A synthetic tripeptide as organogelator: Elucidation of gelation mechanism

Article abstract of DOI:10.1039/b111598g

A new terminally protected synthetic tripeptide with noncoded amino acids, tert-butyloxycarbonyl-β-alanylα-aminoisobutyryl-β-alanyl methyl ester, is synthesized. The product is characterized by ¹H NMR and DQF COSY NMR data. This tripeptide produces thermoreversible gels in 1,2-dichlorobenzene (DCB) at room temperature (30 °C). The morphology of the dried gels is studied using optical microscopy, scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The micrographs show the presence of both rod like morphology and liquid-liquid phase separated morphology in the gels. Thermal study by differential scanning calorimeter (DSC-7) indicates the presence of a reversible first order phase transition during heating and cooling processes. Wide-angle X-ray scattering (WAXS) and electron diffraction experiments indicate the presence of polycrystalline materials in the gel and the crystal structure of the gel is almost the same as that of the pure tripeptide. Solvent subtracted Fourier transformed infra-red (FT-IR) study indicates that in the sol state there are free N-H and intramolecular hydrogen bonded N-H stretching vibrations at 3440 and 3375 cm^-1, respectively. In the gel state both these vibrations shifted to 3325 and 3309 cm^-1, respectively. The phase diagram of these gels, determined from both the DSC method and visually, indicates the tripeptide-DCB complexation with singular point characteristics. An attempt is made to understand the gelation mechanism of this system by measuring the gelation rate (t_gel^-1) using the test tube tilting method. The gelation rate (t_gel^-1) is expressed as t_gel^-1 α f(C)f(T) and the analysis of the concentration function f(C) at a particular temperature indicates that the gelation process obeys the three dimensional percolation mechanism. The microscopic mechanism of the gelation process is explored from the temperature function f(T) of the gelation rate at a particular concentration expressing f(T) in two different forms: (I) for fibrillar (rod like) crystallization and (II) for spinodal decomposition. A tentative structure of the tripeptide in the gel state has been proposed by molecular modeling using the MMX program.

Full text of DOI:10.1039/b111598g

View Article Online / Journal Homepage / Table of Contents for this issue
A synthetic tripeptide as organogelator: elucidation of gelation
mechanism†
Sudip Malik,a Samir K. Maji,b Arindam Banerjee*b and Arun K. Nandi*a
2
a Polymer Science Unit, Indian Association for the Cultivation of Science, Jadavpur,
Calcutta-700032, India. E-mail: psuakn@mahendra.iacs.res.in
b Department of Biological Chemistry, Indian Association for the Cultivation of Science,
Jadavpur, Calcutta-700032, India. E-mail: bcab@mahendra.iacs.res.in;
Fax: ϩ91-33-473 2805; Tel: ϩ91-33-473 4971
Received (in Cambridge, UK) 19th December 2001, Accepted 2nd April 2002
First published as an Advance Article on the web 22nd April 2002
A new terminally protected synthetic tripeptide with noncoded amino acids, tert-butyloxycarbonyl-β-alanyl-
α-aminoisobutyryl-β-alanyl methyl ester, is synthesized. The product is characterized by 1H NMR and DQF COSY
NMR data. This tripeptide produces thermoreversible gels in 1,2-dichlorobenzene (DCB) at room temperature
(30 ЊC). The morphology of the dried gels is studied using optical microscopy, scanning electron microscopy (SEM)
and transmission electron microscopy (TEM). The micrographs show the presence of both rod like morphology and
liquid–liquid phase separated morphology in the gels. Thermal study by differential scanning calorimeter (DSC-7)
indicates the presence of a reversible first order phase transition during heating and cooling processes. Wide-angle
X-ray scattering (WAXS) and electron diffraction experiments indicate the presence of polycrystalline materials in
the gel and the crystal structure of the gel is almost the same as that of the pure tripeptide. Solvent subtracted Fourier
transformed infra-red (FT-IR) study indicates that in the sol state there are free N–H and intramolecular hydrogen
bonded N–H stretching vibrations at 3440 and 3375 cmϪ1, respectively. In the gel state both these vibrations shifted
to 3325 and 3309 cmϪ1, respectively. The phase diagram of these gels, determined from both the DSC method and
visually, indicates the tripeptide–DCB complexation with singular point characteristics. An attempt is made to
understand the gelation mechanism of this system by measuring the gelation rate (tϪge1l) using the test tube tilting
method. The gelation rate (tϪge1l) is expressed as tϪge1l ∝ f(C)f(T ) and the analysis of the concentration function f(C) at
a particular temperature indicates that the gelation process obeys the three dimensional percolation mechanism. The
microscopic mechanism of the gelation process is explored from the temperature function f(T ) of the gelation rate at
a particular concentration expressing f(T ) in two different forms: (I) for fibrillar (rod like) crystallization and (II) for
spinodal decomposition. A tentative structure of the tripeptide in the gel state has been proposed by molecular
modeling using the MMX program.
occurring proteins. Cyclic β-peptides with β-substituted, chiral
Introduction
β-amino acids and cyclic peptides composed of alternating
The formation of organogels is one of the finest examples of
- and -α-amino acids can adopt flat ring conformations
a supramolecular self-assembly process.1,2 Ghadiri and his
co-workers established that self-assembly of cyclic β-peptides
(oligomers of β-amino acids) leads to the formation of nano-
and they can stack in the solid state by backbone–backbone
intermolecular hydrogen bonding to form hollow tubular
ensembles.16 The main forces responsible for supramolecular
tubular structures which function as artificial transmembrane
self-assembly are (1) hydrogen-bonding,1,2 (2) polar and van der
ion channels.3 Supramolecular self-assembly formation com-
Waals interactions,1,2 (3) π–π stacking17 and (4) crystalliz-
pletely depends upon reversible non-covalent interactions.2,4,5
ation.7,18 The control of gelation phenomena and the design of
A main problem for supramolecular assembly is the mainten-
ance of a rapid equilibrium between the assembled and dis-
novel types of organogelators are challenging, particularly in
the field of short peptides.
From the above-mentioned approach we may design a new
type of organogelator (peptide containing noncoded amino
assembled structure. In order to overcome the problem, many
groups offer various solutions like crystal engineering,6,7 prep-
aration of soluble or solvent-stable aggregates1,2,8–12 and
acids) which can satisfy the criteria of self-aggregation and gel-
hydrogen-bonding interactions.1,2 Peptides containing β-amino
ation. The geometry of the building blocks and their arrange-
acids are excellent candidates for foldamers12 due to their sur-
ments in space, and the nature of the hydrogen-bonding are
prising secondary structure forming propensities in organic
mainly responsible for the structure and properties of macro-
solvents, in water and in the solid state. Seebach et al. and
Gellman et al. showed that acyclic β-peptides (oligomers of
molecular self assembled ensembles. Rigid peptides using con-
formationally constrained amino acid residues give rise to a
particular arrangement which remains unaltered in solution
with change in temperature, solvent and composition, but a
homo/hetero β-amino acids) can adopt spectacular secondary
and super-secondary structures like 12-helices,13 14-helices,14
and new β-hairpins15 which are unprecedented in the naturally
molecule with only flexible residues would not yield a definite
structure in solution. However, if one designs a peptide having
† Electronic supplementary information (ESI) available: the 500 MHz
both conformationally rigid and conformationally flexible
1
1
1-D H NMR spectrum, the 500 MHz H–1H DQF COSY spectrum
of the tripeptide in CDCl3 and the MALDI-MS spectrum of the
tripeptide. See http://www.rsc.org/suppdata/p2/b1/b111598g/
residues it may produce varieties of structures depending on
solvent, composition, temperature, etc.19 So in this last category
DOI: 10.1039/b111598g
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186
This journal is © The Royal Society of Chemistry 2002
1177

View Article Online
Table 1 1H NMR parameters for the tripeptide in CDCl3 (the 1H–1H
DQF COSY spectrum for the tripeptide is submitted as supplementary
material)
micrographs it appears that another type of network structure
(marked by arrow) is superimposed on the rod like network
structure. In Fig. 1(c) it is also apparent that some portions of
the network structure are not exactly connected through rods.
In Fig. 1(d) the polarized optical micrograph of the undried gel
is shown. Here also two types of network morphology are
clearly observed. So it may be argued that gel morphology stud-
ied in the dried state is the same as in the gel state. To elucidate
the cause of the coexistence of two different types of network
structures we have performed an X-ray diffraction study and a
phase equilibrium study.
Residue
NH δ (ppm)
CαH δ (ppm)
CβH δ (ppm)
β-Ala(1)
Aib(2)
β-Ala(2)
5.59
6.27
6.94
3.38–3.40
—
3.49–3.52
2.40–2.42
1.51
2.57–2.59
of samples it may be possible to design structures of different
types and our interest is in constructing such a structure which
can self assemble to produce gel. Here we design a peptide with
two conformationally contrasting nonprotein (i.e., noncoded)
amino acids namely α-aminoisobutyric acid (Aib) and a β-
amino acid, β-alanine (β-Ala). The former has a conformation-
ally restricted nature and the latter has sufficient flexibility. This
combination may form well-defined structures like foldamers.
This type of acyclic peptide is a potent candidate to create a
network structure through intermolecular aggregation in
organic solvents due to the presence of hydrogen bond forming
amide linkages. The intermolecular aggregation may also
depend on the nature of the solvent. In this report we present
such a study of aggregation/gelation for a terminally protected
tripeptide, Boc-β-Ala-Aib-β-Ala-OMe in 1,2-dichlorobenzene
(DCB). The N-terminal protection has been made with tert-
butoxycarbonyl (Boc) and the C-terminal is made with methyl
ester to solubilize the compound in organic solvents.
C. Structure
In Fig. 2 the WAXS pattern of dried gel (diffractogram a) is
shown and here sharp crystalline peaks are clearly seen. To
compare it with that of the pure tripeptide the WAXS pattern
of pure tripeptide (diffractogram b) is also included in Fig. 2.
From the two diffractograms it is clear that almost all the peaks
(except a small peak at 12.9Њ) are the same for both the cases,
indicating the crystal structure is the same both in the gel state
and in the solid state. Thus it may be proposed that the
rods seen in the micrographs consist of the tripeptide crystals
and the crystallites are responsible for the formation of cross-
linking junctions. To elucidate the nature of the cross-linking
junctions we have also performed electron diffraction experi-
ments, which gave the pattern shown in Fig. 3 where diffraction
rings are present.22 This indicates that crystallites producing the
cross-linking junctions are polycrystalline in nature.23
The mechanism of organogel formation is still poorly under-
stood. The main objective of this study is, therefore, to establish
a satisfactory gelation mechanism in this tripeptide gel. For
this purpose we have undertaken morphological, structural,
thermodynamic and kinetic approaches to elucidate the gel-
ation mechanism. In the thermoreversible polymer gels it is now
established that a three-dimensional percolation mechanism is
satisfactorily obeyed and we want to investigate whether the
percolation model is obeyed in this gel also. Besides, what is the
structure of the peptide molecule in the gel—is it different from
that in its solution state or in its solid state? How is the network
structure formed? In this paper we want to present a com-
prehensive study of gelation in this system to address the above
questions. Finally molecular modeling is also used to present
a tentative structure of the tripeptide molecule in the gel. The
tripeptide sample gels in 1,2-dichlorobenzene, monochloro-
benzene and benzene, however it does not gel in chloroform,
acetonitrile, isobutanol, N,N-dimethylformamide, diethyl suc-
cinate etc. Here we want to present a detailed mechanistic aspect
of gelation taking 1,2-dichlorobenzene as a representative
solvent.
D. Thermal study
To understand the nature of the other cross-linking pieces we
performed a phase equilibrium study for a composition (1%
w/v) where no gelation takes place, i.e., below the critical gel-
ation concentration. The solution was heated to 110 ЊC to make
it homogeneous and kept at 25 ЊC for 12 h. A distinct phase
separation into two layers was observed. Since there are some
reports that liquid–liquid phase separation may also act as a
cross-linking junction for polymers,25,26 so it may be argued here
that the liquid–liquid phase separation may also be responsible
for the gelation process of this system. The different (non-rod)
type of network structure observed in Fig. 1 (marked by arrow)
may, therefore, occur due to liquid–liquid phase separation.
Liquid–liquid phase separation may also be responsible for the
cloudiness of the gel. So it appears from this result that it is a
unique system where both liquid–liquid phase separation and
crystallization of the tripeptide may be responsible for pro-
ducing the gel. A detailed study of the phase diagrams will be
presented in the following section.
In Fig. 4(a) thermograms of gels of differing tripeptide
concentration for the heating process are shown. It is apparent
from the figure that first order phase transition (i.e., melting)
peaks are observed in all the compositions. One of the import-
ant criteria of thermoreversible gelation is the existence of
reversible first order phase transition.8,27 In Fig. 4(b) the
dynamic cooling thermograms of gels indicate the presence of
exotherms in all the cases. So the presence of fibrillar net-
work structure and reversible first order transition confirms
that the tripeptide produces thermoreversible gel in 1,2-
dichlorobenzene. However, the exotherms in Fig. 4(b) become
broader with the increase in tripeptide concentration and this is
probably due to the difficulty in the diffusion process of the
tripeptide molecule with increasing concentration. For the solid
tripeptide molecule the exothermic peak is absent because of a
very slow crystallization rate due to the same reason. However,
the pure sample crystallizes after keeping for a longer time at
50 ЊC. In Fig. 5 the gel melting temperatures (T gm) and the
gelation temperature (T gel) are plotted with the weight fraction
of the tripeptide (Wpeptide) in the gels. The nature of the phase
diagram is almost the same for both the processes; however,
Results and discussion
A. Characterization of the tripeptide
From 1H NMR and 2D DQF-COSY experiments sequence
specific assignments were done successfully. Important NMR
parameters for the peptide are listed in Table 1. From the mass
spectroscopic results it is shown that the observed mass at
m/z 360 corresponds to a molecular ion peak (MHϩ) which
completely agrees with the calculated mass value (359). (Mass
spectrum and NMR spectrum are supplied as supplementary
material.)
B. Morphology of the gel
The tripeptide gels in 1,2-dichlorobenzene are turbid in all the
compositions studied here. In Fig. 1(a–c) the morphology of the
dried gel is shown from the optical microscopy, scanning elec-
tron microscopy and transmission electron microscopy, respect-
ively. It is apparent from Fig. 1(a)–(c) that the network consists
of rod like structures. However, from a careful look at these
1178
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186

View Article Online
Fig. 1 Photomicrographs of dried gels of the tripeptide in 1,2-dichlorobenzene: (a) optical micrograph (5% w/v), (b) SEM (10% w/v), (c) TEM
(1% w/v) and (d) polarized optical micrograph of undried gel (9% w/v). [Arrow in each micrograph indicates morphologies different from the rod
like network structure].
Fig. 2 WAXS pattern of (a) dried tripeptide gel (10% w/v) and (b)
tripeptide crystal.
Fig. 4 (a) DSC thermograms of tripeptide–DCB gels prepared at
15 ЊC for the indicated weight fractions of the tripeptide (Wpeptide) in
the gel (heating rate 10 ЊC minϪ1). (b) DSC thermograms for cooling of
the tripeptide–DCB sols from 110 ЊC at the scan rate 5 ЊC minϪ1 for the
indicated weight fractions of the tripeptide (Wpeptide).
Fig. 3 Electron diffraction pattern of the dried tripeptide gel (1% w/v,
camera length 0.8 m).
there is a hysteresis effect (60–20 ЊC) which varies for varying
tripeptide concentration, the lower tripeptide content gel hav-
ing a higher hysteresis effect. In order to observe the gel melting
phenomena visually we have also performed the experiment in
sealed glass tubes. In the experiment it has been observed that at
peptide concentrations, Wpeptide ≤ 0.27 the gel breaks (i.e., melts)
before it becomes transparent. The sol becomes transparent
at temperatures (T s) which are a few degrees higher than the
respective gel melting temperatures. The difference between
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186
1179

View Article Online
Fig. 5 Phase diagram of the tripeptide–DCB gel, obtained from (a)
DSC, heating experiment (T gm), (b) DSC, cooling experiment (T gel) and
(c) visual method, (᭹) gel melting temperature (T Јgm) and (᭝) turbidity
disappearance temperature (T s) (see text).
Fig. 6 (a) ∆H vs. Wpeptide plot for the gel melting process in DSC.
(b) Deviation (∆) vs. Wpeptide plot for the tripeptide–DCB gel.
T Јgm and T s decreases with increase in Wpeptide and at Wpeptide
≥
0.40 the T Јgm and T s are almost the same. Thus it appears that
if a sol is cooled, it first passes the liquid–liquid phase separ-
ation boundary and then it becomes gel, particularly for lower
concentrations (Wpeptide ≤ 0.27).
Now we analyze the ∆H values obtained during heating. The
∆H vs. Wpeptide plot is shown in Fig. 6(a) and it shows a positive
deviation from linearity joining the line between the two
components. The enthalpy of gel melting may be considered to
consist of three contributions:7,18b
phase diagrams are drawn from the endotherms or exotherms
which show a single peak. No separate peak for the compound
formation occurs in thermograms for any of the compositions.
So it may be considered that the exotherms or endotherms are
composed of two indistinguishable peaks, the first of which
corresponds to the transformation of
compound
solid ϩ solvent
(3)
and is nonvariant with composition. This is shown by a dotted
line (Fig. 5). The second peak is due to the melting of the
∆H(per gm) = w1∆H1 ϩ w2∆H2 ϩ ∆Hc
(1)
Where w represents the weight fraction and subscripts 1 and
2 represent the solvent, DCB, and the tripeptide, respectively.
∆Hi represents the enthalpy of each component and ∆Hc is the
enthalpy of melting of any tripeptide–solvent assembled (i.e.,
complex) structure, if formed during gelation. Since DCB does
not crystallize at the gelation temperature (15 ЊC) so ∆H1 may
be considered as zero. Consequently,
solid
liquid
(4)
and it varies with composition as shown in the diagrams.18
Thus the thermal study indicates that there are both liquid–
liquid phase separation and tripeptide–solvent complex form-
ation in this system.
E. FT-IR study
∆H Ϫ w2∆H2 = ∆Hc
(2)
In Fig. 7 the FT-IR spectra of solvent subtracted gel (a) and
also the pure solid tripeptide (b) are shown. It is apparent from
the figure that both the spectra are almost identical and this
indicates that the gel structure and the solid crystal structure
are the same. This is in accordance with the WAXS results
discussed earlier. So it proves that during drying of the gel
tripeptide crystal structure (i.e. polymorphic nature) doesn’t
change. The solvent subtracted spectrum of a 0.3% (w/v) solu-
tion of tripeptide in 1,2-dichlorobenzene (c) is shown. This
spectrum is completely different from that of the pure tripep-
tide or its gel. In this spectrum (c) there are N–H stretching
peaks at 3440 and 3375 cmϪ1. The former peak is due to the free
N–H vibration whereas the 3375 cmϪ1 peak is for the hydrogen
bonded (intramolecular) N–H stretching vibration.30,31 How-
ever, in FT-IR spectra (a) and (b) these peaks are completely
The left hand side of eqn. (2) is the positive deviation (∆)
from linearity (Fig. 6a). The deviation (∆) is plotted with Wpeptide
(Fig. 6b) and it shows a maximum which indicates that there is
tripeptide–DCB complex formation in this gel.7,8,18,28 The max-
imum (Wpeptide = 0.37) corresponds to the composition of the
tripeptide–solvent complex formed during gelation. The molar
ratio of DCB and tripeptide comes out as 4 : 1 at this com-
position. The nature of the complex is not known, but it may be
postulated that it occurs through the physical interaction
between four >C᎐O groups of the tripeptide and the benzene
᎐
ring of DCB. The phase diagrams shown in Fig. 5 may be
explained in analogy with the phase diagrams of poly(vinyl-
idene fluoride)–solvent gels.7,18 All the phase diagrams are
characteristic in that they have a singular point.28,29 The above
1180
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186

View Article Online
Fig. 8 FT-IR spectra of 1% (w/v) solution in DCB at different time
intervals: (a) just after preparation, (b) after 15 minutes, (c) after 1 hour
and (d) after 2 hours.
F. Gelation mechanism
Kinetic study of the gelation process is an important tool for
understanding the gelation mechanism and both the macro-
scopic and microscopic mechanism of the gelation process can
be elucidated7,21 from this study. The gelation rate (tϪge1l) is
usually expressed as a combination of two functions: (I) the
concentration function [f(C)] and (II) the temperature function
[f(T )].7,20c,21
Fig. 7 FT-IR spectra of (a) solvent subtracted tripeptide gel in DCB
(10% w/v), (b) pure tripeptide sample and (c) tripeptide solution in
DCB (0.3% w/v).
absent and new peaks at 3325 and 3309 cmϪ1 appear. So in gel
or in the solid crystalline state the molecular arrangement
makes all the N–H bonds bound and it probably occurs
through the intermolecular hydrogen bonding. This inter-
molecular association through H-bonding makes the system
crystallize in the form of rods.
tϪge1l ∝ f(C)f(T )
(5)
The macroscopic mechanism of the gelation process can be
obtained from the analysis of f(C) whereas the analysis of f(T )
yields the microscopic (molecular) mechanism of the gelation
process.7,21 A plot of the gelation rate (tϪge1l) with the tripeptide
concentration is shown in Fig. 9 for different gelation temper-
From the FT-IR spectrum (c) it is apparent that there are two
peaks (1712 and 1683 cmϪ1) for the >C᎐O stretching character-
᎐
izing the ester and amide carbonyl groups, respectively.30,31
However, in the gel or in the pure solid form these peaks are not
observed, rather the peaks at 1737, 1680 and 1650 cmϪ1 are
observed. The peak at 1712 cmϪ1 belongs to the ester carbonyl
group and is shifted to higher energy due to intermolecular
hydrogen bonding whereas the peak at 1683 cmϪ1 splits into
doublets, retaining the original 1680 cmϪ1 and also creating a
new peak at 1650 cmϪ1. No definite reason for the splitting of
the peptide (amide) >C᎐O group is known; however, a probable
᎐
explanation is that some of the >C᎐O groups experience more
᎐
potential ionic character due to intermolecular aggregation.
In Fig. 8 the FT-IR spectrum of 1% (w/v) solution is shown
for different intervals of time after making the solution homo-
geneous at 100 ЊC. It is apparent from the figure that the inten-
sity of the 3440 cmϪ1 peak gradually decreases with increase
in time at 27 ЊC and the intensity of the peaks at 3325 and
3309 cmϪ1 gradually increases. One important observation is
that initially there was a peak at 3371 cmϪ1 in the solution,
however it disappears when the peaks at 3325 and 3309 cmϪ1
begin to develop. This indicates that with time intermolecular
aggregation is occurring gradually at this concentration and
hence, intensity of the free N–H peaks slowly reduces.
Fig. 9 Plot of gelation rate (tϪge1l) with tripeptide concentration in the
gel at indicated temperatures.
Ϫ1
gel
atures (T Ј ). At a particular T Ј , the t increases in a non-
gel
gel
linear way for the lower peptide concentration but gradually it
levels up at higher concentration. The extrapolation of each
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186
1181

View Article Online
Ϫ1
gel
Table 2 Exponent ‘n’ values from least squares slopes of log t vs.
curve to zero gelation rate yields the critical gelation concen-
tration (Ct*= ∞ ) which is the minimum concentration of tripep-
tide required to produce the gel at that gelation temperature.
log Φ plot of tripeptide–DCB gel
Temperature/ЊC
C*/g dlϪ1
‘n’
The concentration function: macroscopic mechanism. At a
particular gelation temperature
15
17.5
20
22.5
25
27.5
30
35
Av
SD
2.5
3
0.44
0.49
0.52
0.49
0.50
0.52
0.54
0.51
0.50
0.03
3.3
3.6
3.8
4.3
4.6
5.1
—
tϪge1l ∝ f(C) ∝ Φn
(6)
where Φ is the reduced overlapping concentration and is
expressed as,
—
(7)
solution when cooled may form an ordered arrangement
(foldamer) by changing the conformation of the bonds and
these foldamers then aggregate to produce the crystallites. Also,
as evidenced from the phase study as well as from morpho-
logical study, liquid–liquid phase separation may also be
responsible for the gelation. So the scheme may be represented
as:
The ‘n’ is an exponent and its value can give us an idea about
the macroscopic mechanism of the gelation process. According
to the percolation theory the gel fraction (G, ratio of gel
macromolecules/total number of gel and sol molecules) is
expressed as32,33
G ∝ (p Ϫ pc)β
(8)
where p is the conversion factor during the gelation process
and pc is its critical value, β is a critical exponent and has a value
of 0.45 for a three-dimensional lattice. According to the theor-
ies of chemical kinetics the reaction rate is proportional to the
fraction of molecules overcoming the activation energy barrier.
Considering this theory is applicable to the gelation process the
gelation rate is directly proportional to the gel fraction.7,21 An
attempt is usually made to measure gelation time as the onset of
gelation, but at the gelation threshold the gel fraction (percol-
ation probability) rises very sharply with p.32,33 So within the
error limit of gelation time measurement the gel fraction has
some very small but finite value. Consequently:
Scheme 1
Crystallites can produce the gel because of rod like morph-
ology, which can easily form network junctions entrapping the
solvent. However, the liquid–liquid phase separation is also
capable of producing gel as evidenced from the SEM and TEM
pictures.25,26 Scheme 1 indicates the gel formation process of the
tripeptide is analogous to the concurrent reaction in chemical
kinetics,34 where the overall rate of disappearance of reactant is
a composite of different reaction paths. Assuming this concept
of chemical kinetics is applicable to the physical process we
have:
tϪge1l ∝ G ∝ (p Ϫ pc)β
(9)
Comparing (6) and (9), ‘n’ may be equal to β (since C < C* no
gelation occurs and p < pc no percolation occurs).
‘n’ values are determined from the least squares slope of
Ϫ1
gel
double-logarithmic plots of t and Φ (Fig. 10) and are pre-
tϪge1l = crystallization rate ϩ liquid Ϫ liquid phase separation rate
In the following section an attempt is made to show how the
temperature coefficient analysis of the gelation rate supports
Scheme 1.
Crystallization. The growth rate (G) of fibrillar crystals can
be expressed as:35
(10)
Where G0 is a pre-exponential constant, Ga is the free energy
0
Ϫ1
Fig. 10
t
vs. Φ (in double logarithmic scale) of tripeptide-gels at
gel
of transport, l is the length of rod, T M is the equilibrium melt-
ing temperature, σ is the surface energy, ∆H0u is the enthalpy of
indicated temperature.
0
fusion for a perfect crystal, ∆T = T M Ϫ T . Since according to
sented in Table 2. From the table it is apparent that the ‘n’
values vary from 0.44 to 0.54 and they are close to the theor-
etical value of β, 0.45. So it may be argued that the tripeptide
gel in DCB obeys the percolation mechanism in a three-
dimensional lattice.
Scheme 1 both the crystallization and liquid–liquid phase sep-
aration processes are involved in the gelation process we can
equate gelation rate (tϪge1l) with the growth rate ‘G’ of eqn. (10).
Then taking the logarithm and rearranging both the equations
we have,
Microscopic mechanism. The microscopic mechanism of
gelation of a tripeptide may be elucidated from the analysis of
the temperature coefficient of the gelation rate.7,21 A tripeptide
(10a)
1182
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186

View Article Online
0
Now to analyze eqn. (10a) we have to measure ∆T gm and T gm
where L is the interaction length, T s is the spinodal temper-
0
0
where T gm is the equilibrium gel melting temperature. The T gm
may be measured by the Hoffman–Weeks extrapolation pro-
cedure36 and it is shown for 10% (w/v) gel (Fig. 11). In Fig. 12 a
ature. Under the approximation that liquid–liquid phase
separation originates from spinodal decomposition and the
frequency of the most probable concentration fluctuation is
directly proportional to the gelation rate, eqn. (11a) may be
expressed as
(12)
where ‘c’ is the proportionality constant. Transforming into
the logarithmic form we have:
Fig. 11 Hoffman–Weeks plot for the 10% (w/v) tripeptide–DCB gel.
[TЈgel is the isothermal gelation temperature.]
(12a)
Thus a plot of log tϪge1l vs. log (∆T /T s) should be a straight line
Ϫ1
gel
with a slope value of 0.5. In Fig. 13 plots of log t vs. log
Ϫ1
gel
Fig. 13 Log t vs. log (∆T /T s) plot for tripeptide–DCB gels for
different tripeptide concentrations.
(∆T /T s) are shown. In this plot T s is taken from the phase
diagram (Fig. 5) determined visually with the approximation
that the turbidity disappearance temperature and spinodal
temperature are the same. It is clear from the figure that straight
lines are obtained as expected from the theory, primarily sup-
porting that spinodal decomposition is also a process of gel-
ation in this system. The least squares slopes of the plots (Fig.
13) are 4.4, 3.9, 3.4, 3.7 and 3.3 for 5%, 7%, 9%, 11% and 15%
(w/v) tripeptide gels respectively. So, the slope values are ∼7–8
times larger than the theoretical value of 0.5 and the higher
slope values may be attributed to the crystallization contri-
bution, indicating that the crystallization mechanism is pre-
dominant over the spinodal decomposition for gelation in this
system. This proposition appears to be true as evident from the
micrographs of Fig. 1 where larger proportions of the fibrillar
(crystallite) network than the phase separated network are
clearly shown.
Fig. 12 Ln tϪge1l vs. T g0m/T ∆T gm plot for the 10% (w/v) tripeptide–DCB
gel.
plot of ln tϪge1l has been made with T g0m/T ∆T gm. The straight-line
nature of the plot supports crystallization being involved in the
gelation process.
Liquid–liquid phase separation. From the SEM and TEM
pictures it is apparent that liquid–liquid phase separation is also
responsible for the gelation in this system. The liquid–liquid
phase separation may occur through spinodal decomposition
due to local concentration fluctuation at the gelation temper-
ature and this mechanism is widely accepted for the thermo-
reversible gelation in certain polymeric systems.25,26,37 Cahn
derived an expression for the most probable wave number of the
concentration fluctuation (βm),38
Molecular modeling. In this section we want to predict a
tentative structure of the tripeptide in the gel. The structure has
been developed from a molecular modeling with the MMX
program40 and this is shown in Fig. 14. The structure A has been
derived by energy minimizing a molecular structure drawn with
the idea that folding may occur at the Aib junction due to the
restriction of bond rotation by the presence of bulkier geminal
dimethyl groups. The energy minimization results in negative
MMX energy and enthalpy values with an intramolecular
(11)
Where k is related to the solute–solvent interaction par-
ameter, (∆Fm/V) is the free energy of mixing of a homogeneous
solution per unit volume, ꢀ1 is the local concentration and ꢀ10
is the average concentration of the solvent. Van Aartsen39
deduced the temperature dependency of βm as:
hydrogen bond between >C᎐O and H–N< groups near the fold
᎐
position. The presence of intramolecular hydrogen bonding
in the gel state or in the solid state is also evidenced from the
FT-IR study discussed earlier. This H-bonding may provide
(11a)
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186
1183

View Article Online
NaOH (40 ml) was stirred and cooled in an ice–water bath. Di-
tert-butyl pyrocarbonate (9.6 g, 44 mmol) was then added and
the stirring was continued at room temperature for 6 h. Then
the solution was concentrated in vacuo to about 40 to 60 ml,
cooled in an ice–water bath, covered with a layer of ethyl
acetate (about 50 ml) and acidified with a dilute solution of
KHSO4 to pH 2–3 (Congo red). The aqueous phase was
extracted with ethyl acetate and this operation was done repeat-
edly. The ethyl acetate extracts were pooled, washed with water,
dried over anhydrous Na2SO4 and evaporated in vacuo. A white
crystalline material was obtained.
Yield: 6.718 g (35.54 mmol, 88.85%). Anal. Calcd. for
C8H15NO4 (189): C, 50.79; N, 7.40; H, 7.9. Found: C, 50.9;
N, 7.35; H, 8.12%.
Fig. 14 Molecular modeling of a probable structure of the tripeptide
gel in DCB (using MMX program).
(2) Boc-ꢀ-Ala1-Aib2-OMe. 5.67 g (30 mmol) of Boc-β-Ala-
OH were dissolved in 30 ml dichloromethane (DCM) in an ice–
water bath. H-Aib-OMe was isolated from 9.21 g (60 mmol) of
the corresponding methyl ester hydrochloride by neutralization
using 1 M sodium carbonate solution, with subsequent extrac-
tion with ethyl acetate. It was concentrated and 10 ml of it were
added, followed immediately by 6.18 g (30 mmol) of dicyclo-
hexylcarbodiimide (DCC). The reaction mixture was allowed to
come to room temperature and stirred for 24 h. DCM was
evaporated, the residue was taken in ethyl acetate (60 ml), and
dicyclohexylurea (DCU) was filtered off. The organic layer was
washed with 2 M HCl (3 × 50 ml), brine and then again with
1 M sodium carbonate (3 × 50 ml) and brine (2 × 50 ml). It was
dried over sodium sulfate, and evaporated in vacuo.
additional strength to stabilize the foldamer. The structure B is
the mirror image of structure of A. During gelation inter-
molecular aggregation through hydrogen bonding occurs, as
evidenced from the FT-IR study. We therefore propose that
structures B and A rest on each other in alternating fashion and
hydrogen bonding may occur through the >C᎐O and >N–H
᎐
groups. This type of intermolecular aggregation may produce
fibrillar or rod like crystal morphology. However, we are unable
to give any definite support of this structure from X-ray diffrac-
tion data since we have failed to produce any single crystal of
this tripeptide.
Conclusion
Yield: 7.2 g (25 mmol, 83.33%). Anal. Calcd. for C13H24N2O5
(288): C, 54.16; H, 8.33; N, 9.72. Found: C, 54.3; H, 8.4; N, 9.7%.
It appears from the results that the above tripeptide gel in DCB
is a unique system where gelation is simultaneously caused by
liquid–liquid phase separation and crystallization. The morph-
ology study indicates the simultaneous presence of rod-like and
liquid–liquid phase separated networks in the gels. The WAXS
study confirms the presence of tripeptide crystallites in the gel
state. The thermal study indicates a reversible first-order phase
transition and it also predicts the formation of tripeptide–DCB
complexation during the gelation. The phase diagrams are also
characteristic of liquid–liquid phase separation and tripeptide–
solvent complex formation with singular point. The FT-IR
study indicates that the gelation occurs due to intermolecular
hydrogen bonding. Analysis of the concentration function of
the gelation rate of the tripeptide shows that three-dimensional
percolation is a suitable model for this system also. The analysis
of the temperature function of the gelation rate supports the
proposal that (1) crystallization and (2) spinodal decom-
position are the probable molecular mechanisms responsible for
the gelation.
1
300 MHz H NMR (CDCl3, δ ppm): 6.3[Aib(2)NH,1H,s];
5.2[β-Ala(1)NH,1H,t]; 3.74[OCH3,3H,s]; 3.36–3.42[β-Ala(2)-
CαHs,2H,m]; 2.37–2.41[β-Ala(2)CβHs,2H,m]; 1.53[Aib(2)Cβ-
Hs,6H,s]; 1.43 [Boc-CH3,9H,s].
(3) Boc-ꢀ-Ala1-Aib2-OH. To 6.62 g (23 mmol) of Boc-β-Ala1-
Aib2-OMe, 75 ml MeOH and 30 ml of 2 M NaOH were added
and the progress of saponification was monitored by thin layer
chromatography (TLC). The reaction mixture was stirred. After
10 h methanol was removed under vacuo, the residue was taken
in 50 ml of water, and washed with diethyl ether (2 × 50 ml).
Then the pH of the aqueous layer was adjusted by 2 using 2 M
HCl and it was extracted with ethyl acetate (3 × 50 ml). The
extracts were pooled, dried over sodium sulfate, and evaporated
in vacuo to yield 5.48 g of the pure material.
Yield: 5.48 g (20 mmol, 86.9%). Anal. Calcd. for C12H22N2O5
(274): C, 52.5; H, 8.02; N, 10.2. Found: C, 52.8; H, 8.12;
N, 10.02%.
300 MHz 1H NMR [(CD3)2SO, δ ppm]: 12.01[–COOH,1H,s];
7.92[Aib(2)NH,1H,s]; 6.55[β-Ala(1)NH,1H,t]; 2.93–2.99[β-
Ala(1)CαHs,2H,m]; 2.06–2.11[β-Ala(1)CβHs,2H,m]; 1.25[Boc-
CH3s,9H,s]; 1.19[Aib(2)CβHs,6H,s].
Experimental
(A) Synthesis of Boc-ꢀ-Ala-Aib-ꢀ-Ala-OMe
(4) Boc-ꢀ-Ala1-Aib2-ꢀ-Ala3-OMe. 4.11 g (15 mmol) of Boc-β-
Ala-Aib-OH in 20 ml of DMF was cooled in an ice–water bath.
H-β-Ala-OMe was isolated from 4.18 g (40 mmol) of the
corresponding hydrochloride by neutralization, with subsequent
extraction with ethyl acetate and it was concentrated. 10 ml of
the concentrated material were added, followed immediately by
3.09 g (15 mmol) DCC and 2.0 g (15 mmol) of 1-hydroxy-
benzotriazole (HOBT). The reaction mixture was stirred for
three days. The residue was taken in ethyl acetate (60 ml) and
the DCU was filtered off. The organic layer was then washed
with 2 M HCl (3 × 50 ml), brine, 1 M sodium carbonate (3 ×
50 ml) and again with brine (2 × 50 ml), dried over anhydrous
sodium sulfate and evaporated in vacuo to yield 3.59 g
(10 mmol) of white solid. Purification was done by silica gel
column (100–200 mesh size) using ethyl acetate as eluant. Yield
68.3%.
The peptide was synthesized by conventional solution phase
methods by using a fragment condensation strategy. The Boc
group was used for N-terminal protection and the C-terminus
was protected as a methyl ester. Deprotection was performed
using saponification. Couplings were mediated by dicyclo-
hexylcarbodiimide–1-hydroxybenzotriazole
(DCC–HOBT).
All the intermediates were characterized by 1H-NMR (300 and
500 MHz) and thin layer chromatography (TLC) on silica gel
and were used without further purification. The final product
was purified by column chromatography using silica (100–200
mesh size) gel as stationary phase and ethyl acetate–toluene
mixture as eluent. The compound was then fully characterized
by 500 MHz 1H-NMR.
(1) Synthesis of Boc-ꢀ-Ala-OH. A solution of β-Ala (3.56 g,
40 mmol) in a mixture of dioxan (80 ml), water (40 ml) and 1 M
1184
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186

View Article Online
Anal. Calcd. for C16H29N3O6 (359): C, 53.8; H, 8.07; N, 11.69.
Found: C, 53.6; H, 8.12; N, 11.52%.
tripeptide in 100 ml of solvent) gels were dried in vacuum for
three days at room temperature and were gold coated. SEM
pictures were then taken in an SEM apparatus (Hitachi
S-415A). The TEM study was done by dropping a 0.3% (w/v)
solution of the tripeptide in 1,2-dichlorobenzene directly on a
carbon coated copper grid which was dried in vacuum at 30 ЊC
for three days. The grid was then observed in a transmission
electron microscope (Hitachi H-600), both in the transmission
mode and in the diffraction mode. The structural investigation
of the gels was done by wide angle X-ray scattering (WAXS)
and Fourier transform infra-red spectroscopy. In the WAXS
study the gels were dried in vacuum at 30 ЊC for 5 days and the
diffractograms were recorded in a Phillips powder diffraction
apparatus (model PW1710). Nickel-filtered Cu-Kα radiation
was used for this purpose. The gel was taken in a glass groove
and the diffractogram was recorded at the step scan with the
step size of 0.02Њ, 2θ and a time per step of 0.8 s. The type of
scan was continuous.
500 MHz 1H NMR (CDCl3, δ ppm): 6.94[β-Ala(3)NH,1H,t];
6.27[Aib(2)NH,1H,s]; 5.59[β-Ala(1)NH,1H,t]; 3.7[–OCH3,
3H,s]; 3.49–3.52[β-Ala(3)CαHs,2H,m]; 3.37–3.40[β-Ala(1)-
CαHs,2H,m];
2.57–2.59[β-Ala(3)CβHs,2H,m];
2.40–2.42-
[β-Ala(1)CβHs,2H,m]; 1.51[Aib(2)CβHs,6H,s]; 1.44[Boc-CH3,
9H,s].
(B) NMR studies
All NMR experiments were performed on Bruker DRX
500 MHz and DPX 300 MHz spectrometers. A phase sensitive
DQF-COSY experiment was done for sequence specific
assignments by collecting 2k data points in f2 and 512 points in
f1. Quadrature detection in f1 was achieved using TPPI. Data
were processed on silicon Graphics using Bruker XWIN NMR
software. Typically, a Qsine-Squared Window function was
applied in both dimensions, with Zero filling in f2 to 512 points.
NMR chemical shifts and line widths were shown to be
independent of tripeptide concentration in the range 40 to
0.25 mM, which strongly suggested that the tripeptide didn’t
aggregate for the above range of concentration.
(G) Thermal study
The thermal studies of the gels were done in a differential
scanning calorimeter (DSC-7, Perkin-Elmer). The tripeptide
and DCB in the required amounts were taken in Large Volume
Capsules (LVCs) fitted with an O-ring and these were tightly
sealed with the help of a quick press. They were kept at 110 ЊC
for 10 minutes to make them homogeneous and cooled at 15 ЊC
and equilibrated for 3 h to produce the gels. The LVC pans were
then heated at the rate of 10 ЊC minϪ1 and thermograms were
recorded. For cooling, after equilibration at 110 ЊC for 10 min-
utes, the samples were cooled at the rate of 5 ЊC minϪ1 and
thermograms were recorded. For the isothermal gelation at
different temperatures the samples were cooled from 110 ЊC
to the desired isothermal temperatures and kept there for
3 h. They were heated at the scanning rate of 10 ЊC minϪ1 from
that temperature without cooling. The melting point, gel-
ation temperature and the corresponding enthalpy changes
were measured using a computer attached to the instrument.
The DSC was calibrated with indium before each set of
experiments.
(C) Mass spectroscopy
The mass spectrum of the tripeptide was taken using the
MALDI-MS spectroscopic technique. The spectrum was
scanned over the mass range of m/z 100–1000 by 0.1 amu, with
dwell time 1 ms per step.
(D) FT-IR spectroscopy
The FT-IR spectra were taken using a Shimadzu (Japan) model
FT-IR spectrophotometer with a sample-shuttle device, aver-
aging over 40 scans. Solvent (DCB) spectra were obtained
under the same conditions using a cuvette with 1 mm
path length. A Nicolet FT-IR instrument [Magna IR-750
spectrometer (series II)] was used to obtain the solid state and
the gel state FT-IR spectra. The solvent spectrum was sub-
tracted from the gel spectra to obtain tripeptide spectra in the
gel state.
(H) Gelation rate measurement
(E) Preparation of gel
The kinetics of the gelation were studied by measuring the
gelation time using the test tube tilting method.7,20–21 To make
gels at different tripeptide concentrations 0.2 ml of 1,2-
dichlorobenzene (density = 1.304 g mlϪ1) and an appropriate
amount of the tripeptide (7–30 mg) were taken in glass tubes
(6 mm id and 1 mm thickness). They were degassed by a
repeated freeze–thaw technique and were sealed under vac-
uum (10Ϫ3 mm Hg). They were homogenized at 100 ЊC and
were quickly transferred to a thermostatic bath set at a pre-
determined temperature. The gelation time (tgel) was counted
as the time where no flow occurred after the tube was
The gels were prepared in two ways. (1) For morphology and
structural investigations, the gels were prepared in glass tubes
(6 mm id) by taking an appropriate amount of the tripeptide
sample and 1,2-dichlorobenzene (DCB) [S.D. Chemicals, India,
Analytical Reagent Grade]. The tubes were degassed by the
repeated freeze–thaw technique and were then sealed under
vacuum (10Ϫ3 mm Hg.). They were made homogeneous at
110 ЊC and then quickly gelled by quenching at room temper-
ature (30 ЊC). (2) For the thermal investigation, the gels were
prepared in Perkin-Elmer large volume capsules (LVC) by tak-
ing appropriate amounts of tripeptide and the solvent and were
tightly sealed with the help of a quick press. They were made
homogeneous by keeping them at 110 ЊC for 10 minutes and
were gelled at 15 ЊC for 3 h by cooling from 110 ЊC at the rate of
tilted.7,20–21 Accuracy in the rate measurement to tgel
=
5 s
was achieved by a trial and error procedure.7,21 The inverse of
the gelation time was approximated as the gelation rate
Ϫ1
(t ).7,20,21
gel
200 ЊC minϪ1
.
Acknowledgements
(F) Morphology and structural investigations of the gels
One of the authors (A. B.) acknowledges the Department of
Science and Technology, New Delhi, for financial support.
The morphology of the gels was investigated using optical
microscopy (OM), transmission electron microscopy (TEM)
and scanning electron microscopy (SEM). The optical micro-
scopy of the samples was done using a small amount of the gel
on microscopic slides and drying in a stream of air at room
temperature. The sample was observed under the microscope.
The morphology of the undried gel was also observed in the
microscope using polarized light which facilitated observation
of the crystallites. For SEM study the 10% (w/v, i.e., 10 g of
References
1 (a) J. H. van Esch and B. L. Feringa, Angew. Chem., Int. Ed., 2000,
39, 2263; (b) F. S. Schoonbeek, J. H. van Esch, R. Hulst, R. M.
Kellogg and B. L. Feringa, Chem. Eur. J., 2000, 6, 2633; (c) J. H.
van Esch, F. S. Schoonbeek, M. de Loos, H. Kooijman, A.-L. Spek,
R. M. Kellogg and B. L. Feringa, Chem. Eur. J., 1999, 5, 537.
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186
1185

View Article Online
2 P. Terech and R. G. Weiss, Chem. Rev., 1997, 97, 3133.
3 T. D. Clark, L. K. Buehler and M. R. Ghadiri, J. Am. Chem. Soc.,
1998, 120, 651.
4 (a) D. Philip and J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1996,
35, 1154; (b) O. Gronawald and S. Shinkai, J. Chem. Soc., Perkin
Trans. 2, 2001, 1933; (c) J. H. Jung, M. Amaike, K. Nakashima and
S. Shinkai, J. Chem. Soc., Perkin Trans. 2, 2001, 1938.
5 M. S. Vollmer, T. D. Clark, C. Steinem and M. R. Ghadiri, Angew.
Chem., Int. Ed., 1999, 38, 1598.
6 Comprehensive Supramolecular Chemistry, Vol. 6: Solid-State
Supramolecular Chemistry: Crystal Engineering, Eds. D. D.
MacNicol, F. Toda and R. Bishop, Elsevier, Oxford, UK, 1996.
7 S. Mal and A. K. Nandi, Polymer, 1998, 30, 6301.
8 A. K. Dikshit and A. K. Nandi, Macromolecules, 1998, 31, 8886.
9 K. Hanabusa, M. Yamada, M. Kimura and H. Shirai, Angew.
Chem., Int. Ed. Engl., 1996, 35, 1949.
1990, 23, 5041; (c) M. Ohkura, T. Kaniaya and K. Kaji, Polymer,
1992, 33, 5044; (d ) J. W. Cho, H. Y. Song and S. Y. Kim, Polymer,
1993, 34, 1024.
21 (a) S. Mal, P. Maiti and A. K. Nandi, Macromolecules, 1995, 28,
2371; (b) S. Malik, T. Jana and A. K. Nandi, Macromolecules, 2001,
34, 275.
22 In the electronic diffraction pattern some spots are also seen; the
formation of single crystals in the gelation condition is a difficult
situation and a probable reason for the simultaneous occurrence of
rings and spots is the different sizes of the crystals. The upper size
limit giving ring patterns is 103–102 Å but above this size electron
diffraction may give spots23,24
.
23 K. W. Andrews, D. J. Dyson and S. R. Keown, Interpretation of
Electron Diffraction Patterns, Plenum Press, New York, 1964.
24 T. Jana and A. K. Nandi, Langmuir, 2000, 16, 3141.
25 H. Berghmans, in Integration of Fundamental Polymer Science and
Technology, Eds. P. J. Lemstra and L. A. Kleintjens, Elsevier Applied
Science, London, 1988, Vol. 2, p. 296.
10 R. J. H. Hafkamp, M. C. Feiters and R. J. M. Nolte, J. Org. Chem.,
1999, 64, 412.
11 Y.-C. Lin, B. Kachar and R. G. Weiss, J. Am. Chem. Soc., 1989, 111,
5542.
12 S. H. Gellman, Acc. Chem. Res., 1998, 31, 173.
26 R. M. Hiknet, S. Callister and A. Keller, in Integration of
Fundamental Polymer Science and Technology, Eds. P. J. Lemstra and
L. A. Kleintjens, Elsevier Applied Science, London, 1988, Vol. 2,
p. 306.
27 C. Daniel, C. Dammer and J. M. Guenet, Polym. Commun., 1994,
35, 4243.
13 (a) D. H. Appella, L. A. Chistianson, D. A. Klein, D. R. Powell,
Y. Huang and S. H. Gellman, Nature, 1997, 387, 381; (b) X. Wang,
J. F. Espinosa and S. H. Gellman, J. Am. Chem. Soc., 2000, 122,
4821.
14 (a) S. Abele, D. Guichard and D. Seebach, Helv. Chim. Acta, 1998,
81, 2141; (b) D. H. Appella, L. A. Chistianson, I. L. Karle, D. R.
Powell and S. H. Gellman, J. Am. Chem. Soc., 1999, 121, 6206.
15 (a) Y. J. Chung, B. R. Huck, L. A. Christianson, H. E. Stanger,
S. Krauthauser, D. R. Powell and S. H. Gellman, J. Am. Chem. Soc.,
2000, 122, 3995; (b) D. Seebach, S. Abele, K. Gademan and B. Jaun,
Angew. Chem., Int. Ed., 1999, 38, 1595.
28 J. M. Guenet and G. B. Mckenna, Macromolecules, 1998, 21,
1752.
29 A. Reisman, Phase Equilibria, Academic Press, New York, 1970.
30 J. R. Dyer, Application of Absorption Spectroscopy of Organic
Compounds, Prentice Hall of India, New Delhi, 1987.
31 N. B. Colthup, L. H. Daly and S. E. Wilberley, Introduction to
Infrared and Raman Spectroscopy, Academic Press, New York,
1964.
16 (a) M. R. Ghadiri, J. R. Granja, R. A. Milligan, D. E. McRee and
N. Jhazanovich, Nature, 1993, 366, 324; (b) M. R. Ghadiri, J. R.
Granja and L. K. Buchler, Nature, 1994, 369, 301; (c) H. S. Kim,
J. D. Hartgerink and M. R. Ghadiri, J. Am. Chem. Soc., 1998, 120,
4417.
17 (a) D. Raganathan, V. Haridas, S. Kumar, A. Thomas, K. P.
Madhusudanam, R. Nagraj, A. C. Kumar, A. V. S. Sarma and I. L.
Karle, J. Am. Chem. Soc., 1998, 120, 8448; (b) D. Ranganathan,
V. Haridas, R. Gilardi and I. L. Karle, J. Am. Chem. Soc., 1998, 120,
10793.
18 (a) S. Mal and A. K. Nandi, Langmuir, 1998, 14, 2238; (b) A. K.
Dikshit and A. K. Nandi, Macromolecules, 2000, 33, 2616.
19 (a) I. L. Karle, A. Banerjee, S. Bhattacharyya and P. Balaram,
Biopolymers, 1996, 38, 515; (b) A. Banerjee, S. Datta, A. Pramanik,
N. Shamala and P. Balaram, J. Am. Chem. Soc., 1996, 118, 9477.
20 (a) H. M. Tan, A. Hiltner, A. Moet and E. Baer, Macromolecules,
1983, 16, 28; (b) A. Prasad and L. Mandelkern, Macromolecules,
32 D. Stauffer, A. Coniglio and M. Adam, in Advances in Polymer
Science, Ed. K. Dusek, Springer-Verlag: Berlin, 1982, Vol. 44, p. 103.
33 R. Zallen, The Physics of Amorphous Solids, John Wiley & Sons,
New York, 1983, p. 135.
34 S. W. Benson, The Foundation of Chemical Kinetics, McGraw-Hill
Book Co., New York, 1960, p. 26.
35 A. J. Pennings, J. Polym. Sci., Polym. Symp., 1977, 59, 55.
36 J. D. Hoffman and J. J. Weeks, J. Res. Natl. Bur. Stand. (U. S.), 1962,
66, 13.
37 (a) D. Asnaghi, M. Cigilo, A. Bossi and P. G. Righetti, J. Chem.
Phys., 1995, 102, 9376; (b) R. Bansil, J. Lal and B. L. Carvalho,
Polymer, 1992, 33, 2961; (c) H. Takeshita, T. Kanaya, K. Nishida
and K. Kaji, Macromolecules, 1999, 32, 7815.
38 J. W. Cahn, J. Chem. Phys., 1965, 42, 93.
39 J. J. Van Aartsen, Eur. Polym. J., 1970, 6, 919.
40 K. E. GaJewski and M. H. Gillberr, in Advances in Molecular
Modeling, Ed. D. Liotta, JAI Press, Greenerick, CT, 1990, Vol. 2.
1186
J. Chem. Soc., Perkin Trans. 2, 2002, 1177–1186