organic compounds
118.37 (3)ꢀ. (Note: transformation of their triclinic cell yields
ꢀ = 119.11ꢀ, which more closely matches our value.) Their
model, deposited as the `monoclinic form' (refcode
VEXFUZ01) in the Cambridge Structural Database (Allen &
Kennard, 1993) has the molecule on a twofold axis, which
requires a similar disorder in the formyl group. This model
produced worse R values (R = 0.145 and wR = 0.151). Despite
intensity statistics suggesting a centrosymmetric structure,
Feulner et al. (1990) also attempted structure solution in space
group Cc, but were unsuccessful for reasons which are unclear.
Our structure of (II), with Z0 = 1 in space group Cc (Fig. 1),
exhibits none of these troubling features. The formyl group is
ordered, and the H atoms of the B(OH)2 group are ordered
and in sensible positions for intermolecular hydrogen bonds
(Table 2). The packing (Fig. 2) exhibits a pseudocenter near
Acta Crystallographica Section C
Crystal Structure
Communications
ISSN 0108-2701
Space-group revision for 4-formyl-
phenylboronic acid
Frank R. Fronczek,* Nadia N. St Luce and Robert M.
Strongin
Department of Chemistry, Louisiana State University, Baton Rouge, LA 70803-1804,
USA
Correspondence e-mail: fronz@chxray.chem.lsu.edu
1 1 1
( , , ) and a pseudo-twofold axis near (12,y,43), running along
2 2 2
Received 10 September 2001
Accepted 24 September 2001
the long axis of the molecule. The two molecules near the
center of the cell are related by the c glide, and are approxi-
mately related by the pseudocenter. Treating the center as
exact rather than the glide leads to the P1 model, while
treating both the center and the glide as exact leads to the C2/c
model. The cause of the disordered formyl group and boronic
acid H atoms in the P1 model can be seen in Fig. 2 by
examination of the relative orientations of the two glide-
The space group of the title compound, C7H7BO3, previously
reported to be P1, is properly Cc. There is no disorder of the
formyl group or in the H atoms of the B(OH)2 group.
Molecules lie on approximate twofold axes and are related by
approximate centers, which relate all but the formyl O atom
and boronic acid H atoms. The BÐO distances are 1.363 (2)
1 1 1
related molecules about ( , , ). The formyl O and boronic acid
2 2 2
Ê
and 1.370 (2) A.
H atoms do not conform to the inversion, while the remainder
of the molecule nearly does. The pseudosymmetry does not
lead to exceptionally high correlations, with the largest being
0.63, between displacement parameters of atoms related by
the approximate twofold axis.
We have ruled out the possibility that a phase change on
cooling causes the difference between the Cc structure which
we observe at 120 K and that reported by Feulner et al. (1990)
at room temperature. Using the same crystal, we collected
intensity data at 296 K and obtained the same Cc structure,
with cell dimensions a = 11.1932 (4), b = 9.8820 (5) and c =
Comment
During the course of studying the structure and mechanism of
formation of colored products in resorcinarene solutions
(Davis et al., 1999; Lewis et al., 2000), the model compound (I)
was investigated. Thermolysis of (I) led to the formation of the
title compound, (II), and its structure was determined to
ascertain its identity.
ꢀ
Ê
7.3373 (3) A, ꢀ = 119.336 (3) and R = 0.043. Using this data
set, we were also able to reproduce the results of Feulner et al.
(1990), re®ning their P1 model to R = 0.099.
The structure of the molecule itself is unremarkable. The
formyl group is essentially coplanar with the phenyl ring, while
the B(OH)2 group is rotated by 20.6 (3)ꢀ out of the phenyl
plane. One hydroxyl H atom is syn to the phenyl group, while
the other is anti, as is typical for phenylboronic acids (Bradley
et al., 1996; Gainsford et al., 1995; Pilkington et al., 1995; Shull
The published crystal structure of (II) (Feulner et al., 1990)
is in space group P1 with Z0 = 1 at 293 K, and has some
unsettling features. In this model, the CHO group has a
twofold disorder which superimposes its CÐH and C
O
bonds. The boronic acid H-atom positions are not sensible for
the expected hydrogen bonding, and form impossibly short
intermolecular HÁ Á ÁH contacts. The model ®ts the data poorly
(R = 0.097 and wR = 0.181), despite the fact that this
compound forms high quality crystals. Furthermore, the
triclinic cell can be transformed (011,011,100) to a C-centered
cell with a monoclinic metric. Feulner et al. (1990) recognized
this transformation, and attempted a structure solution in C2/c
with Z0 = 12. Their reported C-centered cell has dimensions a =
Figure 1
A view of the molecule of (II) with the atom-numbering scheme and
displacement ellipsoids at the 50% probability level. H atoms are shown
as small spheres of arbitrary radii.
Ê
11.177 (5), b = 9.891 (4) and c = 7.339 (4) A, and ꢀ =
ꢁ
Acta Cryst. (2001). C57, 1423±1425
# 2001 International Union of Crystallography
Printed in Great Britain ± all rights reserved 1423