organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
respectively. Relevant torsion angles in the related structures
(II), 5-n-butyl-3-hydroxymethyl-6-methylcyclohexen-4-ol (Bat-
ey et al., 1999), (+)-(1S,2S,3S,6R,10S)-methyl-2-(1-hydroxy-
ethyl)-3-hydroxymethyl-6-methyl-4-cyclohexene-1-carboxyl-
ate and (+)-(1S,2S,3S,6R,10S,100R)-methyl-2,3-bis(1-hydroxy-
ethyl)-6-methyl-4-cyclohexene-1-carboxylate (Ainsworth et
al., 1995) range from 5.9 to 25.4ꢁ. Notably, the DFT-calculated
torsion angles C2ÐC1ÐC6ÐC5 (13.9ꢁ) and C1ÐC2ÐC3Ð
C4 (13.9ꢁ) of cyclohexene, (III), fall in the middle of this range.
Several statistically signi®cant differences are observed in
the chemically equivalent bond lengths and torsion angles of
ISSN 0108-2701
The lack of C2 molecular symmetry in
(1R,2R,3S,6S)-3,6-dibenzyloxycyclo-
hex-4-ene-1,2-diol
Ê
Ê
(I). The O2ÐC4 distance [1.428 (2) A] is 0.010 A longer than
Ê
Robert W. Clark, Ilia A. Guzei,* Sergei A. Ivanov,
Steven D. Burke and William T. Lambert
the related O3ÐC5 distance [1.418 (2) A]. In 4574 relevant
compounds containing 8321 Csp3ÐOH bonds reported to the
Cambridge Structural Database (CSD; Allen & Kennard,
Chemistry Department, University of Wisconsin±Madison, 1101 University Avenue,
Madison, WI 53706, USA
Ê
Ê
1993), the corresponding values averaged 1.424 (15) A.
Ê
Additionally, the O4ÐC14 distance [1.430 (2) A] is 0.016 A
Correspondence e-mail: iguzei@chem.wisc.edu
Ê
longer than O1ÐC7 [1.414 (2) A], and the torsion angles
Received 2 March 2001
Accepted 10 April 2001
O1ÐC7ÐC8ÐC13 [ 31.6 (3)ꢁ] and O4ÐC14ÐC15ÐC16
[ 56.5 (3)ꢁ] are substantially different. To account for these
discrepancies, several DFT geometry optimizations were
performed on (I) and (III). The results of calculations for one
molecule of (III) verify its optimal geometry to be C2
symmetric. However, this is not observed in the case of (I).
The DFT-calculated molecular parameters of (I) are in close
agreement with the experimentally observed values. One
exception to this is that the calculated O1ÐC7ÐC8ÐC13 and
O4ÐC14ÐC15ÐC16 torsion angles are 41.0 and 69.4ꢁ,
respectively. Although ꢀ-stacking interactions are not
observed, other crystal-packing forces likely contribute to this
difference. To test the hypothesis that the C2-symmetric
geometry of (I) is not the lowest in energy, DFT calculations
were carried out for (I) starting from the symmetrical
conformation and consecutively lifting all of the symmetry
constraints. In the progress of optimization, the molecule
departed from the symmetrical conformation. Additionally,
The results of a single-crystal X-ray experiment and density
functional theory calculations performed for the title
compound, C20H22O4, demonstrate that the lowest energy
conformation of this molecule does not contain C2 molecular
symmetry.
Comment
We have synthesized the chiral title molecule, (I), during the
course of our work on the synthesis of natural products related
to marine sponge extracts. We report herein the structure of
(I), discuss its molecular symmetry and present the results of
density functional theory (DFT) calculations (Schrodinger
Inc., 1998; all geometry optimizations were performed with the
B3LYP hybrid functional and Pople basis set 6±3116\*\*).
È
DFT calculations were performed on (I) with two additional
1
water molecules ®xed at the observed O3Á Á ÁO2(1 x, y
,
2
z) and O2Á Á ÁO3(1 x, y + 12 , 12 z) distances to simulate
1
2
possible hydrogen bonding in the structure. This structure
optimization did not fully converge. [The maximum displace-
3
The absolute stereochemistries of the chiral centers were
assigned as 1R, 2R, 3S and 6S from knowledge of the synthesis.
The benzyloxy groups occupy pseudo-equatorial positions
while the hydroxyl substituents are located in axial positions.
Unfavorable steric interactions are minimized when the bulky
substituents occupy pseudo-equatorial positions and this
feature is similarly observed in the related compounds
3,5-dicyano-6-(2-methoxy-1,1,2-trimethylpropyl)cyclohexene,
(II), cis-1,3-dicyano-4-(2-methoxy-1,1,2-trimethylpropyl)cyclo-
hexene and cis-1,5-dicyano-4-(2-methoxy-1,1,2-trimethyl-
propyl)cyclohexene (Borg et al., 1984).
ment (2.13 Â 10 2) and r.m.s. displacement (7.88 Â 10
)
3
values were above the standard threshold values of 1.8 Â 10
and 1.2 Â 10 3, respectively, which in turn is indicative of a ¯at
minimum on the potential energy surface.] Consequently, it is
concluded that hydrogen bonding probably does not contri-
bute signi®cantly to this symmetry lowering.
Weak hydrogen-bonding interactions between the hydroxyl
substituents of symmetry-related molecules in the lattice of (I)
are likely. An intermolecular hydrogen-bonding interaction is
observed between donor atom O3 and acceptor atom
1
1
O2(1 x, y
,
z) (Table 2). The corresponding distances
and angles for 2222 compounds with 3998 similar hydrogen
2
2
The conformation of the cyclohexene ring in (I) is a half-
Ê
chair. Atoms C1, C2, C3 and C6 are planar within 0.02 A.
Ê
bonds in structures reported in the CSD were 2.79 (9) A and
Atoms C4 and C5 are located 0.372 (4) and 0.388 (4) A above
and below this plane, respectively. The C2ÐC1ÐC6ÐC5 and
C1ÐC2ÐC3ÐC4 torsion angles are 18.7 (3) and 18.1 (3)ꢁ,
166 (7)ꢁ. The longer OÁ Á ÁO separation in (I) is indicative of a
weaker hydrogen bond. Interestingly, the chemically equiva-
lent intermolecular O2Á Á ÁHÐO3 hydrogen-bonding interac-
Ê
ꢀ
844 # 2001 International Union of Crystallography
Printed in Great Britain ± all rights reserved
Acta Cryst. (2001). C57, 844±845
Clark
