Macrocycles. 19. Cyclization in the nematic phase? Polyesters derived from hydroquinone 4-hydroxybenzoate and aliphatic dicarboxylic acids

Article abstract of DOI:10.1021/ma020170r

Silylated hydroquinone 4-hydroxybenzoate (HHB) was polycondensed with aliphatic dicarboxylic acid dichlorides (ADADs) in bulk at temperature up to 240 °C. These polycondensations proceeded in the nematic phase, and both transesterification and backbiting degradation were avoided as evidenced by ¹H NMR spectroscopy. The nematic LC phases were characterized by optical microscopy and DSC measurements. The MALDI-TOF mass spectra revealed the formation of even-membered cycles (in addition to linear oligomers), whereas odd-membered cycles were almost absent. When the reaction mixture was slightly diluted with 1-chloronaphthalene, so that the polycondensation proceeded in the isotropic phase, odd-membered cycles with a predominance of the cyclic trimer were found. When free HHB was polycondensed with ADADs in 1-chloronaphthalene at 220 °C in 1,2-dichlorobenzene at 190 °C or in the presence of pyridine at 5-10 °C, again odd-membered cycles were found. Obviously, the nematic phase enforces a parallel alignment of polyester chains with strong electronic interaction of neighboring mesogens, so that cyclization can only occur in the case of even-membered chains via hairpin conformations of the aliphatic spacers.

Full text of DOI:10.1021/ma020170r

Macromolecules 2002, 35, 5449-5453
5449
Macrocycles. 19. Cyclization in the Nematic Phase? Polyesters Derived
from Hydroquinone 4-Hydroxybenzoate and Aliphatic Dicarboxylic
Acids†
Ha n s R. Kr ich eld or f,* Ma r tin Rich ter , a n d Ger t Sch w a r z
Institut fu¨r Technische und Makromolekulare Chemie, Bundesstrrasse 45,
D-20146 Hamburg, Germany
Received February 1, 2002; Revised Manuscript Received April 24, 2002
ABSTRACT: Silylated hydroquinone 4-hydroxybenzoate (HHB) was polycondensed with aliphatic
dicarboxylic acid dichlorides (ADADs) in bulk at temperature up to 240 °C. These polycondensations
proceeded in the nematic phase, and both transesterification and “backbiting” degradation were avoided
1
as evidenced by H NMR spectroscopy. The nematic LC phases were characterized by optical microscopy
and DSC measurements. The MALDI-TOF mass spectra revealed the formation of even-membered cycles
(in addition to linear oligomers), whereas odd-membered cycles were almost absent. When the reaction
mixture was slightly diluted with 1-chloronaphthalene, so that the polycondensation proceeded in the
isotropic phase, odd-membered cycles with a predominance of the cyclic trimer were found. When free
HHB was polycondensed with ADADs in 1-chloronaphthalene at 220 °C in 1,2-dichlorobenzene at 190 °C
or in the presence of pyridine at 5-10 °C, again odd-membered cycles were found. Obviously, the nematic
phase enforces a parallel alignment of polyester chains with strong electronic interaction of neighboring
mesogens, so that cyclization can only occur in the case of even-membered chains via hairpin conformations
of the aliphatic spacers.
In tr od u ction
Numerous so-called LC-main chain polymers forming
a nematic melt have been synthesized and studied over
the past three decades.1-3 More than 95% of these
nematic LC-polymers contain ester bonds, because the
ester linkage introduces flexibility and favors a linear
conformation of the polymer backbones. The nematic
LC-polyesters may be subdivided in two groups: first,
fully aromatic polyesters, and second so-called semirigid
polyesters containing flexible, aliphatic spacers. The
fully aromatic LC-polyesters are typically random co-
polyesters such as the commercial products Vectra and
“Sumicosuper” (Xydar). In the case of semirigid LC-
polyesters, random copolymers, alternating copolymers,
and homopolymers were synthesized. These structural
differences are important, when a characterization by
mass spectrometry is considered, because random co-
polymers are particularly unfavorable for MALDI-TOF
mass spectrometry (complex peak pattern and low peak
intensities).
The purpose of the present work was to find out if
semirigid nematic polyesters can form cyclic oligo- and
polyesters in the nematic phase, and such a study
requires a MALDI-TOF mass spectrometric character-
ization of the reaction products. The scientific back-
ground of this study has two aspects. First, we have
recently demonstrated4-7 for various kinetically con-
trolled syntheses of isotropic polyesters and other poly-
condensates that cyclization competes with propagation
at any concentration and any conversion as expressed
in eq 1. In contrast to the classical theory of Carothers8
and Flory,9 cyclization is an inherent property of all
kinetically controlled polycondensations as predicted by
Stepto et al.10 and Gordon et al.11 Second, it is well-
known from theoretical calculations and experimental
studies that polymer chains and chain segments have
a high tendency to adopt a nearly parallel alignment
inside a domain of a nematic phase. Upon shearing
(stirring, extrusion) a parallelization of most domains
will occur and many domains may even merge into a
large (mono-) domain. A characteristic consequence of
this particular supermolecular structure is a sharp drop
of the melt viscosity upon shearing (e.g., upon stirring
or extrusion), the so-called “shear thinning”. From these
properties of the nematic melt and from the highly
anisotropic mechanical properties of LC-polymers, it has
been concluded that entanglements are absent or at
least extremely rare. In this connection, it was of
particular interest to find out if the nematic phase
allows cyclization in a kinetically controlled polycon-
densation process.
The polyesters suited for such a study should meet
the following requirements. First, they should be ho-
mopolymers to simplify the mass spectroscopy. Second,
a broad nematic phase should exist below 240 °C,
because at higher temperatures transesterification and
formation of cycles by backbiting are unavoidable. Third,
these LC-polyesters should be soluble in volatile sol-
vents to ease the preparation of irradiation targets for
the MALDI-TOF mass spectroscopy. Only very few LC-
polymers meet these requirements, and the polyesters
of structure 1 were selected for the present study. An
additional advantage of these polyesters is the regular
sequence of three different monomer units, so that
transesterification resulting in a scrambling of their
† Dedicated to Prof. R. W. Lenz on the occasion of his 75th
birthday.
10.1021/ma020170r CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/05/2002

5450 Kricheldorf et al.
Macromolecules, Vol. 35, No. 14, 2002
Ta ble 1. Yield s a n d P r op er ties of Nem a tic P olyester s P r ep a r ed by P olycon d en sa tion of Bistr im eth ylsilyl Hyd r oqu in on e
4-Hyd r oxyben zoa te w ith Va r iou s Alip h a tic Dica r boxylic Acid Dich lor id es in Bu lk a t 235-240 °C
lit. data
Tme (°C) Tmf (°C) formula (fw)
elemental analyses
a
yield
ηinh
elemental
dichlorides of
(%) (dL/g) Tmc (°C) Tic (°C) Tid (°C)
C
H
hexane-1,6-dicarboxylic acid
85.0
92.0
0.52
0.67
0.90
253
227
209
319
264
240
338-342 150-170
287-291 140-170
246-250 145-170
211
206
180
C
21H20O6
(368.38)
23H24O6
(396.43)
25H28O6
(424.49)
calcd
found 68.41 5.48
calcd 69.69 6.10
found 69.24 6.10
calcd 70.74 6.65
found 70.21 6.66
68.47 5.47
octane-1,8-dicarboxylic acid
C
decane-1,10-dicarboxylic acid 97.0
C
a
b
Measured at 20 °C with c ) 2 g/L in CH2Cl2/TFA (volume ratio 4:1). From 100.4 MHz 13C NMR spectra recorded in CDCl3/TFA
(volume ratio 4:1). c From DSC measurements with a heating rate of 20 °C/min. Isotropization temperature as observed by optical
d
microscopy. e From ref 13. f From ref 14.
Ta ble 2. P olycon d en sa tion of Bistr im eth ylsilyl
Hyd r oqu in on e 4-Hyd r oxyben zoa te w ith
Deca n e-1,10-d ica r bon yl Ch lor id e
molar
excess
of the
no. dichloride (°C)
b
expt
temp time yield
ηinh
(h)
(%) DPa (dL/g)
comment
sequence should be detectable by 1H NMR and 13C NMR
spectroscopy. These polyesters were previously synthe-
sized and characterized by two different research
groups.12-14 Broad nematic phases were reported and
melting temperatures below 212 °C.
1Ac
1Bc
1Cc
1Dc
1Ec
2
3
4
5
0
0
0
0
0
0
1
2
3
160c 0.5
180c 1.0
200c 1.5
220c 2.0
235c 3.0
6
0.39 LC, no cycles
12 0.43 LC, no cycles
60 0.70 LC, no cycles
80 0.75 LC + cycles
120 0.82 LC + cycles
135 0.86 LC + cycles
235
235
235
235
4.0
4.0
4.0
4.0
96
97
99
88
-
-
-
0.95 LC + cycles
1.20 LC + cycles
0.93 LC + cycles
Exp er im en ta l Section
Mon om er s. Hydroquinone 4-hydroxybenzoate (HHB) was
a gift of Bayer AG (Leverkusen, Germany). It had a gray color.
The white product obtained after washing with methanol (and
drying over P4O10) was used for all experiments (mp 241-243
°C; in ref 15, mp 242-245 °C). The dicarboxylic acid chlorides
(ADADs) were purchased from Aldrich Co. (Milwaukee, WI)
and used after distillation in vacuo. Pyridine was distilled over
freshly powdered CaH2. CH2Cl2 and toluene were distilled over
P4O10. 1,2-Dichlorobenzene was distilled over P4O10 in vacuo.
1-Chloronaphthalene was dried azeotropically and distilled in
vacuo. All these solvents were purchased from Aldrich Co.
Chlorotrimethylsilane and triethylamine were also gifts of
Bayer AG and used as received.
Silyla tion of HHB. HHB (0.5 mol) and chlorotrimethylsi-
lane (1.05 mol) were dissolved in dry toluene (1.2 L) and
triethylamine (1.05 mol) was added dropwise with heating and
stirring. Afterward, the reaction mixture was refluxed for 4
h, cooled with ice and filtered under exclusion of moisture. The
filtrate was concentrated in vacuo and the product was
distilled in portions of 50 g over a short-path apparatus in a
vacuum of 10-3 mbar. Yield: 83%. Mp: 173-175 °C.
Average degree of polymerization as determined by 1H NMR
a
b
spectroscopy. Measured at 20 °C with c ) 2 g/L in CH2Cl2/TFA
(volume ratio 4:1). c Samples were taken from a larger batch.
Ta ble 3. P olycon d en sa tion s of Deca n e-1,10-d ica r bon yl
Ch lor id e w ith Silyla ted HHBa or fr ee HHB in Solu tion
molar
excess
of the
b
expt
no.
temp time yield ηinh
reaction medium dichloride (°C)
(h) (%) (dL/g)
1
2
3
4
5
6
7
8
9
1-chloronaphthaline
1-chloronaphthaline
1-chloronaphthaline
1-chloronaphthaline
1,2-dichlorobenzene
1,2-dichlorobenzene
CH2Cl2 + pyridine
CH2Cl2 + pyridine
CH2Cl2 + pyridine
0
0
1
2
0
1
0
1
2
240
220
220
220
180
180
5
5
5
79
82
89
0.82
0.61
0.48
0.57
0.62
0.36
0.33
0.28
24 84
24 74
0-5/20 1/20 80
0-5/20 1/20 72
0-5/20 1/20 77
a
Silalyted HHB was used in experiment no. 1 and free HHB
in all other experiments. Measured at 20 °C with c ) 2 g/L in
Anal. Calcd for C19H26O4Si2 (374.6): C, 60.92; H, 7.01.
Found: C, 60.65; H, 6.87%.
b
CH2Cl2/TFA (volume ratio 4:1).
P olycon d en sa tion s. (A) Silyl Meth od in Bu lk (Ta ble
1). Silylated HHB (20 mmol), an ADAD (20 mmol), and
benzyltriethylammonium (TEBA) chloride (10 mg) were weighed
into a cylindrical glass reactor equipped with a mechanical
stirrer and gas-inlet, and gas-outlet tubes. The reaction vessel
was placed into an oil bath preheated to 120 °C and the
temperature was rapidly raised to 160 °C where the polycon-
densation started. The temperature was then raised in steps
of 20 °C to 240 °C according to the progress of the polycon-
densation (monitored by the evolution of ClSiMe3 and increas-
ing viscosity of the melt). In the case of butane-1,4-dicarbonyl
and hexane-1,6-dicarbonyl, solidification was unavoidable.
The experiments yielding samples nos. 1-5 of Table 2 were
conducted with 50 mmol quantities of each monomer. The
reaction vessel was opened after the mentioned time intervals,
and 0.5 g samples were removed for characterization.
(B) Silyl Meth od in 1-Ch lor on a p h th a len e (Ta ble 3).
Silylated HHB (15 mmol), decane-1,10-dicarbonyl chloride (15
mmol), TEBA chloride (10 mg), and dry 1-chloronaphthalene
(5 g) were weighed into a cylindrical glass reactor and
polycondensed as described above.
(C) Meth od w ith F r ee HHB in 1-Ch lor on a p h th a len e
(Ta ble 3). HHB (15 mmol) and decane-1,10-dicarbonyl chlo-
ride were dissolved in dry 1-chloronaphthalene (50 mL) and
heated to 220 °C for 5 h with stirring. The eliminated HCl
was removed with a slow stream of nitrogen. The cold reaction
mixture was poured into dry diethyl ether, and the precipitated
polyester was twice washed with dry diethyl ether and dried
at 80 °C in vacuo.
Analogous polycondensations were conducted in refluxing
1,2-dichlorobenzene.
(D) P yr id in e Meth od . HHB (20 mmol) and decane-1,10-
dicarbonyl chloride (20 mmol) were dissolved (or partially
suspended) in dry dichloromethane (150 mL). After the reac-
tion was cooled to approximately 1-2 °C, dry pyridine (80
mmol) was added dropwise with stirring. The reaction mixture
was then stirred at room temperature for 4 d. The reaction
mixture was concentrated to a volume around 100 mL and
precipitated into diethyl ether. The isolated polyester was dried
at 60 °C in vacuo whereby a remainder of pyridine hydrochlo-
ride sublimed.

Macromolecules, Vol. 35, No. 14, 2002
Cyclization in the Nematic Phase? 5451
F igu r e 2. 400 MHz 1H NMR spectrum of the aromatic protons
of polyester 1c.
F igu r e 1. DSC measurements (1st heating, heating rate 20
°C/min): (A) polyester 1c; (B) 1b; (C) 1a .
Mea su r em en ts. The inherent viscosities were measured
with an automated Ubbelohde viscometer thermostated at 20
°C. The DSC measurements were conducted with a PERKIN
ELMER DSC-7 in aluminum pans under nitrogen at a heating
rate of 20 °C/min.
ular weights (indicated by low viscosity data) may be
responsible for part of this difference. The molecular
weights reported by Krigbaum’s group14 were as high
as those obtained in this work. However, both groups
prepared their polyesters by transesterification of the
diphenol bisacetate at temperatures up to 285 °C (eq
3). It is absolutely unlikely that a perfectly alternating
The 400 MHz 1H NMR spectra and the 100.4 MHz 13C NMR
spectra were obtained on a Bruker “Advance” FT spectrometer
in 5 mm o.d. sample tubes. A 4:1 (by volume) mixture of CDCl3
and trifluoroacetic acid (TFA) containing TMS was used as
solvent and shift reference.
The MALDI-TOF mass spectra were recorded with a
Bruker Biflex III mass spectrometer equipped with a nitrogen
laser (λ ) 337 nm) in the reflectron mode. An acceleration
voltage of 20 kV and a cutoff range of 800 Da were used for
all measurements. The irradiation targets were prepared from
hexafluoro-2-propanol solution with dithranol as matrix and
K-trifluoroacetate as dopant.
Resu lts a n d Discu ssion
P olycon d en sa tion in Bu lk . To study the possibility
of kinetically controlled cyclizations in the nematic
phase, polycondensations free of transesterifications had
to be conducted in bulk. The only method meeting these
requirements was the polycondensation of silylated
HHB with ADADs catalyzed by chloride ions (eq 2).
sequence was obtained under these conditions, and
neither research group checked the structure of their
polyesters. In a more or less randomized sequence,
building blocks such as 2, 3, and 4 will be present. The
building blocks 2 and 4 maintain the liquid crystalline
character of these copolyesters, but the (partial) ran-
domization of the sequence will lower the Tms and
broaden the temperature range of the isotropization (as
reported13,14).
From previous studies it was known16,17 that this
approach avoids transesterifications up to reaction
temperatures around 240 °C. The ADADs were used in
an excess of 1 mol % to compensate for the loss by
distillation. In this way the three polyesters 1a -1c were
prepared, having the properties listed in Table 1.
The DSC revealed as expected two endotherms in the
heating curve (Figure 1) and the corresponding two
exotherms in the cooling trace. The optical microscopy
revealed the characteristic mobile Schlieren texture of
a nematic phase in the temperature range defined by
both endotherms. As typical for most semirigid main
chain LC-polymers increasing length of the spacer
reduces both melting temperature (Tm) and isotropiza-
tion temperature (Ti) with a narrowing of the LC phase.
The phase transitions found in this work were consider-
ably higher than those reported by other authors (Table
1). In the case of the Strzeleckis work13 the low molec-
The synthetic approach used in this work was known
to avoid transesterification below 250 °C. None the less,
the structure of the isolated polyesters 1b and 1c was
also checked by 1H and 13C NMR spectroscopy. As
1
demonstrated in Figure 2 the H NMR spectra of the
aromatic protons revealed two doublets originating from
the hydroquinone units. This signal pattern is charac-
teristic for the alternating sequences of 1b and 1c where
all hydroquinone units possess two different neighbors.
In the case of symmetrical substitution (2 and 3) sharp
singlet signals should appear. This assumption was
proven by synthesis and 1H NMR spectroscopy of
hydroquinone bispropionate and hydroquinone bisben-
zoate. The hydroquinone signal of the bisbenzoate
appeared downfield of the propionate signal, and on the

5452 Kricheldorf et al.
Macromolecules, Vol. 35, No. 14, 2002
basis of the observation the doublets “c” and “d” were
1
assigned in Figure 2. Therefore, the H NMR spectra
of the polyesters 1b and 1c suggest that no significant
transesterification has taken place. This suggestion was
confirmed by the 13C NMR which displayed the three
CO signals expected for the alternating structure of 1a -
1c (176.7, 175.7, 167.0 ppm recorded in CDCl3/TFA
volume ratio 4:1) and only one signal for all four CH
carbons of hydroquinone (130.0 ppm).
The unexpectedly high phase transitions (above all
the high Tms) prevented exhaustive polycondensations
with high conversions in the nematic phase for the
polyesters 1a , because the maximum reaction temper-
ature was limited to 240 °C, to avoid transesterification.
Therefore, all further syntheses were focused on poly-
ester 1c. Two questions should be answered. First, at
F igu r e 3. MALDI-TOF mass spectrum of polyester 1b
prepared by the “silyl method” in bulk.
which temperature and at which degree of polymeriza-
tion (DP) does the LC phase appear, and does cycliza-
tion really occur in the nematic phase? Second, how does
the molecular weight respond to variation of the feed
ratio of both monomers? Experiments nos. 1A-1E of
Table 2 were conducted in such a way that samples of
approximately 0.5 g were taken from the reaction
mixture after certain temperature and time intervals.
These samples were characterized by inherent viscosi-
ties and by DP values which were determined from the
1H NMR signals of the Me3SiO end groups (assuming
on the average one end group per chain). Optical
microscopy revealed the typical mobile “Schlieren tex-
ture” of a nematic phase even for the sample obtained
at the lowest reaction temperature (no. 1, Table 1). This
means that the formation of low oligomers sufficed to
stabilize a nematic phase. The characterization by
MALDI-TOF mass spectroscopy revealed the presence
of cycles only for samples polymerized at 220 or 235 °C.
This observation is in perfect agreement with our
concept of kinetically controlled polycondensations,
where the cycles slowly pile up at low and moderate
conversions (<95%) but dramatically increase at very
high conversions. In contrast, the “backbiting degrada-
tion” of thermodynamically controlled polycondensations
favors a rapid formation of cycles even at low to
moderate conversions due to the large number of end
groups.
F igu r e 4. MALDI-TOF mass spectrum of the polyester 1c
prepared by the “silyl method” in bulk (no. 1, Table 2).
confirms the absence of backbiting, because this process
should also yield odd-membered cycles. Second, the
parallel chain packing in the nematic phase obviously
favors a conformation, where the mesogens are in close
contact (via dipole-dipole and dipole-π interactions) and
the kinetically controlled cyclization proceeds via a
hairpin conformation of the aliphatic space (Scheme 1).
Such a conformation is clearly unfavorable for the
formation of odd-membered cycles.
P olycon d en sa tion s in Solu tion . To check the above
formulated hypothesis of cyclization in the nematic
phase (Scheme 1) several polycondensations yielding
polyester 1c were conducted in the isotropic phase. In
the first experiments of this series (no. 1, Table 3), the
mixture of the monomers was diluted with a small
volume of 1-chloronaphthalene and the polycondensa-
tion was again performed at 240 °C (method B in the
Experimental Section). The reaction mixture was trans-
parent at any time indicating its isotropic character. The
MALDI-TOF mass spectrum of the resulting polyester
proved the formation of odd-membered cycles in quanti-
ties similar to those of the even-membered ones (Figure
5). Furthermore, free HHB was polycondensed with
decane 1,10-dicarbonyl chloride in 1-chloronaphthalene
at 220 °C with elimination of HCl (method C in the
Experimental Section, nos. 2-4, Table 3). Analogous
experiments were conducted in refluxing 1,2-dichlo-
robenzene (nos. 5 and 6, Table 3). All the polyester
samples isolated from these five experiments have two
characteristic properties in common. The 1H NMR
For experiments nos. 2-5, heating from 160 to 240
°C was accelerated (3 h instead of 5 h), but the final
temperature was maintained for 1 h more. Variation of
the feed ratio revealed an optimum of the molecular
weight for an 2 mol % excess of the dichloride. Obvi-
ously, selective distillation of the more volatile dichloride
causes a slight imbalance of the stoichiometry, when a
1:1 feed ratio is used. A higher molecular weight means
a higher and more perfect conversion of functional
groups, and thus, on the basis of previous results4-7
higher fraction of cycles was expected.
a
The MALDI-TOF mass spectra revealed the presence
of cyclic oligoesters in all samples of the polyesters 1a -
1c listed in Tables 1 and 2. Particularly interesting is
the finding that almost exclusively even-membered
cycles were formed as illustrated in Figures 3 and 4.
This finding allows two important conclusions. First, it

Macromolecules, Vol. 35, No. 14, 2002
Cyclization in the Nematic Phase? 5453
Sch em e 1
of these polyesters has the advantage that the absence
of transesterification and “backbiting” can be checked
by NMR spectroscopy. The MALDI-TOF mass spectra
revealed the formation of even-membered cyclic oli-
goesters in the nematic phase. The formation of odd-
membered cycles was found, regardless of the synthetic
method, whenever the polycondensation was conducted
in an isotropic phase. These results allow the conclusion
that kinetically controlled cyclizations may occur in the
nematic phase based on a special type of supermolecular
order involving parallelization and electronic contacts
of the mesogens along with hairpin conformations of the
aliphatic spacers. Further studies of this phenomenon
are currently in progress.
Refer en ces a n d Notes
(1) Ciferri, A.; Krigbaum W. R.; Meyer R. B. In Polymer Liquid
Crystals; Ciferri, A., Krigbaum W. R., Meyer R. B., Eds.;
Academic Press: New York, 1982.
(2) Weis R. A.; Ober C. K. In ACS Symposium Series 435; Weis
R. A., Ober C. K., Eds.; American Chemical Society: Wash-
ington, DC, 1990.
(3) Kricheldorf H. R. Advances in Polymer Science; Springer
Publishers: Berlin, Heidelberg, Germany, and New York,
1999; Vol. 141.
(4) Kricheldorf H. R.; Lorene, A.; Spickermann, J .; Markos, M.
J . Polym. Sci., Part A, Polym. Chem. 1999, 37, 3861.
(5) Kricheldorf H. R.; Rabenstein, M.; Markos, M.; Schmidt, M.
Macromolecules 2001, 34, 713.
F igu r e 5. MALDI-TOF mass spectrum of the polyester 1c
prepared by the “silyl method” in concentrated 1-chloronaph-
thalene (no. 1, Table 3).
spectra proved the absence of transesterification and the
mass spectra demonstrated the presence of odd-mem-
bered cycles (quite analogous to Figure 5).
Finally, three polycondensations of free HHB were
conducted in CH2Cl2 at 5-10 °C. Pyridine served as HCl
acceptor and the feed ratio of both monomers was
slightly varied (nos. 7-9, Table 3). Regardless of the feed
ratio, only rather low molecular weights were obtained.
None the less, the mass spectra revealed the formation
of odd-membered cycles and even-membered cycles. In
summary, three different polycondensation methods
were studied proceeding in an isotropic phase and all
these “isotropic polycondensations” yielded odd-mem-
bered cycles in quantities comparable to those of the
even-membered cycles. Therefore, these experiments
clearly confirm that the exclusive formation of even-
membered cycles in the nematic melt is the result of a
chain packing and conformation that is characteristic
for this LC phase.
(6) Kricheldorf H. R.; Bo¨hme, S.; Schwarz, G. Macromolecules
2001, 34, 8879.
(7) Kricheldorf H. R.; Bo¨hme, S.; Schwarz, G.; Kru¨ger, H.-P.;
Schulz, G. Macromolecules 2001, 34, 8886.
(8) Collected Papers of W. H. Carothers on Polymerization; Mark,
H., Whitby, G. S., Eds.; Interscience: New York, 1940.
(9) Flory, P. J . Principles of Polymer Science; Cornell University
Press: Ithaca, NY 195X; Chapter VIII.
(10) Stanford, J . L.; Stepto, R. F. T.; Waynell, D. R. J . Chem. Soc.,
Faraday Trans. 1975, 1, 1308.
(11) Gordon, M.; Temple, W. B. Macromol. Chem. 1972, 152, 277.
(12) van Layen, D.; Strzelecki, L. Eur. Polym. J . 1980, 16, 303.
(13) Strzelecki, L.; Liebert, L. Eur. Polym. J . 1981, 17, 1271.
(14) Krigbaum, W. R.; Kottek, R.; Ishikawa, T.; Hakeni, H. Eur.
Polym. J . 1984, 20, 225.
(15) Carfagna, C.; Acierno, D.; Di Palma, V.; Amendola, E.;
Giamberini, M. Macromol. Chem. Phys. 200, 201, 2631.
(16) Kricheldorf, H. R.; Eggerstedt, S. Macromolecules 1998, 31,
6403.
Con clu sion
The results of this work demonstrate that it is feasible
to synthesize the polyesters of structure 1 in the nematic
phase, so that the LC phase is formed in a very early
stage of the polycondensation. The chemical structure
(17) Shaik, A. A.; Richter, M.; Kricheldorf, H. R.; Kru¨ger, R.-P. J .
Polym. Sci., Part A, Polym. Chem. 2001, 39, 3371.
MA020170R