Multicyclic polyethers derived from 1,4-dicyanotetrafluorobenzene and flexible diphenols

Article abstract of DOI:10.1021/ma060895l

1,4-Dicyanotetrafluorobenzene (DCTB) was polycondensed with various diphenols in DMF or DMSO using K₂CO₃ as catalyst and HF acceptor. At concentrations of 0.1 or 0.2 mol/L for DCTB, gels were obtained with 1,4-bis[2-(4-hydroxyphenyl)

Full text of DOI:10.1021/ma060895l

Macromolecules 2006, 39, 6445-6450
6445
Multicyclic Polyethers Derived from 1,4-Dicyanotetrafluorobenzene
and Flexible Diphenols
Hans R. Kricheldorf,* Johann Schellenberg, and Gert Schwarz
Institut fu¨r Technische und Makromolekulare Chemie, Bundesstr. 45, D-20146 Hamburg, Germany
ReceiVed April 21, 2006; ReVised Manuscript ReceiVed July 19, 2006
ABSTRACT: 1,4-Dicyanotetrafluorobenzene (DCTB) was polycondensed with various diphenols in DMF or
DMSO using K2CO3 as catalyst and HF acceptor. At concentrations of 0.1 or 0.2 mol/L for DCTB, gels were
obtained with 1,4-bis[2-(4-hydroxyphenyl)-2-propyl]benzene, with 1,3-bis[(4-hydroxyphenyl)-2-propyl]benzene
and with R,ω-bis(4-hydroxyphenoxy) alkanes as comonomers. Yet soluble multicyclic polyethers were formed
with R,ω-bis(3-hydroxyphenoxy) alkanes. The lengths of the alkane chains did not play any role for the successful
syntheses of soluble multicycles. Furthermore, a model reaction with p-cresol was conducted, proving the feasibility
of a quantitative tetrasubstitution of all four F-atoms in DCTB. The multicyclic polyethers were characterized by
elemental analyses, 13C NMR spectroscopy, MALDI-TOF mass spectrometry, DSC, viscosity, and SEC
measurements. As expected, high polydispersities (up to 8) were found. The DSC measurements indicated an
amorphous character of all multicycles with glass-transition temperatures (Tgs) between 75 and 161 °C.
Introduction
Multicyclic polymers represent a new architecture which may
be described as a three-dimensional arrangement of mutually
connected smaller and larger cycles. Multicyclic polymers may
also be considered as soluble nanogels. Multicyclic polymers
may, in principle, be synthesized by a2 + bn polycondensations,
where a2 means a bifunctional and bn a multifunctional
monomer. For a proper understanding of such “three-dimen-
sional polycondensation” (a term coined by Flory),1,2 it is useful
to distinguish between polycondensations based on equimolar
monomer mixtures and on equifunctional monomer mixtures.
Using equimolar monomer mixtures, (hyper)branched polymers
are the most likely reaction products (at least at high monomer
concentrations). Equifunctional polycondensations meaning feed
ratios of 1.5/1.0 for “a2 + b3” or 2.0/1.0 for “a2 + b4” systems
will most likely result in insoluble gels when conducted in bulk.
However, when conducted in moderately concentrated solutions
(i.e., 0.02-0.2 mol/L) and when the structure of the monomers
favors cyclization, gelation may be avoided and soluble mul-
ticyclic polymers will be the only reaction product. Whereas
experimental and theoretical studies of a2 + b2 polycondensa-
tions have a long tradition,2-5 little is known about “a2 + b4
polycondensations”. Recently, we have published6-8 a few
examples of reversible, thermodynamically controlled polycon-
densations involving tetrafunctional monomers, which yielded
soluble spirocycles and higher multicyclic oligomers. Further-
more, one example of an irreversible (kinetically controlled) a2
+ bn polycondensation was reported9 using tetra(bromomethyl)
methane as “b4 monomer”. In this context, the present work
was aimed at studying an irreversible a2 + b4 polycondensation
based on 1,4-DCTB as an electrophilic “b4-monomer” and its
reaction with diphenols. It should be elucidated how structure
and flexibility of the diphenols influences the competition
between cyclization and chain growth. Using an equifunctional
stoichiometry (feed ratio 2:1), it should be found out if gelation
can be avoided and soluble multicyclic polyethers can be isolated
(eq 1). Polycondensations with equimolar stoichiometry yielding
(hyper)branched polyethers will be described in a future
publication. Finally, it should be mentioned that numerous
syntheses of polyimides, poly(benzoxazole)s, poly(benzimida-
zole)s, poly(quindine)s, or poly(quinoxaline)s may also be
considered as “a2 + b4” polycondensations. However, in those
cases, a rapid and quantitative formation of five- or six-
membered heterocycles is intended (and achieved) so that neither
hyperbranched nor multicyclic polymers are formed.
Experimental Section
Materials. 1,4-Dicyanotetrafluorobenzene (DCTB) was kindly
supplied by Bayer AG (Leverkusen, Germany) and used after drying
over P4O10 in vacuo (mp 199-201 °C, Aldrich Catalog: mp 197-
199 °C). Resorcinol, hydroquinone, 1,3-dibromopropane, 1,4-
dibromobutane, 1,5-dibromopentane, 1,6-dibromohexane, and 1,10-
dibromodecane were purchased from ACROS Chemicals (Geel,
Belgium) and used as received. 4-Cyanotetrafluoropyridine (CTFP)
and 1,3-bis[2-(4-hydroxy phenyl)-2-propylene]benzene (1,3-BH-
PPB, 1a) and 1,4-bis[2-(4-hydroxy phenyl)-2-propylene]benzene
(1,4-BHPPB, 1b) were purchased from Aldrich Co. (Milwaukee,
WI) and used as received.
* Corresponding author. E-mail: kricheld@chemie.uni-hamburg.de.
10.1021/ma060895l CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/17/2006

6446 Kricheldorf et al.
Macromolecules, Vol. 39, No. 19, 2006
Table 1. Yields and Properties of r,ω-Bis(4-hydroxyphenyl) and r,ω-Bis(3-hydroxyphenyl) Alkanes
elemental analyses
C
formula
no.
yield
(%)
mp
(°C)
mass
(Da)
1H NMR chemical shifts δ (ppm)
H
(in CDCl3/TMS)
2a
2b
3a
3b
3c
3d
3e
20
18
15
22
17
15
15
209-211
187-189
117-119a
107-109
104-106
89-91
274.3
302.4
260.3
274.3
288.3
302.4
358.5
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
calcd
found
70.06
69.66
70.06
69.66
69.22
68.77
70.06
69.57
70.81
70.40
71.50
71.40
73.71
73.36
6.61
6.50
6.61
6.5
1.90 (m, 4H), 3.97 (m, 4H), 6.76 (m, 8H), 7.83 (s, 2H)
1.42 (m, 4H), 1.66 (m, 4H), 3.83 (t, 4H), 6.53 (d, 4H),
6.72 (d, 4H), 8.85 (s, 2H)
1.91 (m, 2H), 3.98 (t, 4H), 6.28 (m, 6H), 6.92 (m, 2H,
6.11 (s, 2H)
1.92 (m, 4H), 4.02 (m, 4H), 6.41 (m, 6H), 7.06 (m, 2H),
8.27 (s, 2H)
164 (m, 4H), 1.82 (m, 4H), 3.97 (t, 4H), 6.40 (m, 6H),
7.05 (m, 2H), 8.23 (s, 2H)
1.42 (m, 4H), 1.69 (m, 4H), 3.87 (t, 4H), 6.30 (m, 6H),
7.01 (m, 2H), 9.33 (s, 2H)
1.41 (m, 12H), 1.75 (m, 4H), 3.93 (t, 4H), 6.40 (m, 6H),
7.05 (m, 2H), 8.23 (s, 2H)
6.20
6.36
6.61
6.89
6.99
7.21
7.33
7.34
8.44
8.39
80-82
Table 2. Yields and Properties of Multicyclic Polyethers Prepared from DCTB
mass peaks (Da)c of multicyclic polyethers
B2C2, B3C3, B4C4, B5C5, B6C6, B7C7, B8C8, B9C9, B10C10, B11C11, B12C12,
B13C13, B14C14
a
b
yield
(%)
ηinh
Tg
exp no.
diphenol
(dL/g)
(°C)
1
2
3
4
5
3a
3b
3c
3d
3e
95
94
95
97
90
0.18
0.14
0.29
0.36
0.41
161
128
127
120
75
1320, 1961, 2601, 3242, 3882, 4522, 5162, 5802, 6442, 7082, 7722, 8362,9002
1376, 2044, 2713, 3381, 4049, 4717, 5386, 6053, 6721, 7389, 8057, 8724, 9392, 10059
1432, 2188, 2825, 3521, 4218, 4914, 5482, 6306, 7002,7697, 8394, 9090, 9785
1489, 2214, 2939, 3664, 4389, 5114, 5350, 6564, 7288, 8013, 8737, 9462, 10188, 10911
1713, 2550, 3387, 4224, 5061, 5898, 6735, 7572, 8409, 9246
a Measured at 20 °C with c ) 2 g/L in CH2Cl2/TFA (volume ratio 4:1). b DSC measurements (2nd heating) conducted with a heating rate of 20 °C/min.
c Incl. doping with a Kx ion.
1,4-Bis(4-hydroxyphenoxy)butane.10 Hydroquinone (1.0 mol)
and 1,4-dibromobutane (0.1 mol) were dissolved in ethanol (200
mL), and sodium dithionite (50 mg) was added. This mixture was
heated to gentle reflux, and the solution of KOH (2.2 mol) in ethanol
(100 mL) was added rapidly with stirring. After reflux for 3 h, the
cold reaction mixture was acidified with 20% H2SO4. The ethanol
was removed by means of a rotatory evaporator, and the residue
was poured into water (2 L) with vigorous stirring. After 1 h, the
crystallized crude product was isolated by filtration. The dry product
was dissolved in acetone and stirred with charcoal for 2 h. The
acetone was then evaporated and the residue twice extracted with
a hot 3:1 mixture (by volume) of water and acetone. The pure
monomer 2a was crystallized by slow evaporation of the combined
extracts. The diphenol 2b was prepared analogously. The charac-
terization of the biphenols 2a and b is summarized in Table 1.
1,6-Bis(3-hydroxyphenoxy)hexane. Resorcinol (1.0 mol) and
1,6-dibromohexane (0.1 mol) were dissolved in ethanol (200 mL),
and sodium dithionite (50 mg) was added. This solution was gently
refluxed, and a solution of ROH (2.2 mol) in ethanol was rapidly
added from a dropping funnel with stirring. This mixture was
refluxed for 3 h, cooled, and acidified with 20% H2SO4 under
cooling with ice. The ethanol was removed by means of a rotatory
evaporator, and the residue was poured into water (2 L) with
vigorous stirring. After storage for 20 h at 10-20 °C, the
crystallized crude product was isolated by filtration and dried in
vacuo at 20 °C. The dry product was dissolved in acetone and stirred
with charcoal for 2 h. After filtration, the acetone was evaporated
and the residue three times extracted with hot water. The combined
extracts were concentrated in vacuo to approximately one-half of
the original volume, and the pure monomer 3d was isolated by
filtration. The diphenols 3a-c and 3d were synthesized analogously.
The diphenol 3a has been mentioned in the literature.11 The
characterization of the diphenols is summarized in Table 1.
Model Reaction. p-Cresol (41 mmol), DCTB (10 mmol), and
K2CO3 (22 mmol) were weighed into a 50 mL Erlenmeyer flask
equipped with a magnetic stirring bar, and dry DMF (20 mL) was
added. This mixture was stirred for 72 h at 75 °C and poured after
cooling into water (200 mL). The precipitated product was isolated
by filtration and recrystallized from dioxane. Yield 73%, mp 226-
228 °C, FAB: m/z ) 553 Da.
13C NMR (CDCl3) δ: 20.60, 110.38, 110.62, 115.84, 130.00,
133.52, 147.15, 154.30 ppm.
Polycondensations of DCTB with R,ω-Bis(3-hydroxy phe-
noxy) Alkanes. DCTB (2.0 mmol) and a diphenol (4.0 mmol) were
dissolved in dry DMF (23 mL), and K2CO3 (4.5 mmol) was added.
The reaction mixture was magnetically stirred in a closed 50 mL
Erlenmeyer flask for 72 h at 70 °C. Afterward, it was poured into
water, and the precipitated polymer was isolated by filtration and
dried at 60 °C in vacuo. Yields and properties of the polyethers
are compiled in Table 2.
An analogous polycondensation was conducted with 4-cyanotet-
rafluoropyridine (2.0 mmol) and 1,3-bis(3-hydroxy phenoxy)-
propane (4.0 mmol), and a soluble polyether was isolated in a yield
of 97% with ηinh ) 0.15 dL/g in CH2Cl2.
Analyses Calcd for C36H28N2O4 (616.6): C, 70.12; H, 4.58; N,
4.54%. Found: C, 69.51; H, 4.69; N 4.41%. Masses (Da) calcd
for multicycles incl. Kx: B1C1, 656; B2C2, 1272; B3C3, 1889; B4C4,
2503; B5C5, 3122; B6C6, 3739; B7C7, 4355; B8C8, 4972; B9C9, 5589;
B10C10, 6205.
Measurements. The inherent viscosities were measured in CH2-
Cl2 using an automated Ubbelohde viscometer thermostated at 20
°C. The 100.4 MHz 13C NMR spectra were recorded on a Bruker
Avance 400 FT spectrometer in 5 mm o.d. sample tubes. CDCl3
(containing TMS) served as solvent. The MALDI-TOF mass spectra
were measured with a Bruker Biflex III spectrometer equipped with
a nitrogen laser (λ ) 337 nm). All mass spectra were recorded in
the reflection mode with an acceleration voltage of 20 kV. The
irradiation targets were prepared from chloroform solutions using
dithranol as matrix and K-trifluoroacetate as dopant. The fast atom
bombardment (FAB) mass spectra were measured with a VG/70-
205F spectrometer from VG Analytical using m-nitrobenzyl alcohol
as matrix (Figure 1).
Results and Discussion
Syntheses of Flexible Diphenols. Commercial diphenols such
as bisphenol A or bisphenol-S(4,4′-dihydroxy diphenyl sulfide)
were assumed to be not flexible enough to favor cyclization to
such an extent that gelation could be avoided in polyconden-
sations with DCTB. 1,3-BHPPB (1a) and 1,4-BHPPB (1b)
seemed to be more favorable due to two bond angles of 110°
at the isopropyl groups and two bond angles of approximately
Elemental Analyses Calcd for C36H28N2O4 (552.64): C, 78.2 H;
H, 5.11; N, 5.07. Found: C, 77.88; H, 5.04; N 5.08%.

Macromolecules, Vol. 39, No. 19, 2006
Multicyclic Polyethers Derived from DCTB 6447
120° at the O atoms upon formation of ether groups. The 1,3
substitution at the central benzene ring seemed to be an
additional advantage of 1a. However, it was foreseeable that
further more flexible diphenols were needed, and thus the
diphenols of structure 2 and 3 should be synthesized. The
preparation of such diphenols derived from hydroquinone (2)
was described in the literature.10,11 Two diphenols having four
or six methylene groups (2, n ) 4 or 6) were synthesized
according to the published procedure. However, the diphenols
mentioned in the literature were not characterized with respect
to their purity. Neither measurements of melting points, chro-
matograms, nor mass spectra were reported. However, such a
characterization is necessary because the syntheses are based
on a condensation of R,ω-dibromoalkanes with a large excess
of hydroquinone. Yet, even a 10-fold excess of hydroquinone
does not completely prevent formation of oligomers (5), and
their presence in small quantities is not detectable by elemental
analyses, IR, 1H NMR, or 13C NMR spectra. Using FAB mass
spectrometry, it was now detected that the diphenols 2a and b
indeed contained oligoethers when isolated as described in the
literature. Extraction of the crude product by hot water proved
to be an efficient purification method due to the low solubility
of the oligoethers. In this way, pure monomers were obtained
at the expense of high yields.
Figure 1. FAB mass spectrum of 1,6-bis(3-hydroxy phenoxy)hexane
prior to the extraction of the monomer (M ) monomer, D ) dimer, T
) trimer).
was characteristic for that time, identity and purity of these
diphenols were checked by elemental analyses and melting
temperatures. The diphenols 3a-e were synthesized in this work
via the same procedure used for hydroquinone,10 and the similar
purification process was necessary to remove the oligoethers 6
(eq 2). The yields and properties of the purified diphenols 2a,b
and 3a-e are summarized in Table 1.
Model Reaction. The only kind of polycondensations DCTB
has been used for as electrophilic monomer are syntheses of
ladder polymers such as 7 (eq 3).12-16 In this case, a complete
Two analogous diphenols derived from resorcinol (3, n ) 2,
3) were mentioned in the literature as early as 1923.11 As it
substitution of all four atoms is favored due to the absence of
any steric hindrance and because the ring closure is entropically
favorable. To find out if a “tetrasubstitution” is also feasible
under mild conditions with four phenols as reaction partners, a
model reaction with p-cresol was performed. The reaction
conditions which had been proven to allow for clean syntheses
of ladder polymers (7) from tetrahydroxy tetramethyl spirobis-
indane were also used for the model reaction. This means that

6448 Kricheldorf et al.
Macromolecules, Vol. 39, No. 19, 2006
the condensation was conducted in DMF at 80 °C with K2CO3
as catalyst and HF acceptor. The model compound 8 was indeed
isolated in high yield and characterized by elemental analyses,
13C NMR spectroscopy, and FAB mass spectrometry. The
elemental analyses agreed satisfactorily with the calculated
values (see Experimental Section). The 13C NMR spectrum
proved the formation of a symmetrical substitution product, and
the FAB mass spectrum exhibited one signal with the correct
mass.
Polycondensations. On the basis of this successful model
reaction, the polycondensations of DCTB with diphenols were
conducted analogously in DMF at 80°C. Using a DCTB
concentration of 0.08 mol/L and the diphenols 1a or 1b as
comonomers, gelation occurred in both experiments. Lowering
the monomer concentrations by a factor of 2 gave the same
negative results. Therefore, the more flexible diphenols 2a and
2b were used as “a2” monomers in subsequent experiments. A
DCTB concentration of 0.08 mol/L was again used in the first
two experiments, which resulted in gelation. Lowering the
monomer concentration by a factor of 2 proved again unsuc-
cessful. A further reduction of the concentration was not
considered because it was learned from previous studies12 that
it is difficult to achieve high conversions (i.e., >99%) at
concentrations e0.02 mol/L and because polycondensations with
the diphenols 3a-3e were successful at higher concentrations.
When the resorcinol-based diphenols 3a-3e were polycon-
densed with DCTB at a concentration of 0.08 mol/L, all reaction
products were completely soluble in CH2Cl2 or CHCl3 contain-
ing 20 vol % of trifluoroacetic acid (TFA) regardless of the
spacer length. In two initial experiments with 3b as monomer,
the reaction time was reduced from the standard value of 72 h
to 48 h and 24 h. However, the MALDI-TOF evidenced that
the conversions were not complete. Multicycles having C-F
and/or C-OH (end)groups were detected in the mass spectra,
as illustrated in Figure 2. For instance, the most intensive peak
of byproducts (labeled X in Figure 2) corresponds to oligomers
having two C-F groups, as exemplarily illustrated by formula
9. However, a reaction time of 72 h sufficed for almost complete
Figure 2. MALDI-TOF mass spectrum of the multicyclic polyether
prepared from DCTB and 1,4-bis(3-hydroxy phenoxy)butane with a
reaction time of 48 h.
Figure 3. MALDI-TOF mass spectrum of the multicyclic polyether
prepared from DCTB and 1,3-bis(3-hydroxy phenoxy)propane with a
reaction time of 72 h (no. 1, Table 2).
Figure 4. MALDI-TOF mass spectrum of the multicyclic polyether
prepared from DCTB and 1,6-bis(3-hydroxy phenoxy)hexane with a
reaction time of 72 h (no. 4, Table 2).
conversion, as demonstrated by the MS presented in Figures 3
and 4. These results show that polycondensations at low
concentrations (e0.02 mol/L), indeed, raise the problem of
reaching high conversions. The MS presented in Figures 3 and
4 also confirm that the formation of multicyclic polyethers is
not limited to low oligomers.
means cycle and the figure behind C gives the total number of
repeat units (i.e., the degree of polymerization calculated as
combination of one a2 and one b4 unit). B means bridging unit
resulting from incorporation of additional a2 units. These B units
either bridge a cycle or connect two cycles. The total number
of cycles, which can be drawn in a two-dimensional formula,
equals Bn + 1. If n equals N, the multicycles are free of
At this point, the terminology used for a simple denomination
of these multicycles should be explained. In the term BnCN, C

Macromolecules, Vol. 39, No. 19, 2006
Multicyclic Polyethers Derived from DCTB 6449
Scheme 1. Selected Isomers of BnCn (BNCN) Multicycles
Figure 5. MALDI-TOF mass spectrum of the multicyclic polyether
10 prepared from 4-cyanotetrafluoropyridine and 1,3-bis(3-hydroxy
phenoxy)propane with a reaction time of 72 h.
Scheme 2. Selected examples of multicyclic oligomers BnCN
with n < N
“endgroups”, and examples of their structure are outlined in
Scheme 1. It is obvious that the number of potential isomers
exponentially increases with N. For n < N, the multicycles
contain “end groups” resulting from the functional groups of
“b4 monomers” (C-F in the present work). Scheme 2 presents
four examples for such functional multicycles. Multicycles
bearing functional groups may, of course, be of interest for
further modifications, and they can be synthesized by using feed
ratios of “a2”/”b4” below 2:1. However, the purpose of the
present work is to find out if a perfect “tetrasubstitution” of
DCTB is feasible so that BnCn (or BNCN) multicycles become
the main products.
Figure 6. Distances of OH groups calculated for the energy minimum
conformation of the diphenols 2 (curve A) and the diphenols 3 (curve
B) plotted vs the number of CH2 groups.
substantiate this suggestion, one polycondensation of the diphe-
nol 3a with 4-cyanotetrafluoropyridine was performed, and a
soluble reaction product mainly consisting of the expected
multicyclic polyethers (exemplarily illustrated by formula 10)
was indeed obtained. The MS presented in Figure 5 confirms
the success of this polycondensation (the K-doped masses of
the multicycles were listed in the Experimental Section).
Computer Modeling. The enormous difference between the
results obtained from the “hydroquinone diols” 2 on one hand,
and the “resorcinol diols” 3, on the other, prompted us to
simulate and compare their low-energy conformations using the
“Spartan 04” program of Wave Function, Inc.
The characterization of the multicyclic polyethers by classical
spectroscopic methods proved difficult. The multicyclic struc-
tures are, in principle, not questionable, and the presence or
absence of small amounts of C + F or OH end groups can
1
neither be checked by IR, H, or 13C NMR spectroscopy. The
NMR signals are broad due to the presence of numerous isomers
with slightly differing conformations. The signals of end groups
are obscured by the feet of the broad main signals. Furthermore,
reliable assignments of C-F groups are difficult to achieve
without pertinent model compounds. Furthermore, SEC were
not feasible because the multicyclic polyethers were neither
completely soluble in neat THF, nor in neat chloroform, nor in
neat DMAc. DSC measurements were, of course, feasible and
confirmed the expected amorphous character of all multicyclic
polyethers. The glass-transition temperatures (Tgs) displayed a
strong dependence on the length of the aliphatic spacers (Table
2).
The distances of the OH groups were calculated for the
conformation of the lowest energy, expecting that this confor-
mation possesses the highest frequency in the reaction mixture.
The calculated OH distances were plotted against the number
of CH2 groups in the spacers of the diphenols 2 and 3. The
resulting diagram presented in Figure 6 allows two useful
conclusions. First, the distances of the OH groups indicating
the cyclization tendency is lower for the diphenols of structure
3, regardless of “n”. This trend corresponds to the higher
cyclization tendency found for the polycondensations of these
diphenols. Second, the distances of the OH groups in diphenols
2 and 3 are quite similar for six or more CH2 groups. Therefore,
these calculations would not allow for reliable prediction of the
experimental results obtained in this work. This aspect is not
surprising because this computer modeling of conformations
The successful polycondensation of DCTB with flexible
diphenols suggested that soluble multicyclic polyethers may also
be obtained from other activated tetrafluoroaromatics. To

6450 Kricheldorf et al.
Macromolecules, Vol. 39, No. 19, 2006
was performed in a vacuum and does not consider the strong
influence the solvation of OH groups and phenoxide ions may
have. Nonetheless, the computer modeling of conformations is
helpful for better understanding of given experimental data.
Further examples of similar “a2 + b4” polycondensations will
be presented in a future publication.
Acknowledgment. We wish to thank Dr. Detlef Fritsch
(Institut fu¨r Polymerforschung, GKSS, Geesthacht, Germany)
for a generous gift of chemicals.
Conclusion
References and Notes
The model experiment and the polycondensation experiments
of this work clearly demonstrate that a complete substitution
of all four F atoms in DCTB by phenoxide ions can be achieved
under the given reaction conditions. These results allowed for
an evaluation of the structure of diphenols with regard to their
cyclization tendency. Diphenols containing para-hydroxy phenyl
groups favored chain growth to such an extent that cyclization
was not efficient enough to prevent gelation. Even flexibilization
of the diphenols with alkylene groups did not suffice to enhance
the cyclization tendency sufficiently when the functional groups
were in the para position. In contrast, m-hydrophenoxy groups
attached to alkylene chains yielded soluble multicyclic poly-
ethers, regardless of the length of the alkylene chains. In
summary, slight changes in the structure of the diphenols may
have a strong influence on the competition between cyclization
and chain growth. The successful syntheses of soluble multi-
cycles from the diphenols 3a-e suggest that further classes of
multicyclic polymers can be synthesized by variation of the
electrophilic “b4” monomer. A first confirmation for this
suggestion comes from the polycondensation of 4-cyanotet-
rafluoropyridine with the diphenol 3a. As illustrated by the MS
of Figure 5, multicyclic polyethers of structure 10 were obtained.
(1) Flory, P. J. Chem. ReV. 1946, 39, 137.
(2) Flory, P. J., Principles in Polymer Chemistry; Cornell University
Press: Ithaca, New York, 1953.
(3) Kienle, R. H.; van der Meulen, F. A.; Petke, F. E. J. Am. Chem. Soc.
1939, 61, 2258.
(4) Kienle, R. H.; Petke, F. E. J. Am. Chem. Soc. 1940, 62, 1053.
(5) Kienle, R. H.; Petke, F. E. J. Am. Chem. Soc. 1941, 63, 481.
(6) Kricheldorf, H. R.; Fechner, B. Biomacromolecules 2002, 3, 691.
(7) Kricheldorf, H. R.; Rost, S. Macromolecules 2004, 37, 7955.
(8) Kricheldorf, H. R.; Rost, S. Polymer 2004, 45, 3205.
(9) Kricheldorf, H. R.; Schellenberg, J.; Schwarz, G. Macromol. Chem.
Phys., in press.
(10) Yu, S. C.; Hou, S.; Chan, W. K. Macromolecules 2000, 33, 3259.
(11) Kohn, M.; Wilhelm, F. Monatsh. Chem. 1923, 43, 545.
(12) Budd, P. M.; Chanem, B. S.; McKweon, N. B.; Msayib, K. J.;
Tattershall, C. F. Chem. Commun. 2004, 230.
(13) Budd, P. M.; Elabas, E. S.; Chanem, B. S.; Maksheed, S.; MacKweon,
N. B.; Msayib, K. J.; Tattershall, C. E.; Wang, O. AdV. Mater. 2004,
16, 456.
(14) Kricheldorf, H. R.; Fritsch, D.; Vakhtangishvili, L.; Schwarz, G.
Macromol. Chem. Phys. 2005, 206, 2239.
(15) Kricheldorf, H. R.; Fritsch, D.; Lomadze, N.; Schwarz, G. J. Polym.
Sci., Part A.: Polym. Chem., in press.
(16) Kricheldorf, H. R.; Lomadze, N.; Polefka, C.; Schwarz, G. Macro-
molecules 2006, 39, 2107.
MA060895L