Chain-growth polycondensation for aromatic polyethers with low polydispersities: Living polymerization nature in polycondensation

Article abstract of DOI:10.1021/ma021553s

Polycondensation normally proceeds in a step-growth reaction manner to give polymers with a wide range of molecular weights. However, the polycondensation of potassium 5-cyano-4-fluoro-2-propylphenolate (1) proceeded at 150 °C in a chain polymerization manner from an initiator, 4-fluoro-4′-trifluoromethylbenzophenone (2a), to give aromatic polyethers having controlled molecular weights and low polydispersities (M_w/M_n ≤ 1.1). The resulting polycondensation of 1 had all of the characteristics of living polymerization and displayed a linear correlation between molecular weight and monomer conversion, maintaining low polydispersities. The MALDI-TOF mass spectrum of poly1 revealed that this polycondensation did not include conventional step-growth polycondensation which gave the polymer without initiation unit and macrocycles. The poly1 with low polydispersity showed higher crystallinity than that with broad molecular weight distribution, obtained by the conventional polycondensation of 1 without 2a.

Full text of DOI:10.1021/ma021553s

4756
Macromolecules 2003, 36, 4756-4765
Chain-Growth Polycondensation for Aromatic Polyethers with Low
Polydispersities: Living Polymerization Nature in Polycondensation
Yu k im itsu Su zu k i,† Sh u ich i Hir a ok a ,† Ak ih ir o Yok oya m a ,† a n d
Tsu tom u Yok oza w a *,†,‡
Department of Applied Chemistry, Kanagawa University, Rokkakubashi, Kanagawa-ku,
Yokohama 221-8686, J apan, and “Synthesis and Control”, PRESTO, J apan Science and
Technology Corporation (J ST)
Received October 4, 2002; Revised Manuscript Received March 31, 2003
ABSTRACT: Polycondensation normally proceeds in a step-growth reaction manner to give polymers
with a wide range of molecular weights. However, the polycondensation of potassium 5-cyano-4-fluoro-
2-propylphenolate (1) proceeded at 150 °C in a chain polymerization manner from an initiator, 4-fluoro-
4′-trifluoromethylbenzophenone (2a ), to give aromatic polyethers having controlled molecular weights
and low polydispersities (Mw/Mn e 1.1). The resulting polycondensation of 1 had all of the characteristics
of living polymerization and displayed a linear correlation between molecular weight and monomer
conversion, maintaining low polydispersities. The MALDI-TOF mass spectrum of poly1 revealed that
this polycondensation did not include conventional step-growth polycondensation which gave the polymer
without initiation unit and macrocycles. The poly1 with low polydispersity showed higher crystallinity
than that with broad molecular weight distribution, obtained by the conventional polycondensation of 1
without 2a .
In tr od u ction
We expect two approaches to the chain-growth poly-
condensation of 1: electrophilic site propagation and
nucleophilic site propagation (Scheme 1). In the elec-
trophilic site propagation by using reactive initiator 2a
bearing an electron-withdrawing group, 1 would react
with 2a to yield ether 3 faster than with the aromatic
fluorine of another 1 having the strong electron-donat-
ing phenoxide group. Monomer 1 would now react with
3 to yield a dimeric ether faster than with another 1
because the ether linkage of 3 is a weaker electron-
donating group than the phenoxide group of 1, and the
aromatic fluorine of 3 would be more reactive than that
of the monomer. Growth would continue in a chain
polymerization manner with the conversion of the
strong electron-donating phenoxide group of 1 to the
weak electron-donating ether linkage in polymer. In the
nucleophilic site propagation, phenoxide is the propa-
gating end group instead of fluorine. Thus, when potas-
sium 4-methoxyphenoxide (4) having an electron-
donating group is used as an initiator, 4 would react
with 1, and the ether linkage formed would make the
polymer terminal phenoxide more reactive than the
phenoxide of monomer because ether linkage has stron-
ger electron-donating character than fluorine. Both
approaches depend on whether there are enough dif-
ference of substituent effects between monomer and
polymer. It is not easy to predict which approaches could
achieve chain-growth polycondensation, but it seems
that the electrophilic site propagation of 1 would be at
least more difficult to be achieved than the successful
electrophilic site propagation for polyamides.5 This is
because the substituent effects on the electrophilic site
are not expected to change much between the phenoxide
in monomer and the ether linkage in polymer in the
polycondensation of 1, in comparison with the change
of substituent effects between the aminyl anion in
monomer and the amide linkage in polymer in the
chain-growth polycondensation for polyamides.5
Poly(arylene ether)s are one of the important classes
of high-performance engineering plastics and possess
excellent chemical and thermal stabilities as well as
good mechanical properties.1 These polymers are gener-
ally produced by polycondensation such as oxidation of
phenol derivatives2 and aromatic nucleophilic substitu-
tion,3 and the nature of polycondensation that proceeds
in a step-growth reaction manner results in polymers
with broad molecular weight distributions. If poly-
(arylene ether)s with low polydispersities were pro-
duced, further novel physical properties could be added
to these high-performance functional polymers. Percec
conducted the oxidative poymerization of 4-bromo-2,6-
dimethylphenol in the presence of chain initiators to
obtain poly(arylene ether)s with low polydispersities.4
The polymers, however, were obtained after precipita-
tion, and the crude polymerization mixture had a broad
molecular weight distribution.
We have been recently successful in the synthesis of
polyamides with defined molecular weights and low
polydispersities5 by chain-growth polycondensation,6
where the monomers react with the polymer end group
selectively, not with other monomers. The principle of
this polycondensation is different substituent effects
between monomer and polymer; strong electron-donat-
ing ability of the aminyl anion in the monomer deacti-
vates the phenyl ester moiety in the monomer, whereas
the amide linkage in the polymer activates the polymer
end phenyl ester moiety in comparison with that of the
monomer. Therefore, the monomer is prevented from
the reaction with other monomers and reacts with only
the polymer end group, resulting in chain-growth po-
lymerization like living polymerization. We now apply
this chain-growth polycondensation to the synthesis of
poly(arylene ether)s with low polydispersities and then
have designed monomer 1.7
† Kanagawa University.
‡ J apan Science and Technology Corporation (J ST).
10.1021/ma021553s CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/28/2003

Macromolecules, Vol. 36, No. 13, 2003
Polycondensation for Aromatic Polyethers 4757
Sch em e 1
Sch em e 2
In this paper, we report the chain-growth polycon-
densation of 1 for well-defined poly(arylene ether)s
through the electrophilic site propagation. The polym-
erization behavior has all of the characteristics of living
polymerization and displays a linear correlation be-
tween molecular weight and monomer conversion, main-
Resu lts a n d Discu ssion
Syn th esis a n d Con ven tion a l P olycon d en sa tion
of 1. Monomer 1, potassium 5-cyano-4-fluoro-2-propyl-
phenolate, was prepared as shown in Scheme 2. Fluoro-
phenol 5 was reacted with allyl bromide to give allyl
ether 6, which was then subjected to Claisen rearrange-
ment with Et2AlCl to yield 7. After hydrogenation of 7
the hydroxy group of 8 was protected with tert-bu-
tyldimethylsilyl (TBDMS) group. Regioselective lithia-
tion of 9 was carried out with sec-butyllithium at -78
°C, followed by the reaction with N,N-dimethylfolm-
amide (DMF) to give aldehyde 10.8 The formyl group of
10 was converted to the cyano group with hydroxyl-
amine hydrochoride and sodium formate in formic acid.9
Phenol 11 obtained was then treated with potassium
hydroxide in methanol to yield monomer 1.
1
taining low polydispersities. Furthermore, the H NMR
spectrum of poly1 is studied in detail by comparing with
the spectra of unimer and dimer connected with initiator
2 unit, indicating that the polymer without 2 unit is not
produced from conventional step-growth polycondensa-
tion. The MALDI-TOF mass spectrum of poly1 also
reveals that this polycondensation does not include
conventional polycondensation which gives macrocycles
and the polymer without 2 unit. On the basis of all of
the results, it has proved that the polycondensation of
1 possesses a real living polymerization nature.

4758 Suzuki et al.
Macromolecules, Vol. 36, No. 13, 2003
F igu r e 2. (a) Poly1 molecular weight and molecular weight
distribution (PDI) and the ratios of end group to initiator unit
in poly1 as a function of monomer conversion in the polym-
erization of 1 with 2a (sulfolane, 150 °C, [1]0 ) 0.17 M; [2a ]0
) 11.7 mM). (b) Poly1 molecular weight and as a function of
the feed ratio of 1 to 2a (sulfolane, 150 °C, [1]0 ) 0.17 M; [2a ]0
) 8.3-167 mM; conversion ) 100%).
F igu r e 1. (a) Monomer conversion as a function of time for
the polymerization of 1 in the presence of initiator (open
squares) 2a and (open triangles) 4 in sulfolane at 150 °C: [1]0
) 0.17 M; [initiator]0 ) 11.7 mM. (b) GPC profile of poly1
obtained in the polymerization of 1 with 2a for 40 min and
for 120 min.
Before study of chain-growth polycondensation of 1,
conventional polycondensation of 1 was carried out in
the absence of initiator 2a or 4 in sulfolane at 220 °C
for 4 h to see the polymerizability of 1 and the solubility
of poly1 in organic solvents. The reaction mixture was
poured into 3% hydrochloric acid, and precipitate was
washed with water and then dried in vacuo to give poly1
in 93% yield. The Mn and Mw/Mn of the polymer,
estimated by gel permeation chromatography (GPC)
with polystyrene standards, were 15 500 and 3.03,
respectively, indicating that 1 possesses sufficient po-
lymerizability. The polymer obtained was soluble in
DMF, dichloromethane, and chloroform at room tem-
perature. Monomer without the propyl group was also
attempted to polymerize, but the polymer obtained was
not soluble in any organic solvents.
Ch a in -Gr ow th P olycon d en sa tion of 1. According
to the two approaches to chain-growth polycondensation
of 1 mentioned in the Introduction, the polymerizations
of 1 with 7 mol % of initiator 2a for electrophilic site
propagation and with 7 mol % of initiator 4 for nucleo-
philic site propagation were carried out at 150 °C in
sulfolane. Surprisingly, the polymerization of 1 with 4
did not proceed at all, whereas the polymerization with
2a took place and the GPC chromatogram of polymer
shifted toward the higher molecular weight region with
increase of monomer conversion (Figure 1). This result
implies that not only the reaction of 4 with 1 but also
the reaction of monomers 1 with each other did not take
place under this condition and that the polymerization
of 1 in the presence of 2a did not involve step polym-
erization but was initiated with 2a .
To elucidate whether chain-growth polymerization
takes place in this polycondensation, the Mn values,10
the Mw/Mn ratios (PDI), and the ratios of end group to
initiator unit in polymer were plotted against monomer
conversion in the polymerization of 1 with 7 mol % of
2a (Figure 2a). In general polycondensations that
proceed in a step polymerization manner, the molecular
weight does not increase much in low conversion of
monomer and is accelerated in high conversion, and the
Mw/Mn ratios increase up to 2.0. The ratios of end group
to initiator unit would be larger than 1.00 because
monomers react with not only an initiator but also other
monomers. As shown in Figure 2a, the Mn values
increased in proportion to conversion, and the Mw/Mn
ratios were less than 1.1 over the whole conversion
range. The ratios of end group to initiator unit, which
were easily determined by the 19F NMR spectra of
polymer, were constantly about 1.00 irrespective of
conversion. Consequently, Figure 2a shows that the
polycondensation of 1 proceeds in a chain-growth po-
lymerization manner like living polymerization. In
another series of experiments, 1 was polymerized with
varying feed ratio ([1]0/[2a ]0). As shown in Figure 2b,

Macromolecules, Vol. 36, No. 13, 2003
Polycondensation for Aromatic Polyethers 4759
F igu r e 3. 1H NMR spectra of (a) poly1 from conventional polycondensation, (b) poly1 from chain-growth polycondensation with
2a ([1]0/[2a ]0 ) 7.5), (c) unimer 12, and (d) dimer 13.
Sch em e 3
the observed Mn values of polymers were in good
agreement with those calculated with the assumption
that one initiator molecule forms one polymer chain. The
Mw/Mn ratios were less than 1.1 when the [1]0/[2a ]0
ratios were more than 1.0.11 This also agrees with the
features of chain-growth polymerization.
multiplet between 7.90 and 7.83 and two doublets at δ
7.76 and 6.99 (H8). The signal of 1 unit was assigned
1
on the basis of the H-19F coupling: the doublet of H9
at δ 7.22 (J H9-F ) 5.5 Hz) and the doublet of H10 at δ
7.18 (J H10-F ) 9.2 Hz).
The 19F NMR spectrum of 13 also showed the signals
of the trifluoromethyl group as a singlet at δ 99.3 and
the signal of the aromatic fluorine as a doubled doublet
at δ 50.9 (J H14-F ) 5.4 Hz and J H15-F ) 9.2 Hz). In the
1H NMR spectrum of 13 (Figure 3d), the signals of
initiator 2a unit appeared as a multiplet between 7.91
and 7.85 and two doublets at δ 7.77 and 7.03 (H11). The
signals of the internal aromatic ether unit were ob-
served as two singlets at δ 7.30 (H12) and 6.78 (H13).
The signals of the chain end group appeared as two
1
1H NMR Sp ectr a of P oly1. The H NMR spectrum
of poly1 from conventional polycondensation was simple,
as shown in Figure 3a: two singlets of the polymer
repeating unit at δ 7.18 and 6.88 ppm as aromatic
proton signals. In the 1H NMR spectrum of poly1
obtained by chain-growth polycondensation from 2a ,
however, several signals other than those of the repeat-
ing unit were obtained (Figure 3b). To assign those
signals, model compounds were prepared. Thus, 1
reacted with 2 equiv of 2a in sulfolane at 150 °C to give
1:1 reactant 12 (unimer) and dimer 13 bonded to 2a
residue in 47% and 41% yields, respectively (Scheme
3).
doublets at δ 7.19 (J H15-F ) 9.2 Hz) and 7.10 (J H14-F
5.5 Hz).
)
On the basis of the spectra of 12 and 13, the signals
of poly1 obtained by chain-growth polycondensation at
the feed ratio [1]0/[2a ]0 ) 7.5 could be assigned as shown
in Figure 3b. Thus, a doublet at δ 7.07 is H1 of the
initiator unit. Two singlets at δ 7.30 and 6.76 are H2
and H3 of the 1 unit adjacent to the initiator unit 2a ,
respectively. The doublet at δ 7.13 is H6 of the polymer
end group. The intensity ratio of signals of the end group
The 19F NMR spectrum of 12 showed two signals: a
singlet corresponding to the trifluoromethyl group at δ
99.3 and a doubled doublet corresponding to the aro-
matic fluorine at δ 50.9 (J H9-F ) 5.4 Hz and J H10-F
)
9.2 Hz). The 1H NMR spectrum of 12 is shown in Figure
3c. The signals of initiator 2a unit were observed as a

4760 Suzuki et al.
Macromolecules, Vol. 36, No. 13, 2003
F igu r e 4. MALDI-TOF mass spectrum of poly1 obtained in the presence of 2a ([1]0/[2a ]0 ) 7.5) in sulfolane at 150 °C.
Sch em e 4
proton H6 to the initiator unit proton H1 was 1:2,
indicating that there was no polymer without initiator
2a unit from conventional polycondensation. In addition,
the 19F NMR spectrum of the polymer also showed that
one initiator molecule forms one polymer chain.
MALDI-TOF Ma ss Sp ectr u m of P oly1. It is well-
known that some polycondensations for poly(arylene
ether)s are equilibrium polymerization. At high tem-
peratures, and in the presence of fluoride or phenolate
ions, transetherification can take place and is greatly
facilitated by the presence of an electron-withdrawing
group in either the ortho or para positions relative to
the electrophilic site of monomers.12,13 Since the poly-
condensation of 1, which has the electron-withdrawing
cyano group at the ortho position of fluorine, forms
potassium fluoride as a byproduct, one might think that
the reversible transetherfication takes place also in the
polycondensation of 1 as shown in Scheme 4. If a linear
polymer bearing phenoxide and fluorine at each end was
formed, the ratios of end group to initiator unit would
be more than 1.0, which is not consistent with the
results of Figure 2. However, if macrocycles were
formed, it would not affect the ratios of end group to
initiator unit of linear polymer, and the 1H NMR spectra
of polymer could not rule out the possibility of the
formation of macrocycles.
For this reason, we have examined the MALDI-TOF
mass spectrum of poly1. MALDI-TOF mass spectrom-
etry is an ideal technique for the detection of small
quantities of side reaction products. Under the MALDI
conditions used, poly1 is ionized to [M + Ag]+ ions. The
mass spectrum obtained from the product of the polym-
erization of 1 with 2a at feed ratio [1]0/[2a ]0 ) 7.5 is
shown in Figure 4. It contains a major distribution
whose m/z values correspond to the Ag+ adducts of poly1
with initiator 2a . For example, the 8-mer of this
distribution is expected to produce a signal at m/z 8 ×
159.06 (repeat unit) + 268.06 (2a ) + 106.90 (107Ag+) )
1647.44, as indeed is observed as the highest signal and
in good agreement with the feed ratio [1]0/[2a ]0 ) 7.5.
Macrocycles with no 2a unit (-268.06 Da) is absent (no
m/z 1379 detected in Figure 4). The linear polymer
without 2a unit from conventional polycondensation
(-268.06 Da + 20.01 Da (HF)) is also absent (no m/z
1399 detected in Figure 4). The minor distribution
between signals of the Ag+ adducts of poly1 with
initiator 2a is due to the K+ adducts of the same
polymer. For example, m/z 2374.8 detected in Figure 4

Macromolecules, Vol. 36, No. 13, 2003
Polycondensation for Aromatic Polyethers 4761
F igu r e 5. 19F NMR spectra of 2a and 1 (470 MHz DMSO-d6 at 25 °C) and of the end group of poly1 prepared with 2a ([1]0/[2a ]0
) 7.5) in sulfolane at 150 °C (470 MHz, CDCl3 at 25 °C).
Sch em e 5
is identical with the expected signal of the 13-mer at
m/z 13 × 159.06 (repeat unit) + 268.06 (2a ) + 38.96 (K+)
) 2374.8. Consequently, this polycondensation does not
contain conventional step-growth polycondensation giv-
ing the linear polymer having no initiator unit and
macrocycles.
Rea ctivity of Ar om a tic F lu or in es. This successful
chain-growth polycondensation would be based on the
different reactivity of aromatic fluorines between the
polymer end group and monomer 1; initiator 2a and the
polymer end group are more reactive than 1. It has been
reported that more reactive aromatic fluorines show the
signals in lower field in the 19F NMR spectra,14 so the
19F NMR spectra of 1, 2a , and the polymer end group
were measured to estimate their reactivities (Figure
5).15 The signal of 2a appeared in the lowest field, and
differences in the polymerization with 2c can be ac-
the signal of the polymer end group and that of 1
counted for by slow initiation relative to propagation.
appeared in higher field in this order. Especially, the
Ridd and co-workers have reported the rates of the
reactions of substituted 4-halogenobenzophenones with
signal of 1 was about 24 ppm upfield for that of the
polymer end group, which was only 5 ppm upfield for
the signal of 2a . This result indicates that the negative
charge of phenoxide in 1 effectively deactivates the
the potassium salts of substituted 4-hydroxybenzophe-
nones.16 These reactions obey the Hammett equation
using normal σ values for substitution; the order of the
reactivities of aromatic fluorines of 4-X-4′-fluoroben-
zophenone is X ) CF3 > H > MeO. Consequently, slower
initiation in the polymerization of 1 with 2c is identical
with the reported substituent effects. It should be also
noted that 4-fluorobenzophenone 2b is a sufficient
initiator for the chain-growth polycondensations of 1 as
well as 2a . This means that an electron-withdrawing
group such as the CF3 group at the 4′-position of
4-fluorobenzophenone is not necessary for the initiator
of this polymerization and that bifunctional modified
fluorobenzophenone could be used for applications to
block copolymerization.
fluorine of 1 and that there is the difference of reactivity
between the monomer and the polymer end group
enough to attain chain-growth polycondensation, con-
trary to our first expectation described in the Introduc-
tion. Consequently, an approach to successful chain-
growth polycondensation will be to use monomers
having negative charged nucleophilic site.
Effect of In itia tor 2. Monomer 1 was polymerized
with 7 mol % of some 4-X-4′-fluorobenzophenone (X )
CF3, H, MeO) initiators 2 in sulfolane at 150 °C (Scheme
5). All the polymerizations were completed in 2-4 h
(Figure 6), implying that the polymerization proceeded
in a chain-growth polymerization manner because the
step polymerization of 1 did not proceed in this time as
shown in Figure 1. Linear increase of the Mn against
conversion also showed chain-growth polymerization,
and the Mn values were in good agreement with those
calculated ones (Figure 6b). The polymerization behav-
iors of 1 with 2a and 2b were almost the same: similar
polymerization rate and similar Mw/Mn ratios. In the
polymerization of 1 with 2c having an electron-donating
group, however, the initiation was slower than the
polymerization with 2a and 2b, and the Mn values were
slightly higher than the calculated ones in the initial
stage and became closer to them as the polymerization
proceeded. The Mw/Mn ratios were a little broader than
the polymerization with 2a and 2b. All these observed
Cr ysta llin ity of P olym er s Effected by P olyd is-
p er sity. We found that the polyether having initiator
2a unit with low polydispersity was less soluble in
organic solvent than the polyether obtained by conven-
tional polycondensation without 2a . Furthermore, the
powder X-ray diffraction (XRD) pattern of both polymers
with similar molecular weights showed that the crystal-
linity of the polymer obtained by chain-growth polycon-
densation (Mn ) 2450, Mw/Mn ) 1.13) was higher than
that of the polymer obtained by conventional polycon-
densation (Mn ) 2240, Mw/Mn ) 1.36) (Figure 7). The
differential scanning calorimetry (DSC) profile of the
polymer obtained by conventional polycondensation
showed no peak of both the first run and second run

4762 Suzuki et al.
Macromolecules, Vol. 36, No. 13, 2003
weights and low polydispersities. The MALDI-TOF
mass spectrum of poly1 reveals that this polyconden-
sation does not include conventional step-growth poly-
condensation which gives macrocycles and the polymer
without initiation unit. The poly1 with low polydisper-
sity showed higher crystallinity than that with broad
molecular weight distribution, obtained by the conven-
tional polycondensation of 1. Since not only the previous
monomer having the acyl group for polyamides but also
the monomer without acyl group for polyethers undergo
chain-growth polycondensation like living polymeriza-
tion, we are confident that deactivation of the electro-
philic site in monomers with the negative charged
nucleophilic site is an effective approach to chain-growth
polycondensation. A lot of condensation polymers with
low polydispersities would be produced by chain-growth
polycondensation with this approach, and their novel
physical properties could be discovered. Experiments
along these lines are in progress.
Exp er im en ta l Section
Gen er a l. 1H, 13C, and 19F NMR spectra were obtained on a
J EOL A-500 and FX-200 operating in the pulsed Fourier
transfer (FT) modes, using tetramethylsilane (1H and 13C
NMR) as an internal standard and C6F6 (19F NMR) as an
external standard, respectively. IR spectra were recorded on
a J ASCO FT/IR-410. The Mn and Mw/Mn of polymers were
measured with a TOSOH HLC-8020 gel permeation chroma-
tography (GPC) unit [eluent: N,N-dimethylformamide (DMF)
containing LiBr (1.05 g L-1), calibration: polystyrene stan-
dards] using two TSK-gel columns (2 × Multipore HXL-M).
MALDI-TOF mass spectra were recorded on a Shimazu/
Krotos Kompact MALDI IV tDE in the reflectron mode using
laser (λ ) 337 nm). A sample solution of polymer, dithranol
as a matrix, and silver trifluoromethanesulfonate as a cation-
izing salt was made in chloroform. X-ray diffraction was
performed on a Rigaku RAD-rA diffractometer using Cu KR
radiation, a diffracted beam monochromator, and counting
equipment. Differential scanning calorimertry (DSC) was
performed on a Seiko Instruments Inc. DSC 6000 at a heating
rate of 10 °C min-1
p-Fluorobenzophenone was commercially available and used
without further purification. Sulfolane was vacuum-distilled
over CaH2.
.
F igu r e 6. (a) Monomer conversion as a function of time for
the polymerization of 1 in the presence of initiator (filled
diamonds) 2a , (filled squares) 2b, and (filled triangles) 2c in
sulfolane at 150 °C: [1]0 ) 0.17 M; [2]0 ) 11.7 mM. (b) Poly1
molecular weight and PDI as a function of monomer conver-
sion: (filled diamonds) 2a , (filled squares) 2b, and (filled
triangles) 2c (sulfolane, 150 °C, [1]0 ) 0.17 M; [2]0 ) 11.7 mM).
Syn th esis of Mon om er 1. Allyl 4-F lu or op h en yl Eth er
(6). A mixture of 4-fluorophenol (22.4 g, 200 mmol), allyl
bromide (20.2 g, 220 mmol), potassium carbonate (33.2 g, 220
mmol), and 18-crown-6 (5.28 g, 20 mmol) in dry acetone (200
mL) was heated at reflux and stirred vigorously for 8 h. The
mixture was then poured into water and extracted with diethyl
ether three times. The combined organic layer was washed
with water and dried over anhydrous MgSO4. The solvent was
removed in vacuo, and the residue was distilled under reduced
pressure to give 28.6 g of 6 (94%); bp 54-55 °C/4.2 mmHg (lit.17
46-50 °C/4.2 mmHg). IR (neat): 3082, 2921, 2862, 1601, 1506,
1245, 1121, 828, 768, 726 cm-1. 1H NMR (200 MHz, CDCl3) δ:
6.94 (m, 4H), 6.03 (ddt, J ) , 9.3, 20.5, and 5.3 Hz, 1H), 5.38
(dd, J ) 1.5 and 20.5 Hz, 1H), 5.32 (dd, J ) 1.5 and 9.3 Hz,
1H), 4.48 (d, J ) 5.3 Hz, 2H). 13C NMR (50 MHz, CDCl3) δ:
157.3 (d, J C-F ) 247.2 Hz), 155.0, 133.3, 121.2, 117.0 (d, J C-F
) 9.3 Hz), 115.7 (d, J C-F ) 7.6 Hz), 69.5.
2-Allyl-4-flu or op h en ol (7). A three-necked flask, equipped
with a dropping funnel and a gas inlet tube, was purged with
argon. To a solution of 6 (27 g, 178 mmol) in dry hexane (800
mL), 0.94 M diethlyaluminum chloride in hexane (270 mL, 254
mmol) was added dropwise with stirring at ambient temper-
ature under nitrogen. The mixture was stirred at that tem-
perature for 2 h. The solution was poured into ice-cooled 1 M
hydrochloric acid (1 L). The organic layer was separated and
treated with a solution of 4 M sodium hydroxide. The aqueous
layer was acidified with 6 M hydrochloric acid and extracted
with diethyl ether three times. The combined organic layer
(Figure 8a). On the other hand, the DSC profile of the
polymer obtained by chain-growth polycondensation
showed the endothermic peak at 255 °C (Tm), where the
heating fusion (∆H) was evaluated to be 20 J g-1 (Figure
8b). The complete melting sample was glassified by
rapidly cool quenching to room temperature. The second
heating DSC profile of the polymer showed the endot-
hermic peak at 78 °C (Tg) and the exothermic peak at
172 °C (cold crystallization) with ∆H of 28 J g-1 and
the endothermic peak at 255 °C (Tm) with ∆H of 21 J
g-1. These results distinctly also indicate that the
polymer with low polydispersity obtained by chain-
growth polycondensation possesses higher crystallinity.
This implies that the crystallinity of condensation
polymers could be controlled by polydispersity.
Con clu sion
Our present results demonstrate that potassium
fluorophenolate derivative 1 undergoes chain-growth
polycondensation via electrophilic site propagation to
yield aromatic polyethers having defined molecular

Macromolecules, Vol. 36, No. 13, 2003
Polycondensation for Aromatic Polyethers 4763
F igu r e 7. XRD pattern: (a) poly1 obtained in the presence of 2a at 150 °C in sulfolane: [1]0 ) 0.17 M; [2a ]0 ) 11.7 mM. (b)
poly1 obtained by conventional polycondensation in sulfolane at 150 °C: [1]0 ) 0.17 M.
117.1, 116.5 (d, J C-F ) 13.0 Hz), 116.5 (d, J C-F ) 1.5 Hz), 113.9
(d, J C-F ) 16.0 Hz), 34.9.
4-F lu or o-2-p r op ylp h en ol (8). Into a solution of 7 (14.2 g,
93 mmol) in ethanol (450 mL) was added platinum(IV) oxide
(0.113 g, 0.5 mmol) with stirring at ambient temperature. The
air in the flask was replaced with hydrogen, and the mixture
was stirred under a hydrogen atmosphere for 2 h. Platinum-
(IV) oxide was filtered off, and the solvent was removed in
vacuo to give 14.2 g of 8 (99%). IR (neat): 3407, 2963, 2873,
1620, 1506, 1340, 1222, 1181, 865, 709 cm-1 1H NMR (200
.
MHz, CDCl3) δ: 6.86-6.67 (m, 3H), 5.18 (s, 1H), 2.53 (t, J )
7.7 Hz, 2H), 1.62 (m, 2H), 0.92 (t, J ) 7.4 Hz, 3H). 13C NMR
(50 MHz, CDCl3) δ: 157.2 (d, J C-F ) 236.0 Hz), 149.5, 130.3
(d, J C-F ) 7.6 Hz), 116.4 (d, J C-F ) 22.9 Hz), 115.7 (d, J C-F
25.0 Hz), 112.9 (d, J C-F ) 25.0 Hz), 32.0, 22.7, 13.9.
)
4-ter t-Bu t yld im et h ylsiloxy-3-p r op yl-1-flu or ob en zen e
(9). A solution of 8 (13.6 g, 87.6 mmol) in DMF (35 mL) was
added to a solution of imidazole (7.15 g, 105 mmol) and tert-
butyldimethylsilyl chloride (15.8 g, 105 mmol) in DMF (25 mL),
and the mixture was stirred at ambient temperature for 6 h.
The solution was diluted with water and then extracted with
diethyl ether. The organic layer was washed with 10% sodium
carbonate and water and then dried over Na2SO4. The solvent
was removed in vacuo, and the residue was purified by flash
chromatography on silica gel (hexane) to give 22.6 g of 9 (96%).
IR (neat): 2959, 2859, 1612, 1494, 1226, 1199, 866, 806, 780
1
cm-1. H NMR (200 MHz, CDCl3) δ: 6.88-6.78 (m, 3H), 2.55
(t, J ) 8.0 Hz, 2H), 1.71-1.51 (m, 2H), 1.03 (s, 9H), 0.96 (t, J
) 7.5 Hz, 3H), 0.23 (s, 6H). 13C NMR (50 MHz, CDCl3) δ: 157.3
(d, J C-F ) 238.0 Hz), 149.6, 134.9 (d, J C-F ) 7.6 Hz), 118.9 (d,
J C-F ) 9.2 Hz), 116.4 (d, J C-F ) 22.9 Hz), 112.5 (d, J C-F
22.9 Hz), 32.7, 25.9, 23.1, 18.3, 14.0, -4.2.
)
5-Cya n o-4-flu or o-2-p r op ylp h en ol (11).8,9 A three-necked
flask, equipped with a dropping funnel and a gas inlet tube,
was purged with argon. To a solution of 9 (16.8 g, 62.6 mmol)
in dry THF (300 mL), 0.99 M sec-butyllithium in cyclohexane
(75 mL, 74 mmol) was added dropwise with stirring at -78
°C under dry nitrogen. The mixture was stirred at that
temperature for 1 h. A solution of DMF (15 mL) in dry THF
(30 mL) was added dropwise at -78 °C with stirring, and the
reaction mixture was stirred at that temperature for 1 h. After
10% aqueous HCl (50 mL) was added at -78 °C, the mixture
was warmed to ambient temperature with stirring and then
extracted with diethyl ether three times. The combined organic
layer was washed with water and dried over anhydrous
MgSO4. The solvent was removed in vacuo to give aldehyde
10. A mixture of 10, hydroxylammonium chloride (5.80 g, 66.6
mmol), sodium formate (7.66 g, 111 mmol), and formic acid
(80 mL) was heated at reflux for 2 h. The solution was
F igu r e 8. DSC profiles of poly1: (a) poly1 obtained by
conventional polycondensation in sulfolane at 150 °C: [1]0 )
0.17 M (Mn ) 2240, Mw/Mn ) 1.36). (b) poly1 obtained in the
presence of 2a at 150 °C in sulfolane: [1]0 ) 0.17 M; [2a ]0 )
11.7 mM (Mn ) 2450, Mw/Mn ) 1.13).
was washed with water and dried over anhydrous MgSO4. The
solvent was removed in vacuo, and the residue was distilled
under reduced pressure to give 26.8 g of 7 (100%); bp 97-98
°C/8.4 mmHg. IR (KBr): 3320, 2959, 2859, 1612, 1494, 1254,
1226, 1199, 866, 806, 780 cm-1. 1H NMR (200 MHz, CDCl3) δ:
6.86-6.70 (m, 3H), 6.09-5.89 (m, 1H), 5.19 (dd, J ) 1.5 and
9.8 Hz, 1H), 5.15 (dd, J ) 1.5 and 18.7 Hz, 1H), 4.98 (s, 1H),
3.38 (d, J ) 6.3 Hz, 2H). 13C NMR (50 MHz, CDCl3) δ: 157.3
(d, J C-F ) 238.8 Hz), 149.9, 135.6, 127.1 (d, J C-F ) 6.9 Hz),

4764 Suzuki et al.
Macromolecules, Vol. 36, No. 13, 2003
extracted with ethyl acetate three times. The combined organic
layer was washed with water and dried over anhydrous
MgSO4. The solvent was removed in vacuo, and the residue
was dissolved with diethyl ether again. The etherial layer was
washed with water and dried over anhydrous MgSO4. The
solvent was removed in vacuo, and the residue was purified
by flash chromatography on silica gel (hexane/ethyl acetate )
10/1) and then recrystallized from hexane to give 9.04 g of 11
(91%) as a colorless solid; mp 75-76 °C. IR (KBr): 3386, 2961,
sulfolane (6 mL). The tube was degassed and sealed in vacuo,
followed by heating for 2-5 h at 150 °C. After the tube was
cooled, the solution was added dropwise into vigorously stirred
1 M HCl (50 mL) to afford precipitated polymer. The polymer
obtained was dissolved with dichloromethane, and the solution
was washed with water and dried over anhydrous MgSO4.
After removal of CH2Cl2 in vacuo, the residue was again
dissolved with a small amount of CH2Cl2, followed by pouring
into diethyl ether (100 mL) with vigorous stirring. The
precipitated polymer was collected and dried in vacuo.
Syn th esis of Mod el Com p ou n d s 12 a n d 13. In a glass
tube was placed 1 (0.211 g, 0.97 mmol), 2a (0.638 g, 2.38
mmol), and sulfolane (6 mL). The tube was degassed and
sealed in vacuo, followed by heating for 4 h at 150 °C. After
the tube was cooled, the solution was added dropwise into
vigorously stirred 1 M HCl (50 mL) to afford precipitated
oligomer. The oligomer obtained was dissolved with ethyl
acetate, and the solution was washed with water and dried
over anhydrous MgSO4. After removal of ethyl acetate in
vacuo, the residue was purified by high-performance liquid
chromatography to give 0.128 g of 12 (47%) as a colorless liquid
and 0.087 g of 13 (41%) as a colorless solid; mp 100.4-103.6
°C.
1
2231, 1615, 1189, 911, 798 cm-1. H NMR (200 MHz, CDCl3)
δ: 7.04 (d, J H-F ) 5.1 Hz, 1H), 6.96 (d, J H-F ) 9.3 Hz, 1H),
6.88 (s, 1H), 3.61 (t, J ) 7.6 Hz, 2H), 1.68-1.53 (m, 2H), 0.96
(t, J ) 7.3 Hz, 3H). 13C NMR (50 MHz, CDCl3) δ: 157.5 (d,
J C-F ) 251 Hz), 150.6 (d, J C-F ) 5.4 Hz), 139.0 (d, J C-F ) 6.8
Hz), 117.6 (d, J C-F ) 20.6 Hz), 117.5 (d, J C-F ) 5.4 Hz), 114.5,
97.0 (d, J C-F ) 16.5 Hz), 32.4, 22.1, 13.8. 19F NMR (470 MHz,
CDCl3) δ: 41.7. Anal. Calcd for C10H10FNO: C, 67.03; H, 5.62;
N, 7.82. Found: C, 67.06; H, 5.26; N, 8.05.
P otassiu m 5-Cyan o-4-flu or o-2-pr opylph en olate (1). Into
a solution of 11 (2.00 g, 11.2 mmol) in dry methanol (20 mL)
was added a solution of KOH (0.736 g, 10.9 mmol) in dry
methanol (50 mL). Following the removal of methanol in vacuo,
the residue was washed with dry THF and then dried under
reduced pressure to give 2.37 g (88%) of 1 as a colorless solid.
IR (KBr): 2922, 2228, 1654, 1189, 850. 1H NMR (500 MHz,
DMSO-d6) δ: 6.74 (d, J H-F ) 10.5 Hz, 1H), 6.23 (d, J H-F ) 5.5
Hz, 1H), 2.40 (t, J ) 7.5 Hz, 2H), 1.54-1.45 (m, 2H), 0.88 (t,
J ) 7.5 Hz, 3H). 19F NMR (470 MHz, DMSO-d6) δ: 26.7 (brs,
1F).
12. IR (KBr): 2964, 2235, 1661, 1598, 1494, 1325, 1277,
1226, 1170, 1130, 1065, 1017, 930, 860, 772 cm-1 1H NMR
.
(500 MHz, CDCl3) δ: 7.90-7.83 (m, 4H), 7.76 (d, J H-H ) 8.2
Hz, 2H), 7.22 (d, J H-F ) 5.5 Hz, 1H), 7.18 (d, J H-F ) 9.2 Hz,
1H), 6.99 (d, J H-F ) 8.8 Hz, 2H), 2.64 (t, J ) 7.8 Hz, 2H), 1.68-
1.62 (m, 2H), 0.96 (t, J ) 7.6 Hz, 3H). 13C NMR (50 MHz,
CDCl3) δ: 194.1, 161.3, 160.0 (d, J C-F ) 257.3 Hz), 149.5, 143.8
In itia tor 2a .16 A solution of 4-trifluoromethylbenzoyl chlo-
ride (2.97 mL, 20 mmol), trifluoromethanesulfonic acid (0.02
mL, 0.02 mmol), and fluorobenzene (7.51 mL) was refluxed
for 7 days and then concentrated in vacuo. The residual solid
was purified by recrystallization from ethanol to give 3.57 g
(66%) of 2a as a colorless solid; mp 97-99 °C. IR (KBr): 1597,
1329, 1172, 1130, 1067, 845 cm-1. 1H NMR (200 MHz, CDCl3)
δ: 7.82-7.70 (m, 4H), 7.24-7.13 (m, 4H). 13C NMR (50 MHz,
CDCl3) δ: 194.1, 165.1 (d, J C-F ) 256 Hz), 140.7, 132.8 (d, J C-F
) 9.2 Hz) 132.0, 130.3 (d, J C-F ) 388 Hz), 130.0, 125.5, 125.4
(d, J C-F ) 6.8 Hz), 115.8 (d, J C-F ) 21.3 Hz). 19F NMR (470
MHz, DMSO-d6) δ: 98.8 (s, 3F), 57.1 (m, 1F).
(d, J C-F ) 7.3 Hz), 142.4 (q, J C-F ) 264.9 Hz), 133.7 (d, J C-F
)
33.2 Hz), 132.9, 132.0, 130.0, 125.5, 125.4 (d, J C-F ) 4.1 Hz),
124.4, 116.8 (d, J C-F ) 21.7 Hz), 116.8, 113.4, 99.7, 32.5, 22.7,
13.9. 19F NMR (470 MHz, CDCl3) δ: 99.3 (s, 3F), 50.9 (dd, J F-H
) 5.4 and 9.2 Hz, 1F). Anal. Calcd for C10H10FNO: C, 69.62;
H, 4.47; N, 4.78. Found: C, 69.71; H, 4.39; N, 4.72.
13. IR (KBr): 2963, 2236, 1664, 1598, 1490, 1206, 1190,
1126, 1066, 931, 860 cm-1 1H NMR (500 MHz, CDCl3) δ:
.
7.91-7.85 (m, 4H), 7.77 (d, J H-H ) 8.2 Hz, 2H), 7.30 (s, 1H),
7.19 (d, J H-F ) 9.2 Hz, 1H), 7.10 (d, J H-F ) 5.5 Hz, 1H), 7.03
(d, J H-F ) 8.9 Hz, 2H), 6.78 (s, 1H), 2.71 (t, J ) 7.8 Hz, 2H),
2.58 (t, J ) 7.6 Hz, 2H), 1.73-1.67 (m, 2H), 1.60-1.54 (m, 2H),
0.99 (t, J ) 7.3 Hz, 3H), 0.91 (t, J ) 7.3 Hz, 3H). 13C NMR (50
MHz, CDCl3) δ: 194.0, 161.1, 159.8 (d, J C-F ) 256.3 Hz), 154.7,
149.8, 149.3, 143.1 (d, J C-F ) 14.4 Hz), 141.9 (q, J C-F ) 274.9
Hz), 132.8, 132.6, 131.9, 130.0, 129.9, 125.3, 125.0, 122.1,
119.6, 118.6 (d, J C-F ) 20.7 Hz), 117.0, 116.9, 114.5, 113.3,
102.7, 99.7, 32.6, 32.5, 22.8, 22.5, 13.8, 13.7. 19F NMR (470
MHz, CDCl3) δ: 99.3 (s, 3F), 50.9 (dd, J F-H ) 5.4 and 9.2 Hz,
1F). Anal. Calcd for C24H17F4NO2: C, 67.45; H, 4.01; N, 3.28.
Found: C, 67.32; H, 4.10; N, 2.86.
In itia tor 2c.16 Into a mixture of 4-fluoro-4′-hydroxyben-
zophenone (0.216 g, 1.0 mmol) and potassium carbonate (0.166
g, 1.2 mmol) in dry DMF (2 mL) was added a solution of
iodomethane (75 µL, 1.2 mmol) in dry DMF (0.5 mL) with
stirring at ambient temperature for 6 h. The solution was
poured into water and then extracted with ethyl acetate three
times. The combined organic layer was washed with water and
dried over anhydrous MgSO4. The solvent was removed in
vacuo, and the residue was purified by recrystallization from
ethanol to give 0.154 g of 2c (68%) as a colorless solid; mp
90.3-92.7 °C (lit.16 90-92 °C). IR (KBr): 2985, 1641, 1598,
1259, 1151, 858, 765 cm-1. 1H NMR (500 MHz, CDCl3) δ: 7.78
Ack n ow led gm en t. This work was supported in part
by a Grant-in-Aid (12450377) for Scientific Research
from the Ministry of Education, Science, and Culture,
J apan.
(dd, J H-H ) 8.5 Hz and J H-F ) 5.5 Hz, 2H), 7.78 (d, J H-H
)
8.5 Hz, 2H), 7.13 (t, J H-H ) 8.5 Hz and J H-F ) 8.5 Hz, 2H),
6.95 (d, J H-H ) 8.5 Hz, 2H), 3.87 (s, 3H). 13C NMR (50 MHz,
CDCl3) δ: 194.1, 165.0 (d, J C-F ) 254.3 Hz), 163.2, 134.4, 132.3,
132.2 (d, J C-F ) 8.7 Hz) 130.0, 115.3 (d, J C-F ) 21.6 Hz), 113.6,
55.4.
Refer en ces a n d Notes
Con ven tion a l P olym er iza tion of 1. In a glass tube was
placed 1 (0.217 g, 1.0 mmol) and sulfolane (3 mL). The tube
was degassed and sealed in vacuo, followed by heating for 4 h
at 220 °C. After the tube was cooled, the solution was added
dropwise into vigorously stirred 1 M HCl (50 mL) to afford
precipitated polymer. The polymer obtained was dissolved with
dichloromethane, and the solution was washed with water and
dried over anhydrous MgSO4. After removal of CH2Cl2 in
vacuo, the residue was again dissolved with a small amount
of CH2Cl2, followed by pouring into diethyl ether (100 mL) with
vigorous stirring. The precipitated polymer was collected and
dried in vacuo. IR (KBr): 2925, 2239, 1621, 1223, 889, 864,
(1) Cotter, R. J . Engineering Plastics: A Handbook of Polyaryl
Ethers; Gordon and Breach Publishers: Amsterdam, 1995.
(2) Haym, A. S. J . Polym. Sci., Part A: Polym. Chem. 1998, 36,
505.
(3) (a) Maiti, S.; Mandal, B. Prog. Polym. Sci. 1986, 12, 111. (b)
Bruma, M.; Fitch, J . W.; Cassidy, P. E. J . Macromol. Sci.,
Rev. Chem. Phys. 1996, C36, 119.
(4) (a) Percec, V.; Shaffer, T. D. J . Polym. Sci., Part C: Polym.
Lett. 1986, 24, 439. (b) Percec, V.; Wang, J . H. J . Polym. Sci.,
Part A: Polym. Chem. 1991, 29, 63. (c) Percec, V.; Wang, J .
H. Polym. Bull. (Berlin) 1990, 24, 493. (d) Wang, J . H.; Percec,
V. Polym. Bull. (Berlin) 1991, 25, 33.
(5) Yokozawa, T.; Asai, T.; Sugi, R.; Ishigooka, S.; Hiraoka, S.
1
749 cm-1. H NMR (500 MHz, CDCl3): δ 7.16 (s, 1H), 6.83 (s,
J . Am. Chem. Soc. 2000, 122, 8313.
1H), 2.67-2.64 (br, 2H), 1.63-1.59 (br, 2H), 0.99-0.85 (br, 3H).
Ch a in -Gr ow th P olycon d en sa tion of 1 w ith 2. In a glass
tube was placed 1 (0.217 g, 1 mmol), 2 (2-100 mol %), and
(6) For examples of polycondensation in which the reaction of
monomer with polymer end group is faster than that of
monomers with each other, although the polymerization

Macromolecules, Vol. 36, No. 13, 2003
Polycondensation for Aromatic Polyethers 4765
behavior does not show the character of living polymerization,
see: (a) Lenz, R. W.; Handlovits, C. E.; Smith, H. A. J . Polym.
Sci. 1962, 58, 351. (b) Newton, A. B.; Rose, J . B. Polymer
1972, 13, 465. (c) Risse, W.; Heitz, W. Makromol. Chem. 1985,
186, 1835. (d) Robello, D. R.; Ulman, A.; Urankar, E. J .
Macromolecules 1993, 26, 6718. (e) Yokozawa, T.; Shimura,
H. J . Polym. Sci., Part A: Polym. Chem. 1999, 37, 2607.
(7) For a preliminary report, see: Yokozawa, T.; Suzuki, Y.;
Hiraoka, S. J . Am. Chem. Soc. 2001, 123, 9902.
(12) Baxter, I.; Ben-Haida, A.; Colquhoun, H. M.; Hodge, P.;
Kohnke, D. J .; Williams, D. J . Chem. Commun. 1998, 2213.
(13) (a) Ercolani, G.; Mandolini, L.; Mencarelli, P.; Roelens, S. J .
Am. Chem. Soc. 1993, 115, 3901. (b) Colquhoun, H. M.; Zhu,
Z.; Williams, D. J . Org. Lett. 2001, 3, 4031.
(14) (a) Lozano, A. E.; J imeno, M. L.; Abajo, J .; Camp, J . G.
Macromolecules 1994, 27, 7164. (b) Carter, K. R. Macromol-
ecules 1995, 28, 6462.
(15) The 19F NMR spectrum of polymer was measured in CDCl3
because the polymer was not soluble in DMSO. However, it
was confirmed that there was little difference of the chemical
shift of the signal b of 2a between in DMSO-d6 and in CDCl3.
(16) (a) Ridd, J . H.; Yousaf, T. I.; Rose, J . B. J . Chem. Soc., Perkin
Trans. 2 1988, 1729. (b) Lovering, J . R.; Ridd, J . H.; Parker,
D. G.; Rose, J . B. J . Chem. Soc., Perkin Trans. 2 1988, 1735.
(8) (a) Furlano, D. C.; Calderon, S. N.; Chen, G.; Kirk, K. L. J .
Org. Chem. 1988, 53, 3145. (b) Kirk, K. L.; Olubajo, O.;
Buchhold, K.; Lewandowski, G. A.; Gusovsky, F.; McCulloh,
D.; Daly, J . W.; Creveling, C. R. J . Med. Chem. 1986, 29, 1982.
(9) van Es, T. J . Chem. Soc. 1965, 1564.
(10) The Mn values of polymer were estimated by the 1H NMR
spectra based on the ratios of signal intensities of the
repeating units to the initiator unit.
(17) St. Greorgiev, V.; Mullen, G. B. U.S. Patent 42727157 A, 1985.
(11) When the [1]0/[2a ]0 ratios were 25 or above, the polymer was
precipitated during polymerization.
MA021553S