1
212
Russ.Chem.Bull., Int.Ed., Vol. 50, No. 7, July, 2001
Salazkin et al.
in the ortho-position to the Ñ=Î group in molecule 1b
is more than 2.5-fold more active than the second
halogen atom in the para-position in molecule 1a. It
follows from this that 1a containing almost no
The molecular weight distribution and mean molecular
weight of polymers were determined by GPC using a previously
1
,3
described procedure.
sured by the light scattering method on
The mean molecular weight was mea-
Fica photo-
a
goniodiffusimeter at 25 °Ñ and λ = 546 nm in the vertically
polarized light using the double extrapolation according to
2
,4´-isomer should be used for the synthesis of high-
molecular PAEK.
Simm. The instrument was calibrated by benzene.
A comparison of the glass transition temperatures
1Í and 13Ñ NMR spectra for compounds 1à,b and PAEK
(
Ò ) and some physicomechanical properties of the
were recorded on an Bruker ÀÌÕ-400 instrument (400.13 and
100.61 MHz, respectively) for solutions in CDCl3 using Me4Si
as the internal standard. Signals in the Í and Ñ NMR
spectra were assigned according to calculated data obtained by
the additive scheme. 19F NMR spectra for 1à,b were recorded
on a Bruker WP-200-SY spectrometer (188.31 MHz) for solu-
g
films (the forced elasticity limit (σ ), strength at the
rupture moment calculated per initial cross section of
the sample (σ), deformation at which the forced elastic-
0
1
13
ity limit appears (ε ), relative elongation upon rupture
0
(
ε ), strength at the rupture moment calculated per true
tions in CDCl3 using ÑF COOH as the external standard.
3
cross-section of the sample (σ ), and high-elasticity
t
IR spectra were recorded on a PerkinElmer-457 spec-
trometer.
The reduced viscosity (ηred) of PAEK solutions was mea-
sured in CHCl3 (at a concentration of 0.5 g of the polymer per
modules (E )) of the synthesized amorphous homo- and
co-PAEK shows that the presence of fragments 1b in
them has no substantial effect on the studied properties
of these polymers compared to the 1a-based ho-
mopolymers.
1
00 mL of the solvent) at 25 °Ñ.
Thermochemical tests of polymers were carried out accord-
ing to a previously described procedure.4
The conformational properties of the macromol-
ecules containing fragments 1a and 1b differ most sig-
nificantly. This is expressed, first, in a decrease in ηred
of polymer solutions at the same MW (see Table 3);
second, in underestimated Mw values of the 1b-based
polymers determined by GPC and calculated using the
calibration for the 1a-based polymers (see Table 3)
compared to the absolute Mw values measured by the
light scattering method; third, in an increase in the
content of cyclic oligomers in reaction mixtures (cyclic
dimer, from 2.6 to 3.8% and trimer, from 1.3 to 1.8%).
The results of the detailed study of cycle formation in
these PAEK will be published elsewhere. The data pre-
sented in this work indicate a more closed conformation
of the chains of the 1a-based PAEK macromolecules,
which should be expected from a comparison of the
chemical structure of PAEK based on 1a and 1b. It can
be assumed that a higher affinity to cycle formation
during the synthesis of the 1b-based polymers and a
more closed conformation of the chains of the PAEK
macromolecule prevent, along with the low activity of
the î-halogen atom in 1b, the preparation of PAEK
with higher MW.
4
,4´-Difluorobenzophenone (1à) was synthesized by a pro-
cedure5 modified by us, its m.p. was 107.5108.0 °Ñ after
3
sublimation (cf. Ref. 6: m.p. 107.5108.0 °Ñ). The isomeric
purity was monitored by HPLC (content of 1b << 0.3 wt.%).
IR, ν/cm1: 1646 (C=O). 1H NMR (CDCl3), δ: 7.10 (dd, 4 Í,
Í(4), H(6), 3J
= 8.8 Hz, 3JF,Í = 8.8 Hz); 7.76 (dd, 4 Í,
Í,Í
Í(3), H(7), 3J
= 8.8 Hz, 3
J
= 5.2 Hz). 13Ñ NMR
F,Í
Í,Í
2
(CDCl ), δ: 115.26 (Ñ(4), C(6), JÑ,F = 21.7 Hz); 132.24
3
(
1
Ñ(3), C(7), JÑ,F = 9.2 Hz); 133.41 (Ñ(2), 4JÑ,F = 2.9 Hz);
3
66.35 (Ñ(5), 1JÑ,F = 254.6 Hz); 193.43 (Ñ(1)). 19F NMR
(
CDCl ), δ: 28.03 (s, 1 F, Ñ(5)F) (for the numeration of
3
atoms, see Scheme 1).
2,4´-Difluorobenzophenone (1b) was synthesized by the re-
action of 2-fluorobenzoyl chloride with fluorobenzene in the
3
presence of AlCl3 and FeCl3 by a procedure modified by us,
m.p. 22.022.5 °Ñ (cf. Ref. 7: m.p. 22.823.6 °Ñ). IR, ν/cm1:
1
1
3
664 (C=O). H NMR (CDCl ), δ: 7.13 (d, 2 Í, Í(4), H(6),
3
3
JÍ,Í = 8.8 Hz, JF,Í = 8.8 Hz); 7.21 (t, 1 Í, Í(10),
3J
= 8.8 Hz, 3J
= 8.8 Hz); 7.26 (t, 1 Í, Í(12),
F,Í
= 8.8 Hz); 7.53 (m, 2 Í, Í(11), H(13)); 7.88 (dd, 2 Í,
= 5.6 Hz). 13Ñ NMR
Í,Í
3J
Í,Í
Í(3), H(7), 3JÍ,Í = 8.8 Hz, 3
J
F,Í
2
(CDCl3), δ: 115.37 (Ñ(4), C(6), JC,F = 22.1 Hz); 115.99
2
4
(
(
(
1
1
Ñ(10), J
= 21.3 Hz); 124.15 (Ñ(12), J
= 3.6 Hz); 126.44
= 2.8 Hz); 132.19
C,F
C,F
Ñ(8), 2JC,F = 14.9 Hz); 130.57 (Ñ(13), 3J
C,F
Ñ(3), C(7), 3JC,F = 9.2 Hz); 132.97 (Ñ(11), 3JC,F = 8.0 Hz);
33.46 (Ñ(2), 4J
= 2.4 Hz); 159.62 (Ñ(9), 1JC,F = 252.2 Hz);
C,F
65.65 (Ñ(5), 1JC,F = 255.8 Hz); 191.50 (Ñ(1)). 19F NMR
Experimental
(
CDCl ), δ: 33.34 (s, 1 F, Ñ(9)F); 26.62 (s, 1 F, Ñ(5)F) (for
3
the numeration of atoms, see Scheme 1).
An admixture of 1b in 1à was analyzed by normal-phase
HPLC using a Milikhrom-1 instrument with a spectrophoto-
metric detector on a microcolumn (5 cm ½ 2 mm) packed with
the Silasorb-600 sorbent in a mixture of methylene chlo-
ridehexane (3 : 1 or 2 : 1). The sample volume was 510 µL,
and the concentration was 12 mg mL . The method allows
the separation of the peaks of 1à and 1b to the basic line with
the resolution coefficient R > 1 and the quantitative determina-
tion of the concentration of each component. The peaks were
identified by the specially synthesized pure isomers, and the
response coefficients were calibrated at different λ. The average
reproducibility of the results of determination of the content of
2,2-Bis(4´-hydroxyphenyl)propane (2à) (trade mark A, Ufa
PO Khimprom, Russia) was additionally purified according to a
previously described procedure.8
3,3-Bis(4´-hydroxyphenyl) phthalide (phenolphthalein) (2b)
("farmakopeinyi" trade mark, Moscow Alkaloid Plant) with
m.p. 260.5261.0 °C was dried for 4 h at 120 °C.
1
Synthesis of PAEK. Argon was passed through a four-
necked flask with a stirrer, a pipe for argon supply, and a
system for azeotropic water distillation. Then compound 1à or
1b (21.8 g, 0.1 mol), compound 2a (22.8 g, 0.1 mol), prelimi-
narily powdered and calcined Ê ÑÎ (18 g, 0.13 mol), N,N-di-
2
3
methylacetamide (200 mL), and PhCl (100 mL) were loaded
into the flask. The temperature of the oil bath was gradually
(within ∼0.5 h) increased to 185 °Ñ. The duration of the
synthesis after the completion of the azeotropic water distilla-
1
b by two measurements in a mixture was 0.2%. The procedure
allowed the measurement of the concentration of 1b to 0.3%
and detection of admixtures up to <0.3%.