Synthesis and self-assembling properties of 4,6?O-benzylidene acetal protected D-glucose and D-glucosamine β?1,2,3?triazole derivatives

Article abstract of DOI:10.1016/j.carres.2018.02.011

Sugar based low molecular weight gelators (LMWGs) are useful small molecules that can form reversible supramolecular gels with many applications. Selective functionalization of common monosaccharides has resulted in several classes of effective LMWGs. Rec

Full text of DOI:10.1016/j.carres.2018.02.011

Accepted Manuscript
Synthesis and self-assembling properties of 4,6−benzylidiene acetal protected D-
glucose and D-glucosamine β−1,2,3−triazole derivatives
Anji Chen, Ifeanyi Okafor, Consuelo Garcia, Guijun Wang
PII:
S0008-6215(17)30694-8
10.1016/j.carres.2018.02.011
CAR 7527
DOI:
Reference:
To appear in: Carbohydrate Research
Received Date: 22 September 2017
Revised Date: 17 February 2018
Accepted Date: 17 February 2018
Please cite this article as: A. Chen, I. Okafor, C. Garcia, G. Wang, Synthesis and self-assembling
properties of 4,6−benzylidiene acetal protected D-glucose and D-glucosamine β−1,2,3−triazole
derivatives, Carbohydrate Research (2018), doi: 10.1016/j.carres.2018.02.011.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Synthesis and self-assembling properties of 4,6-benzylidiene acetal protected
D-glucose and D-glucosamine β-1,2,3-triazole derivatives
Anji Chen, Ifeanyi Okafor, Consuelo Garcia, Guijun Wang*
Department of Chemistry and Biochemistry, Old Dominion University, Norfolk, VA 23529, USA.
*
Corresponding Author:
Guijun Wang, PhD
Professor of Chemistry and Biochemistry
Old Dominion University
4
541 Hampton Boulevard
Norfolk, VA 23529-0126
Email: g1wang@odu.edu
Telephone: (757) 683-3781
Fax: (757) 683-4628
1

ACCEPTED MANUSCRIPT
Abstract:
Sugar based low molecular weight gelators (LMWGs) are useful small molecules that can form
reversible supramolecular gels with many applications. Selective functionalization of common
monosaccharides has resulted in several classes of effective LMWGs. Recently we found that
certain peracetylated sugars containing anomeric triazole functional groups were effective
gelators. In this study we synthesized two series of 4,6-O-benzylidene acetal protected β-1,2,3-
triazolyl glycoside of D-glucose and N-acetyl D-glucosamine derivatives and evaluated their
self-assembling properties in a few solvents. Several gelators were obtained and the gelation
properties of these compounds rely on the structures of the 4-triazolyl substituents. Typically,
alkyl derivatives resulted in effective gelation in organic solvents and aqueous mixtures of
ethanol and dimethyl sulfoxide. But further acetylation of these compounds resulted in loss of
gelation properties. The gels were characterized using optical microscopy, rheology, and FTIR
1
spectroscopy. We also analyzed the molecular assemblies using H NMR spectroscopy to probe
the influences of the hydroxyl, amide, and triazole functional groups. Naproxen was used as a
model drug and it formed co-gels with compound 25 in DMSO water mixtures. Using UV
spectroscopy, we found that naproxen was slowly released from the gel to aqueous solution. The
general structure and gelation trend obtained here can be useful in designing sugar based
biomaterials. We expect that further structural optimization can lead to more effective gelators
that are compatible with different drug molecules for encapsulation and sustained release.
Keywords: D-Glucose; D-Glucosamine; Organogelator; Hydrogelator; Triazole; Self-
assembling; Benzylidene acetal
2

ACCEPTED MANUSCRIPT
Introduction
Carbohydrate based low molecular weight gelators (LMWGs) have received increasing attention
1
-5
over recent years due to their interesting physical chemical properties and many applications.
The gelators typically self-assemble and form fibrous supramolecular architectures that can
immobilize certain solvents and form reversible gels. The driving forces for the formation of
supramolecular gel networks are non-covalent forces such as hydrogen bonding, hydrophobic
interactions, π-π stacking, CH-π interactions, and van der Walls forces, etc. Low molecular
weight organogelators (LMOGs) have shown numerous applications for the formation of new
6
-8
supramolecular materials. They have been found to be useful for oil spill recovery by several
9
-22
research groups.
Organogels were also have studied for drug delivery and biomedical
2
3,24
applications.
For optical electronic device applications the solvents used are typically
hydrocarbons and volatile aromatic solvents such as toluene and xylenes. Polar organic solvents
such as ethanol, glycerol, ethylene glycol, dimethyl sulfoxide, etc. have been studied for
transdermal drug delivery applications using lectin and other gelators. However it is obvious for
toxicity and biocompatibility reasons, hydrogels would be more applicable for drug delivery and
other biomedical related applications. Supramolecular gels have also been utilized in synthesis
7
,25-29
and catalysis and environmental applications.
Some proline containing gelators were shown
3
0-33
to function as organocatalysts with enhanced catalytic activities in the gel state.
The structures of LMWGs encompass a broad range of functionalities and natural products
2
,3,34-36
including amino acids, peptides, sugars, lipids etc.
Carbohydrates are naturally abundant,
renewable resources and they contain multiple chiral centers. Selective functionalization of
sugars can afford new soft materials which are likely to be biocompatible and biodegradable. For
3

ACCEPTED MANUSCRIPT
instance, several readily available monosaccharides such as D-glucose and D-glucosamine have
3
7-43
been modified to form supramolecular gels with many different applications.
Despite the
many advancements on low molecular weight gelators, the rational design of small molecules to
function as effective gelators is still challenging. From a chemical structure point of view, we are
interested in understanding how the modification of simple sugars can lead to effective
molecular gelators and stimuli responsive materials. Previously, Shinkai et al have studied
gelation behaviors of 4,6-O-benzylidene acetal protected glucose, galactose, mannose, etc.
extensively and obtained useful structure and gelation properties relationship, the orientation of
4
4-48
the free 2,3-diol plays an important role.
We have also systematically studied the
functionalization of monosaccharides including D-glucose and D-glucosamine to obtain effective
4
9-53
LMWGs.
Further functionalization of hydroxyl groups at certain 4,6-O-benzylidene acetal
5
4-57
protected D-glucose afforded functional LMWGs.
As shown in Figure 1, certain derivatives of 4,6-O-benzylidene acetal protected D-glucose and
D-glucosamine derivatives 1 and 2 were found to be effective low molecular weight gelators
(
LMWGs). Recently we studied the peracetylated D-glucosyl triazoles and D-glucosamine
triazole derivatives (3 and 4) too and screened their gelation properties in several organic
5
1,52
solvents and aqueous mixtures.
Similarly to the benzylidene acetal derivatives, the
peracetylated triazoles are effective organogelators and are able to gelate alcohols, DMSO and
water mixtures, ethanol and water mixtures. Many D-glucosamine triazole derivatives are
5
2,53
effective low molecular weight hydrogelators too.
The N-acetyl glucosamine derivatives 4
are more effective at forming hydrogels when R group is long chain alkyl or alkyl with polar
functional groups. Based on the discussions of the structural features, we envision that
4

ACCEPTED MANUSCRIPT
combining the core structures of the two systems may lead to more effective gelators.
Incorporating triazole to the anomeric position of the 4,6-O-benzylidene acetal protected glucose
or glucosamine will give compounds I and II (Figure 1), the study of these new compounds may
allow us to better understand the structural requirement for effective gelation. The introduction
of benzylidene acetal functional groups may prevent gelation in certain solvents since the
additional intermolecular forces may be too overwhelming and leads to crystallization or
precipitation. From a structural point of view, it is useful to find out whether the popular
benzylidene acetal protective group is important for molecular self-assembly and gelation when
combined with triazole functional groups. As part of our effort in understanding structure and
gelation property relationships and rational design of effective supramolecular gelators that can
respond to different physical and chemical environment, we designed and synthesized a series of
monosaccharide derivatives containing 4,6-O-benzylidene acetal and anomeric β-(1,2,3)-triazole
functional groups. In addition to the triazole function, the phenyl group can lead to enhanced π-π
interactions, therefore these new molecules may be more suitable for nonpolar organic solvents
and aromatic solvents, and oils rather than polar solvents or aqueous mixture.
Figure 1. Structures of various protected D-glucose and D-glucosamine derivatives and the
hybrid compounds I and II.
5

ACCEPTED MANUSCRIPT
Results and Discussions:
In order to further explore the self-assembling properties of glycosyl triazole derivatives, we
introduced 4,6-O-benzylediene acetal protecting group to the glycosyl triazoles to afford
compounds with the general structures I and II. As mentioned earlier, the phenyl group is
expected to enhance π-π interactions and perhaps enhance gelation tendencies for aromatic
solvents. As shown in Schemes 1 and 2, we synthesized a series of triazole derivatives with the
general structures of I and II from the corresponding azido peracetate of D-glucose and D-
glucosamine (5 and 20). The selection of R group was based on our previous findings of
effective gelators. The headgroup compound 7 was synthesized by treating glycosyl azide 5 with
sodium methoxide in methanol, followed by protection with benzylidene acetal protective group.
Then terminal alkynes with different functional groups were selected for the click reaction using
5
8-61
copper sulfate and sodium L-ascorbate as the catalyst.
A series of new triazole derivatives 8-
1
9 with the general structure I were then obtained in good yields.
After these compounds were synthesized, their gelation properties in several solvents were tested
and the gelation results are shown in Table 1. Besides the solvents listed in the table, we also
analyzed hexane and THF, all compounds were insoluble in hexane, and all were soluble in THF
except 16, which was insoluble. All gels obtained were opaque except toluene gels, which were
translucent. For the D-glucose triazole derivatives, the 4-triazolyl alkyl derivatives are generally
more effective gelators than the derivatives with polar substituents. The n-butyl analog 8 formed
stable gels in isopropanol, pump oil and unstable gels in toluene and engine oil. The chlorobutyl
analog 9 formed gels in ethanol/water mixture and DMSO/water mixture. As the chain length
increases, the compounds are more effective for oils. The hexyl derivative 10 was the most
6

ACCEPTED MANUSCRIPT
efficient among these gelators, forming a gel in engine oil at 6.7 mg/mL concentration. The octyl
analog 11 was the most versatile gelator, forming gels in six of the tested solvents, including
isopropanol, ethanol/water, and DMSO/water mixtures. When the alkyl chains are longer (>8
carbons), the compounds are no longer effective gelators for the tested solvents. D-glycosyl
triazole 15 containing a phenyl group with an alkyl spacer can form gels in DMSO/water
mixtures. However, the hydroxyl bearing triazole analogues 17-19 didn’t form gels in any of the
tested solvents, possibly because the systems are too polar to function as a gelator, they are more
soluble for polar organic solvents. In general, these series of derivatives I are less effective
comparing to their peracetylated precursors 3. However, we were pleased to find that all of the
alkyl triazole derivatives 8, 10-13 were effective gelators for pump oil.
Scheme 1. Synthesis of benzylidene acetal protected β-triazolyl glucosides 8-19.
Table 1. Gelation test results of the partially protected D-glucosyl triazole derivatives I.
7

ACCEPTED MANUSCRIPT
EtOH: DMSO
:H
Comp
ound
Pum Engin Toluen i-
Structure R=
EtOH EG
H
2
O
2
O
H
2
O
p oil e oil
e
PrOH
(
1:2)
(1:2)
P
8
9
G20 UG
UG
P
G 20
P
S
P
S
S
S
P
S
S
S
P
P
I
S
S
G10
P
G10
G 6.7
G 20
I
1
1
1
1
0
1
2
3
G10 G 6.7
G20 G 20
P
P
I
P
G 20
S
UG G10
I
G20
I
P
S
P
P
P
I
G10 UG
P
P
I
I
1
4
I
I
I
P
S
S
P
S
I
1
1
5
6
UG
I
P
S
P
P
S
P
S
S
S
P
S
G 6.7
S
I
UG
P
1
1
7
8
I
I
I
I
I
P
S
S
S
S
S
S
P
S
P
S
I
P
1
9
I
I
I
S
S
S
S
S
p
All compounds were tested starting from 20 mg/mL. G, stable gel at room temperature, the
number is minimum gelation concentration (MGC) in mg/mL; UG, partial gel or unstable gel,
the concentration is 20 mg/mL unless otherwise specified. P, precipitation; S, soluble; I,
insoluble.
Previously we have shown that the peracetylated glycosyl triazole derivatives from glucosamine
5
2
were also effective gelators. To probe the effect of changing protective groups, here we have
synthesized the 4,6-O-benzylidene acetal protected glycosyl triazole derivatives II. As shown in
8

ACCEPTED MANUSCRIPT
Scheme 2, using the same alkynes as in the synthesis of glycosyl triazoles I, the derivatives II
were synthesized. Next their gelation properties in several solvents were tested, the results are
shown in Table 2. Besides the solvents shown in Table 2, we also found that at 20mg/mL, these
glucosamine derivatives were insoluble in hexane, but soluble in THF and ethylene glycol (EG).
Scheme 2. Synthesis of bezylidene acetal protected β-triazolyl glycosides 23-32.
For the D-glucosamine based triazoles 23-32, in contrast to the ester series I, these compounds
were considerably less effective as organogelators. The hexyl and octyl derivatives were the
most efficient gelators in this series; forming gels in toluene, ethanol/water and DMSO/water
mixtures. The compounds II were effective gelators for toluene when the triazolyl-4-substituents
were aliphatic or phenyl; the shorter alkyl chain derivatives were effective gelators towards
aqueous mixtures of DMSO and ethanol. It is interesting to observe that none of these
glucosamine triazole compounds were effective for oils. The gels were typically opaque while
some of them were clear gels. The representative gels are shown in Figure 2. In general, the D-
9

ACCEPTED MANUSCRIPT
glucosamine triazole series II are less likely to function as gelators in comparison to the
benzylidene acetal glucosamine series 2 and the peracetylated glycosyl β-triazole derivatives 4.
Table 2. Gelation properties of the triazole derivatives (II) synthesized from N-acetyl D-
glucosamine
EtOH:
DMSO:
Pum Engine
p oil oil
No.
Structure R=
Toluene i-PrOH EtOH H O
H O
2
H O
2
2
(
1:2)
(1:2)
2
2
3
4
I
I
I
I
I
I
S
S
S
S
UG20
S
O
G12
S
O
I
I
2
2
2
2
2
3
3
5
6
7
8
9
0
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
G 20
G 10
G 20
S
C
C
C
S
S
S
S
I
S
S
S
S
P
S
S
G 5.0
O
G 4.0
O
O
I
G 6.7
G 3.3
I
O
P
P
I
P
P
I
I
I
G 20
I
I
T
S
S
S
S
S
S
S
S
I
3
2
I
I
P
S
S
S
S
S
All compounds were tested starting from 20 mg/mL. G, stable gel at room temperature, the
numbers are MGCs in mg/mL; UG, partial gel or unstable gel; P, precipitation; S, soluble; I,
insoluble. The subscribed letters denote the gel appearances: C, clear or transparent; T,
translucent; O, opaque.
1
0

ACCEPTED MANUSCRIPT
(b)
(c)
(a)
Figure 2. Gel photos for a) compound 9 in EtOH:H O (v/v 1:2) at 10 mg/mL; b) compound 25 in
2
DMSO:H O (v/v 1:2) at 4 mg/mL; c) compound 26 in EtOH:H O (v/v 1:2) at 6.7 mg/mL.
2
2
Next we characterized the gels formed by compounds 9, 25, and 26 using an optical microscope.
As shown in Figure 3, in EtOH:H O (v/v 1:2), compound 9 formed long thin fibers as the major
morphology together with certain areas that were semi crystalline (left side of the image, Fig.
a). It is possible that the thinner fibers or wires formed rapidly without reaching equilibrium and
the larger tubules were composed of thinner tubes rearranged to form clusters. In DMSO:H O
2
3
2
(v/v 1:2) the gel formed by compound 9 showed uniform fibrous morphology (Fig. 3b). The
glucosamine derivative 25 in EtOH:H O (v/v 1:2) mixture formed long fibers in uniform
2
diameter (~1 µm) and up to 200 µm in length (Fig. 3c). And it formed long and narrower fibrous
2
structures in DMSO:H O (v/v 1:2) (Fig. 3d), the diameter was about 0.5 µm and up to 200 µm in
length, the fibers were more interconnected and curved. For both solvent systems, the fibers were
more uniform and didn’t appear to have different domains of morphology. The EtOH:H O (v/v
2
1
:2) gel of compound 11 showed uniform birefringent fibrous networks (Fig. 3e). The compound
6 formed a mixed morphology similarly to that of compound 9, the majority of the aggregates
2
appeared liquid crystalline and birefringent sheets or ribbons (Fig. 3f).
1
1

ACCEPTED MANUSCRIPT
Figure 3. Optical micrographs of dried gels formed by compound 9, 25, 11 and 26; a) compound
9
in EtOH:H
mg/mL; c) compound 25 in EtOH:H
v/v 1:2) at 4.0 mg/mL; e) compound 11 in EtOH:H
in EtOH:H O (v/v 1:2) at 6.7 mg/mL.
2
O (v/v 1:2) at 10.0 mg/mL; b) compound 9 in DMSO:H
O (v/v 1:2) at 5.0 mg/mL; d) compound 25 in DMSO:H
O (v/v 1:2) at 10.0 mg/mL; f) compound 26
2
O (v/v 1:2) at 10.0
2
2
O
(
2
2
1
2

ACCEPTED MANUSCRIPT
Next we analyzed the rheological properties of the gels formed by several compounds in
DMSO:H O (v/v 1:2). The storage modulus G′ and loss modulus G″ at different angular
2
frequencies of the gels formed by compounds 9, 15, 25, and 26 were shown in Figure 4. The
storage moduli for all four compounds were greater than their loss moduli in all frequency
ranges, with compound 9 having the largest storage moduli in the four, the gels were more stable
than the others.
2
Figure 4. The rheological measurement of the gels in DMSO:H O (v/v 1:2) formed by 9 at 10
mg/mL, 15 at 6.7 mg/mL, 25 at 4.0 mg/mL, and 26 at 3.3 mg/mL concentrations, the applied
strain was 5% for all samples.
1
3

ACCEPTED MANUSCRIPT
Among the different classes of glycosyl β-triazole derivatives, some of the benzylidene acetal
derivatives I and II were able to form gels. In order to understand the effect of protective groups
towards the self-assembling properties, several effective gelators in each series were selected to
prepare the fully protected peracetate of the benzylidene acetal. Full acetylation of the
corresponding glucose derivatives 8 and 10 afforded compounds 33 and 34 (Scheme 3).
Compounds 36-38 are the acetates of the glucosamine based gelators 23, 25, and 26. They can be
synthesized by direct acetylation of the precursors or from the azidosugar 35. As shown in
Scheme 3, they were prepared by acetylation of the azide headgroup 22 first and followed by the
click reaction. Five fully protected compounds (33-34, 36-38) were synthesized and tested and
the results are shown in Table 3, interestingly most of the acetylated compounds lost gelation
tendencies.
Scheme 3. Synthesis of the acetylated β-triazolyl glycosides 36-38.
1
4

ACCEPTED MANUSCRIPT
Table 3. Gelation properties of the acetylated β-triazolyl glycosides
No.
Precursor Pump oil Engine
oil
Toluene i-PrOH EtOH EtOH:
DMSO:
2
DCM
H
2
O (1:2) H O (1:2)
3
3
3
3
3
3
4
6
7
8
8
I
S
S
S
P
P
P
I
I
I
I
I
I
I
I
I
I
I
I
S
S
I
10
23
25
26
G 20
UG 20
P
I
P
P
P
I
I
I
I
P
P
P
P
I
S
P
All compounds were tested starting from 20 mg/mL. G, stable gel at room temperature, the
numbers are MGC in mg/mL; UG, partial gel or unstable gel; P, precipitation; S, soluble; I,
insoluble. All were insoluble in hexane, and water.
As we have seen before that the 3-hydroxy group on the sugar ring plays an important role in
5
0
forming gels. Here we noticed the same effect, after blocking the 3-OH position, the gelation
tendency diminished or disappeared. Compound 25 was an effective gelator for toluene,
ethanol/water (1:2) mixture and DMSO/water (1:2) mixture; but after introducing the acetyl
group at the 3-OH position, the resulting compound 37 no longer formed gels in these tested
solvents. This indicated that the 3-OH group was important towards gelation, the hydrogen
bonding donor is critical for effective gelation for this class of compounds. The same trend was
observed for other analogs as well. The diacetate of the glucosyl triazole derivatives 8 and 10
give compounds 33 and 34, which exhibited diminished gelation tendencies. And for the
glucosamine derivatives, all three derivatives lost gelation properties. This response towards
simple acetylation could be useful in application of stimuli-responsive gelation either with a base
or an esterase to cleave the acetate, which will be studied in future research.
1
5

ACCEPTED MANUSCRIPT
1
We rationalized the observation using H NMR spectroscopy studies at different temperatures for
1
compound 25 and its acetate 37. The H NMR spectra at different temperatures for compound 25
are shown in Figure 5. Upon increasing temperature from 30 °C to 60 °C, the triazole aromatic
C-H chemical shift moved upfield noticeably from 7.91 to 7.86 ppm (∆ 0.05 ppm); the amide
signal changed upfield from 7.93 to 7.81 ppm (∆ 0.12 ppm). And the proton of the 3-OH signal
shifted from 5.58 to 5.43 ppm (∆ 0.15 ppm); this was the largest change of the chemical shift
which indicates that hydrogen bonding using the free 3-hydroxyl group is very important for this
system and contributes significantly towards gelation. The amide hydrogen bond is also an
important contributing factor towards gelator self-assembling process. The anomeric hydrogen
showed a small downfield change upon increasing temperature. This slight downfield change can
be caused by enhanced intramolecular hydrogen bonding at higher temperature, though the
intermolecular hydrogen bonding is weakened.
After forming the acetate which blocks the hydrogen bonding donor at 3-position, the triazole C-
H signal in compound 37 changed from 8.00 to 7.93 (∆ 0.07 ppm). While the amide signal
changed from 8.06 to 7.92 ppm (∆ 0.14 ppm) upon increasing temperature; these are similar or a
little stronger than those for compound 25, indicating that after acetylation, the amide and
triazole functional groups didn’t lose their intermolecular interactions. Interestingly, the
anomeric proton in this case didn’t show meaningful chemical shift change, and the 3-CH
changed slightly downfield from 5.39 to 5.41 ppm. The NMR spectra indicated that the amide
hydrogen bonding, triazole and phenyl π-π interactions were important intermolecular forces in
molecular self-assembly. But these stronger interactions didn’t induce gelation necessarily. The
variable temperature study of this model system allows us to understand the effect of different
1
6

ACCEPTED MANUSCRIPT
derivatization to the sugar template. Moreover, a delicate balance of hydrophilicity and
hydrophobicity, intermolecular forces are necessary for effective gelation. Too strong
interactions will result in precipitation or crystallization but poor gelation.
1
Figure 5. The H NMR spectra of compound 25 from 30 °C to 60 °C in d -DMSO (10 mg/mL).
6
1
7

ACCEPTED MANUSCRIPT
1
Figure 6. The H NMR spectra of compound 37 from 30 °C to 60 °C in d -DMSO (10 mg/mL).
6
1
8

ACCEPTED MANUSCRIPT
2
We carried out additional temperature dependent experiments by adding D O to DMSO-d6 for
compounds 9 and 25. This solvent combination mimics the gelation solvents and also allows us
6
2
to observe the change of exchangeable protons after adding D O. We acquired the NMR
2
spectrum of the compounds in pure DMSO-d6, then D O was added to the NMR tube and the
2
solvent was mixed through vortex, the 3-OH and the amide signals disappeared, at higher
temperatures, only the water signals changed significantly and other signals didn’t change much.
These are shown in the Supporting Information-I Figures S9 and S10. In addition, we also
2
carried out the time dependent NMR study for compound 25 after adding D O and acquire the
1
spectrum right away. The time dependence H NMR spectra are shown in Figure 7. As what to
be expected, the proton signal of 3-OH disappeared upon addition of D O right away, but the
2
amide signal gradually exchanged with deuterium within half an hour. The rate of the exchange
with deuterium for amide functional group is fairly fast (SI-I Figure S10c and Table S5).
Addition of D O also led to chemical shift changes for several signals including the NH and
2
anomeric proton. The most significant change observed was for the water absorptions, which
shifted from 3.30 to 3. 90 ppm after equilibrium. These results indicated that both the hydroxyl
group and the amide functional group are involved in intermolecular hydrogen bonding which
leads to the formation of molecular assembly and gelation.
1
9

ACCEPTED MANUSCRIPT
-
OH
H-1
H -6, H-2
a
-
NH
0
5
min
min
1
0 min
5 min
1
2
0 min
5 min
2
3
0 min
6
0 min
1
Figure 7. Time dependent H NMR spectra of compound 25 (1.6 mg in 0.40 mL d -DMSO) after
6
2
addition of 0.10 mL D O to the sample. The NMR spectra were all acquired at 30 °C.
FTIR spectroscopy can shed some light on the structural information of functional groups and
6
3,64
help us to understand the structures in the gel phase.
Therefore we acquired the FTIR spectra
for triazole derivative 25 in its solid form and the gel in ethanol water mixture. The expanded
-1
regions for the FTIR spectra are shown in Figure 8. In the gel phase, a new peak at 3375 cm
was detected, which can be attributed to enhanced intermolecular hydrogen bonding between the
free 3-OH with other gelators and solvents, this could be the amide NH stretching vibration too.
2
0

ACCEPTED MANUSCRIPT
-1
An enhanced signal at 3062 cm in the gel phase than the solid indicates that the gel state has
stronger aryl and triazole CH stretch vibration, possibly due to stronger π-π interactions at the
-1
gel phase. The C-H stretch signals in 2924, 2869, and 2854 cm have small red shifts in the gel
phase, this implies that the C-H stretching are weaker (lower in energy) at the gel phase, possibly
due to the stronger hydrophobic interactions in the gel. The amide I band (the C=O stretching
band) in the solid form absorbed at 1661 cm-1, and the amide II band (the N-H bend) frequency
-1
-1
was 1536 cm . In the gel state, the amide I band shifted to 1668 cm and the amide II shifted to
-1
-
1
530 cm with stronger relative intensities. The triazole C=C stretch signal appeared at 1643 cm
1
for the gel state, which was also stronger than the signal in the pure solid, this indicated stronger
intermolecular triazole packing in the gel state. Likewise, the aryl carbon-carbon stretch shifted
-1
-1
from 1454 cm in solid to 1458 cm in the gel state with stronger intensity. The FTIR analysis
clearly indicated that hydrogen bonding and other intermolecular forces are important in the
gelation process. We also carried out the FTIR experiments for the gel of compound 10 in
DMSO:water (v/v 1:2) at 6.7 mg/mL, the DMSO water gel showed less changes in their solid
and gel phase comparing to the system discussed above. The FTIR spectra of several gels formed
by compounds 9, 10, 11, 25 and 26 are shown in Supporting Information I Figures S11-S20.
2
1

ACCEPTED MANUSCRIPT
(
A)
(B)
-1
-1
Figure 8. FTIR spectra for compound 25 in 2700-3500 cm and 1400-1700 cm (expanded
regions). The green curve corresponds to the solid form, the red curve corresponds to the gel
form of 25 in EtOH:H O (v/v 1:2) at 5.0 mg/mL, and the blue curve is the solvent standard of
2
EtOH:H O (v/v 1:2).
2
For potential drug delivery or other applications using the gel matrix, it is more desirable if the
gelator can form gels in water or with minimum amount of DMSO. We tested the gelation
properties of compounds 10 and 25 in different ratios of DMSO and water, the results are shown
in supporting information Table S6. The glucosamine derivative 25 was able to form a stable gel
2
in DMSO:H O at 1:5 ratio. We selected this compound and formed a co-gel with naproxen in
DMSO:H O (v/v 1:5) at 6.7 mg/mL and analyzed the sustained release of naproxen from the gel
2
phase to water phase, the results are shown in Figure 9. Naproxen was released rapidly to the
aqueous phase, in 2 hours about half of the naproxen was detected in the aqueous phase and it
was fully released to the water phase within a day. With different gelator concentrations, we
2
2

ACCEPTED MANUSCRIPT
expected that the timed release of naproxen could be controlled for longer or shorter duration.
This compound could be used to encapsulate other drug compounds and may be able to form
more compatible and stable co-gels at different conditions, from which the drug will be released
more slowly.
3
2
1
0
.5
3
1
00
Naproxen
9
8
7
6
5
4
3
2
0
0
0
0
0
0
0
0
0
1
1
2
.5 h
h
.5
2
.5 h
h
.5
1
3 h
4
6
7
2
.5 h
h
.5 h
4 h
.5
0
10
0
0
10
20
3
00
320
340
360
380
400
Time (h)
Wavelength (nm)
a
b
Figure 9. Release of naproxen from gel to aqueous phase. a) UV spectra of the naproxen timed
release, the gel was formed by 2.0 mg of compound 25 in 0.3 mL DMSO:H O (v/v 1:5) with 1.0
2
mg of naproxen, then 3 mL of water was added on top of the gel; naproxen control was prepared
by dissolving 1.0 mg naproxen in 3 mL of water (pH 7); b) Absorbance at 330 nm at different
times versus the standard was used to calculate the % release.
Conclusions:
We have synthesized and studied a series of β-triazolyl glycoside of 4,6-O-benzylidene acetal
protected D-glucose and D-glucosamine systematically. A few low molecular weight
organogelators were obtained and the gels were characterized using optical microscopy, rheology
2
3

ACCEPTED MANUSCRIPT
and FTIR spectroscopy. The D-glucose derivatives are more effective gelators for organic
solvents especially for pump oil and engine oil, whereas the D-glucosamine derivatives are less
effective for oils but more effective towards polar solvents and their aqueous mixtures. However,
the presence of the 4,6-O-benzylidene acetal functional group seems to weaken the gelation
tendencies for some triazolyl glycosides in comparisons to the corresponding tetra-acetyl
glycosyl triazoles perhaps due to the structural rigidity of the bicyclic ring system. A few
effective gelators were further acetylated and the resulting fully protected diacetates lost gelation
abilities. We noticed that the hydrogen bonding of the 3-OH group plays a critical role in the
1
observed gelation properties. H NMR spectra at different temperatures for compound 25 and its
acetate 37 revealed the importance of the free 3-hydroxyl group in the molecular self-assembly
and gelation process. Furthermore, we analyzed the FTIR spectra for compound 25 in its solid
and gel states, new absorption signals were observed for the gel phase, which supported the
NMR study results. Compound 25 was also used in a test for naproxen co-gel formation and the
release of naproxen from the gel matrix was monitored by UV-vis spectroscopy. We expect that
further structural optimization of these derivatives could lead to more effective gelators or
gelators that are compatible to different drug molecules for encapsulation and sustained release.
The general trend of these compounds can be used to design systems with or without the desire
to form stable gels, which have implications in biological or materials applications.
Experimental sections
General method and materials: Reagents and solvents were used as they were received from
the suppliers. All purification was conducted by flash chromatography using 230-400 mesh
2
4

ACCEPTED MANUSCRIPT
1
13
silica gel with a gradient of solvent systems. H NMR and proton-decoupled C NMR spectra
were obtained with Bruker 400 MHz spectrometers in DMSO-d or CDCl . The chemical shifts
6
3
were reported using CDCl /DMSO-d as internal standard at 7.26/2.50 ppm and at 77.00/39.50
3
6
ppm, respectively. 2D NMR experiments (HSQC, COSY) were also conducted using a 400 MHz
Bruker NMR spectrometer to assist the proton and carbon signal assignment. Melting point
measurements were carried out using a Fisher Jones melting point apparatus. Rheology
experiment was done using a HR-2 Discovery Hybrid Rheometer from TA instrument and a 25
mm Peltier Plate. FTIR experiments were conducted using liquid or solid samples directly. The
molecular mass was measured using LCMS on an Agilent 6120B Single Quad Mass
Spectrometer and LC1260 system.
Optical Microscopy: A small amount of the gels was placed on a clean glass slide and this was
observed under an Olympus BX60M optical microscope and the Olympus DP73-1-51 high
performance 17MP digital camera with pixel shifting and Peltier cooled. The imaging software
for image capturing was CellSens 1.11. For aqueous DMSO mixtures, the gel was left air dried
for a day or so, for other solvents the slides were observed directly.
Gel Testing: About 2.0 mg of dried compound was placed in a one dram glass vial and the
corresponding solvent was added to obtain a concentration of 20 mg/mL. The mixture was then
heated until the solid was fully dissolved, sometimes sonication was needed to dissolve the
sample, and then the solution was allowed to cool to room temperature and left standing for 30
minutes. If a stable gel formed (observed by inverting the vial), it was then serially diluted till
2
5

ACCEPTED MANUSCRIPT
the minimum gelation concentration (MGC), which is the concentration prior to unstable
gelation, was obtained.
Rheological Analyses: The rheological behavior of the gel was investigated with an HR-2
Discovery hybrid rheometer from TA Instruments with TRIOS software. A sample
(
approximately 1mL of the gel) was placed on the steel plate of the rheometer. The cone
geometry is a 25 mm Peltier plate with a gap of 100 ꢀm. The experimental temperature was
5 °C, and the sample was subjected to amplitude sweep for oscillation strain from 0.125 to
0%. A frequency sweep was then performed for the sample in the range of 0.1 to 150
2
1
rad/s for the angular frequency. The results were expressed as the storage modulus (G′)
and loss modulus (G″) as a function of angular frequency.
FTIR Spectroscopy: FTIR measurements were performed on a Bruker ALPHA Platinum ATR
FTIR Spectrometer, the operating software is OPUS. The absorption spectra of all samples were
-1
obtained at room temperature in the range of 400-4000 cm .
Liquid chromatography-mass spectrometer (LC-MS) conditions: The molecular mass
acquisition was performed on a liquid chromatography-mass spectrometer system which consists
of an Agilent 1260 Infinity LC system coupled to a G6120B single quadrupole mass
spectrometer with an atmospheric pressure ionization electrospray (API-ES). The analytical
column is an Agilent poroshell 120 EC-C18 4.6 mm × 50 mm column with 2.7 ꢀm particle size.
The mobile phase is 0.1% formic acid in acetonitrile/water (70/30, v/v). Ejection volume is 5 µL.
Total runtime is 8 min. The flow rate is 0.40 mL/min and the column temperature is set at 25 °C.
2
6

ACCEPTED MANUSCRIPT
Wavelength of diode-array detector (DAD) is 254 nm. Data was collected and processed using
Agilent OpenLAB CDS ChemStation (Version C.01.05). The ESI spray voltage was set at 4000
V. The source temperature is 350 °C. The nebulizer gas setting is 35. Range of mass
spectrometer detector is set from 0 to 2000 Dalton.
Naproxen trapping and release: A gel was prepared in a one dram via using 2.0 mg of
2
compound 25 and 1.0 mg of naproxen sodium and 0.3 mL DMSO:H O (v1:5), after a stable gel
was formed and the gel was left undisturbed for 2 hours, 3 mL of water at pH 7 was added to the
top of the gel carefully. Naproxen release from the gel was monitored by UV absorption at
certain time by transferring the supernatant with a pipet to a cuvette, after each measurement the
aqueous phase was carefully transferred back to the vial and placed on top of the gel again till the
next measurement. The UV spectra of the pure naproxen and pure gelator in the test solvent
system were also recorded.
Synthesis of sugar headgroup azide 7:
The headgroup azide 7 was synthesized following literature method with slight modifications
6
5
using D-glucose as the starting material. Compound 5 (3.80 g, 10.15 mmol, 1 equiv) was
dissolved in MeOH (15 mL) and NaOMe (273 mg, 5.08 mmol, 0.5 equiv) were added to 50 mL
round bottom flask. The reaction mixture was stirred for 12 hours at rt, neutralized with
Amberlite IR 120 hydrogen form, filtered, concentrated and vacuum drying overnight to give
2
.08 g (10.15 mmol, quantitative) brown semi-solid as the desired product. Compound 6 was co-
distilled with toluene (3 x 10 mL) and directly used for the next step without further purification.
Toluene treated compound 6 (2.08 g, 10.15 mmol, 1 equiv), dimethyl benzaldehyde acetal (1.83
2
7

ACCEPTED MANUSCRIPT
mL, 12.18 mmol, 1.2 equiv), PTSA monohydrate (196 mg, 1.02 mmol, 0.1 equiv) were added in
the given order and followed by adding 10 mL of DMF (anhydrous) as the solvent. The resulting
mixture was stirred for 8 hours at 60 °C. After neutralizing the reaction mixture by adding
NaHCO (170 mg, 2.04 mmol, 0.2 equiv) DMF was removed by the rotavap and workup was
3
,
performed using DCM (3x40 mL) /water (20 mL) to give the crude product, which was purified
by column chromatography using eluent from 1% MeOH/DCM to 5% MeOH/DCM to give
f
white solid (2.32 g, 7.91 mmol, 78% over two steps) as the desired product (R = 0.3 in 3%
1
MeOH/DCM). mp 136-138 °C; H NMR (400 MHz, CDCl
3
) δ 7.52-7.46 (m, 2H), 7.41-7.35 (m,
3
2
H), 5.54 (s, 1H), 4.67 (d, J = 8.6 Hz, 1H), 4.38 (dd, J = 10.4 Hz, 4.3 Hz, 1H), 3.85-3.75 (m,
1
3
H), 3.59-3.51 (m, 2H), 3.44 (t, J = 8.6 Hz, 1H), 3.00-2.45 (br s, 2H); C NMR (400 MHz,
CDCl ) δ 136.7, 129.4, 128.4, 126.3, 102.0, 90.6, 80.1, 74.1, 73.5, 68.41, 68.36.
3
General procedure for the synthesis of triazoles 8-19:
To a scintillation vial, azide 7 (100 mg, 0.340 mmol, 1 equiv) and alkyne (0.400 mmol, 1.2
equiv) were dissolved in the 6 mL of solvent mixture of THF: H O: t-BuOH (v/v 1:1:1). CuSO
2
4
(11.1 mg, 0.068 mmol, 0.2 equiv) and L-ascorbic acid sodium salt (27.2 mg, 0.136 mmol, 0.4
equiv) were added to the reaction mixture. Reaction was stirred for 24 hours at rt. Solvent was
removed under reduced pressure to give the crude product, which was purified by column
chromatography to afford the desired triazole product 8-19.
Synthesis of compound 8:
2
8