R. B. Sandor, B. M. Foxman / Tetrahedron 56 (2000) 6805±6812
6811
colorless crystalline needles. The product was then recrys-
tallized from 15 mL of benzene giving 1.50 g (6.68 mmol,
68.8%) of 3,5-dichloro-4-aminobenzoylchloride 1, m.p.
1628C. (m.p. 3,5-dibromo-4-aminobenzoylchloride 2,
1808C; m.p. 3,5-diiodo-4-aminobenzoylchloride 3, 1908C
(decomp.)). The melting point test must be done rapidly
because these products tend to polymerize and the polymer-
ized products do not melt. 3,5-Dibromo- and 3,5-dichloro-4-
aminobenzoylchloride crystallize as needles elongated
along the c-axis, but 3,5-diiodo-4-aminobenzoylchloride
crystallizes as thin plates. The 3,5-dichloro- derivative
crystallizes in two polymorphic modi®cations. The crystal
structure of modi®cation 1b is isomorphous with the deter-
mined structure of the 3,5-dibromo-derivative. The other
polymorph 1a has a similar unit cell to that of 1b, but
with a equal to one-half of the value for 1b and 2. Both
modi®cations were crystallized from benzene, and the over-
whelming majority of crystallizations yielded modi®cation
1b; production of crystals of 1a appeared to be a fortuitous
occurrence. No reliable synthesis for the preparation of
modi®cation 1a could be created. Cell constants for
modi®cation 1b were determined from 11 re¯ections and
dichloro-p-benzamide) derived from the solid-state reaction
of 1a indicate that the polymers are partially crystalline.
The powder pattern of the polymer has only four lines (at
6.37, 5.45, 3.73 and 2.75 A), but these are distinct from the
powder pattern of the monomer (vide supra).
Ê
X-Ray structure determination
After preliminary studies of 1a, 2 and 3 by oscillation and
Weissenberg techniques, fresh crystals of each were
mounted on a Pyrex ®ber af®xed to a brass pin, then trans-
ferred to a Supper No. 455 goniometer and optically
centered on a Syntex P21 diffractometer. Routine operations
were performed as described previously;20 other operations
are described below. The structures were solved using the
heavy-atom method; preliminary computational work was
carried out using the Enraf-Nonius molen software
package.21 Final re®nement was carried out using the
Oxford University crystals package;22 drawings were
produced using cameron.23 The structure analysis of 1a
was complicated by disorder of the molecule about a
pseudo-mirror plane at z0 in Pna21. Atom Cl(2) was
sited on the pseudo-mirror plane, and re®ned at full occu-
pancy. Other atoms in each component were constrained to
have the same occupancy, with the occupancies of both
components constrained to sum to 1.0; the ®nal value for
the major component was 0.753(8). The minor component
was constrained to be related to the major component by the
pseudo-mirror; isotropic displacement parameters in the
major and minor components were constrained to be
equal. The C(7)±Cl(2) distance was restrained to
re®ned 2u, v, x values in the range 9#u2uu#138 (l(Mo
Ê
Ka0.71073 A):
a25.178(5),
b17.708(4),
c
Ê
Ê 3
3.785(1) A, and V1687.79 A ; for 1a, the cell constants
were (12^(hkl) and re®ned 2u, v, x values in the range
Ê
Ê
11#u2uu#238): a12.599(3), b17.691(4), c3.791(1) A,
Ê 3
and V845.0 A . X-Ray powder diffraction for 1b: (A)
6.03m, 5.18w, 4.32m, 4.22s, 4.07s, 3.99w, 3.50s, 3.31ms,
3.20ms, 3.14s, 3.02m, 2.92m, 2.82ms, 2.75ms, 2.54m,
2.46s, 2.40m, 2.33w, 2.23w, 2.03s, 1.87w, 1.80w.
Ê
1.79(3) A, and the C±C distances in the phenyl moiety
Thermal polymerization of crystalline 3,5-dihalo-4-
aminobenzoylchlorides
Ê
were restrained to 1.397(50) A. The phenyl group was
Ê
restrained to be planar within 0.05 A. The ®nal re®nement
included anisotropic displacement parameters for the major
component Cl atoms, and isotropic displacement parameters
for other nonhydrogen atoms. The coordinates of the H atom
attached to N were re®ned (Uiso ®xed at 0.05); phenyl H
atoms were ®xed at calculated positions (494 data for which
I.1.96s(I); 70 parameters). The enantiopole parameter
could not be successfully re®ned using the present data
set. The ®nal re®nement for 2 was carried out using aniso-
tropic displacement parameters for Br, N, O and Cl; C atoms
were re®ned using isotropic displacement parameters (1256
data for which I.1.96s(I); 156 parameters). Phenyl hydro-
gen atoms were included at ®xed positions. H atoms
The crystalline solid was placed in a 1 cm test tube equipped
with a side arm. The test tube was heated in a bath of
silicone oil and maintained at a constant temperature
under dry nitrogen. Depending on the particular experiment
the temperature was kept between 95 and 3008C. The
contents of the tube became opaque but did not melt. The
solid polymer retained the habit of the monomer crystals. In
the case of the 3,5-diiodo-derivative, the contents also
turned brown. The reason for the color change in the case
of the 3,5-diiodo-derivative may indicate decomposition,
with partial release of iodine. The reaction could be
followed by changes in infrared spectra. The monomer IR
spectra show the characteristic doublets of a primary amine.
One absorption band of the doublet is near 3500 cm21, and
the other is near 3400 cm21. A strong absorption near
1740 cm21 due to the CvO stretching frequency is typical
of conjugated acid chlorides. The spectra of the polymers
show reduced intensity of the primary amine doublet and the
growing in of a broad absorption due to the N-arylated
amide proton of the polymer, along with replacement of
the strong CvO stretch of the acid chlorides near
Ê
attached to N were restrained (N±H, 0.88(5) A; H±N±H,
109.5(5), DUiso (N(1)±H(1)0.0(5)). Only one of the two H
atoms attached to N(1) could be located. The absolute polar-
ity of the crystal could not be established; a value of 0.43(8)
for the enantiopole parameter was obtained. For 3, a linear
decay correction (42.4% maximum) was applied to the data.
The ®nal re®nement was carried out using anisotropic
displacement parameters for all nonhydrogen atoms (1445
data for which I.1.96s(I); 118 parameters). Phenyl hydro-
gen atoms were included at ®xed positions, while H atoms
1740 cm21 by
a band characteristic of amides at
,1660 cm21. The polymerization could be followed more
quantitatively by weighing a heated sample after speci®c
times. Owing to the loss of HCl(g), the weight difference
may be used to calculate percent conversion. Table 3
illustrates time±temperature-conversion data for com-
pounds 1 and 3. X-Ray powder patterns taken of poly(3,5-
Ê
attached to N were restrained (N±H, 0.88(5) A; H±N±H,
109.5(5)). Coordinates for the structures of 1a, 2 and 3 have
been deposited with the Cambridge Crystallographic Data
Centre. The coordinates can be obtained on request from the
Director, Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK.
Sandor, Robert B
