The effect of a para substituent on the conformational preference of 2,2-diphenyl-1,3-dioxanes: Evidence for the anomeric effect from X-ray crystal structure analysis

Article abstract of DOI:10.1021/jo970742j

The molecular structures of 2,2-di(para-substituted phenyl)-1,3-dioxanes were elucidated for the first time by X-ray crystallographic analysis, which revealed two important structural features: (1) These compounds have the chair conformation in which electron-withdrawing aryl groups [viz. p-nitro-or p-(trifluormethyl)phenyl] are always axial and electron-donating aryl groups (viz. p-methoxyphenyl) are always equatorial. (2) In these compounds as well as in symmetrically substituted 2,2-diphenyl-1,3-dioxane the axial C₂-aryl bond is longer than the equatorial C₂-aryl bond. The axial preference of the electron-withdrawing aryl group was also demonstrated in solution by ¹H and ¹³C NMR spectroscopy. The anomeric carbon substituted with an electron-withdrawing aryl group resonates at an unusually high field, as does the aromatic carbon bearing the electron-withdrawing substituent. The observed ¹³C NMR data clearly indicate enhanced electron density at these carbons due to the anomeric effect. Semiempirical molecular orbital calculations by the MOPAK PM3 method reproduced the bond lengthening for axial C₂-aryl, while the heat of formation derived from this calculation failed to support the axial preference of electron-withdrawing aryl groups. The X-ray crystallographic data on the conformational preference and bond lengths at the anomeric carbon, as well as the solution NMR spectroscopic data, clearly indicate the anomeric effect that is best rationalized in terms of stabilizing interaction between the lone-pair electrons on the ring oxygens and the antibonding orbital of the axial C₂-aryl bond.

Full text of DOI:10.1021/jo970742j

1
436
J . Org. Chem. 1999, 64, 1436-1441
Articles
Th e Effect of a P a r a Su bstitu en t on th e Con for m a tion a l
P r efer en ce of 2,2-Dip h en yl-1,3-d ioxa n es: Evid en ce for th e
An om er ic Effect fr om X-r a y Cr ysta l Str u ctu r e An a lysis1
†
,†,‡
†
§
Fumiaki Uehara, Masayuki Sato,* Chikara Kaneko, and Hiroyuki Kurihara
Pharmaceutical Institute, Tohoku University, Aobayama, Sendai 980-77, J apan, and Tsukuba Research
Laboratories, Yamanouchi Pharmaceutical Co. Ltd., 21 Miyukigaoka, Tsukuba, Ibaraki 305, J apan
Received April 25, 1997
The molecular structures of 2,2-di(para-substituted phenyl)-1,3-dioxanes were elucidated for the
first time by X-ray crystallographic analysis, which revealed two important structural features:
(1) These compounds have the chair conformation in which electron-withdrawing aryl groups [viz.
p-nitro- or p-(trifluormethyl)phenyl] are always axial and electron-donating aryl groups (viz.
p-methoxyphenyl) are always equatorial. (2) In these compounds as well as in symmetrically
substituted 2,2-diphenyl-1,3-dioxane the axial C
2
-aryl bond is longer than the equatorial C -aryl
2
bond. The axial preference of the electron-withdrawing aryl group was also demonstrated in solution
1
13
by H and C NMR spectroscopy. The anomeric carbon substituted with an electron-withdrawing
aryl group resonates at an unusually high field, as does the aromatic carbon bearing the electron-
withdrawing substituent. The observed 13C NMR data clearly indicate enhanced electron density
at these carbons due to the anomeric effect. Semiempirical molecular orbital calculations by the
MOPAK PM3 method reproduced the bond lengthening for axial C -aryl, while the heat of formation
2
derived from this calculation failed to support the axial preference of electron-withdrawing aryl
groups. The X-ray crystallographic data on the conformational preference and bond lengths at the
anomeric carbon, as well as the solution NMR spectroscopic data, clearly indicate the anomeric
effect that is best rationalized in terms of stabilizing interaction between the lone-pair electrons
on the ring oxygens and the antibonding orbital of the axial C -aryl bond.
2
-7
In tr od u ction
The tendency of electronegative substituents X at C
from experimental and theoretical viewpoints.5 The
best evidence for n -σ* delocalization is crystallographic
data showing that species with O-C -X (axial) frag-
ments have shorter C -O bonds and longer C
O
2
2
of tetrahydropyranes 1 to occupy an axial position is
8
,9
2
2
2
-X bonds.
termed the anomeric effect. Its origin may be found in
These changes in bond length cannot be rationalized on
the basis of electrostatic effects, which suggests that
delocalization alone sufficiently accounts for the observa-
tions.2
destabilization of equatorial substituents due to unfavor-
3
able dipole-dipole (electrostatic) interactions. An alter-
native view is stabilization due to favorable overlap
between the orbital carrying a ring-oxygen lone-pair of
a,d
During the course of our study on the use of chiral 1,3-
diheterocyclic enones in asymmetric synthesis,1 we found
that π-facial selectivity in the conjugate addition of
electrons (n
between C
O
) and the antibonding orbital (σ*) of the bond
and X. It is not easy to find which of the
0
4
2
interactions is sufficient to account for the observations,
and this point has been a subject of extensive studies
(
6) (a) Salzner, U.; Schleyer, P. v. R. J . Am. Chem. Soc. 1993, 114,
1
0231-10236. (b) Salzner, U.; Schleyer, P. v. R. J . Org. Chem. 1994,
*
To whom correspondence should be addressed. E-mail: msato@
59, 2138-2155. (c) Salzner, U. J . Org. Chem. 1995, 60, 986-995.
(7) (a) Carballeira, L.; Fernandez, Mosquera, R. A. THEOCHEM
1990, 105, 235-244. (b) Carballeira, L.; Fernandez, B.; Ri’os, M. A.
THEOCHEM 1990, 206, 277-285.
(8) (a) Brigs, A. J .; Glenn, R.; J ones, P. G.; Kirby, A. J .; Ramaswamy,
P. J . Am. Chem. Soc. 1984, 106, 6200-6206. (b) Romers, C.; Altona,
C.; Buys, H. R.; Havinga, E. Topics in Stereochemistry; J ohn Wiley &
Sons: New York, 1969; Vol. 4, pp 39-97. (c) Luger, P.; Kothe, G.;
Paulsen, H. Chem. Ber. 1976, 109, 1850-1855. (d) J effery, G. A.; Pople,
J . A.; Binkley, J . S.; Vishveshwara, S. J . Am. Chem. Soc. 1978, 100,
373-379. (e) Cosse-Barbi, A.; Dubois, J .-E. J . Am. Chem. Soc. 1987,
109, 1503-1511.
(9) (a) Allen, F. H.; Kirby, A. J . J . Am. Chem. Soc. 1984, 106, 6197-
6200. (b) Hilling, K. W., II; Lattimer, R. P.; Kuczkowski, R. L. J . Am.
Chem. Soc. 1982, 104, 988-983. (c) Narasimhamurthy, N.; Manohar,
H.; Samuelson, A.; Chandrasekhar, J . J . Am. Chem. Soc. 1990, 112,
2937-2941.
(10) For reviews, see: (a) Sato, M.; Sunami, S.; Kaneko, C. Hetero-
cycles 1995, 42, 861-883. (b) Kaneko, C.; Sato, M.; Sakaki, J .; Abe, Y.
J . Heterocycl. Chem. 1990, 27, 25-30.
u-shizuoka-ken.ac.jp.
†
Tohoku University.
Present address: School of Pharmaceutical Sciences, University
‡
of Shizuoka, 52-1 Yada, Shizuoka, 422-8526, J apan.
§
Yamanouchi Pharmaceutical Co. Ltd.
(
1) Use of 1,3-Dioxin-4-ones and Related Compounds in Synthesis.
8. For part 47, see: Sato, M.; Sunami, S.; Sugita, Y.; Kaneko, C.
Heterocycles 1995, 41, 1435-1444.
2) (a) Kirby, A. J . The Anomeric effect and Related Stereoelectronic
4
(
Effects at Oxygen; Springer-Verlag: Berlin, 1983. (b) Tvaroska, I.; Bleca,
T. Adv. Carbohydr. Chem. Biochem. 1989, 47, 45-123. (c) Box, V. G.
S. Heterocycles 1990, 31, 1157-1181. (d) J uaristi, E.; Cuevas, G.
Tetrahedron 1992, 48, 5019-5087. (e) J uaristi, E.; Cuevas, G. The
Anomeric Effect; CRC Press: Boca Raton, 1995.
(
(
(
3) Edward, J . T. Chem. Ind. 1995, 1102-1104.
4) Lucken, E. A. C. J . Chem. Soc. 1959, 2954-2960.
5) (a) Perin, C. L.; Armstrong, K. B.; Fabian, M. A. J . Am. Chem.
Soc. 1994, 116, 715-722. (b) Wiberg, K. B.; Marquez, M. J . Am. Chem.
Soc. 1994, 116, 2197-2198.
1
0.1021/jo970742j CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/18/1999

Anomeric Effect from X-ray Crystal Structure Analysis
J . Org. Chem., Vol. 64, No. 5, 1999 1437
and 1,3-dithianes23 from conformation analysis and has
been rationalized by the anomeric effect. The influence
of aryl groups on the conformation of anomeric com-
Sch em e 1
2
4
pounds has been also studied. K o¨ hler et al. observed
the axial preference of the 2-aryl group in 1,3-dithianes
and attributed it to the anomeric effect. Some data
2
5
concerning conformational preference and X-ray crystal
structures of 2-aryl-1,3-dioxanes2
6,27
have been also re-
ported. However, the electronic effects of the 2-aryl group
on the conformational behaviors of 1,3-dioxanes have
remained obscure, because the electronic character of the
aryl group varies significantly with the substituent and
its position on the aryl ring. In addition, it is not easy to
estimate the purely electronic effect of the aryl group on
the conformation because the conformation varies sig-
nificantly with the steric interactions involving the aryl
group, whose effective bulk varies with the positions
organometallic reagents to 1,3-dioxin-4-one derivatives
is greatly affected by pseudoaxial substituents on the
anomeric carbon. Thus, addition to 2a (X ) H11,12 or Me )
(
axial or equatorial) and with the torsion angle of the aryl
13
2
8
ring.
occurs selectively from the more hindered top face, while
To prove our hypotheses on the origins of the π-facial
selectivity in 2 and the conformational preference in 3
as well as to study the relative importance of the two
origins of the anomeric effect, we carried out a more
detailed study of the stereoelectronic effect of 2-aryl
substituents in 1,3-dioxanes.
To study the purely electronic effects of the aryl group,
we used a series of 2,2-di(para-substituted phenyl)-1,3-
dioxane 4 as a probe. The two aryl substituents are
sterically equivalent but electronically nonequivalent
when an electron-withdrawing or -donating substituent
is introduced to the para position. Moreover, even when
the two aryl groups were electronically equivalent, we
would be able to study the magnitude of the anomeric
addition to 2b (X ) Ph)14 and 2c (X ) CO
R)
15,16
proceeds
2
selectively from the less hindered bottom face. While
other explanations have been proposed for the top face
attack,1
2,13,17
we rationalized both selectivities by invoking
1
4-16
two competing stereoelectronic effects.
Namely, the
top face attack to 2a is best rationalized by the Cieplack
effect;18 the lone-pair electrons of O
interact with the
1
antibonding orbital (σ‡*) of the incipient bond facilitating
the top face attack. In 2b and 2c, the lone-pair electrons
of O
1
interact strongly with the antibonding orbital of
-X bond and the Cieplack effect would
the pseudoaxial C
2
become less effective, allowing the attack from the bottom
face. We also found a characteristic boat conformation
in the crystal structure of 2,2-diaryl-1,3-oxazine-4,6-
diones 3 where the more electronegative aryl group
occupies a pseudoaxial position (Scheme 1). We rational-
ized this conformational preference in terms of the
stabilizing interaction between the lone-pair electrons on
effect by comparing the lengths of the C
the C -equatorial-aryl bonds via X-ray crystallographic
analysis.
2
-axial-aryl and
2
Thus, we analyzed the conformation and molecular
structure of 2-(4-methoxyphenyl)-2-(4-nitrophenyl)-(4a ),
N
3
and/or O
1
and the σ* orbital of the pseudoaxial C -
2
2
-(4-methoxyphenyl)-2-[(4-trifluoromethyl)phenyl]-(4b),
1
9
aryl bond.
and 2,2-diphenyl-1,3-dioxanes (4d ) by X-ray crystal-
The axial preference of an alkoxycarbonyl group has
been observed in tetrahydropyranes,20 1,3-dioxanes,
lography (Scheme 2). The conformational preference of
21,22
1
13
4a -d in solution was studied by H and C NMR
spectroscopy. Finally, semiempirical molecular orbital
calculations (MOPACK PM3 method) were done on the
geometry and heat of formation for 4 where an electron-
withdrawing aryl group occupies the axial position and
(
11) (a) Zimmermann, J .; Seebach, D. Helv. Chim. Acta 1987, 70,
1
6
104-1114. (b) Seebach, D.; Zimmermann, J . Helv. Chim. Acta 1986,
9, 1147-1152.
(12) Seebach, D.; Zimmermann, J .; Gysel, U.; Ziegler, R.; Ha, T.-K.
J . Am. Chem. Soc. 1988, 110, 4763-4772.
13) Organ, M. G.; Froes, R. D. J .; Taylor, N. J .; Lange, G. L. J .
Am. Chem. Soc. 1994, 116, 3312-3323.
14) Sato, M.; Uehara, F.; Kamaya, H.; Murakami, M.; Kaneko, C.;
Furuya, T.; Kurihara, H. Chem. Commun. 1966, 1063-1064.
15) Sato, M.; Murakami, M.; Sunami, S.; Kaneko, C.; Furuya, T.;
Kurihara, H. J . Am. Chem. Soc. 1995, 117, 4279-4287.
16) Sato, M.; Murakami, M.; Kaneko, C.; Furuya, T. Tetrahedron
993, 49, 8529-8540.
17) Huang, X. L.; Dannenberg, J . J . J . Am. Chem. Soc. 1993, 115,
017-6024.
18) (a) Cieplak, A. S. J . Am. Chem. Soc. 1981, 103, 4540-4522. (b)
J ohnson, C. R.; Tait, B. D.; Cieplak, A. S. J . Am. Chem. Soc. 1987,
09, 5875-5876. (c) Cieplak, A. S.; Tait, B. D.; J ohnson, C. R. J . Am.
Chem. Soc. 1989, 111, 8447-8462.
19) Sato, M.; Nagashima, S.; Murakami, M.; Kaneko, C.; Furuya,
T. Tetrahedron 1993, 49, 6575-6580.
20) (a) Eliel, E. L.; Hargrave, K. D.; Pietrusiewicz, K. M.; Mano-
(
(23) (a) Eliel, E. L.; Hartman, A. H.; Abatjoglou, A. G. J . Am. Chem.
Soc. 1974, 96, 1807-1816. (b) J ualisti, E.; Tapia, J .; Mendez, R.
Tetrahedron 1986, 42, 1253-1264. (c) J uaristi, E.: Gonz a´ lez, E. A.;
Pinto, B. M.; J ohnston, B. D.; Nagelkerke, R. J . Am. Chem. Soc. 1989,
111, 6745-6749. (d) Tschieerske, C.; K o¨ hler, H.; Zaschke, H.; Klein-
peter, E. Tetrahedron 1989, 45, 6987-6998. (e) K o¨ hler, H. Tschierske,
H.; Kleinpeter, E. Tetrahedron 1990, 46, 4241-4246. (f) Eliel, E. L.;
Hartman, A. H.; Abatjoglou, A. G. J . Am. Chem. Soc. 1974, 96, 1807-
1816.
(24) K o¨ hler, H.; Tschierske, C.; Zaschke, H.; Kleinpeter, E. Tetra-
hedron 1990, 46, 4241-4246.
(25) (a) Eliel, E. L.; Knoeber, M. C. J . Am. Chem. Soc. 1968, 90,
3444-3458. (b) Nader, F. W.; Eliel, E. L. J . Am. Chem. Soc. 1970, 92,
3050-3055. (c) Gung, B. W.; Zhu, Z.; Mareska, D. A. J . Org. Chem.
1993, 58, 1367-1371.
(
(
(
1
(
6
(
1
(
(26) (a) De Kok, A. J .; Romers, C. Rec. Trav. Chim. 1970, 89, 313-
320. (b) Kellie, G. M.; Murray-Rust, P.; Riddell, F. G. J . Chem. Soc.,
Perkin Trans. 2 1972, 2384-2387. (c) Collins, P. M.; Travis, A. S.;
Tsiquaye, K. N. J . Chem. Soc., Perkin Trans. 1 1974, 1895-1901.
(27) (a) Nader, W. F. Tetrahedron Lett. 1975, 1207-1210. (b) Nader,
F. W. Tetrahedron Lett. 1975, 1591-1594. (c) Eliel, E. L.; Bailey, W.
F.; Wiberg, K. B.; Connon, H.; Nader, F. W. Liebigs Ann. Chem. 1976,
2240-2259.
(
haran, M. J . Am. Chem. Soc. 1982, 104, 3635-3643. (b) Booth, H.;
Dixon, J . M.; Khedhair, K. A. Tetrahedron 1992, 48, 6161-6174.
(21) For reviews on the conformational aspects in 1,3-dioxanes,
see: (a) Antenis, M. J . O.; Tavernier, D.; Borremans, F. Heterocycles
1
5
976, 4, 293-371. (b) Eliel, E. L. Pure Appl. Chem. 1971, 25, 509-
25. (c) Eliel, E. L. Angew. Chem., Int. Ed. Engl. 1972, 11, 739-860.
(22) (a) Bailey, W. F.; Eliel, E. L. J . Am. Chem. Soc. 1974, 96, 1798-
(28) Allinger, N. L.; Tribble, M. T. Tetrahedron Lett. 1971, 3259-
3262.
1
806.

1
438 J . Org. Chem., Vol. 64, No. 5, 1999
Uehara et al.
Sch em e 2
The data for bond lengths of 4 are almost comparable
to those for 5-7. However, the bonds at the anomeric
carbon of 4 are slightly longer than those of 5-7 as seen
from Table 1. This difference in bond lengths at the
anomeric carbon arises from a steric reason; a repulsive
interaction due to the steric congestion for the tertiary
anomeric carbon in 4 results in the longer bonds com-
pared to the 2-monoaryl analogues 5-7. This rationaliza-
tion is well evidenced by the remarkable bond lengthen-
ing in the sterically much congested 3 where all of the
bonds at the anomeric carbon are markedly lengthened
(
1
axial C
.524; N
The observed conformations for 4a and 4b and the
remarkable bond lengthening at the axial C -aryl in
a ,b,d strongly suggest an anomeric effect that is best
2
-aryl, 1.524; equatorial C
2
-aryl, 1.527; O
1
2
-C ,
-C
2
, 1.465 Å).19
its conformational isomer 4′ where an electron-donating
aryl group occupies the axial position, and the results
were compared with the observed conformational prefer-
ence and molecular structures in X-ray crystallography
and NMR spectroscopy.
3
2
4
rationalized in terms of the n
tions.
O
-σ* stabilizing interac-
1H a n d 13C NMR Sp ectr oscop ic An a lyses. To study
the conformational behaviors of 4 in solution, we exam-
Resu lts a n d Discu ssion
1
13
X-r a y Cr ysta llogr a p h ic An a lysis. Compounds 4a -d
were prepared by condensation of diaryl ketones with 1,3-
propanediol under an acid catalysis.29 We analyzed the
molecular structures of 4a and 4b by X-ray crystal-
lography. For comparison purposes, the structure of
symmetrical derivative 4d was also analyzed by X-ray
crystallography. Figure 1 contains molecular structures
of 4a , 4b, and 4d , and Table 1 contains selected bond
distances at the anomeric carbon and ring oxygens and
selected bond angles. As expected, the 4-nitrophenyl
group that is much more electron-withdrawing than the
ined H and C NMR spectroscopy. Figure 2 contains
selected data obtained at 25 °C in CDCl . The signals
for C5 methylene protons are greatly affected by the
3
introduction of para substituents on the phenyl ring; the
signals for diphenyl derivative 4d and 2-(4-methoxy-
phenyl)-2-phenyl derivative 4c appeared as quintets at
1.77 and 1.81 ppm (each 2H), respectively, indicating a
fast exchange of two conformers 4 and 4′, while that for
4a and 4b appeared as well-separated multiplets (4a ,
1.70 and 1.95 ppm; 4b, 1.75 and 1.88 ppm). The observed
nonequivalence of the two protons at C strongly suggests
that 4a and 4b are conformationally fixed to a great
5
4
-methoxyphenyl group occupied an axial position in 4a .
This conformation is not incidentally caused by crystal
packing, because compound 4b also took the conforma-
tion having the electron-withdrawing 4-(trifluoromethyl)-
phenyl group at the axial position. In the crystal of 4a ,
extent in solution due to the anomeric effect. This
assumption is strongly supported by 13C NMR data. The
chemical shifts of anomeric carbons in 4a (100.19 ppm)
and 4b (100.50 ppm) appeared at higher fields than in
diphenyl derivative 4c (101.03 ppm) by 0.84 and 0.53
ppm, respectively. In addition, the nitrated carbon (147.32
ppm) and trifluoromethylated carbon (129.83 ppm) also
appeared at higher fields by 0.98 and 1.01 ppm than
those for nitrobenzene 8 (148.30 ppm) and R,R,R-tri-
fluorotoluene 9 (130.84 ppm), respectively. In good con-
trast, the chemical shift of methoxylated carbons for
4a -c (159.25-159.68 ppm) is almost the same as that
for anisole itself (159.74 ppm). Though compounds 8-11
are not ideal models for 4, the observed high-field shifts
for the anomeric carbon and the para carbon of the axial
aryl group strongly indicates enhanced electron density
by the stabilizing interaction between the lone-pair
electrons of O and O and antibonding orbital of the axial
the axial C
than the equatorial C
which is close to the accepted average value for the Csp -
2
-nitrophenyl bond (1.544 Å) is much longer
2
-methoxyphenyl bond (1.526 Å),
3
2
30
Csp bond (1.504 Å). Bond lengths for O
and O -C (1.413 Å) are shorter than those for O
1.440 Å) and O -C (1.444 Å) as has been generally
observed in 1,3-dioxanes.
anomeric carbon for 4b are comparable to those for 4a ;
1
-C
2
(1.405 Å)
3
2
1
-C
6
(
3
4
9
a,26,27
The bond lengths at the
the axial C
equatorial C
2
-aryl bond (1.539 Å) is longer than the
-aryl bond (1.520 Å). It is of interest to note
2
that these characteristic features in the bond length at
the anomeric carbon are essentially retained in the
molecular structure of 2,2-diphenyl derivative 4d ; the
axial C
equatorial C
2
-phenyl bond (1.540 Å) is longer than the
-phenyl bond (1.521 Å).
1
3
2
4
2
C -aryl bond. K o¨ hler et al. found that in conforma-
2
X-ray crystal structure analyses of 2-monoaryl-1,3-
dioxanes with equatorial and axial aryl groups have been
reported.2 The bond length at the anomeric carbon and
ring oxygens and selected bond angles for equatorial
tionally locked 2-phenyl-1,3-dithiane derivatives the para
carbon of the axial phenyl diastereoisomer appears at
higher field than that of the equatorial phenyl diastereo-
isomer, and they interpreted this observation in terms
O
6,27
4
4
-chlorophenyl (5), axial 4-chlorophenyl (6), and axial
-(trifluoromethyl)phenyl derivatives (7) are given in
of the n -σ* stabilizing interaction.
Com p u ta tion a l An a lysis
Table 1, respectively. The axial C
2
-aryl bond in 6 and 7
-aryl bond (1.50
(
1.532 Å) is longer than the equatorial C
2
The conformational preference of 4a ,b over 4′a ,b is now
clear from the above X-ray crystallographic and NMR
spectroscopic studies. Thus, we finally examined semi-
empirical molecular orbital calculations on the confor-
mational preference and molecular structure of 2,2-
diaryl-1,3-dioxanes 4a -d .
2
7b
Å) in 5. Nader et al. suggested the anomeric effect as
one of the reasons for the axial bond lengthening in 7
(Chart 1).
(
29) Ceder, O. Arkiv. Kemi. 1954, 6, 523-535; Chem. Abstr. 1954,
8, 7406b.
30) Bastiansen, O. Skancke, P. N. Adv. Chem. Phys. 1961, 3, 323-
62.
4
Geometries for 4a -d and their conformational isomers
4′a -c were calculated by MOPAK PM3 on a Tektronix
(
3

Anomeric Effect from X-ray Crystal Structure Analysis
J . Org. Chem., Vol. 64, No. 5, 1999 1439
F igu r e 1. Molecular structures for 4a , 4b, and 4d from X-ray crystallographic analysis.
Ta ble 1. Selected Bon d Len gth s a n d An gles for 1,3-Dioxa n es 4-7 Obser ved in X-r a y Cr ysta llogr a p h y
substituent
bond length (Å)
O1-C2 O3-C2
bond angle (deg)
compd
no.
X
Y
C2-X
C2-Y
O1-C6
O3-C4 O1-C2-O3 C2-O3-C4 C2-O1-C6 ref
4
4
4
5
6
7
a
b
d
C6H4NO2(p) C6H4OMe(p) 1.544(3) 1.526(3) 1.405(3) 1.413(3) 1.440(3) 1.444(3) 112.0(2)
C6H4CF3(p) C6H4OMe(p) 1.539(4) 1.520(4) 1.412(3) 1.539(4) 1.440(4) 1.442(4) 111.4(2)
112.5(2)
113.0(2)
114.0(2)
111
113.7
113.9
113.1(2)
113.2(2)
113.7(2)
111
113.5
113.7
C6H5
C6H5
1.540(2) 1.521(3) 1.418(2) 1.413(2) 1.443(3) 1.436(3) 111.6(2)
H
C6H4Cl(p)
1.50
1.41
1.40
1.42
1.45
111
26a
27c
27a
C6H4Cl(p)
C6H4CF3(p)
H
H
1.532
1.532
1.408
1.409
1.412
1.412
1.450
1.452
1.442
1.447
112.5
112.3
Ch a r t 1
result was compared with the data obtained by semi-
empirical molecular orbital calculations. The X-ray crys-
tallographic analysis of 4a and 4b provided excellent
evidence for the anomeric effect of the aryl group; the
more electron-withdrawing aryl group always occupies
an axial position and the axial C -aryl bond is remark-
2
ably lengthened. It should be emphasized that the
anomeric effect was also observed in the 2,2-diphenyl
derivative as revealed by the comparable bond lengthen-
CAChe work system using EF (eigenvector-following)
method. The heat of formation and selected bond dis-
tances and angles for 4a -d are listed in Table 2. The
ing of the axial C
analogues with a strong electron-withdrawing aryl group.
The nonequivalence of the two protons at C in 4a and
b is attributable to the conformational fixation by the
2
-phenyl bond with those of the
axial C
the equatorial C
negativity of the axial aryl group results in the increased
bond length of the axial C -aryl. The calculated bond
lengths of axial C -aryl and equatorial C -aryl for
a ,b,4d are relatively close to those observed in their
2
-aryl bonds are somewhat longer compared to
2
-aryl bonds, and increased electro-
5
4
anomeric effect caused by the electron-withdrawing aryl
group. The high-field shifts of the anomeric carbons and
nitrated and trifluoromethylated carbons of 4a and 4b
in their 13C NMR spectra clearly show enhanced electron
density at these carbons. Though calculations by the
MOPAK PM3 method could not reproduce these experi-
mental results except for the bond lengthening for axial
2
2
2
4
crystal structures listed in Table 1. In 4′a and 4′b where
the electron-donating p-methoxyphenyl group occupies
an axial position and the electron-withdrawing aryl group
occupies an equatorial position, the axial C
are shortened and equatorial C -aryl bonds are length-
ened compared to those in 4a and 4b, respectively.
However, the bond lengths for O -C (1.428-1.429 Å)
are longer and those for O -C (1.415-1.417 Å) are
2
-aryl bonds
C
2
-aryl, the crystallographic and NMR spectroscopic
data clearly show n -σ* stabilizing interaction in 2-aryl-
,3-dioxanes. Thus, we conclude that the delocalization
2
O
1
1
2
is more substantiative than the electrostatic explanation
for the anomeric effect. Moreover, the result well supports
our hypothesis concerning operation of the anomeric
effect in 2b, 2c, and related compounds including 1,3-
oxazine-4,6-dione 3.
1
6
shorter compared to the data obtained by the crystal-
lographic analysis. In addition, calculated heat of forma-
tion indicated 4a and 4b are less stable by 0.78 and 0.57
kcal/mol than 4′a and 4′b, respectively. Thus, it became
apparent that this semiempirical calculation is not suf-
ficient to account for the conformational preference of 2,2-
diaryl-1,3-dioxanes, while this method clearly indicates
Exp er im en ta l Section
that increased electronegativity of axial C
in increased bond lengthening of the axial C
2
-aryl results
-aryl.
Gen er a l Meth od s. All melting points were recorded on a
Yanagimoto micro-hot stage and are uncorrected. IR spectra
2
1
were measured on a J ASCO A-102 spectrophotometer. H and
1
3
C NMR spectra were measured at 25 °C on a Varian Gemini
000 spectrometer with tetramethylsilane as an internal
2
Con clu sion
1
3
standard. The C NMR spectra were measured using 0.5 M
sample solutions. High-resolution mass spectra were recorded
on a J EOL J MS-DX-303 or J MS-AX-500 spectrometer. Wako-
gel (C-200) silica gel and Merck aluminum oxide 90 (neutral,
activity stage I) were used in the column chromatographies.
We analyzed the conformation and molecular structure
of a series 2,2-diaryl-1,3-dioxanes by X-ray crystal-
lography and 1H and C NMR spectroscopy, and the
13

1
440 J . Org. Chem., Vol. 64, No. 5, 1999
Uehara et al.
F igu r e 2. Selected H and 13C NMR data (ppm, CDCl
1
, 25 °C) for 4a -d .
3
Ta ble 2. Hea t of F or m a tion a n d Selected Geom etr y P a r a m eter s for 4 a n d 4′ Ca lcu la ted by P M3 Meth od
heat of
substituent
bond length (Å)
bond angle (°)
compd
no.
formation
(kcal/mol) C2-X C2-Y O1-C2, O3-C2 O1-C6, O3-C4 O1-C2-O3 C2-O3-C4 C2-O1-C6
X
Y
4
4
4
4
4
a
b
d
′a
′b
C6H4NO2(p) C6H4OMe(p)
-70.12 1.534 1.525
1.428
1.428
1.429
1.428
1.429
1.416
1.415
1.415
1.417
1.415
105.8
106.0
105.5
105.1
106.0
114.6
115.2
115.0
113.9
115.0
115.5
115.2
115.1
115.3
115.0
C6H4CF3(p)
C6H5
C6H4OMe(p) -219.78 1.533 1.526
C6H5
-23.53 1.530 1.528
-70.90 1.529 1.529
-220.35 1.528 1.529
C6H4OMe(p) C6H4NO2(p)
C6H4OMe(p) C6H4CF3(p)
The ratios of solvent mixtures for chromatography are shown
as volume/volume.
temperature. After being stirred for 2 h, 2-propanol (5 mL)
was added to the mixture. The whole was extracted with
dichloromethane. The organic layer was washed with satu-
Gen er a l P r oced u r e for th e Syn th esis of 2,2-Dia r yl-1,3-
d ioxa n es (4a -d ). Compounds 4a -d were prepared by the
3 4
rated NaHCO solution, dried over anhydrous MgSO , and
2
9
reported procedure. A solution of diaryl ketone (1.0 mmol),
evaporated. Recrystallization of the residue from a mixture of
hexane and dichloromethane afforded the 4,4′-disubstituted
benzophenone (814 mg, 57%, two steps) as prisms, mp 123-
124 °C (lit.33 mp 123-124 °C). Following the general procedure,
4-methoxy-4′-(trifluoromethyl)benzophenone (266 mg, 0.95
mmol) was condensed with 1,3-propanediol (228 mg, 3.0 mL).
Purification of the crude product by alumina column chroma-
tography (hexane/chloroform 3:1) gave 4b (210 mg, 65%) as a
1
,3-propanediol (3.0 mmol), and a catalytic amount of p-
toluenesulfonic acid (30 mg) in benzene (60-75 mL) was
refluxed with a Dean-Stark trap for removal of water. After
1
3
0-30 h, saturated NaHCO solution was added, and the
mixture was extracted with ether. The organic layer was
washed with brine and dried over anhydrous MgSO . The
4
solvent was removed in vacuo, and the residue was purified
by alumina column chromatography.
solid. Recrystallization from ether afforded colorless prisms:
1
2
-(4-Meth oxyph en yl)-2-(4-n itr oph en yl)-1,3-dioxan e (4a).
mp 118-119 °C; H NMR (CDCl
3
) δ 1.72-1.79 (1H, m), 1.82-
Following the general procedure, 4-methoxy-4′-nitrobenzophe-
1.90 (1H, m), 3.78 (3H, s), 3.95-4.11 (4H, m), 6.85-6.91 (2H,
3
0
13
none (257 mg, 1.0 mmol) was condensed with 1,3-propanediol
228 mg, 3.0 mmol). Purification of the crude product by
3
m), 7.39-7.45 (2H, m), 7.56-7.66 (4H, m). C NMR (CDCl )
(
δ 25.44, 55.19, 61.61, 100.50, 114.11, 124.27 (q, J ) 271.0 Hz),
125.48 (q, J ) 3.4 Hz), 126.72, 127.96, 129.83 (q, J ) 31.9 Hz),
alumina column chromatography (hexane-chloroform, 3:1)
-
1
gave 4a (222 mg, 71%) as a solid. Recrystallization from ether
133.67, 147.33, 159.55; IR (CHCl
3
)1605, 1505, 1320 cm
338.1130, found 338.1150. Anal.
17 3 3
H F O : C, 63.90; H, 5.06. Found: C, 64.15; H,
;
1
afforded pure 4a as yellowish prisms: mp 161-163 °C; H
HRMS calcd for C18
17 3 3
H F O
NMR (CDCl
3
) δ 1.65-1.74 (1H, m), 1.88-1.99 (1H, m), 3.77
Calcd for C18
5.24.
(3H, s), 3.95-4.11 (4H, m), 6.85-6.90 (2H, m), 7.38-7.43 (2H,
13
m), 7.65-7.70 (2H, m), 8.11-8.17 (2H, m); C NMR (CDCl
δ 25.30, 55.18, 61.60, 100.19, 114.27, 123.65, 126.99, 127.98,
32.67, 147.32, 150.76, 159.68; IR (CHCl ): 1610, 1520, 1530
315.1107, found 315.1110.
: C, 64.75; H, 5.43; N, 4.44. Found:
3
)
2-(4-Met h oxyp h en yl)-2-p h en yl-1,3-d ioxa n e (4c). Fol-
lowing the general procedure, 4-methoxybenzophenone (424
mg, 2.0 mmol) was condensed with 1,3-propanediol (456 mg,
6.0 mmol). Purification of the crude product by alumina column
chromatography (hexane/chloroform 5:1) gave 4c (95 mg, 18%)
and 4-methoxybenzophenone (316 mg, 69%). Recrystallization
1
3
-
1
cm ; HRMS calcd for C17
Anal. Calcd for C17
C, 64.85; H, 5.64; N, 4.45.
-(4-Met h oxyp h en yl)-2-[4-(t r iflu or om et h yl)p h en yl]-
,3-d ioxa n e (4b). 4-Methoxy-4′-(trifluoromethyl)benzophe-
H17NO
5
H
17NO
5
3
4
2
of 4c from ether afforded colorless prisms: mp 53-54 °C (lit.
1
1
mp 52-53 °C); H NMR (CDCl
s), 4.01 (4H, t), 6.82-6.87 (2H, m), 7.19-7.53 (7H, m); C NMR
(CDCl
3
) δ 1.77 (2H, quintet), 3.74 (3H,
13
none was prepared as follows. 4-(Methoxyphenyl)magnesium
bromide (9.0 mmol, 1.2 M ether solution, 7.5 mL) was added
dropwise to a solution of 4-(trifluoromethyl)benzaldehyde (894
mg, 5.14 mmol) in THF (10 mL) at 0 °C. The mixture was
stirred at room temperature for 12 h. A solution of 5% HCl
was added, and the mixture was extracted with ether. The
organic layer was washed with brine, dried over anhydrous
3
) δ 25.54, 55.16, 61.55, 101.03, 113.80, 126.53, 127.73,
127.91, 128.48, 134.82, 142.75 159.25.
2,2-Dip h en yl-1,3-d ioxa n e (4d ). Following the general
procedure, benzophenone (911 mg, 5.0 mmol) was condensed
(
(
(
31) Auwers, K. Ber. 1903, 36, 3893-3902.
32) Kelly, D. P.; J enkins, M. J . J . Org. Chem. 1984, 49, 409-413.
33) Leigh, W. J .; Arnold, D. R.; Humphreys, W. R.; Wong, P. C.
MgSO , and evaporated in vacuo to afford crude R-(4-meth-
4
32
oxyphenyl)(4-trifluoromethyl)benzyl alcohol as a solid. A
solution of 8 M J ones reagent (1.0 mL, 8.0 mmol) was added
to the solution of the benzyl alcohol in acetone (40 mL) at room
Can. J . Chem. 1980, 58, 2537-2549.
34) Ueno, S.; Oshima, T.; Nagai, T. Bull. Chem. Soc. J pn. 1986,
59, 305-306.
(

Anomeric Effect from X-ray Crystal Structure Analysis
J . Org. Chem., Vol. 64, No. 5, 1999 1441
Ta ble 3. Cr ysta llogr a p h ic Da ta for Com p ou n d s 4a ,b,d
The structure was solved by direct method with SHELX-
635 and expanded with DIRDIF 92. Positions and anisotropic
36
8
compound
4a
4b
4d
displacement parameters were refined for all non-hydrogen
atoms by the full-matrix least-squares technique. Hydrogen
atoms were placed at the calculated positions with an isotropic
parameter equal to 1.2 times that of the attached atom and
were fixed but included in the refinement. All calculations were
carried out using teXsan software package of Molecular
chemical formula
C17H17NO5
26.157(1)
7.947(2)
7.380(1)
93.15(1)
1531.7(7)
4
C18H17F3O3
26.871(1)
8.017(2)
7.471(2)
90.81(1)
1609.2(8)
4
C16H16O2
8.238(2)
13.239(2)
6.249(2)
111.67(2)
635.9(2)
2
a, Å
b, Å
c, Å
â, °
3
V, Å
Z
37
Structure Corp. All of the X-ray analytical data are deposited
as Supporting Information.
space group
P21/a
0.055
P21/a
0.067
P21
0.039
a
R
Rw
0.105
0.123
0.069
a
R ) Σ//Fo/ - Fc//Σ/Fo/, Rw ) [Σw(/Fo/ - /Fc/) /ΣwFo ]1/2.
2
2
Ack n ow led gm en t. We are grateful to Mr. K. Kawa-
mura and Mrs. M. Inada for mass spectral measure-
ments and to Mrs. A. Sato for elemental analyses.
with 1,3-propanediol (1.14 g, 15.0 mmol). Purification of the
crude product by alumina column chromatography (hexane/
dichloromethane 3:1) gave 4d (795 mg, 62%). Recrystallization
2
9
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallogra-
phy data for compounds 4a ,b. This material is available free
of charge via the Internet at http://pubs.acs.org.
from ether afforded colorless prisms: mp 113-115 °C (lit.
1
mp 111.5-112.5 °C); H NMR (CDCl
3
) δ 1.81 (2H, quintet),
.05 (4H, t), 7.21-7.55 (10H, m); 1 C NMR (CDCl
3
4
6
3
) δ 25.48,
1.54, 100.97, 126.49, 127.75, 128.48, 142.48.
X-r a y Cr ysta llogr a p h ic An a lyses of 4a , 4b, a n d 4c.
J O970742J
Diffraction data collections were made on a Rigaku AFC-5R
for 4a and 4b and a Rigaku AFC-7R for 4d with Cu KR
radiation (λ ) 1.541 78 Å) at room temperature. Lorentz,
polarization, empirical absorption, and secondary extinction
corrections were applied to all data. Crystal data and selected
parameters for all three compounds are shown in Table 3. The
number of measured and observed (I > σ(I)) reflections are
(
35) SHELX86: Sheldrick, G. M. In Crystallographic Computing 3;
Sheldrick, G. M., Kruger, C., Goddard, R., Eds.; Oxford University
Press: Oxford, 1985; pp 175-189.
(36) DIRDIF92: Beurskens, P. T.; Admiraal, G.; Beurskens, G.;
Bosman, W. P.; de Gelder, R.; Smits, J . M. M. The DIRDIF program
system, Technical Report of the Crystallography Laboratory, Univer-
sity of Nijmengen, The Netherlands, 1994.
2
1
691 and 2073 for 4a , 2811 and 2157 for 4b, and 1326 and
263 for 4d , respectively.
(37) teXsan: Crystal Structure Analysis Package, Molecular Struc-
ture Corp., 1985 and 1992.