Liquid crystalline polyamines containing side dendrons: Toward the building of ion channels based on polyamines

Article abstract of DOI:10.1016/j.polymer.2013.07.027

In this article, we modified poly[2-(aziridin-1-yl)ethanol] (PAZE) with the dendron 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoic acid, to obtain liquid crystalline columnar polyamines. The chemical modification reaction was first tuned on a model compound, N,N-dimethyl-2-hydroxyethylamine. The best results were obtained by the esterification method with N,N'- dicyclohexylcarbodiimide at room temperature, in the presence of 4-dimethylaminopyridine. The obtained copolymers showed higher char yield than starting PAZE. In all cases they exhibited small crystalline portions after annealing and columnar mesophases, as inferred by DSC, XRD and POM. The dimension of the unit cell resulted slightly narrower than in the case of the copolyethers bearing the same dendron. This is probably due to the presence of a longer spacer in PAZE, which allows better accommodating of the side tapered group.

Full text of DOI:10.1016/j.polymer.2013.07.027

Polymer 54 (2013) 5133e5140
Contents lists available at SciVerse ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Liquid crystalline polyamines containing side dendrons: Toward the
building of ion channels based on polyamines
a
b
a,
ꢀ
*
_
Asta Sakalyte , José Antonio Reina , Marta Giamberini
a Departament d’Enginyeria Química, Universitat Rovira i Virgili, Av. Països Catalans, 26, 43007 Tarragona, Spain
b Departament de Química Analítica i Química Orgànica, Universitat Rovira i Virgili, Carrer Marcel$lí Domingo s/n, 43007 Tarragona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
In this article, we modified poly[2-(aziridin-1-yl)ethanol] (PAZE) with the dendron 3,4,5-tris[4-(n-dodecan-
1-yloxy)benzyloxy]benzoic acid, to obtain liquid crystalline columnar polyamines. The chemical modifi-
cation reaction was first tuned on a model compound, N,N-dimethyl-2-hydroxyethylamine. The best results
were obtained by the esterification method with N,N’-dicyclohexylcarbodiimide at room temperature, in
the presence of 4-dimethylaminopyridine. The obtained copolymers showed higher char yield than starting
PAZE. In all cases they exhibited small crystalline portions after annealing and columnar mesophases, as
inferred by DSC, XRD and POM. The dimension of the unit cell resulted slightly narrower than in the case of
the copolyethers bearing the same dendron. This is probably due to the presence of a longer spacer in PAZE,
which allows better accommodating of the side tapered group.
Received 10 May 2013
Received in revised form
4 July 2013
Accepted 9 July 2013
Available online 15 July 2013
Keywords:
Chemical modification
Liquid crystalline polymers
Columnar mesophase
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
In our previous studies [10,11], we reported the chemical
modification of poly(epichlorohydrin) (PECH) with tapered meso-
The design, synthesis, and study of supramolecular assemblies
capable of transporting ions across membranes by the channel
mechanism have been the subject of increasing interest [1e3].
A self-assembly event can be dominated by a single molecular
recognition process based on directional and attractive forces such
as hydrogen bonding, van der Waals, electrostatic, and hydrophobic
interactions. However, these processes are cooperative, and their
cooperative contribution could be equally important not only in the
aggregation process but also particularly in the stabilization of a
particular architecture.
Liquid crystals (LCs) present an attractive option to design
functional materials with a well-ordered internal structure. Fitie
and coworkers [4] reported the structure of mixed LC phases
formed by orthogonal self-assembly between acid-modified dis-
cotic and various polyamines and demonstrated that the principle
of orthogonal self-assembly can be applicable to create liquid
crystalline superlattices of polyamines.
genic groups to yield high-molecular-weight polyethers with
different degrees of modification. Since we obtained good modifi-
cation degrees and detected no dehydrochlorination side reactions
in the chemical modification of PECH [12,13], we were encouraged
to use the same strategy and reaction conditions also for
poly(epichlorohydrin-co-ethylene oxide) [P(ECH-co-EO)]. In a very
recent paper [14], we have reported the preparation of a new family
of liquid crystalline columnar polyethers by chemically modifying
P(ECH-co-EO) with the dendron 3,4,5-tris[4-(n-dodecan-1-yloxy)
benzyloxy]benzoate under different conditions. Modification de-
grees ranged from 57 to 67%. These copolymers can be used to
prepare oriented membranes for small cation transport, in agree-
ment with the results that we obtained on oriented membranes
based on PECH modified with the same dendron, where proton
permeability comparable to NafionÒ 117 was found [15]. In those
polymers, the polyether main chain forms a channel in the inner
part of the columns, while the hydrophobic side-chain dendrons lie
on the outer part. The presence of the polar ether linkages in the
inner channel favors the interaction with proton and other cations,
in the same way as crown ethers would do [16]. For this reason, the
inner polyether chain can work as an ion channel. If this concept is
transferred to a polyamine main chain, the occurrence of basic ni-
trogen atoms in the inner helical backbone structure will probably
confer to these materials the capability to transport proton ions by
the channel mechanism.
Percec and coworkers [5e9] have comprehensively investigated
the self-organization of supramolecular monodendrons and sty-
rene-, methacrylate-, or oxazoline-based polymers for the design of
ion-active nanostructured supramolecular systems.
* Corresponding author. Tel.: þ34 977558174; fax: þ34 977559621.
E-mail address: marta.giamberini@urv.net (M. Giamberini).
0032-3861/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.polymer.2013.07.027

ꢀ
_
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
5134
Hydroxylated polymers can be regarded as powerful precursors
The synthesis of trifluoroacetic 3,4,5-tris[4-(n-dodecan-1-yloxy)
benzyloxy]benzoic anhydride (3) was performed according to the
following procedure: to a solution of 1 g (1 mmol) of 2 in 20 ml of
chloroform, 0.2 g (1 mmol) trifluoracetic anhydride was added. The
reaction mixture was stirred at room temperature and monitored
by TLC with toluene as an eluent. After 15 min the solvent and
excess of acid were vacuum distilled and the obtained product was
recrystallized from hexane to yield of 66%.
in the preparation of new polymers through chemical modification,
which can finally lead to properly functionalized macromolecules
for specific applications. In our previous work [17] we tackled
the polymerization of a hydroxylated aziridine, 2-(aziridin-1-yl)
ethanol (AZE). In that study we deepened cationic polymerization
of this monomer by using several initiators and reaction conditions
in order to minimize the participation of the hydroxyl group and to
get a polyamine with a structure as linear as possible and molecular
weight suitable for future applications. This functionalized polymer
(PAZE) resulted interesting as a versatile starting material for
chemical modification reactions in order to obtain side-chain liquid
crystalline polyamines. Nematic side-chain polyamines were ob-
tained by chemical modification of polyethyleneimine (PEI) [18]. In
this case, LC phases were obtained for modification degrees higher
than 69%; however, polymers containing about 80% of mesogenic
groups were crystalline. Moreover, the authors showed that
ammonium quaternization occurred to some extent, which is un-
desirable in our case, since the presence of basic nitrogen atoms is
crucial to determine proton transport. Therefore, in this article, we
modified PAZE with the dendron 3,4,5-tris[4-(n-dodecan-1-yloxy)
benzyloxy]benzoic acid, to obtain liquid crystalline columnar
polyamines. To the best of our knowledge, this is the first example
of LC columnar polyamine.
1H NMR [CDCl3, TMS,
d, ppm]: 7.30 (s, 2H, ArHeCOOe), 7.26
(d, 4H, ArH meta to eCH2eOe in lateral benzylic units), 7.18 (d, 2H,
ArH meta to eCH2eOe in central benzylic units), 6.69 (m, 6H, ArH
ortho to eCH2eOe), 4.95 (m, 6H, PheCH2eO in lateral benzylic
units), 3.86 (m, 6H, eCH2eOePh), 1.66 (m, 6H, eCH2eCH2eOPh),
1.37 (m, 6H, eCH2(CH2)2eOePh), 1.19 (m, 48H, e(CH2)8e), 0.81
(t, 9H, eCH3).
13C NMR (CDCl3, TMS,
d, ppm): 167.3 (COCF3), 159.2 (ArC para to
eCH2eOePh), 153.0 (COOCOCF3), 152.7 (ArC meta to COO), 143.2
(ArC para to COO),130.5 (ArC ipso to COO),129.4 (ArC ortho to CH2e
OePh in central benzylic unit), 128.6 (ArC ortho to eCH2eOePh in
lateral benzylic units), 125.6 (ArC ipso to eCH2-O-Ph), 114.6 (ArC
meta to eCH2-O-Ph in lateral benzylic units), 114.3 (ArC meta to e
CH2eOePh in central benzylic unit), 114.2 (CF3), 109.7 (ArC
ortho to COO), 74.8 (PheCH2eOePh in central benzylic unit), 71.1
(PheCH2eOePh in lateral benzylic units), 68.2 (ReCH2eOePh),
32.4 (eCH2CH2CH3), 30.1e28.3 (C2 and C4 to C9 in aliphatic chains),
26.2 (PheOeCH2CH2CH2e), 22.8 (eCH2CH3), 14.3 (eCH2CH3).
2. Experimental
19F NMR (CDCl3, TMS,
19F NMR (CDCl3, TMS,
[21].
d
, ppm): ꢁ75.0 (CF3).
2.1. Materials
d
, ppm): ꢁ76.0 (trifluoracetic anhydride)
Model compound N,N-dimethyl-2-hydroxyethylamine (DMHEA)
(98%) was supplied from Aldrich and used as received (MW ¼ 89.1 g/
2.1.3. Synthesis of 2-(dimethylamino)ethyl 3,4,5-tris[4-(n-dodecan-
1-yloxy)benzyloxy]benzoate (4) (Scheme 1)
mol,
r
¼ 0.89 g/ml).
All organic and inorganic reagents were supplied by Fluka or
Aldrich and used as received.
N,N-dimethylformamide (DMF) was purified according to the
literature [19].
2.1.3.1. Method with thionyl chloride (SOCl2) in the presence of DBU.
To a solution of 1 g (1 mmol) of 2 in 20 ml of dry 1,2-dichloroethane,
0.3 g (2 mmol) of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) and
few drops of DMF were added. The mixture was then cooled to 0 ꢀC
and 0.1 g (1 mmol) of SOCl2 was subsequently added slowly. The
reaction was monitored by TLC with toluene as an eluent. After
10 min 0.1 g (1 mmol) of the model compound N,N-dimethyl-2-
hydroxyethylamine (DMHEA) was added and then the reaction
mixture was stirred at room temperature for 20 h. The resulting
solid was purified by column chromatography using toluene as an
eluent to yield of 32%.
2.1.1. Synthesis of the poly[2-(aziridin-1-yl)ethanol] (PAZE)
The synthesis of the polymer PAZE was reported in our previous
paper [17]. The polymer used for modification reactions was obtained
with 1 mol% ethylamine trifluoroborane (BF3$EtNH2) in the absence
of solvent at 45 ꢀC: in this case, a polymerization degree of about 44
monomeric units was found (as determined by size exclusion
chromatography-multi-angle laser light scattering (SEC-MALLS)).
2.1.3.2. Method based on the GareggeSamuelsson reaction [22].
To a solution of 2 g (8 mmol) I2 in dry dichloromethane (30 ml), 2 g
(8 mmol) triphenylphosphine (PPh3) was added, giving a brown
2.1.2. Synthesis of the tapered mesogenic compounds (1e3)
Methyl
3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoate
(1) was prepared as described elsewhere [10].
The synthesis of 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]
benzoic acid (2) was performed following a slight modification of a
reported procedure [20] that involved easier workup and higher
yield:
Potassium hydroxide (0.8 g, 14 mmol) in 5 ml of water was
added to a solution of 1 (5 g, 5 mmol) in a mixture of 10 ml of
tetrahydrofuran (THF) and 40 ml of ethanol. The reaction mixture
was refluxed and monitored by thin layer chromatography (TLC)
with toluene as an eluent and stopped after 30 min by pouring into
1 l of ice water. It was then acidified carefully with concentrated
hydrochloric acid (HCl) (5 ml), refluxed for an additional 15 min
period and the precipitate was filtered. The resulting solid was
redissolved in chloroform and concentrated HCl was added. The
organic layer was separated and washed several times with water,
dried over anhydrous magnesium sulfate (MgSO4) and filtered.
The solvent was vacuum distilled and the obtained solid was
recrystallized from isopropanol to yield 91% of the white product.
Scheme 1. Reaction with the model compound.

ꢀ
_
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
5135
Table 1
yellow solution. Then, 1.2 g (20 mmol) imidazole was added, and
the color changed to light yellow. Subsequently, 5 g (5 mmol) of 2
was added and the solution was stirred for 5 min at room tem-
perature. Then 0.7 g (8 mmol) DMHEA was added. After 3 days the
solvent was vacuum distilled and the resulting solid was purified by
column chromatography using toluene as an eluent to yield of 47%
of compound 4.
Chemical modification conditions, resulting yields and modification degrees of co-
polymers PA1ePA6.
Copolymer
T (ꢀC)
Polymer/COOH
molar ratio
RCOOH
(mmol)
Yield (%)
Modification
degree (%)
PA1
PA2
PA3
PA4
PA5
PA6
25
25
25
60
25
25
1:0.2
1:0.4
1:0.5
1:1
1:0.7
1:1
1.5
2.7
2.3
2.2
2.4
2.2
79
77
70
58
69
62
16
39
46
55
67
2.1.3.3. Transesterification reaction using sodium methoxide (NaOCH3).
To a solution of 1.5 g (1.5 mmol) of 1 and 0.3 g (3 mmol) of DMHEA in
20 ml of chloroform, 0.004 g (0.08 mmol) of NaOCH3 was added. The
mixture was then stirred for 5 days at 60 ꢀC. The reaction was
monitored by TLC with toluene as an eluent. Then the solvent was
vacuum distilled and the obtained product was recrystallized twice
from methanol to yield 58% of compound 4.
100
2.2. Characterization and measurements
2.2.1. Thermogravimetric analysis (TGA)
2.1.3.4. Esterification method with N,N0-dicyclohexylcarbodiimide
(DCC)/DMAP. In a round-bottomed flask 0.2 g (2 mmol) of DMHEA
and 0.1 g (0.8 mmol) of 4-dimethylaminopyridine (DMAP) were
dissolved in chloroform (20 ml) by stirring for 15 min at 0 ꢀC and
then 0.4 g (2 mmol) of N,N’-dicyclohexylcarbodiimide (DCC) was
added. After additional 15 min 1.7 g (2 mmol) of 2 was added slowly
and the mixture was magnetically stirred at room temperature for 5
days. The reaction mixture was filtered in order to eliminate un-
desirable N,N0-dicyclohexylurea. The solvent was vacuum distilled
and the obtained solid was purified by column chromatography
using toluene as an eluent to yield 70% of the white product.
Melting point: 43e45 ꢀC.
Thermal stability studies were carried out in ALU OXIDE
crucibles of 70 l (ME-24123) with a Mettler TGA/SDTA851e/LF/
m
1100 device at temperatures ranging from 30 to 600 ꢀC with a
heating rate of 10 ꢀC/min using about 10 mg of sample in a ni-
trogen atmosphere (100 ml/min). The equipment was previously
calibrated with indium (156.6 ꢀC) and aluminum (660.3 ꢀC)
pearls.
2.2.2. Differential scanning calorimetry (DSC)
Calorimetric studies were performed in aluminum standard
40 ml crucibles without pin (ME-26763) with a Mettler DSC822e
thermal analyzer at the heating rate of 10 ꢀC/min using about 5 mg
of sample, nitrogen as a purge gas (100 ml/min) and liquid nitrogen
for the cooling system. The equipment was previously calibrated
with indium (156.6 ꢀC) and zinc (419.58 ꢀC) pearls.
1H NMR [CDCl3, tetramethylsilane (TMS),
d, ppm]: 7.30 (s, 2H,
ArHeCOOe), 7.26 (d, 4H, ArH meta to eCH2eOe in lateral benzylic
units), 7.18 (d, 2H, ArH meta to eCH2eOe in central benzylic units),
6.69 (m, 6H, ArH ortho to eCH2eOe), 4.94 (s, 4H, PheCH2eO in
lateral benzylic units), 4.93 (s, 2H, PheCH2eO in central benzylic
units), 4.30 (m, 2H, eOeCH2eCH2eN(CH3)2), 3.85 (m, 6H, eCH2e
OePh), 2.62 (t, 2H, eCH2eN(CH3)2), 2.25 (s, 6H, eN(CH3)2), 1.69 (m,
6H, eCH2eCH2eOPh), 1.36 (m, 6H, eCH2(CH2)2eOePh), 1.19 (m,
48H, e(CH2)8e), 0.81 (t, 9H, eCH3).
2.2.3. Nuclear magnetic resonance (NMR) spectroscopy
Compounds were characterized by 1H, 19F and 13C NMR spectra,
which were recorded in deuterated chloroform (CDCl3) with a
Varian Gemini 400 MHz spectrometer (1H e 400 MHz, TMS; 13C e
100 MHz, TMS; 19F e 377 e MHz) at room temperature using a
pulse delay time of 5 s for 1H NMR spectra.
13C NMR (CDCl3, TMS,
d, ppm): 166.9 (COO), 159.6 (ArC para to e
CH2eOePh), 152.7 (ArC meta to COO), 142.4 (ArC para to COO),
130.4 (ArC ipso to COO), 129.4 (ArC ortho to CH2eOePh in central
benzylic unit), 128.8 (ArC ortho to eCH2eOePh in lateral benzylic
units), 125.3 (ArC ipso to eCH2eOePh), 114.7 (ArC meta to eCH2e
OePh in lateral benzylic units), 114.2 (ArC meta to eCH2eOePh in
central benzylic unit), 109.3 (ArC ortho to COO), 74.7 (PheCH2eOe
Ph in central benzylic unit), 71.3 (PheCH2eOePh in lateral benzylic
units), 67.4 (ReCH2eOePhe), 63.3 (eOeCH2eCH2eN(CH3)2), 57.9
(eCH2eN(CH3)2), 46.2 (eN(CH3)2), 32.1 (eCH2CH2CH3), 30.1e29.5
(C2 and C4 to C9 in aliphatic chains), 26.9 (PheOeCH2CH2CH2e),
22.9 (eCH2CH3), 14.6 (eCH2CH3).
2.2.4. Size exclusion chromatography (SEC)
Average molecular weights of PA1ePA6 were determined in
tetrahydrofuran by SEC. Analyses were carried out in an Agilent
1200 series system with three serial columns (PLgel 3
mm MIXED-E,
PLgel 5 m MIXED-D and PLgel 20 m MIXED-A), and equipped
m
m
with an Agilent 1100 series refractive-index detector. Calibration
curves were based on polystyrene standards with low poly-
dispersities. THF was used as an eluent at a flow rate of 1.0 ml/min,
the sample concentrations were 5e10 mg/ml, and injection vol-
umes of 100 mL were used.
IR (cmꢁ1): 2917e2850 (CeH aliphatic); 1721 (C]O); 1514 (C]C
aromatic); 1334 (CeC aliphatic); 1244 (CeN).
2.2.5. Polarized optical microscopy (POM)
Clearing temperatures were roughly estimated using polarized
optical microscopy (POM); textures of the samples were observed
with an Axiolab Zeiss optical microscope equipped with a Linkam
TP92 hot stage.
2.1.4. Synthesis of copolymers PA1ePA6
In a round-bottomed flask 0.7 g (8 mmol) of polymer PAZE
and 0.02 g (0.16 mmol) of DMAP were dissolved in 20 ml of
chloroform by stirring for 15 min at 0 ꢀC and then 0.3 g
(1.5 mmol) of DCC was added. After additional 15 min, the
necessary amount of 2 was added slowly and the mixture was
magnetically stirred at different temperatures (25 ꢀC or 60 ꢀC) for
10 days. The reaction mixture was filtered in order to eliminate
undesirable N,N’-dicyclohexylurea and precipitated twice from
chloroform into methanol. The different modification conditions
are reported in Table 1.
2.2.6. Fourier transform infrared spectroscopy (IR)
Fourier transform infrared spectra were recorded on an FTIR-
ATR 680 PLUS spectrophotometer from JASCO with a resolution
of 4 cmꢁ1 in the absorbance mode. This device is equipped with
an attenuated total reflection accessory (ATR) with thermal
control and
a diamond crystal (Golden Gate heated single
reflection diamond ATR from Specac-Teknokroma). The spectra
were recorded at room temperature from the solid state pure
compounds.
IR (cmꢁ1): 3364 (OH); 2920e2852 (CeH aliphatic); 1712 (C]O);
1513 (C]C aromatic); 1330 (CeC aliphatic); 1244 (CeN).

ꢀ
_
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
5136
2.2.7. X-ray diffraction (XRD)
XRD patterns were recorded in a Bruker-AXS D8-Discover
diffractometer equipped with parallel incident beam (Göbel mirror),
temperatures and solvents. In the reaction with the model com-
pound we obtained compound 4 with 58% yield in the best case.
However, when we applied this methodology to the chemical
modification of PAZE, even using higher temperatures, longer re-
action times and different solvents, we could not achieve modifi-
cations degrees higher than 20%.
Morcillo and coworkers [22] developed a method for the
esterification/amidation of carboxylic acids based on the Garegge
Samuelsson reaction [24], which proposes the formation of an
alkoxyphosphonium intermediate that suffers the nucleophilic
attack to generate an iodo compound. In our case, by reaction with
the model compound we were able to obtain compound 4 with 47%
yield. However, when we performed the reaction with PAZE,
modification degrees as low as 15% were obtained, probably due to
the steric repulsion between the substituents of the phosphorous
atom and the substituents of the alcohols, which prevented the
nucleophilic attack.
vertical
qeq goniometer, and XYZ motorized stage. The GADDS
(general area diffraction system) detector was an HI-STAR (multiwire
proportional counter of 30 ꢂ 30 cm2 with a 1024 ꢂ 1024 pixel). The
X-ray diffractometer was operated at 40 kV and 40 mA to generate Cu
Ka radiation. SamplesweredissolvedinTHFand placedon the sample
holderfortransmissionmodeandthenTHFwasleft toevaporate. Two
analytical conditions were used.
For low 2
30 cm; the collected frame (2D XRD pattern) covers a range from
0.9 up to 9.2ꢀ 2
. The diffracted X-ray beam traveled through a He-
q range: collimator, 100 mm; distance sample-detector,
q
filled chamber (SAXS attachment) to reduce the air scattering at
low angles. The direct X-ray beam was stopped by a beam stop
placed directly on the detector face. The exposition time was of
1800 s per frame and it was first chi-integrated to generate the
conventional 2
tegrated to generate a Chi vs. intensity diffractogram.
For medium 2 range: collimator, 500 m; distance sample-
detector, 9 cm; the collected frame (2D XRD pattern) covers a
q
vs. intensity diffractogram and after it was 2
q
-in-
Stacey and coworkers [25] suggested a method of esterification
using trifluoroacetic anhydride. However, though we were able to
obtain trifluoroacetic 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]
benzoic anhydride, some cleavage of the benzyl ether moieties was
observed in 1H NMR spectrum of compound 4.
q
m
range from 3.0 up to 25.50 2
q. The direct X-ray beam was stopped by
a beam stop placed behind the sample with an aperture of 40. The
exposition time was of 300 s per frame and it was first chi-integrated
The use of N,N’-dicyclohexylcarbodiimide (DCC) as a promoter
represents one of the most versatile esterification methods [26].
Although this reagent is irritant to skin and stoichiometric dosage
or more is necessary, this procedure enjoys various advantages. The
reaction usually proceeds at room temperature, and the conditions
are so mild that substrates with various functional groups can be
employed. The reaction is not sensitive to steric hindrance of the
reactants, allowing production of esters of tertiary alcohols. As
such, a wide range of applications have been achieved in the field of
natural products, peptides, nucleotides, etc. however, unfortunately
this procedure has some drawbacks: yields are not always high, and
undesirable N-acylureas are occasionally formed. These drawbacks
can be overcome by addition of strong acids such as p-toluene-
sulfonic acid [27,28]. Alternatively, addition of a catalytic amount of
p-aminopyridines is more effective [29,30].
to generate the conventional 2q vs. intensity diffractogram and after
it was 2 -integrated to generate a Chi vs. intensity diffractogram.
q
For the recording of XRD patterns on oriented samples, the co-
polymers were mechanically oriented by shearing just below
clearing temperature on a silicon single-crystal wafer surface cut
parallel to the (510) plane. The exposition time was of 300 s per
frame and it was first chi-integrated to generate the chi vs. intensity
diffractograms. In this case XRD patterns were recorded in reflec-
tion mode.
3. Results and discussion
The aim of this work was to obtain polyamines bearing the
tapered
3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoate
To check the feasibility of this reaction with the model com-
pound, we applied the esterification method using DCC and DMAP
at room temperature. The reaction took place for 5 days and finally
the desired compound 4 was obtained with 82% yield. This
encouraged us to apply this reaction to the chemical modification of
PAZE using a higher temperature and longer reaction time. How-
ever, high reaction temperatures gave poor results, since the degree
of modification was low and we faced to the difficulties to eliminate
unreacted acid. This could be explained by the occurrence of side
reactions, for example, NeO intramolecular acyl migration reaction,
which reduce the coupling yields, and also by the formation of
acylureas, which usually exhibit very similar solubilities to the
desired products, and therefore make their removal difficult to
achieve. The amount of the side product is significantly affected by
the experimental conditions: N-acylurea formation is less likely at
lower temperatures and when solvents with low polarity are used
[30]. So, when we performed the reaction with PAZE using DCC/
DMAP at room temperature with chloroform as a solvent, good
modification degrees were achieved and the final copolymers (PA1-
PA6) were easy to purify (Scheme 2).
group in order to induce the formation of columnar mesophases.
3.1. Modification of model compound DMHEA and polymer PAZE
In order to choose the best reaction conditions to proceed with
the chemical modification of PAZE, first we tested several esterifi-
cation procedures with the commercially available model com-
pound DMHEA, which was structurally very similar to our polymer
(Scheme 1).
Thionyl chloride (SOCl2) is probably the most attractive reagent
for the chlorination of carboxylic acids. Thus, the chlorination of
3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoic acid with
SOCl2 in the presence of a catalytic amount of DMF was tested.
However, 1H NMR spectrum showed that HCl induced cleavage of
the highly reactive benzyl ether moieties also occurred. We also
tried to use oxalyl chloride for the chlorination, but the same
cleavage reaction was observed. We carried out the chlorination
also with SOCl2 in the presence of an excess of 1,8-diazabicyclo
[5.4.0]undec-7-ene (DBU), since Balagurusamy and coworkers
[23] observed no cleavage products in the presence of an excess of a
non-nucleophilic base scavenger. In our case, no cleavage products
were observed, but compound 4 was obtained in low yield (32%)
due to difficulties to eliminate excess of DBU.
Chemical modification conditions, the resulting degrees of
modification and yields are reported in Table 1.
The structure and composition of the copolymers were char-
acterized by NMR spectroscopy. Fig.1 reports the 1H NMR spectrum
of polyamine PA2 as an example.
Compound 4 can also be prepared directly from compound 1 by
transesterification reactions. As we observed the cleavage of benzyl
ether moieties in the presence of hydrogen chloride, we decided to
try only base catalyst (NaOCH3) for transesterification, different
All 1H NMR spectra are characterized by broad signals in three
regions. The aromatic region shows four partially overlapped sig-
nals at 7.3, 7.2, 6.8 and 6.7 ppm. Considering the relative integration

ꢀ
_
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
5137
Scheme 2. Modification reaction of PAZE with 3,4,5-tris[4-(n-dodecan-1-yloxy)benzyloxy]benzoic acid (2) using esterification method with DCC/DMAP.
areas and by comparison with the spectrum of methyl 3,4,5-tris[4-
(n-dodecan-1-yloxy)benzyloxy]benzoate, the signals at 7.3 and
7.2 ppm (8H) can be assigned to the protons of the benzoate group
plus the benzyl aromatic protons ortho to the eCH2Oe. The signals
at 6.8 and 6.7 ppm (4H þ 2H) correspond to the benzyl aromatic
protons meta to the eCH2Oe of the lateral and central alkylox-
ybenzyloxy substituents respectively. The characteristic signals,
corresponding to most protons of the long alkyl chains in the
dendron, can be observed in the high-field region at 1.7, 1.4, 1.2 and
0.8 ppm. The most interesting region lies between 5 and 2.5 ppm in
which five signals can be observed. The signal at 4.3 ppm corre-
sponds to the methylenic protons d in the modified unit; the signal
centered at 3.5 ppm corresponds to methylenic protons d’ in the
unmodified unit; the broad signal at 2.5 ppm corresponds to the
methylenic protons a, b, c, aʹ, bʹ and cʹ in the modified and un-
modified units. The signal at 3.9 ppm corresponds to the methylene
attached to the oxygen in the long alkyl chains. Finally, the signal at
4.9 ppm can be assigned to the benzylic methylenes of the dode-
cyloxybenzyloxy substituent, which is overlapped with OH group.
Fig. 2 shows the 13C NMR spectrum of the polyamine PA2 with
the corresponding assignments. The aromatic and carbonyl carbons
of the moiety introduced appear between 166 and 108 ppm,
whereas carbons 2e12 of the long alkyl chains appear in the ex-
pected region at 32e14 ppm. The carbon 1 of the alkyl chains ap-
pears at 68.1 ppm. The chemical shifts of the benzylic methylenes
depend on their relative position in the aromatic ring. Those in
position 3 and 5 appear at 71.2 ppm, whereas the same carbon in
position 4 appears downfield at 74.8 ppm. Similarly, aromatic sig-
nals IIʹ and IIIʹ are also split. The carbons of the main chain appear at
50e60 ppm: the signals a, aʹ, b, bʹ appear at 52.9 ppm; the signals c,
cʹ appear at 56.6 ppm. The methylenic carbons d and dʹ of the
modified and unmodified units appear resolved at 61.1 and
60.1 ppm, respectively.
The structures of copolymers were characterized and confirmed
by FTIR-ATR technique. In the FTIR-ATR spectrum, a decrease of the
eOH vibration band around 3360 cmꢁ1, coming from PAZE, was
observed; we also detected the appearance of carbonyl group
characteristic band around 1712 cmꢁ1 (C]O) which confirms that
polymer chemical modification occurred.
The modification degrees of copolymers were determined by
NMR spectroscopy, since this methodology gave accurate results.
Quantification was carried out from the 1H NMR spectra by
Fig. 1. 1H NMR spectrum in CDCl3 at room temperature of polyamine PA2.

ꢀ
_
5138
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
Fig. 2. 13C NMR spectrum in CDCl3 at room temperature of the polyamine PA2.
comparing the areas of the methylenic protons d in the modified
unit at 4.3 ppm and the methylenic protons d’ in the unmodified
unit at 3.5 ppm.
keep in mind that molecular weight values were determined by SEC
under the assumption that the copolymers behave like polystyrene
in THF. For this reason, it is not easy to predict a trend of the mo-
lecular weights with the modification degree, since the introduc-
tion of the dendritic groups can lead to significant changes in
copolymer conformation and greatly modify the hydrodynamic
volume of the PA1ePA6 systems with respect to the starting
polymer. Actually, the molecular weights of modified PA1ePA6
copolymers, which were obtained relatively to polystyrene in THF,
were found considerably lower than the molecular weight
of starting PAZE, determined by SEC-MALLS in N,N-
dimethylacetamide. This suggests that a considerable change in
copolymer conformation occurs as a consequence of the intro-
duction of the bulky dendrons.
Yields were calculated on the basis of the polymer recovery and
the expected mass, considering the degree of modification ach-
ieved. PAZE was reacted with different ratios of acid (2). The degree
of modification depended on the reaction conditions and ranged
from 16 to 100%. The reactions performed at room temperature
gave almost quantitative modification degrees and yields around
70%. On the other hand, the reaction performed at 60 ꢀC with
polymer/COOH molar ratio 1:1 gave 55% modification degree and
58% yield (copolymer PA4). As we mentioned before, the modifi-
cation degree decrease found in this case could be explained by the
formation of bigger amount of the side product N-acylurea, which
reduce the coupling yields at higher temperatures.
All values of molecular weight and density are reported in
Table 2.
3.2. Mesomorphic and thermal characterization of PA1ePA6
The molecular weight of the starting polymer PAZE was deter-
mined by SEC-MALLS in N,N-dimethylacetamide as reported [17].
One might expect an increasing trend of the molecular weight with
modification degree, since considerably heavy dendritic groups
were introduced. Actually, it was not the case; however, one should
The mesomorphic characterization was performed on the basis of
differential scanning calorimetry (DSC), optical microscopy between
crossed polars (POM), and X-ray diffraction experiments (XRD). In
most cases, the first calorimetric analysis showed a broad small
endotherm at low temperature and a second sharper and bigger one
centered at higher temperature. This suggested that these samples,
due to their quite regular structure and relatively low molecular
weight, tended to crystallize partially. Therefore, we annealed them
for 2 h at temperatures located slightly below the first peak
maximum and then we performed DSC analysis again. The first
endotherm could be therefore related to a melting, while the second
one was attributed to a clearing transition, as showed by POM. As a
matter of example, Fig. 3 shows the liquid crystalline texture of PA3
at 32 ꢀC and PA4 at 35 ꢀC, respectively, observed by POM on cooling
from the isotropic phase. These images show the typical dendritic
growth aggregates of a discotic columnar mesophase.
Table 2
Molecular weights and densities of PA1-PA6 copolymers determined by SEC in
tetrahydrofuran.
Polymer
Mn$10ꢁ3 (g molꢁ1
)
M
w$10ꢁ3 (g molꢁ1
)
r
(g cmꢁ3
)
PAZEa
PA1
PA2
PA3
PA4
PA5
PA6
3.80
1.61
1.74
1.83
1.41
1.79
1.54
7.70
1.76
1.88
2.10
1.44
2.02
1.57
e
1.09
1.10
1.14
1.02
1.12
1.02
a
Table 3 reports the melting and clearing temperatures, as well as
the clearing enthalpies of PA1ePA6 samples.
Molecular weight values determined by SEC-MALLS in N,N-dimethylacetamide
[16].

ꢀ
_
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
5139
Fig. 3. Optical micrographies between crossed polars on cooling from the isotropic phase of (a) PA3 at 32 ꢀC and (b) PA4 at 35 ꢀC.
In the case of PA4 and PA5, DSC analysis just put into evidence
one endotherm, which could be related to the clearing point. On
increasing the modification degree, clearing points and enthalpies
of the copolymers also increased, similarly to what was reported for
columnar copolyethers obtained by modification of poly(epi-
chlorohydrin) with the same Dendron [11]. As one could expect, the
clearing enthalpies were found lower than in the case of similar low
molecular weight dendrons [31]. On the other hand, clearing
enthalpy values resulted considerably higher than in the case of the
aforementioned copolyethers based on poly(epichlorohydrin)
modified with the same dendron. This can be explained by the
presence of the polyamine backbone which, due to its higher
symmetry with respect to poly(epichlorohydrin), allowed better
ordering of the dendritic side moieties.
XRD analysis was performed at different temperatures in order
to confirm calorimetric features and to identify the copolymers
mesophase. Prior to XRD analysis, copolymers PA1ePA3 and PA6
were annealed as previously described. Below Tm, XRD pattern of
these samples showed the presence of very small diffraction peaks,
which subsequently disappeared on heating above Tm. This
confirmed the presence of crystalline portions in the annealed
sample. These small diffraction peaks could be also foreseen in the
XRD pattern of PA4 and PA5 recorded in the range 0ꢀe20 ꢀC, thus
suggesting also in those cases the existence of crystalline domains,
whose melting could not be seen by DSC.
allowed to calculate the dimensions of the unit cell, while the latter
could be referred to the distance between dendrons [11,32]. Table 4
reports the features of the X-ray patterns of PA1ePA6 copolymers
below clearing temperature. The XRD patterns, recorded on sam-
ples oriented by shearing, showed that dendrons are perpendicular
to the column axis in the whole copolymer series PA1ePA6.
For a hexagonal columnar mesophase, and given the experi-
mental densities
copolymer that are present in a hexagonal prism layer of height c
from the following equation:
r, one can calculate the number of repeat units of
m
2 M
m
3NAa2c
r
¼ pffiffiffi
(1)
where M is the average molecular weight of the repeat unit, NA is
Avogadro’s number, a ¼ 2d100/O3 is the dimension of the hexagonal
unit cell, and c is the prism height (distance between dendrons). By
considering the experimental modification degree of the copol-
ymer
a, we can finally find the number of dendrons contained in a
unit cell,
d
¼ ( )/100. We have not demonstrated a hexagonal
m
ꢂ
a
columnar organization for PA1ePA6 copolymers. In spite of this,
the same calculation can be applied to our samples in order to give
a rough estimation of the number of dendrons, because geometrical
considerations allow the assumption that in a columnar mesophase
(such as shown by PA1ePA6) the columns self-assemble in a
compact hexagonal packing where statistical fluctuations in the
column positions do not produce any of the additional reflections
that are expected in a Colh phase: that is, the instantaneous posi-
tions of the columns fit a hexagonal organization even if the
average positions do not [10].
In all cases, XRD analysis of the copolymers below Tc showed a
sharp reflection, located around 2
q
¼ 2.3ꢀ, and a broad halo at 2
q
about 19.5ꢀ. This diffractogram is compatible with a columnar
mesophase, the lower spacing corresponding to the average dis-
tance between dendrons, which is related to adjacent benzene
rings and to the low ordered alkyl groups; the higher spacing cor-
ꢁ
In all cases the dimension of the unit cell ranged around 45 A,
responds to the lateral distance between columns. The former 2
value corresponded to the d100 plane of a columnar phase and
q
that is, was slightly narrower than in the case of the copolyethers
bearing the same dendron [14]. This is probably due to the shorter
spacer linking the backbone and dendron on copolyethers de-
rivatives, which forced the backbone to adopt a broader helical
Table 3
Calorimetric features of PA1-PA6 copolymers.
Copolymer
Modification
degree (%)
Tm (ꢀC)a
Tc (ꢀC) a
D
Hc
Mesophase
a
Table 4
(kJ/mol)d
X-ray diffraction patterns of PA1ePA6 copolymers below clearing temperature.
PA1
PA2
PA3
PA4
PA5
PA6
16
39
46
55
67
8
3
8
30e50c
37
39
39
43
5.6
5.7
11.8
24.1
24.9
34.9
Col
Col
Col
Col
Col
Col
b
ꢁ
ꢁ
Copolymer
Spacingsa (A)
a (A)
mc
dd
PA1
PA2
PA3
PA4
PA5
PA6
4.5 (b), 40 (s)
4.5 (b), 38 (s)
4.5 (b), 38 (s)
4.5 (b), 41 (s)
4.5 (b), 40 (s)
4.5 (b), 39 (s)
46
44
44
47
46
45
21.9
10.5
9.6
8.5
7.5
3.5
4.1
4.4
4.7
5.0
4.6
n.d.b
n.d.b
55
100
65
a
For PA1ePA3, melting temperature, clearing temperature and enthalpy deter-
mined after 2 h annealing in the range ꢁ10ꢀ to ꢁ5 ꢀC; for PA6 after 2 h annealing at
45 ꢀC.
4.6
a
b ¼ broad halo; s ¼ sharp reflection.
b
b
c
Not determined.
Broad peak.
Expressed per mol of repeat unit.
Dimension of the hexagonal unit cell.
c
Number of repeat units contained in a unit cell.
Number of dendrons contained in a unit cell.
d
d

ꢀ
_
A. Sakalyte et al. / Polymer 54 (2013) 5133e5140
5140
case of the copolyethers bearing the same dendron, probably due to
the longer spacer present in PAZE, which allows better accommo-
dating of the side tapered group. This first example of LC columnar
polyamine can represent a potential candidate for preparing
proton-transport membranes by the channel mechanism, due to
the presence of basic nitrogen atoms in the inner helical backbone
structure.
Acknowledgments
Financial support from MAT2008-00456/MAT (Ministerio de
Ciencia e Innovación) is gratefully acknowledged. The authors are
also grateful to Dr. Francesc Gispert for XRD experiments.
References
[1] McNally BA, Leevy WM, Smith BD. Supramolecular Chemistry 2007;19(1e2):
29e37.
[2] Ye Y-S, Huang Y-J, Cheng C-C, Chang F-C. Chemical Communications
2010;46(40):7554e6.
[3] Beginn U, Zipp G, Mourran A, Walther P, Möller M. Advanced Materials
2000;12(7):513e6.
[4] Fitie CFC, Mendes E, Hempenius MA, Sijbesma RP. Macromolecules
2011;44(4):757e66.
Fig. 4. TGA thermograms of selected copolymers recorded at
a heating rate of
10 ꢀC/min.
[5] Percec V, Imam MR, Peterca M, Leowanawat P. Journal of the American
Chemical Society 2012;134(9):4408e20.
conformation to successfully accommodate the dendrons. It must
be noted that in PECH derivatives, dendrons are not perpendicular
to the column axis, but tilted; this can also be explained on the basis
of its shorter spacer. In the case of PA1ePA6, the number of den-
drons contained in a unit cell is roughly independent (3e5) on the
degree of modification, and consequently the number of repeat
[6] Percec V, Peterca M, Tadjiev T, Zeng X, Ungar G, Leowanawat P, et al. Journal of
the American Chemical Society 2011;133(31):12197e219.
[7] Percec V, Rinaldi PL, Auman BC. Polymer Bulletin 1983;10(9):397e403.
[8] Percec V, Ahn CH, Ungar G, Yeardley DJP, Moller M, Sheiko SS. Nature
1998;391(6663):161e4.
[9] Percec V, Holerca MN, Uchida S, Yeardley DJP, Ungar G. Biomacromolecules
2001;2(3):729e40.
units
m decreases as the degree of modification increases.
[10] Ronda JC, Reina JA, Cádiz V, Giamberini M, Nicolais L. Journal of Polymer
Science Part A: Polymer Chemistry 2003;41(19):2918e29.
[11] Ronda JC, Reina JA, Giamberini M. Journal of Polymer Science Part A: Polymer
Chemistry 2004;42(2):326e40.
[12] Pérez M, Reina JA, Serra A, Ronda JC. Acta Polymerica 1998;49(6):312e8.
[13] Pérez M, Reina JA, Serra A, Ronda JC. Polymer 2000;41(20):7331e7.
[14] Bhosale SV, Rasool MA, Reina JA, Giamberini M. Polymer Engineering & Sci-
ence 2013;53(1):159e67.
The thermal stability of the polymers was studied using TGA. As
an example, TGA scans of selected samples are reported in Fig. 4.
The onset of thermal decomposition (determined as the tem-
perature corresponding to 5% mass loss) ranged between 270 and
300 ꢀC. The onset of thermal decomposition increased on
increasing the copolymers modification degree, probably due to
higher amount of mesogenic groups introduced in the structure.
Copolymer PA6 behaved the best in this respect, exhibiting a mass
loss onset temperature of 293 ꢀC. As far as char yield is concerned,
in the case of starting PAZE it was only 1.4% at 600 ꢀC [17], while all
the copolymers showed more than 10% char yield at 800 ꢀC. Again,
PA6 behaved the best in this respect, exhibiting a char yield of 11.8%.
[15] Tylkowski B, Castelao N, Giamberini M, Garcia-Valls R, Reina JA, Gumí T.
Materials Science and Engineering: C 2012;32(2):105e11.
[16] Dulyea LM, Fyles TM, Robertson GD. Journal of Membrane Science 1987;34:
97e108.
ꢀ
_
[17] Sakalyte A, Reina JA, Giamberini M, Lederer A. Polymer Engineering and Sci-
ence 2013, in press.
[18] Xie M, Zhang C. Liquid Crystals 2007;34(11):1275e83.
́
[19] Furniss BS, Hannaford AJ, Smith PWG, Tatchell AR. Vogel’s textbook of prac-
tical organic chemistry. 5th ed. London: Pearson Education; 1989.
[20] Percec V, Ahn CH, Cho WD, Jamieson AM, Kim J, Leman T, et al. Journal of the
American Chemical Society 1998;120(34):8619e31.
[21] Sarkouhi M, Hassan J, Shamsipur M. Applied Magnetic Resonance 2012;43(3):
377e84.
4. Conclusions
A family of copolyamines bearing the tapered 3,4,5-tris[4-(n-
dodecan-1-yloxy)benzyloxy]benzoate group was synthesized by
chemical modification of poly[2-(aziridin-1-yl)ethanol] (PAZE). The
chemical modification was tackled by different procedures on the
model compound N,N-dimethyl-2-hydroxyethylamine, and we
finally found that the esterification method with N,N0-dicyclohex-
ylcarbodiimide at room temperature, in the presence of 4-dime-
thylaminopyridine, gave the highest yield. In the case of PAZE, the
reactions performed at room temperature gave almost quantitative
modification degrees and yields around 70%. The resulting co-
polymers showed higher char yield than starting PAZE. Moreover,
all of them exhibited small crystalline portions after annealing, as
well as columnar mesophases, as inferred by DSC, XRD and POM.
The dimension of the unit cell resulted slightly narrower than in the
[22] Morcillo SP, Alvarez de Cienfuegos L, Mota AJ, Justicia J, Robles R. The Journal
of Organic Chemistry 2011;76(7):2277e81.
[23] Balagurusamy VSK, Ungar G, Percec V, Johansson G. J. Am. Chem. Soc.
1997;119:1539.
[24] Garegg PJ, Samuelsson B. Journal of the Chemical Society, Chemical Commu-
nications 1979;(22):978e80.
[25] Stacey M, Bourne EJ. Nature 1949;164(4173):705.
[26] Otera J, Nishikido J. Esterification. Wiley; 2009.
[27] Holmberg K, Hansen B. Acta Chemica Scandinavica B 1979;33:410e2.
[28] Neises B, Steglich W. Angewandte Chemie International Edition in English
1978;17(7):522e4.
[29] Hassner A, Alexanian V. Tetrahedron Letters 1978;19(46):4475e8.
[30] Goodman M, Felix A, Moroder L, Toniolo C. Houben-Weyl methods in organic
chemistry. 4th ed. Thieme Publishers; 2003.
[31] Percec V, Cho W-D, Ungar G. Journal of the American Chemical Society
2000;122(42):10273e81.
[32] Percec V, Johansson G, Heck J, Ungar G, Batty SV. Journal of the Chemical
Society Perkin Transactions 1 1993;1:1411e9.
́