Conformational Analysis of 5-Substituted 1,3-Dioxanes. 7. Effect of Lithium Bromide Addition

Article abstract of DOI:10.1021/jo9610117

The position of equilibria, established by means of BF₃, between diastereomeric cis- and trans-5- substituted-2-phenyl-1,3-dioxanes, in solvents THF and CHCl₃, and in the presence of 0, 1, and 10 equiv of LiBr has been determined. The observed ΔG° values show that the addition of salt to the reaction medium influences the position of equilibrium. Lithium bromide effects on the conformational behavior are discussed in terms of lithium ion complexation events that lead to increased stability of the axial isomers when the substituent at C(5) is CO₂H, CO₂CH₃, CONHCH₃, and CH₂- OH. By contrast, disruption of the intramolecular hydrogen bond present in the axial 5-acetamido derivative (cis-9 substituent equal to NHCOCH₃) modifies the preference for the axial conformation in salt-free 9 to a net dominance of the equatorial isomer in the presence of LiBr. Interpretation of the experimental observations was based on models that are apparently supported by semiempirical AM1 calculations. The results derived from the present study contribute to our understanding of the processes involved in molecular recognition and may model salt effects in physiological events.

Full text of DOI:10.1021/jo9610117

J . Org. Chem. 1997, 62, 4029-4035
4029
Con for m a tion a l An a lysis of 5-Su bstitu ted 1,3-Dioxa n es. 7. Effect
of Lith iu m Br om id e Ad d ition †,1
Eusebio J uaristi,*,‡ Francisco D´ıaz,‡,§ Geiser Cue´llar,‡ and Hugo A. J ime´nez-Va´zquez§
Departamento de Quı´mica, Centro de Investigacio´n y de Estudios Avanzados del Instituto Polite´cnico
Nacional, Apartado Postal 14-740, 07000 Me´xico D.F., Me´xico, and Departamento de Qu´ımica Orga´nica,
Escuela Nacional de Ciencias Biolo´gicas, IPN, 11340 Me´xico, D.F., Me´xico
Received May 30, 1996 (Revised Manuscript Received March 31, 1997X)
The position of equilibria, established by means of BF3, between diastereomeric cis- and trans-5-
substituted-2-phenyl-1,3-dioxanes, in solvents THF and CHCl3, and in the presence of 0, 1, and 10
equiv of LiBr has been determined. The observed ∆G° values show that the addition of salt to the
reaction medium influences the position of equilibrium. Lithium bromide effects on the confor-
mational behavior are discussed in terms of lithium ion complexation events that lead to increased
stability of the axial isomers when the substituent at C(5) is CO2H, CO2CH3, CONHCH3, and CH2-
OH. By contrast, disruption of the intramolecular hydrogen bond present in the axial 5-acetamido
derivative (cis-9 substituent equal to NHCOCH3) modifies the preference for the axial conformation
in salt-free 9 to a net dominance of the equatorial isomer in the presence of LiBr. Interpretation
of the experimental observations was based on models that are apparently supported by
semiempirical AM1 calculations. The results derived from the present study contribute to our
understanding of the processes involved in molecular recognition and may model salt effects in
physiological events.
Sch em e 1
In tr od u ction
It has long been known that the course of a chemical
reaction may be influenced by the surrounding medium.
Thus, the solvent can alter reaction rates and yields,
product stereochemistry, and the position of chemical
equilibria.2,3 Furthermore, such changes can also be
produced by addition of chemically inert salts to the
reaction medium. Nevertheless, while salt effects are
well established in organic chemistry, not much is known
as to how the electrolyte influences the mechanism and
kinetics in the course of the reaction.4,5 Of particular
interest are salt effects on conformational equilibria;
however, although solvent effects on conformational
equilibria have received much attention, there are few
available thermodynamic data dealing with salt effects.6
For example, in a qualitative level, Angyal and Davies
found that the conformational equilibrium of â-methyl-
ribopyranose is altered by the presence of CaCl2.7 An-
other remarkable example is provided by the strong
influence of LiCl on the backbone conformation of cyclic
undecapeptide cyclosporin A, a well known immunosup-
pressive agent.8
derivatives continue to serve as excellent probes for the
study of steric, electrostatic, and stereoelectronic interac-
tions.1,10 The present paper describes the preparation
and chemical equilibration of a series of 5-substituted-
1,3-dioxanes 1-9, both in presence and absence of
lithium bromide (Scheme 1).
Resu lts a n d Discu ssion
In a series of classical studies, Eliel demonstrated the
utility of substituted 1,3-dioxanes for the evaluation and
understanding of conformational effects,9 and 1,3-dioxane
A. Syn th eses of Dia ster eom er ic cis- a n d tr a n s-
5-Su bstitu ted -2-p h en yl-1,3-d ioxa n es. Condensation
of benzaldehyde and diethyl bis(hydroxymethyl)ma-
lonate,11 according to the described procedure,12 afforded
2-phenyl-5,5-dicarbethoxy-1,3-dioxane, 10, whose saponi-
fication gave the dicarboxylic acid, which was decarbox-
† Dedicated to Professor Dieter Seebach on the occasion of his 60th
birthday.
‡ Centro de Investigacio´n y de Estudios Avanzados.
§ Escuela Nacional de Ciencias Biolo´gicas.
X Abstract published in Advance ACS Abstracts, May 15, 1997.
(1) For Part 6, see: J uaristi, E.; Antu´nez, S. Tetrahedron 1992, 48,
5991.
(8) (a) Kofron, L. J .; Kuzmic, P.; Kishore, V.; Gemmecker, G.; Fesik,
S. W.; Rich, D. H. J . Am. Chem. Soc., 1992, 114, 2670. (b) Kock, M.;
Kessler, H.; Seebach, D.; Thaler, A. J . Am. Chem. Soc. 1992, 114, 2676.
(c) Carver, J . A.; Rees, N. H.; Turner, D. L.; Senior, S. J .; Chowdhry,
B. Z. J . Chem. Soc., Chem. Commun. 1992, 1682.
(9) (a) Eliel, E. L. Acc. Chem. Res. 1970, 3, 1. (b) Eliel, E. L. Angew.
Chem., Int. Ed. Engl. 1972, 11, 739.
(10) (a) J uaristi, E.; Mart´ınez, R.; Me´ndez, R.; Toscano, R. A.;
Soriano-Garc´ıa, M.; Eliel, E. L.; Petsom, A.; Glass, R. S. J . Org. Chem.
1987, 52, 3806. (b) Gordillo, B.; J uaristi, E.; Mart´ınez, R.; Toscano, R.
A.; White, P. S.; Eliel, E. L. J . Am. Chem. Soc. 1992, 114, 2157.
(11) Block, P. Org. Synth. 1960, 40, 27.
(2) Reichardt, C. Chem. Rev. 1994, 94, 2321.
(3) Reichardt, C. In Solvent and Solvent Effects in Organic Chem-
istry, 2nd ed.; VCH Publishers: Weinheim, 1988.
(4) Loupy, A.; Tchoubar, B. In Salt Effects in Organic and Organo-
metallic Chemistry; VCH Publishers: Weinheim, 1992.
(5) Loupy, A.; Tchoubar, B; Astruc, D. Chem. Rev. 1992, 92, 1141.
(6) Seebach, D.; Bossler, H. G.; Flowers, R.; Arnett, E. M. Helv.
Chim. Acta 1994, 77, 291.
(7) Angyal, S. J .; Davies, K. P. J . Chem. Soc., Chem. Commun. 1971,
500.
(12) Eliel, E. L.; Banks, H. D. J . Am. Chem. Soc. 1972, 94, 171.
S0022-3263(96)01011-0 CCC: $14.00 © 1997 American Chemical Society

4030 J . Org. Chem., Vol. 62, No. 12, 1997
J uaristi et al.
Sch em e 2
free energy differences are summarized in Table 3, which
includes the ∆G° values of interest in the presence of 0.0,
1.0, and 10.0 equiv of LiBr. (Lithium bromide alone does
not induce the equilibration of anancomeric dioxanes
1-9). The last column in Table 3 presents the corre-
sponding salt-free conformational free energy differences
in the 2-isopropyl analogues, studied by Eliel9 some 25
years ago.
Con for m a tion a l P r efer en ces in th e Absen ce of
Lith iu m Br om id e. Comparison, when possible, be-
tween the 2-phenyl- (this work) and 2-isopropyl-5-
substituted-1,3-dioxanes (Eliel’s work9) show some simi-
larities as well as some significant differences in
conformational behavior. For example, the observed ∆G°
values for the 5-hydroxy derivative (X ) OH) are es-
sentially identical in phenyl-substituted 5 (∆G° ) -0.38
kcal/mol, in THF) and the isopropyl analogue (-0.41 kcal/
mol, in Et2O). For this system, Eliel has shown the
absence, in ether solvent, of intramolecular hydrogen
bonding in either the axial or equatorial isomers; thus,
the lack of any salt effect in the equilibria of 5 (see below)
is not surprising; i.e., addition of lithium bromide neither
breaks up a potential hydrogen bond in 5-axial nor gives
rise to a stabilizing complex.
By contrast, significant differences (∆∆G° ∼ 0.5 kcal/
mol) are observed in the conformational preferences of 2
(X ) CO2CH3),18 4 (X ) CH2OH), 6 (X ) OCOCH3), and
8 (X ) NO2) and those described for the 2-isopropyl
analogues.
In the case of the 2-phenyl-5-nitro derivative 8, the
observed axial preference of the nitro group can be
explained in terms of electrostatic attraction between the
partially negative endocyclic oxygens and the positive
nitrogen atom.19
ylated to afford a mixture of cis- and trans-2-phenyl-5-
carboxy-1,3-dioxane (cis- and trans-1) (Scheme 2).
The diastereomeric carboxylic acids cis- and trans-1
were separated by flash chromatography, and converted
to the corresponding methyl esters 2 by means of silver
oxide/methyl iodide (Scheme 2). Next, cis- and trans-2
were treated with methyl amine to yield amides cis- and
trans-3. Furthermore, the reduction of cis- and trans-2
with lithium aluminum hydride produced cis- and trans-
4, with conservation of configuration (Scheme 2). Tables
1
1 and 2 contain H and 13C NMR chemical shifts for cis-
and trans-1-4.
cis- and trans-2-Phenyl-5-hydroxy-1,3-dioxane (cis- and
trans-5) were prepared by condensation of benzaldehyde
and glycerol, according to the described procedure.1
Carbinols cis- and trans-5 were converted to the corre-
sponding acetates with acetic anhydride in pyridine13 or
treated with (2-methoxyphenoxy)acetic acid14 and p-
toluenesulfonyl chloride in pyridine to yield the desired
derivatives 6 and 7, respectively.15 1H and 13C NMR
chemical shifts for compounds 5-7 are included in Tables
1 and 2.
The increased axial preference in going from carbon
tetrachloride (∆G° ) +0.38 kcal/mol) to THF solvent (∆G°
) +0.73 kcal/mol) can be explained in terms of the higher
polarity of the latter solvent, which reduces the well-
known dipole-dipole repulsion in axial, 5-polar substi-
tuted 1,3-dioxanes9 (eq 1). Similar solvent effect was
observed in 2-isopropyl-5-nitro-1,3-dioxane, when the
solvent was changed from CCl4 to CH2Cl2.19
On the other hand, 2-phenyl-5-(hydroxymethyl)-5-
nitro-1,3-dioxane, 12, was prepared by condensation of
commercially available tris(hydroxymethyl)nitromethane
with benzaldehyde. Deformylation with lithium/am-
monia16 furnished 5-nitro-2-phenyl-1,3-dioxane, 8 (Scheme
3). The only isomer obtained from this procedure was
cis-8, but sublimation of this material afforded a 1:1
mixture of cis- and trans-5-nitro-2-phenyl-1,3-dioxane
(cis- and trans-8), which were separated by flash chro-
matography. Each individual isomer was reduced under
catalytic conditions to give cis- and trans-13 which were
acetylated with acetic anhydride and pyridine to provide
amides cis- and trans-9 (Scheme 3). Tables 1 and 2
The increased axial preference of the 5-acetoxy group
in 6 and the diminished equatorial preference of the
5-carbomethoxy group in 2, relative to their 2-isopropyl
analogues (+0.47 versus 0.00 kcal/mol and -0.50 versus
-0.82 kcal/mol, respectively) may be explained if one
assumes that the electronegative 2-phenyl substituent
reduces the effective ring dipole and thus the resulting
dipole-dipole interactions which favor the equatorial
over the axial isomer for electron-withdrawing groups
such as acetoxy and carbomethoxy.
1
contain H and 13C NMR chemical shifts for compounds
8 and 9.
B. Con for m a tion a l An a lysis of 1-9. Equilibration
of diastereomeric 1,3-dioxanes (cis- and trans-1-9) was
readily performed by means of BF3.17 The corresponding
(13) Abraham, R. J .; Banks, H. D.; Eliel, E. L.; Hofer, O.; Kaloustian,
M. K. J . Am. Chem. Soc. 1972, 94, 1913.
(14) Hirota, M.; Hirano, G. Bull. Chem. Soc. J pn. 1972, 45, 1448.
(15) Brewster, J . H.; Ciotti, J . C. J . J . Am. Chem. Soc. 1965, 77,
6214.
(17) Eliel, E. L.; Knoeber, M. C. J . Am. Chem. Soc. 1968, 90, 3444.
(18) We found ∆G° ) -0.52 ( 0.03 kcal/mol for the axial h
equatorial equilibrium of the 2-isopropyl analogue in THF solvent.
(19) Kaloustian, M. K.; Dennis, N.; Mager, S.; Evans, S. A.; Alcudia,
F.; Eliel, E. L. J . Am. Chem. Soc. 1976, 98, 956.
(16) Marei, A. A.; Raphael, R. A. J . Chem. Soc. 1960, 886.

Conformational Analysis of 5-Substituted 1,3-Dioxanes
J . Org. Chem., Vol. 62, No. 12, 1997 4031
Ta ble 1. 1H NMR Ch em ica l Sh ifts (δ) for 1-9 a t 27 °C in CDCl3
compd
H(2)
H(4,6-ax)
H(4,6-eq)
H (5)
H (arom)
CH3
CH2O
other
cis-1a
trans-1
cis-2
trans-2
cis-3
trans-3
cis-4
trans-4
cis-5
trans-5
cis-6
trans-6
cis-7
trans-7
cis-8
5.46
5.40
5.51
5.40
5.54
5.52
5.50
5.40
5.52
5.31
5.53
5.45
5.54
5.43
5.53
5.45
5.54
5.49
4.03
4.02
4.10
3.97
4.18
4.15
4.01
3.68
4.08
3.45
4.13
3.36
4.16
3.70
4.22
4.17
4.10
3.60
4.60
4.50
4.70
4.45
4.42
4.32
4.22
4.26
4.15
4.16
4.26
4.38
4.29
4.38
4.97
4.70
4.10
4.32
2.34
3.25
2.43
3.13
2.39
2.91
1.62
2.32
3.59
3.81
4.68
5.01
4.85
5.12
4.21
4.69
4.00
4.29
7.23-7.40
7.30-7.40
7.31-7.46
7.34-7.46
7.30-7.50
7.31-7.52
7.44-7.47
7.30-7.50
7.35-7.50
7.35-7.46
7.40-7.52
7.34-7.49
6.82-7.49
6.82-7.50
7.24-7.42
7.23-7.47
7.49-7.50
7.39-7.49
3.81
3.70
2.87
2.66
6.90 (NH)
5.91 (NH)
2.20 (OH)
2.00 (OH)
3.20 (OH)
2.90 (OH)
4.01
3.38
2.15
2.05
3.86
3.87
4.85
4.69
trans-8
cis-9
trans-9
2.01
2.01
6.95 (NH)
5.50 (NH)
a
In THF.
Ta ble 2. 13C NMR Ch em ica l Sh ifts (δ) for 1-9 a t 27 °C in CDCl3
compd
C(2)
C(4,6)
C(5)
CdO
Carom
other
cis-1a
trans-1
cis-2
trans-2
cis-3
trans-3
cis-4
trans-4
cis-5
trans-5
cis-6
trans-6
cis-7
trans-7
cis-8
102.1
101.2
101.9
101.3
101.9
101.1
101.9
101.3
101.6
100.9
101.1
101.3
101.2
101.3
101.7
101.4
101.4
100.9
67.9
68.1
67.2
67.9
68.3
68.9
68.8
75.5
72.3
71.5
70.0
68.4
68.9
68.1
66.5
67.1
70.6
69.1
40.7
39.8
40.9
39.8
26.6
26.0
36.5
36.8
63.9
61.0
66.0
62.8
66.5
63.5
77.1
74.2
56.5
45.5
172.9
172.1
171.7
170.1
173.5
170.0
127.0-140.2
125.9-137.8
126.2-138.0
126.0-137. 7
125.8-137.9
125.9-137.7
125.9-138.4
125.9-138.1
125.9-137.9
126.9-137.3
126.0-137.8
126.1-137.3
112.0-149.0
112.0-149.0
126.7-137.0
126.0-136.1
125.7-137.7
126.1-137.3
52.33 (OCH3)
51.70 (OCH3)
41.6 (NHCH3)
42.27 (NHCH3)
61.60 (OCH2)
69.50 (OCH2)
170.9
169.8
169.0
168.2
21.15 (CH3)
20.70 (CH3)
55.9, 66.8 (OC)
55.9, 66.4 (OC)
trans-8
cis-9
trans-9
169.6
170.2
23.2 (CH3)
23.2 (CH3)
a
In THF.
Sch em e 3
hydrogen bonding in the axial isomers, since such
interaction is expected to lower the energy of these
isomers.
On the other hand, the thermodynamic stability of
axial 5-OCOCH2OAr, ∆G°(7) ) +0.56 kcal/mol, is of the
same magnitude of that found in cis-6 (X ) OCOCH3,
∆G° ) +0.47 kcal/mol), which has been interpreted in
terms of the so-called gauche effect operative in O-C-C-O
segments.1,20
The large axial preference of the acetamido group in
dioxane 9 (X ) NHCOCH3; ∆G° ) +0.94 kcal/mol) is
unprecedented. IR spectra for cis- and trans-9 showed
the existence of an intramolecular hydrogen bond in the
former, which accounts for its higher stability.
On the other hand, the conformational behavior of 5-X-
1,3-dioxanes with X ) CO2H, CONHCH3, OCOCH2-
OC6H4-o-OCH3, and NHCOCH3 had not been determined
previously. The substantial predominance of equatorial
5-CO2H (∆G° ) -0.77 kcal/mol) and 5-CONHCH3 (∆G°
) -0.76 kcal/mol) is in line with expected steric and
electrostatic (dipole-dipole, see above) repulsive interac-
tions operative in the axial isomers and argue against
the existence of intramolecular O-H---O or N-H---O
Indeed, the N-H stretching frequency in cis-9 (3300
cm-1, 1 × 10-4 M THF solution) corresponds to intramo-
(20) (a) Wolfe, S. Acc. Chem. Res. 1972, 5, 102. (b) J uaristi, E. J .
Chem. Educ. 1979, 56, 438. (c) J uaristi, E. Introduction to Stereochem-
istry and Conformational Analysis; Wiley: New York, 1991; Chapter
18.

4032 J . Org. Chem., Vol. 62, No. 12, 1997
J uaristi et al.
Ta ble 3. Con for m a tion a l Equ ilibr ia in 5-Su bstitu ted -2-p h en yl-1,3-d ioxa n es 1-9, in Absen ce or P r esen ce of LiBr , a t 25 °C
in THF
∆G° (kcal/mol)
compd
X
0.0 equiv
1.0 equivd
10.0 equive
2-isopropyl analogue17
1
2
3
4
5
6
7
8
9
CO2H
-0.77 ( 0.03
-0.50 ( 0.02
-0.76 ( 0.05
-0.20 ( 0.01
-0.38 ( 0.04
+0.47 ( 0.01
+0.56 ( 0.02
+0.73 ( 0.02
+0.94 ( 0.03
-0.41 ( 0.03
-0.15 ( 0.03
-0.67 ( 0.04
-0.04 ( 0.02
-0.35 ( 0.03
+0.45 ( 0.02
+0.89 ( 0.03
+0.52 ( 0.03
+0.44 ( 0.03
-0.17 ( 0.03
-0.43 ( 0.03
-0.60 ( 0.04
+0.22 ( 0.02
-0.43 ( 0.03
+0.43 ( 0.02
+0.43 ( 0.02
+0.57 ( 0.02
-0.13 ( 0.03
CO2CH3
CONHCH3
CH2OH
-0.82 ( 0.02b
-0.03 ( 0.04
-0.41 ( 0.03
0.00 ( 0.04
OH
OCOCH3
OCOCH2OAra
NO2
0.38 ( 0.04c
NHCOCH3
a
b
d
Ar ) 2-CH3OC6H4. In diethyl ether (30 °C). c In CCl4 (30 °C). Concentration, 8 × 10-2 M. e Concentration, 0.8 M.
Ta ble 4. Con for m a tion a l Equ ilibr ia in 5-Su bstitu ted -2-p h en yl-1,3-d ioxa n es 1, 2, 4, 5, a n d 9, in Absen ce or P r esen ce of
LiBP h 4, a t 25 °C in CHCl3
∆G° (kcal/mol)
compd
X
0.0 equiv of LiBPh4
1.0 equiva of LiBPh4
equivb of LiBPh4
1
2
4
5
9
CO2H
-0.80 ( 0.03
-0.75 ( 0.02
+0.18 ( 0.03
+0.80 ( 0.02
+1.0 ( 0.03
-0.25 ( 0.04
-0.11 ( 0.03
+0.25 ( 0.02
+0.23 ( 0.03
+0.50 ( 0.04
+0.28 ( 0.02
+0.24 ( 0.02
+0.30 ( 0.02
-0.25 ( 0.04
- 0.25 ( 0.03
CO2CH3
CH2OH
OH
HNCOCH3
a
b
Concentration, 5 × 10-2 M. Concentration, 0.5 M.
Ta ble 5. Effect of LiBr on th e 13C Ch em ica l Sh ifts (p p m )
of th e Ca r bon yl Gr ou p OdCsY in th e cis- a n d
tr a n s-Isom er s of 1-3 in THF -d 8 a t 25 °C
lecular hydrogen-bonded species; by contrast, trans-9
present a free N-H group with its IR band at 3480
cm-1 21
.
Sa lt Effects on Con for m a tion a l Equ ilibr ia . As
indicated in the introduction, Seebach, Arnett et al.6
observed large heats of interaction between open-chain
and cyclic peptides with Li+ salts in THF, which sug-
gested strong and specific interactions between the metal
ion and polar groups such as CdO and CsOH. The salt
effects measured in the present work revealed three
tendencies upon salt addition: (1) increased axial prefer-
ence when salt is present in the equilibria of dioxanes
1-4, (2) increased equatorial preference most notably for
9, and (3) no significant salt effect for the conformational
equilibria of 5-8 (Table 3). Nevertheless, lithium ion is
expected to remain in contact with THF solvent,22 and
thus the salt effects tend to be small and/or not monotonic
with LiBr concentration.
In order to confirm the proposed lithium ion effects,
equilibration of dioxanes 1, 2, 4, 5, and 9 was established
in chloroform solvent, a medium where lithium ion is
poorly solvated.2-5 Because LiBr is not sufficiently
soluble in CHCl3, LiBPh4 salt was used instead. The
results are summarized in Table 4 and clearly confirm
the tendencies found in the LiBr/THF system for 1, 2, 4,
and 9. As anticipated, 5-hydroxy derivative 5 exhibits a
strong intramolecular hydrogen bond in CHCl3, which
stabilizes the axial, cis-5, isomer. Addition of LiBPh4
apparently breaks up such hydrogen bond since equilib-
rium shifts toward the equatorial, trans-5, isomer upon
salt addition (Table 4).
compd
Y
0.0 equiv
1.0 equiva
10.0 equivb
cis-1
trans-1
cis-2
trans-2
cis-3
trans-3
OH
OH
OCH3
OCH3
NHCH3
NHCH3
172.9
171.4
171.9
170.6
175.0
170.2
175.1
171.9
172.1
170.6
175.1
171.6
177.5
174.1
172.8
171.6
175.4
172.4
a
b
Concentration, 8 × 10-2 M. Concentration, 0.8 M.
is, it mimicks a change to higher dielectric in the solvent.
This may explain the result with CO2H, where dipole-
dipole interactions should favor the equatorial isomer (eq
1); such interaction is disminished in higher dielectric
and presumably also higher ionic strength solutions.
An alternative interpretation for the increased axial
population of 5-X-1,3-dioxanes with X ) CO2H, CO2CH3,
and CONHCH3 in the presence of LiBr may be that
lithium interacts both with the endocyclic oxygen atoms
and the carbonyl group leading to a stabilization of the
cis (axial) form (cf. D and E). Indeed, in gaseous state,
association between lithium and carbonyl compounds
amounts up to 40-45 kcal/mol.23 In apparent agreement
with this model, examination of the addition of 1.0 and
10.0 equiv of LiBr upon 13C chemical shifts of carbonyl,
shown in Table 5, shows significant downfield shifts for
cis-1 (X ) CO2H). For X ) CO2CH3 and CONHCH3 the
difference between cis and trans is very small, however.
Additional information regarding this point was sought
A plausible explanation for the increased axial prefer-
ence in the presence of lithium salt in dioxanes 1-4 is
that addition of LiBr causes a general salt effect, that
(21) Cf. Baggett, N.; Bukhari, M. A.; Foster, A. B.; Lehmann, J .;
Webber, J . M. J . Chem. Soc. 1963, 4157.
(22) Collum has shown that even TMEDA does not compete with
THF in solvating lithium cations: Colum, D. B. Acc. Chem. Res. 1992,
25, 448.
(23) (a) Wieting, R. D.; Staley, R. H.; Beauchamp, J . L. J . Am. Chem.
Soc., 1975, 97, 924. (b) Murthy, A. S. N.; Bhardwaj, A. P. J . Chem.
Soc., Perkin Trans. 2 1984, 727. (c) Rao, C. H. P.; Balaram, P.; Rao, C.
N. R. J . Chem. Soc. Faraday Trans. 1 1980, 76, 1008.

Conformational Analysis of 5-Substituted 1,3-Dioxanes
J . Org. Chem., Vol. 62, No. 12, 1997 4033
Ta ble 6. Effect of LiBr on 1H Ch em ica l Sh ifts (δ) of cis-
a n d tr a n s-4 in THF a t 27 °C
Ta ble 7. Con for m a tion a l Equ ilibr ia of
2-P h en yl-5-ca r boxy-1,3-d ioxa n es 1, in th e P r esen ce of
LiBr , LiI a n d LiBP h 4, in THF a t 25 °C
∆G° (kcal/mol)
salt
1.0 equiva
10.0 equivb
LiBr
LiI
LiBPh4
-0.41 ( 0.03
-0.20 ( 0.02
-0.26 ( 0.03
-0.17 ( 0.03
-0.11 ( 0.02
-0.17 ( 0.03
compd LiBr equiv H(2) H(4,6-ax) H(4,6-eq) H(5) CH2O OH
a
b
Concentration, 8 × 10-2 M. Concentration, 0.8 M.
cis-4
trans-4
cis-4
trans-4
cis-4
trans-4
0.0
0.0
1.0
1.0
10.0
10.0
5.40
5.35
5.45
5.37
5.50
5.42
4.00
3.65
4.02
3.69
4.12
3.76
4.14
4.19
4.18
4.25
4.21
4.26
1.50 3.83 3.90
2.18 3.34 3.70
1.72 3.90 4.80
2.26 3.41 4.46
1.98 3.96 5.57
2.38 3.52 5.40
the equatorial orientation in the presence of 10 equiv of
LiBr or LiBPh4 (∆G° ) -0.13 kcal/mol or ∆G° ) -0.25
kcal/mol, respectively). This salt effect could be under-
stood when the infrared spectra of cis- and trans-9 were
determined. The cis isomer shows an intramolecular
hydrogen bond (THF, 3330 cm-1, stretching N-H) be-
tween amide hydrogen and ring oxygens, which is absent
in the trans conformer (THF, 3480 cm-1, stretching
N-H). Upon addition of LiBr, the 3330 cm-1 band
disappears and a new band appears at 3480 cm-1 (free
NH). With this information at hand, it seems reasonable
that the axial isomer is intramolecularly hydrogen bonded
in absence of LiBr. When the salt is added, the lithium
ion binds to the carbonyl oxygen.26 Furthermore, it is
known that the anion can bind to amide hydrogen27 or
amide nitrogen,28 so the hydrogen bond is disrupted and
the equilibrium shifts from the axial toward the equato-
rial isomer (eqs 2 and 3).
from semiempirical AM1 modeling studies, as discussed
in section C below.
We hasten to add that the dioxane substrates are likely
to be competing with THF solvent for coordination sites
on lithium.22 Thus species D and E should be in complex
equilibria with uncoordinated heterocycles. Further-
more, chelate substructures D and E are simplified
models for the likely species in the system. In particular,
lithium bromide is known to be aggregated in THF
solutions and at our experimental concentrations prob-
ably dimeric.24
For dioxane 4, X ) CH2OH, the axial isomer is
preferred in high concentration of salt (both THF/LiBr
or CHCl3/LiBPh4). In order to explain this phenomenon
we recorded the 1H NMR spectra for cis- and trans-4 with
1 or 10 equiv of lithium bromide (Table 6). The substan-
tial downfield shifts experienced by the CH2O, H(5), and
OH protons in both the cis and trans isomers support
the idea that the Li+ ion interacts with the hydroxyl
group. Noteworthy is also the observation that H(2) is
more downfield shifted in the cis isomer (∆δ ) 0.1 versus
0.07 ppm). These observations are pertinent in view of
the proposal by Borremans and Anteunis25 that the
conformational preference of 5-(hydroxymethyl)-1,3-di-
oxane is determined by a local electrostatic attraction
between ring oxygens and the partially positive C-H
moiety of the axial 5-CH2Y conformer (F ). AM1 calcula-
tions (Section C) suggested, however, a different picture
(G).
Is Th er e a Cou n ter a n ion Effect? As pointed out
in the previous section, we are dealing not with simple
cations but with ion pairs.29 Thus, the question arose
as to whether the bromide counteranion does have a
specific influence on the observed salt effects. To gain
information on this question, 2-phenyl-5-carboxy-1,3-
dioxane (1) was equilibrated in the presence of lithium
iodide (LiI) and lithium tetraphenylborate (LiBPh4), both
in THF solvent. Comparison with the thermodynamic
data obtained for the same equilibrium in the presence
of LiBr (Table 7) shows a quite good correspondence of
∆G° values, suggesting that salt effects are mainly due
to association with the lithium cation, rather than to the
anion (bromide, iodide, or tetraphenylborate).
C. AM1 Sem iem p ir ica l Ca lcu la tion s. Theorical
modeling of the conformational analysis described in
section B was deemed important in order to increase our
understanding of the equilibria involved, and in order to
The most clear-cut case is with the NHAc substituent
(dioxane 9), where Li salt addition results in a complete
reversal of conformational preference, from a robust axial
predominance in absence of salt (∆G° ) +0.94 kcal/mol
in THF, ∆G° ) +1.00 in CHCl3) to a clear preference for
(26) Bull, W. E.; Madan, S. K. ; Willis, J . E. Inorg. Chem. 1963, 2,
303.
(27) Bufalini, J .; Stern, K. H. J . Am. Chem. Soc. 1961, 83, 4362.
(28) Hinton, J . F.; Amis, E. S.; Mettetal, W. Spectrochim. Acta 1969,
25A, 119.
(24) (a) Abu-Hasanayn, F.; Streitwieser, A. J . Am. Chem. Soc. 1996,
118, 8136. (b) Wong, M. K.; Popov, A. I. J . Inorg. Nucl. Chem. 1972,
34, 3615.
(25) Borremans, F.; Anteunis, J . O. Bull. Soc. Chim. Belg. 1976, 85,
681.
(29) (a) Seebach, D. Angew. Chem., Int. Ed. Engl. 1988, 27, 1624.
(b) J uaristi, E.; Beck, A. K.; Hansen, J .; Matt, T.; Mukhopadhyay, T.;
Simson, M.; Seebach, D. Synthesis 1993, 1271, and references 11-15
therein.

4034 J . Org. Chem., Vol. 62, No. 12, 1997
J uaristi et al.
THF was initially distilled over KOH and then heated to
reflux over sodium/benzophenone (under nitrogen) until the
blue color of the benzophenone ketyl persisted; at this point
the THF was distilled and handled by means of syringes and
cannulas.31 Chloroform (spectroscopic grade) was distilled over
Drierite and further dried over molecular sieves. Flasks,
stirring bars, and hypodermic needles were dried for ca. 12 h
at 120 °C and allowed to cool in a desiccator over anhydrous
calcium sulfate.
Lithium bromide was dried at 190 °C for 4 days. Lithium
tetraphenylborate was prepared by metathesis from a concen-
trate THF solution of NaBPh4 which was added to a concen-
trated THF solution of LiCl.32 The insoluble NaCl precipitate
was removed by filtration, and the lithium salt was recovered
by evaporating the solvent. The salt was recrystalized from
chloroform and then dried over P2O5 under vacuum for 24 h
at room temperature followed by 24 h at 80 °C.
gain support for the interpretations advanced. Table 8
(Supporting Information) collects the differences in heat
of formation (∆H°f) for the axial versus equatorial
conformers of lowest energy, as derived from AM1
semiempirical calculations.30 Table 8 also includes the
corresponding ∆H°f values in the presence of H+, Li+,
Na+, and K+ ions. With very few exceptions, incorpora-
tion of a proton (H+) or cation (Li+, Na+, K+) leads to
substantial stabilization of the axial isomers. While this
observation could be in line with the idea of intramo-
lecular bridging of the positive ion, closer examination
of the modeled species suggests a more complex situation.
Figures 1-8 (Supporting Information) present the
conformers of lowest energy for the 5-substituted 1,3-
dioxanes affording the computed data summarized in
Table 8. Furthermore, modeling of the lithium ion effects
by means of AM1 semiempirical calculations argues
against a unique interaction mechanism; rather, several
different perturbation modes of conformational equilibria
were revealed:
(1) Lithium ion shifts the position of the conformational
equilibria of 5-carboxy- and (N-methylcarbamoyl)-1,3-
dioxane toward the axial side via coordination to the
carbonyl oxygen, which conducts to the formation of
strong intramolecular O-H and N-H hydrogen bonding
with both ring oxygens.
(2) Lithium cation perturbs the equilibrium of 5-car-
bomethoxy- and 5-(hydroxymethyl)-1,3-dioxane via coor-
dination between lithium, the carboxy, or hydroxy group,
and one of the endocyclic oxygens, stabilizing the axial
isomer.
Microanalyses were performed by Galbraith Laboratories,
Inc., Knoxville, TN.
cis- a n d tr a n s-2-P h en yl-5-ca r boxy-1,3-d ioxa n e (cis-
a n d tr a n s-1). These compounds were prepared as described
in the literature.12 Diethyl bis(hydroxymethyl)malonate (11.4
g, 0.06 mol) was condensed with 6.5 g (0.06 mol) of benzalde-
hyde, affording 16.1 g (95.2% yield) of 2-phenyl-5,5-dicarb-
ethoxy-1,3-dioxane, 10, as a yellow liquid, bp 170-175 °C/4
mmHg. Saponification was effected by addition of diester 10
(77.0 g, 0.25 mol) to 60 g (1.07 mol) of KOH in 500 mL of 95%
ethanol and boiling for 1 h. The ethanol was evaporated, and
the resulting solid was dissolved in 400 mL of dichlo-
romethane. The solution was chilled in an ice bath and
acidified with 10% hydrochloric acid. The organic layer was
separated and dried over anhydrous Na2SO4. Concentration
in a rotary evaporator afforded 24.1 g (40.4% yield) of 2-phenyl-
5,5-dicarboxy-1,3-dioxane (11) as a white solid with mp 148-
150 °C. Finally, a magnetically stirred mixture of dicarboxylic
acid 11 (10.0 g, 0.039 mol) and 15 mL of triethylamine was
heated to reflux for 40 min. The triethylamine was distilled,
and the remaining solid was dissolved in 100 mL of dichlo-
romethane. The solution was chilled in an ice bath and
acidified until pH 2 with 10% hydrochloric acid. The reaction
mixture was extracted with ether (2 × 100 mL), and the
combined ether extracts were dried over anhydrous Na2SO4.
Concentration in a rotary evaporator afforded 7.3 g (86.2%
yield) of a mixture of isomers. The mixture was separated by
flash chromatography over silica gel (hexane/ethyl acetate/
dichloromethane/acetone/methanol, 8.0:0.5:0.5:0.5:0.5) to afford
2.5 g (34.2% yield) of cis-1 as a white solid, mp 148-149 °C,
and 3.8 g (52.0% yield) of trans-1 as a white solid, mp 168-
169 °C. 1H and 13C NMR spectra in Tables 1 and 2, respec-
tively. Anal. Calcd for C11H12O4: C, 63.45; H, 5.81. Found
(cis isomer): C, 63.64; H, 6.05. Found (trans isomer): C, 63.49;
H, 5.87.
cis- a n d tr a n s-2-P h en yl-5-ca r bom eth oxy-1,3-d ioxa n e
(cis- a n d tr a n s-2). A mixture of cis-1 (13.4 g, 0.064 mol) and
silver oxide (29.7 g, 0.13 mol) was cooled to 0 °C in an ice bath
and treated with 36.5 g (16 mL, 0.25 mol) of methyl iodide.
The reaction mixture was stirred at 0 °C for 2 h before the
addition of 50 mL of dichloromethane, the solid was filtered,
and the filtrate was concentrated in a rotary evaporator.
Purification of the crude product was accomplished by crystal-
lization from hexane to give 14.3 g (91.8% yield) of cis-2 as a
white solid, mp 75-76 °C. 1H and 13C NMR spectra in Tables
1 and 2, respectively. Anal. Calcd for C12H14O4: C, 64.85; H,
6.35. Found: C, 65.10; H, 6.48.
In similar fashion, trans-1 was converted to trans-2 in 94.2%
yield, mp 79-81 °C. 1H and 13C NMR in Tables 1 and 2,
respectively. Anal. Calcd for C12H14O4: C, 64.85; H, 6.35.
Found: C, 64.82; H, 6.51.
cis- a n d tr a n s-2-P h en yl-5-(N-m eth ylca r ba m oyl)-1,3-
d ioxa n e (cis- a n d tr a n s-3). A solution of 5.0 g (1.87 mmol)
of cis-2-phenyl-5-carbomethoxy-1,3-dioxane (2), 15 mL of
methanol, and 5 mL of 40% aqueous methylamine was left
standing for 12 h at 4 °C. The reaction mixture was filtered,
(3) Quite a different situation was observed for 9, X )
NHCOCH3, where salt addition breaks up intramolecular
hydrogen bonding present in the salt-free axial isomer.
Of course, the results of these calculations have the
limitation that bare cations, rather than solvated ions
and ions pairs, are used.
Con clu sion s
In absence of salt, the conformational behavior of
5-polar-substituted-1,3-dioxanes is generally determined
by an electrostatic dipole-dipole repulsion that favors
the equatorial isomer. Nevertheless, compensating ef-
fects such as intramolecular hydrogen bonding, electro-
static attraction, and stereoelectronic effects may shift
the equilibrium toward the axial isomers.
The conformational preference of 5-substituted-1,3-
dioxanes can be affected by addition of lithium salts, and
this effect is stronger in CHCl3 relative to THF solvent,
in agreement with expectation based on cation solvation
abilities. By contrast, the nature of the counteranion
does not play a significant role on the observed salt
effects.
Exp er im en ta l Section
Gen er a l In for m a tion . Melting points were determined
with open capillary tubes and are uncorrected. 1H and 13C
NMR spectra were recorded in CDCl3, DMSO-d6, or THF-d8
solution with tetramethylsilane (TMS) as an internal standard.
Chemical shifts are given as δ values (ppm) and coupling
constants (J ) in Hz.
(30) (a) Dewar, M. J . S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J . J .
P. J . Am. Chem. Soc. 1985, 107, 3902. (b) Dewar, M. J . S.; Reynolds,
C. H. J . Comput. Chem. 1986, 2, 140. (c) Cramer, C. J ., Truhler, D.
G.; Science 1992, 256, 213. (d) Cramer, C. J .; Truhler, D. G. J . Comput.
Aided Mol. Des. 1992, 6, 69
(31) Brown, H. C.; Kramer, G. W.; Levy, A. B.; Midland, M. M.
Organic Synthesis via Boranes; Wiley: New York, 1975; p 256.
(32) Kunze, R. W.; Fuoss, M. J . Phys. Chem. 1967, 67, 385.

Conformational Analysis of 5-Substituted 1,3-Dioxanes
J . Org. Chem., Vol. 62, No. 12, 1997 4035
and the resulting solid consisted of a mixture of cis-3 and
trans-3 (30:70). Fractional recrystallization afforded 500 mg
(10% yield) of cis-3 as a white solid, mp 118-120 °C. 1H and
13C NMR spectra in Tables 1 and 2, respectively. Anal. Calcd
for C12H15NO3: C, 65.13; H, 6.83. Found: C, 65.38; H, 7.05.
Similarly, trans-2 was converted to trans-3 in 96.4% yield,
mp 193-195 °C. 1H and 13C NMR spectra in Tables 1 and 2,
respectively. Anal. Calcd for C12H15NO3: C, 65.13; H, 6.83.
Found: C, 64.75; H, 6.97.
cis-2-P h en yl-5-n itr o-1,3-d ioxa n e (cis-8). According to the
procedure of Marei and Raphael,16 2-phenyl-5-(hydroxy-
methyl)-5-nitro-1,3-dioxane, 1234 (24.0 g, 0.1 mol), was treated
with a solution of lithium in liquid ammonia (1.2 g of lithium
and 400 mL of ammonia) to furnish 11.7 g (52.6% yield) of
cis-8 as the sole product, mp 127-128 °C (lit.16 mp 127-129
°C). 1H and 13C NMR spectra in Tables 1 and 2, respectively.
The sublimation of cis-8 at 135-140 °C/4 mmHg afforded a
1:1 mixture of cis-8 and trans-8.
cis- a n d tr a n s-2-P h en yl-5-(h yd r oxym eth yl)-1,3-d iox-
a n e (cis- a n d tr a n s-4). A solution of cis-2-phenyl-5-car-
bomethoxy-1,3-dioxane (2) (4.0 g, 0.015 mol) in 50 mL of
anhydrous ether was added (15 min) to a suspension of lithium
aluminum hydride (1.5 g, 0.039 mol) in 50 mL of anhydrous
ether under nitrogen atmosphere. The resulting suspension
was stirred at rt for 1 h and then was heated to reflux for 30
min. The suspension was treated with ethyl acetate (15 mL)
and a 10% solution of NaOH (ca. 100 mL). The reaction
mixture was extracted two times with ether, and the combined
organic extracts were dried over anhydrous Na2SO4 and
concentrated to provide the crude product. Recrystallization
from hexane gave 3.03 g (84.6% yield) of cis-4 as a white solid
with mp 61-62 °C. 1H and 13C NMR spectra in Tables 1 and
2, respectively. Anal. Calcd for C11H14O3: C, 68.02; H, 7.26.
Found: C, 68.21; H, 7.48
The trans isomer was separated by flash chromatography
(hexane/ethyl acetate, 9:1), mp 68-69 °C. 1H and 13C NMR
spectra in Tables 1 and 2, respectively. Anal. Calcd for
C10H11NO4: C, 57.41; H, 5.30. Found: C, 57.36; H, 5.49.
cis- a n d tr a n s-2-P h en yl-5-a ceta m id o-1,3-d ioxa n e (cis-
a n d tr a n s-9). A solution of cis-2-phenyl-5-nitro-1,3-dioxane
(cis-8) (12.3 g 0.06 mol) in 20 mL of ethanol was added to a
suspension of 10% Pd(C) (1.2 g, 1.14 mmol) and then was
exposed to hydrogen (800 psi) with stirring and heating (30
°C) during 3 h. The catalyst was removed by filtration (Celite),
and the filtrate was concentrated at reduced pressure to afford
9.74 g (92.6%, yield) of cis-13 as a colorless oil. This inter-
mediate, cis-13 (10 g, 0.06 mol), was treated with 50 mL of
acetic anhydride and 1 mL of pyridine. The reaction mixture
was stirred at rt for 2 h, and then the excess of acetic
anhydride was distilled at reduced pressure. The solid residue
was crystallized from ethyl acetate to afford 11.5 g (93.5%
yield) of the cis-9 as a white solid with mp 153-154 °C. 1H
and 13C NMR spectra in Tables 1 and 2, respectively. Anal.
Calcd for C12H15NO3: C, 65.13; H, 6.83. Found: C, 64.77; H,
6.93.
Similarly, trans-2 was converted to trans-4 in 94.8% yield,
mp 88-89 °C. 1H and 13C NMR in Tables 1 and 2, respectively.
Anal. Calcd for C11H14O3: C, 68.02; H, 7.26. Found: C, 68.23;
H, 7.57.
cis- a n d tr a n s-2-P h en yl-5-h yd r oxy-1,3-d ioxa n e (cis-
a n d tr a n s-5). These compounds were prepared by condensa-
tion of glycerol (50 g, 0.54 mol) and benzaldehyde (50 g, 0.48
mol) according to a published procedure.1,33 The crude product
(81.7 g) consisted of a mixture of 1,3-dioxanes and 1,3-
dioxolanes. This mixture was separated by flash chromatog-
raphy (hexane/ethyl acetate, 1:1) to afford 9.6 g (11.6% yield)
of trans-5 as a white crystals, mp 63-65 °C (lit.33 mp 64.5-
65.5 °C) and 25.1 g (29.2% yield) of cis-5 as a white crystals,
mp 83-85 °C (lit.33 mp 83-85 °C). 1H NMR spectrum in Table
1. 13C NMR spectrum in Table 2.
cis- a n d tr a n s-2-P h en yl-5-O-a cetyl-1,3-d ioxa n e (cis-
a n d tr a n s-6). These compounds were prepared as described
in the literature.33 Carbinol cis-5 (3.0 g, 0.016 mol) in pyridine
(0.5 g, 0.006 mol) was treated with acetic anhydride (15.0 g,
0.146 mol). The reaction mixture was stirred at rt for 15 h
and then poured into ice-water; the precipitate was collected,
dissolved in chloroform, washed successively with ice-cold 1
N hydrochloric acid, saturated aqueous hydrogen carbonate,
and water, and finally dried over anhydrous Na2SO4. Con-
centration in a rotary evaporator afforded the crude product
which was crystallized from hexane to give 3.38 g (95.3% yield)
of cis-6 as a white solid with mp 98-100 °C (lit.33 mp 99-100
°C). 1H NMR and 13C NMR spectra in Tables 1 and 2,
respectively.
Similarly, trans-8 was converted to trans-9 in 90.1% yield,
mp 202-203 °C. 1H and 13C NMR spectra in Tables 1 and 2,
respectively. Anal. Calcd for C12H15NO3: C, 65.13; H, 6.83.
Found: C, 64.84, H, 6.98.
E q u ilib r a t ion s a n d An a lysis. Equilibrium was ap-
proached from both sides; boron trifluoride etherate was the
catalyst: ca. 60 mg of dioxane and 1 or 10 equiv of Li salt
were placed in a 5 mL ampoule and dissolved in 3 mL of
solvent before the addition of two drops of the catalyst. The
ampoule was sealed and submerged in a constant-temperature
bath until equilibrium was reached. The progress of the
equilibration was conveniently monitored by 1H NMR spec-
troscopy. Quenching was effected by pouring the equilibration
solution into aqueous 30% sodium bicarbonate. The dioxanes
were then extracted with ether, dried over anhydrous Na2SO4,
and evaporated.
Ack n ow led gm en t. We are indebted to Consejo
Nacional de Ciencia y Tecnolog´ıa for financial support
(Grant 3534 E-9311) and to G. Uribe, O. Garc´ıa-
Barradas, V. Gonza´lez, and G. Zepeda for recording the
NMR spectra. F.D. wishes to thank IPN and CoNaCyT
for a scholarship. We are most grateful to Professors
Ernest Eliel and Andrew Streitwieser, and to the
referees for many valuable comments on this work.
Similarly, trans-5 was converted to trans-6 in 90% yield.
This compound formed as white crystals with mp 114-116 °C
(lit.33 mp 115-116 °C).
cis- a n d tr a n s-2-P h en yl-5-[(2-m et h oxyp h en oxy)a ce-
toxy]-1,3-d ioxa n e (cis- a n d tr a n s-7). These compounds
were synthesized according to the published procedure.15
A
solution of (2-methoxyphenoxy)acetic acid14 (1.82 g, 0.01 mol)
in pyridine (50 g, 0.63 mol) was treated with p-toluenesulfonyl
chloride (4 g, 0.02 mol). The solution was cooled in an ice bath
before the addition of 1.8 g (0.01 mol) of carbinol cis-5. The
reaction mixture was stirred at 0 °C for 1 h and then poured
into three to four volumes of ice and water. The resulting solid
esters were collected by filtration to afford 3.1 g (90.1% yield)
of cis-7 as a white solid, mp 91-92 °C. 1H NMR and 13C NMR
spectra in Tables 1 and 2, respectively. Anal. Calcd for
C19H20O6: C, 66.27; H, 5.85. Found: C, 66.45; H, 5.56.
In similar fashion, trans-5 was converted to trans-7 in 92.4%
yield, mp 102-103 °C. 1H NMR and 13C NMR spectra in
Tables 1 and 2, respectively. Anal. Calcd for C19H20O6: C,
66.27, H, 5.85. Found: C, 66.23; H, 6.00.
Su p p or tin g In for m a tion Ava ila ble: Calculated (MO-
PAC/AM1) heats of formation and structures of minimum
energy for axial and equatorial 5-substituted 1,3-dioxanes
(color images available only electronically) (10 pages). This
material is contained in libraries on microfiche, immediately
follows this article in the microfilm version of the journal, and
can be ordered from ACS; see any current masthead page for
ordering information and for Internet access instructions.
J O9610117
(34) 2-Phenyl-5-(hydroxymethyl)-5-nitro-1,3-dioxane (12) was pre-
pared by condensation of 2-(hydroxymethyl)-2-nitro-1,3-propanediol (15
g, 0.1 mol) and benzaldehyde (10 g, 0.94 mol) according to a published
procedure.19
(33) Baggett, N.; Brimacombe, J . S.; Foster, A. B.; Stacey, M. M.;
Whiffen, D. H. J . Chem. Soc. 1960, 2574.