32]. Accord-
ingly, the FTIR spectra of 2 in CDCl3 (solution state, 2.0%)
Figure 4 FTIR spectra of 2/CDCl3 solution and 2/benzene gel.
Figure 3 SEM images of the xerogels of 2 from cyclohexane at different
concentrations. (1) 3.0%; (2) 2.0%; (3) 1.0%; (4) 0.5%.
Figure 5 Temperature-dependent 1H NMR spectra of 2/benzene gel.

Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
481
ence to Figure 5, the signal of the amide proton of 2 shifts to
higher field along with increasing the temperature of the
system, an evidence of disruption of the hydrogen bonds, in
which the amide proton has been involved [33].
As mentioned previously, the gelation behaviors and the
morphologies of the aggregates of 2, 3, 4, and 5 in DMSO
showed a remarkable even-odd effect (c.f. Table 1, and Fig-
ure 2). This effect is closely related with the effect of the
linker length on the packing of both the headgroups and the
linkers of the compounds, and the packing, particularly that
based on hydrogen bond formation, could be investigated
by conducting FTIR studies. Accordingly, the FTIR spectra
of the aggregates of the four compounds from DMSO were
measured, and the characteristic frequencies of the NH and
C=O stretching vibrations, and the NH bending vibration
of the amide groups in the linkers were recorded. The fre-
quencies were plotted as functions of the methylene num-
bers in the linkers (c.f. Figures S4-S6, Supporting Informa-
tion). The figures reveal, as expected, an even-odd phe-
nomenon, which is a fluctuation change for the chain
lengths (1 n 4), an indication of the difference in the
strength of the intermolecular hydrogen-bonds of the four
compounds, which may affect the balance of dissolution
and precipitation and thereby results in different gelation
behaviors.
Figure 6 X-ray diffraction trace of the fresh gel of 2/benzene (3%).
studied except the one containing 5, the spacing of the first
peak is very close to the length of a complex of two
molecules, rather than the length of a molecule itself,
particularly that of compound 4. This demonstrates again
that the linker length has a great effect on the packing mode
of the compounds. It is noteworthy that the intensities of the
XRD signals of the systems containing 2 or 4 are much
higher than those of the systems containing 3 or 5, which is
another even-odd effect, confirming the observations
obtained in the gelation test and in the SEM and FTIR
studies of the compounds. Considering the XRD results
described above and the hydrogen bonding nature of the
linker packing of the compounds as confirmed by FTIR and
1H NMR measurements, a possible packing mode of the
compounds in the gels studied was proposed and is sche-
matically shown in Figure 7.
4.5 XRD study
The XRD pattern of a fresh gel of 2/benzene is shown in
Figure 6. The curve for the fresh gel sample shows four
peaks in the low angle region (2 values, 2.12°, 4.26°, 6.42°,
and 8.54°) corresponding to d values of 4.13 nm, 2.06 nm,
1.37 nm and 1.04 nm, respectively, suggesting a lamellar
packing of the gelator in the gel. Similar XRD measure-
ments were conducted on the fresh aggregates of 2, 3, 4, and
5 from DMSO, and the results are shown in Figures S7–S10
(c.f. Supporting Information). The traces shown in Figures
S7–S9 reveal that each trace possesses three sharp reflection
peaks, and the corresponding d values follows a ratio of 1:
1/2: 1/3, suggesting a lamellar structure of the aggregates. It
is noteworthy that the precipitate of 5 from DMSO did not
show any significant XRD signals in the low angle region,
indicating that 5 aggregated in an amorphic state. Further
inspection of the figures, reveals that for each systems
5 Summary
In summary, it has been demonstrated that the gelation be-
haviors of the novel dimeric cholesteryl derivatives in vari-
ous organic solvents can be regulated by changing the
number of the methylene units of the linker part in the
LS2-type molecules. The compounds containing relatively
shorter linkers can gel more solvents than their analogues
with longer linkers. Increase in the lengths of the linkers of
the compounds results in complete loss of the fibrous
Figure 7 A possible aggregation mode of gelator 2 in benzene.

482
Liu KQ, et al. Sci China Chem March (2011) Vol.54 No.3
immobilize quantum dots. Chem Mater, 2008, 20: 1522–1526
16 Wu JC, Yi T, Xia Q, Zou Y, Liu F, Dong J, Shu TM, Li FY, Huang
CH. Tunable gel formation by both sonication and thermal processing
in a cholesterol-based self-assembly system. Chem Eur J, 2009, 15:
6234–6243
17 Sugiyasu K, Fujita N, Shinkai S. Fluorescent organogels as templates
for sol-gel transcription toward creation of optical nanofibers. J
Mater Chem, 2005, 15: 2747–2754
18 Jong JH, Nakashima K, Shinkai S. Preparation of ultrastable
mesoporous silica using a phenanthroline-appended cholesterol or-
ganogelator as a template. Nano Lett, 2001, 1: 145–148
19 Jong JH, Ono Y, Shinkai S. Novel silica structures which are pre-
pared by transcription of various superstructures formed in organo-
gels. Langmuir, 2000, 16: 1643–1649
structure of the aggregates in the gel sate. The morphologies
of the aggregates are also affected by the solvent nature.
Generally, a less polar solvent favors formation of thin fi-
bers. The gelator concentration is also a significant factor
affecting the morphology of its aggregate in gels. Lower
concentration is favorable for the formation of networks
composed of flexible, thin fibers. FTIR and tempera-
ture-dependent 1H NMR studies reveal that hydrogen bond-
ing between the amide NH and the amide C=O of the ge-
lator is one of the main driving forces for the formation of
2/benzene gel.
20 Jung, JH, Kobayashi H, Masuda M, Shimizu T, Shinkai S. Helical
ribbon aggregate composed of a crown-appended cholesterol deriva-
tive, which acts as an amphiphilic gelator of organic solvents and as a
template for chiral silica transcription. J Am Chem Soc, 2001, 123:
8785–8789
This work was supported by the Ministry of Science and Technology of
China (2007AA03Z349), and the National Natural Science Foundation of
China (20773083, 20902055).
21 Jung JH, Kobayashi H, van Bommel KJC, Shinkai S, Shimizu T.
Creation of novel helical ribbon and double-layered nanotube TiO2
structures using an organogel template. Chem Mater, 2002, 14:
1445–1447
22 Wu JC, YiT, Zou Y, XiaQ, Shu T, Liu F, Yang YH, Li FY, Chen ZG,
Zhou ZG, Huang CH. Gelation induced reversible syneresis via
structural evolution. J Mater Chem, 2009, 19: 3971–3978
23 Li YG, Liu KQ, Liu J, Peng JX, Feng XL, Fang Y. Amino acid de-
rivatives of cholesterol as “latent” organogelators with hydrogen
chloride as a protonation reagent. Langmuir, 2006, 22: 7016–7020
24 Li YG, Fang Y, Liu J, Wang MZ. Synthesis and gelation behavior of
cholesteryl glycinate anthraquinone-2-carboxylamide and cholesteryl
glycinate 9,10-dimethyloxyl anthracene-2-carboxylamide. J Chin
Chem Soc, 2006, 53: 359–366
25 Liu KQ, Yan N, Peng JX, Liu J, Zhang QH, Fang Y. Supramolecular
gels based on organic dioic acid mono-amides of cholesteryl glyci-
nate. J Colloid Interface Sci, 2008, 327: 233–242
1
2
3
4
Terech P, Weiss RG. Low molecular mass gelators of organic liquids
and the properties of their gels. Chem Rev, 1997, 97: 3133–3160
Liu XY. Gelation with small molecules: From formation mechanism
to nanostructure architecture. Top Curr Chem, 2005, 256: 1–37
Žinić M, Vögtle F, Fages F. Cholesterol-based gelators. Top Curr
Chem, 2005, 256: 39–76
Fages F, Vögtle F, Žinić M. Systematic design of amide- and urea-
type gelators with tailored properties. Top Curr Chem, 2005, 256:
77–131
Araki K, Yoshikawa I. Nucleobase-containing gelators. Top Curr
Chem, 2005, 256: 133–165
Brizard A, Oda R, Huc I. Chirality effects in self-assembled fibrillar
networks. Top Curr Chem, 2005, 256: 167–218
5
6
7
Kato T, Mizoshita N, Moriyama M, Kitamura T. Gelation of liquid
crystals with self-assembled fibers. Top Curr Chem, 2005, 256:
219–236
Hirst AR, Smith DK. Dendritic gelators. Top Curr Chem, 2005, 256:
237–273
26 Peng JX, Liu KQ, Liu J, Zhang QH, Feng XL, Fang Y. New dicho-
lesteryl-based gelators: Chirality and linker length effect. Langmuir,
2008, 24: 2992–3000
8
9
27 Zhu GY, Dordick JS. Solvent effect on organogel formation by low
molecular weight molecules. Chem Mater, 2006, 18: 5988–5995
28 Kogiso M, Ohnishi S, Yase K, Masuda M, Shimizu T. Dicarboxylic
oligopeptide bolaamphiphiles: Proton-triggered self-assembly of mi-
crotubes with loose solid surfaces. Langmuir, 1998, 14: 4978–4986
29 Nakazawa I, Masuda M, Okada Y, Hanada T, Yase K, Asai M,
Shimizu T. Spontaneous formation of helically twisted fibers from
2-glucosamide bolaamphiphiles. Langmuir, 1999, 15: 4757–4764
30 Wang TY, Li YG, Liu MH. Gelation and self-assembly of glutamate
bolaamphiphiles with hybrid linkers: Effect of the aromatic ring and
alkyl linkers. Soft Matter, 2009, 5: 1066–1073
31 Shimizu T, Masuda M. Stereochemical effect of even-odd connecting
links on supramolecular assemblies made of 1-glucosamide bolaam-
phiphiles. J Am Chem Soc, 1997, 119: 2812–2818
32 Kogiso M, Ohnishi S, Yase K, Masuda M, Shimizu T. Dicarboxylic
oligopeptide bola-amphiphiles: Proton-triggered self-assembly of
microtubes with loose solid surfaces. Langmuir, 1998, 14: 4978–4986
33 Liu J, Yan JL, Yuan XW, Liu KQ, Peng JX, Fang Y. A novel low- mo-
lecular-mass gelator with a redox active ferrocenyl group: Tuning gel
formation by oxidation. J Colloid Interface Sci, 2008, 318: 397–404
Carretti E, Dei L, Weiss RG. Soft matter and art conservation.
Rheoreversible gels and beyond. Soft Matter, 2005, 1: 17–22
10 Xue PC, Lu R, Li DM, Jin M, Bao CY, Zhao YY, Wang ZM. Rear-
rangement of the aggregation of the gelator during sol-gel transcrip-
tion of a dimeric cholesterol-based viologen derivative into fibrous
silica. Chem Mater, 2004, 16: 3702–3707
11 Li CP, Yu Q, Du M. Structural modulation of Cd(II) supramolecular
frameworks with a versatile 2,4-dipyridyl-type building block and
different dicarboxylate ligands. Sci China Ser B-Chem, 2009, 52:
1470–1478
12 Lin YY, Qiao Y, Yan Y, Huang JB. Thermo-responsive viscoelastic
wormlike micelle to elastic hydrogel transition in dual-component
systems. Soft Matter, 2009, 5: 3047–3053
13 Balzani V, Credi A, Raymo FM, Stoddart JF. Artificial molecular
machines. Angew Chem Int Ed, 2000, 39: 3348–3391
14 de Silva AP, Gunaratne HQN, Gunnlaugsson T, Huxley AJM,
McCoy CP, Rademacher JT, Rice TE. Signaling recognition events
with fluorescent sensors and switches. Chem Rev, 1997, 97: 515–1566
15 Yan XH, Cui Y, He Q, Wang KW, Li JB. Organogels based on
self-assembly of diphenylalanine peptide and their application to