The conformational energies of 2-methyl- and 4-methyl-1,3-dithiane. The breakdown of 1,3-syn diaxial repulsion hypothesis

Article abstract of DOI:10.1016/S0022-2860(03)00357-0

The conformational enthalpies (ΔH) of 2-methyl- (-1.76 kcal mol ^-1) and 4-methyl-1,3-dithiane (-1.75 kcal mol^-1) were obtained by the analysis of ¹³C chemical shifts as a function of temperature. These energies are in excellent agreement both with calculated values (B3LYP/6-31G(d,p)) and with literature values based on 2,4-dialkyl-1,3-dithiane. The results confirm the similarity between the conformational energies of the methyl group at the 2 and 4 positions of the dithiane ring and that of methylcyclohexane, despite the larger distances between ring 1,3-syn-axial hydrogens and the closest methyl axial hydrogen in the dithiane ring. The possibility of a buttressing effect on the 2,4-dialkyl-1,3-dithianes previously studied and the rationale of 1,3-syn-axial steric interaction are discussed. ζ 2003 Elsevier B.V. All rights reserved.

Full text of DOI:10.1016/S0022-2860(03)00357-0

Journal of Molecular Structure 657 (2003) 85–92
www.elsevier.com/locate/molstruc
The conformational energies of 2-methyl- and 4-methyl-1,3-
dithiane. The breakdown of 1,3-syn diaxial repulsion hypothesis
*
Douglas S. Ribeiro , Roberto Rittner
´
Physical Organic Chemistry Laboratory, Instituto de Quımica, UNICAMP, Caixa Postal 6154, 13083-862 Campinas, S.P., Brazil
Received 27 March 2003; revised 20 May 2003; accepted 21 May 2003
Abstract
The conformational enthalpies (DH) of 2-methyl- (21.76 kcal mol21) and 4-methyl-1,3-dithiane (21.75 kcal mol21) were
obtained by the analysis of 13C chemical shifts as a function of temperature. These energies are in excellent agreement both with
calculated values (B3LYP/6-31G(d,p)) and with literature values based on 2,4-dialkyl-1,3-dithiane. The results confirm the
similarity between the conformational energies of the methyl group at the 2 and 4 positions of the dithiane ring and that of
methylcyclohexane, despite the larger distances between ring 1,3-syn-axial hydrogens and the closest methyl axial hydrogen in
the dithiane ring. The possibility of a buttressing effect on the 2,4-dialkyl-1,3-dithianes previously studied and the rationale of
1,3-syn-axial steric interaction are discussed.
q 2003 Elsevier B.V. All rights reserved.
Keywords: Dithianes; VT NMR; Conformational equilibrium; Syndiaxial repulsions
1. Introduction
conformational equilibrium, by the equation DG ¼
2RT lnðNeq=NaxÞ:
The larger energy of the axial conformer has long
been attributed to the repulsive steric interaction
between 1,3-syn-axial hydrogens and the methyl
hydrogens. Such a view is widely found in the
literature [1,3,5–10].
Methylcyclohexane is considered to be a key
compound in the area of conformational analysis [1,
2]. The methyl group in it and in several other methyl-
heterocyclohexanes can adopt two conformations, the
axial or the generally most stable equatorial one. The
differences in these conformational energies have
been determined by a number of experimental and
theoretical methods [1,3,4].
This rationale works quite well in many instances,
as in the drop in energy in 3-methyl-cyclohexanones
[11,19], -pyrans [12,19], -thianes [13], and -piper-
idines [14], which show an almost constant value of
1.4 kcal mol21 compared to the 1.8 kcal mol21 in
methylcyclohexane. This drop in energy results from
the absence of one syn-axial proton. Also, the
increased conformational energies of 2-methylpyran
The experimental determination of these energies
is carried out by measuring the mol fractions of the
axial (Nax) and equatorial (Neq) conformers, which
are related to the free energy of the axial–equatorial
(2.9 kcal mol21
(2.5 kcal mol21) [14] are explained on the basis of
)
[12] and 2-methylpiperidine
*
Corresponding author. Tel./fax: þ55-19-3788-3150.
E-mail address: dribeiro@iqm.unicamp.br (D.S. Ribeiro).
0022-2860/03/$ - see front matter q 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0022-2860(03)00357-0

86
D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
the shorter distances between the syn-axial and the
methyl protons brought about by the shorter C–O and
C–N bond lengths. Accordingly, 2-methylthiane has a
lower conformational energy (1.4 kcal mol21) due to
the longer distance between the methyl and the ring
syn-axial hydrogens stemming from the longer C–S
bond [13]. The conformational energies of 4-methyl-
1,3-dioxane (2.9 kcal mol21) [15] and 2-methyl-1,
3-dioxane (3.6 kcal mol21) can equally well be
understood on this basis. The latter compound has a
larger energy difference supposedly because of two
close syn-axial protons.
[17] would be higher than the true values of the
conformational energies of pure 2-methyl-1,3-
dithiane and 4-methyl-1,3-dithiane. In such a case,
the methyl conformational energies in the dithiane
ring would fit perfectly with the qualitative expec-
tation of
methylcyclohexane.
a
lower energy compared to
With this in mind, the conformational energies of
2-methyl and 4-methyl-1,3-dithiane were determined
directly by a different method. This paper reports the
conformational energies of 2-methyl (I) and
4-methyl-1,3-dithiane (II) (Fig. 1) obtained by the
analysis of the temperature dependence of C-13
chemical shifts. This method has shown to yield
accurate conformational energies for a number of
methyl cycloalkanes [19,20]. The energies and
geometries of the title compounds by DFT/B3LYP/
6-31G(d,p) have also been calculated, to analyse the
distances between the methyl hydrogens and the 1,3-
syn hydrogens of the dithiane ring. The conformation-
al energies for 2,4-dimethyl-1,3-dithiane were also
calculated, in order to find out if there is any
buttressing effect in these molecules [1,18].
The difficulty arises when 2-methyl- and 4-methyl-
1,3-dithiane are considered. In these cases, despite the
longer C–S bond lengths, the energies are the same as
in methylcyclohexane. Eliel et al. [13,16] had
suggested that the dithiane ring has the same distances
between syn-axial protons and methyl protons as in
methylcyclohexane due to a peculiarity of the dithiane
ring. A survey of the literature demonstrates that the
conformational energies of 2-methyl- and 4-methyl-
1,3-dithiane have been measured only twice and by
the same methodology. Eliel and Hutchins [16]
determined the conformational energies of 2-methyl-
1,3-dithiane by the chemical equilibration of cis-4,6-
dimethyl-2-methyl-1,3-dithiane and obtained a value
of 1.77 kcal mol21. The conformational energy of the
2. Theoretical
4-methyl-1,3-dithiane (1.69 kcal mol21
)
was
The method used in this paper for the determi-
nation of DH values of the axial–equatorial equilibria
has been described elsewhere [19] and was shown to
give reliable results for a number of methyl sub-
stituted cycloalkanes. For this reason, only a brief
account will be provided. The chemical shift of
any given carbon of a molecule in a conforma-
tional equilibrium, such as that represented in Fig. 1,
measured by the chemical equilibration of 2-tert-
butyl-4-methyl-1,3-dithiane. Pihlaja [17] also deter-
mined the conformational energy of 4-methyl-1,
3-dithiane by the chemical equilibration of 2-tert-
butyl-4-methyl-1,3-dithiane, but he obtained a
slightly lower value of 1.55 kcal mol21. The 2-methyl
energy (1.92 kcal mol21) was obtained by the equi-
libration of 2,4-dimethyl-1,3-dithiane.
The chemical equilibration method provides an
accurate determination of equilibrium constants
because the populations are usually determined by
gas chromatography. However, the method relies on
the suitability of rigid molecules to model the
conformational energies of monosubstituted com-
pounds and this may not always be very accurate. It
has been pointed out that the buttressing effect of 1,3-
dialkyl groups may increase the energy of the alkyl
substituent [1,18]. If this possibility proved to be a
reality for the 2,4-dialkyl-1,3-dithiane, the values
obtained by Eliel and Hutchins [16] and by Pihlaja
Fig. 1. Conformational equilibrium of 2-methyl- (I) and 4-methyl-
1,3-dithiane (II), including carbon atoms numbering.

D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
87
is a weighted average of the contribution of each
constituent conformer
small [19]. There have been some theoretical
calculations [2] of the conformational entropy of the
methyl group and values up to 0.2 cal mol21 K21
were obtained. Thus, DS is fixed in our calculations as
an upper limit of ^0.2 cal mol21 K21. In this case
DG differs from DH by 0.06 kcal mol21 at room
temperature.
dobs ¼ naxdax þ neqdeq
ð1Þ
where nax and neq are the mole fractions and dax and
deq are the chemical shifts of the axial and equatorial
conformers, respectively. The thermodynamical par-
ameters DH and DS are related to the mole fractions as
given by Eq. (2):
3. Experimental
DH 2 TDS ¼ 2RT lnðneq=naxÞ
ð2Þ
13C and 1H NMR spectra of ca. 40 mg ml21
solutions in CDCl3 referenced to TMS in 5 mm o.d.
tubes were obtained on a Varian Gemini-300
spectrometer operating at 75.45 MHz for carbon and
300.067 MHz for proton. The conditions for 13C
NMR spectra were as follows: 128 transients,
relaxation delay of 2.0 s, spectra width 6802.7 Hz
with 32k data points zero filled to 64k for a digital
The temperature dependence of the observed
chemical shift (dobs) is from two sources:
(1) an intrinsic dependence of the carbons of each
conformer which can be described by a linear function
as follows
dax ¼ a0 þ bT
deq ¼ a þ bT
ð3Þ
1
resolution of 0.21 Hz. The conditions for H NMR
or in some cases a quadratic equation may be more
appropriate
spectra were as follows: 64 transients, relaxation
delay 0.2 s, flip angle 37.18, spectral width 2401.4 Hz,
32k data points zero filled to 64k to yield a digital
resolution of 0.07 Hz. The probe temperature was
checked against a thermocouple and was accurate to
^2 K.
dax ¼ a0 þ bT þ cT2
deq ¼ a þ bT þ cT2
and
ð4Þ
1,3 Dithiane (III) was available from Aldrich. 2,4-
Dimethyl-1,3-dithiane (IV) and 2-methyl-1,3-dithiane
(I) were prepared by the reaction of acetaldehyde in
benzene with 1,3-butanedithiol [16] and 1,3-propane-
dithiol (Aldrich), respectively, following the direc-
tions given by Eliel and Hutchins [16].
(2) a conformational dependence which is brought
about because the populations of the conformers vary
with temperature.
The method is based on the curve fitting of the
observed carbon chemical shifts as a function of
temperature, according to the equation
2,4-Dimethyl-1,3-dithiane (IV): yield 74%, NMR
1
cis form H (300 MHz) d 1.22 (d, 6.6 Hz, 3-H, 4-
dax 2 deq
ꢀ
ꢁ
dobs
¼
þ deq
ð5Þ
methyl), 1.48 (d, 6.9 Hz, 3-H, 2-methyl), 1.48 (m, 1-
H, H-5ax), 2.11 (dm, 1-H, H-5eq), 2.9 (m, 3-H, H-4ax,
6ax, 6eq), 4.14 (q, 6.9 Hz, 1-H, H-2ax).
2-Methyl-1,3-dithiane (I): yield 81%, NMR 1H
(300 MHz) d 1.47 (d, 6.9 Hz, 3-H, methyl), 1.81 [qt,
DH
DS
1 þ exp 2
þ
RT
R
where dax and deq are the chemical shifts of the axial
and equatorial conformers as given by Eq. (3) or (4).
The intrinsic temperature dependence of the
carbon chemical shifts of the individual conformers
(parameters b of Eq. (3) and b and c of Eq. (4)) are
taken from conformationally homogeneous reference
molecules. Thus, the curve fitting of the experimental
data (dobs; T) into Eq. (5) yields values for the
four unknowns, DH; DS and the intercepts a of the
equatorial and axial conformers. We have shown that
DS for a number of methyl substituted cycloalkanes is
Jeq,ax ø Jeq,eq ¼ 3.3 Hz (triplet),
J
geminal ø
Jax,ax ¼ 11.8 Hz (quartet), 1-H, H-5 ax ], 2.11 (dm,
13.9 Hz, 1-H, H-5eq), 2.81 (dt, 2-H, H-4eq, 6eq), 2.91
(m, 2-H, H-4ax, 6ax), 4.13 (q, 6.9 Hz, 1-H, H-2ax).
4-Methyl-1,3-dithiane (II) was prepared as fol-
lows: four grams (0.0328 mol) of 1,3-butanedithiol
[16] were dissolved in 30 ml of dichloromethane with
ca. 0.15 g of p-toluenesulfonic acid. 1.5 g (0.05 mol)
of paraformaldehyde were depolimerised by heating

88
D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
in a flask connected to the reaction flask through
tubing. After three hours, the reaction medium was
washed with two portions of dilute sodium hydroxide
solution and then with water. The solvent was
evaporated and the crude product was distilled in a
Kugelrohr to yield 1.5 g of product. NMR 1H
(300 MHz) d 1.22 (d, 6.9 Hz, 3-H, methyl), 1.67 (m,
1-H, H-5ax), 2.15 (dm, 14.1 Hz, 1-H, H-5eq), 2.83–
2.90 (m, 3-H, H-4ax, 6ax, 6eq), 3.53 (d, 14.1 Hz, 1-H,
H-2eq), 4.07 (d, 14.1 Hz, 1-H, H-2ax).
Table 1
13C chemical shifts for compounds I–IV, in CDCl3, as a function of
the temperature
Compd.
T
C-2
C-4
C-5
C-6
2-Methyl 4-Methyl
(K)
I
323 42.02 30.67 25.37 30.67 21.30
303 42.01 30.66 25.28 30.66 21.27
278 42.00 30.63 25.17 30.63 21.25
263 42.00 30.61 25.09 30.61 21.24
243 41.99 30.59 25.00 30.59 21.23
223 41.99 30.57 24.91 30.57 21.24
323 32.36 38.18 35.34 30.19
13C chemical shift assignments were performed
according to the literature data for all four compounds
[21].
II
21.95
21.97
22.00
22.03
22.06
22.11
303 32.36 38.17 35.27 30.18
278 32.36 38.16 35.16 30.17
DFT calculations were performed with the
B3LYP/6-31G(d,p) basis set in the GAUSSIAN 98
program [22] using the computational facilities of
CENAPAD-SP.
263 32.36 38.15 35.09 30.16
243 32.35 38.15 35.00 30.14
223 32.38 38.15 34.91 30.14
III
IV
323 32.00 29.98 26.68 29.98
303 31.96 29.93 26.61 29.93
The curve fitting program used was described in a
previous paper [19]. Only one minor modification was
introduced to allow for the quadratic term of Eq. (4).
278 31.90 29.85 26.50 29.85
263 31.86 29.80 26.44 29.80
243 31.79 29.72 26.35 29.72
223 31.74 29.66 26.26 29.66
323 43.25 39.63 34.28 31.59 20.70
303 43.18 39.56 34.16 31.50 20.70
278 43.09 39.47 34.01 31.39 20.71
263 43.03 39.41 33.91 31.32 20.73
243 42.97 39.34 33.78 31.22 20.76
223 42.91 39.29 33.66 31.13 20.80
21.69
21.70
21.71
21.72
21.73
21.75
4. Results
4.1. Reference compounds
Fig. 1 shows the conformational equilibria under
study, including the carbon atom numbering. Table 1
presents the 13C chemical shifts of compounds I–IV,
in CDCl3 solutions, as a function of temperature.
The reference compounds used here were 1,3-
dithiane (III) and cis-2,4-dimethyl-1,3-dithiane (IV).
The former undergoes ring inversion, but with two
equivalent chair conformations. Thus, the 13C chemi-
cal shifts are not dependent on conformational effects.
Compound IV is fixed in a single conformation,
because the alternate conformation has two axial
methyl groups, 1,3 to each other, leading to a higher
energy conformer. This makes compound IV a
reference compound, because the dependence of the
carbon chemical shifts on temperature is not related to
any conformational effect.
the carbon atoms that are at a gamma position to
the substituted carbon are preferred, because they
exhibit a large difference in chemical shifts,
between the two conformers, and the temperature
gradients (b and c) are better modelled by the
reference molecule.
The chemical shifts used for the calculations of
2-methyl-1,3-dithiane (I) were those of the methyl
carbon and of C-4/C-6. Therefore, the correspond-
ing methyl carbon of IV and C-4/C-6 of III were
used as the reference carbons, respectively. C-6 and
C-2 chemical shifts were chosen for the equili-
brium of 4-methyl-1,3-dithiane (II). All 13C
chemical shifts of compounds III and IV showed
a linear dependence on T; except the methyl carbon
(bonded to C-2) of IV, whose plot of d versus T is
a curve.
The reference compounds III and IV provide
the parameters b and c of Eqs. (3) and (4). In
principle, any carbon chemical shift may be used
for the calculation of DH and, in fact, more than
one carbon atom has been used to obtain the
conformational energies in earlier papers. However,
Table 2 presents the results of the linear regression
analyses for the chemical shifts of the reference

D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
89
Table 2
Regression analyses for 13C chemical shifts as a function of the temperature
Compd.
Carbon
a (ppm)a
b (104 ppm K21
)
c (105 ppm K22
)
r
sd (ppm)
III
III
IV
IV
C4,6
29.83
31.15
30.10
21.98
32.8
26.6
–
0.9980
0.9960
0.9998
0.9990
0.0086
0.0096
0.0040
0.0018
C2
–
C6
46.2
–
Methyl-C2
282.4
1.325
a
The number of fitted points is six and temperature range is 223–323 K.
carbons as a function of temperature. The correlation
coefficients (r) are greater than 0.995 and the standard
deviations (sd) smaller than the experimental accu-
racy of the chemical shift measurements (0.01 ppm).
The gradient of C-6 for compound IV is consider-
ably larger than that of C-6 for compound III. The
difference of 0.00134 ppm K21 is due to the presence
of two gamma equatorial methyl groups in IV. The
substituent effects of the methyl group in the dithiane
ring is significantly higher than in other rings [21].
The ganti effect of a methyl group in dithiane is ca.
1.2 ppm, compared to ca. 0 ppm in the cyclohexane
before with other methyl substituted cyclic com-
pounds. Thus, from all the possible solutions the one
with the expected (c and a) values is chosen. The
chemical shift differences are known from the
substituent chemical shifts found in the literature.
From Eliel et al. [21], the chemical shift differences
between axial and equatorial 4-methyl-1,3-dithiane
are 8.1 ^ 0.6 ppm for C-2 and 7.0 ^ 0.5 ppm for C-6.
For 2-methyl-1,3-dithiane the differences are
6.8 ^ 0.4 ppm for C-4 and C-6, and 4.2 ^ 0.9 ppm
for the methyl carbon.
The 13C spectrum of 4-methyl-1,3-dithiane at
223 K presents broad signals. Therefore, the curve
fitting was restricted to the 243–323 K interval.
Table 3 presents the results of curve fitting for I
and II, along with the gradients used in the
calculations.
ring.
Consequently,
a
correction
of
þ0.0007 ppm K21 for the gradients was estimated,
to be used in the calculations of DH: In other words,
0.0007 was added to the observed gradients of C-6
and C-2 of III, used in the calculations for 4-methyl-
1,3-dithiane and 2-methyl-1,3-dithiane.
4.3. Calculated energies and geometries
4.2. Conformationally mobile compounds
Table
4 presents the calculated distances
A common feature of the observed carbon
chemical shifts for compounds I and II, as a function
of temperature, is the absence of any significant
curvature. This means that Eq. (4) can only yield two
parameters, not the three unknown parameters needed
(DH; a; c). The true solution can be found, provided
the difference of the chemical shifts between the
conformers is known. This procedure has been used
between the closest methyl proton and the syn-
axial ring protons for I and II in the axial
conformation and for 2ax-4eq-dimethyl-1,3-dithiane
and 2-eq-4ax-dimethyl-1,3-dithiane and the corre-
sponding distances in methylcyclohexane. The
calculated conformational energies, experimental 2
DH values and literature values are also included
for comparison.
Table 3
Gradients b and c; calculated intercepts a and enthalpy difference, as calculated by Eq. (5), for compounds I and II
Compd.
Carbon
b (104 ppm K21
)
c (105 ppm K22
)
a; ax.
a; eq.
2DH (kcal mol21
)
I
Me
282.4
40.0
1.325
26.55–26.91
22.60–23.40
23.00–24.32
21.87–22.92
22.35
1.79 ^ 0.02
1.72 ^ 0.10
1.80 ^ 0.12
1.70 ^ 0.10
I
C-4,6
C-2
C-6
–
–
–
29.80–29.83
31.69–31.73
29.36–29.40
II
II
34.0
40.0

90
D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
Table 4
Selected distances and calculated, experimental and literature conformational energies of dithianes and methylcyclohexane
2DH (DFT)a (kcal mol21
)
2DH (Exp.), average
2DH; lit.
˚
˚
Compd.
rHðMeÞ–HðC6Þ (A)
rHðMeÞ–HðC2Þ (A)
I
2.476
2.382
2.478
2.376
2.381
–
1.71
1.83
1.71e
1.82e
2.15
1.76
1.75
1.77b/1.92c
1.69b/1.55c
II
2.438
2.462d
2.417
2.381
2ax-4eq (IV)
2eq-4ax (IV)
Me-Cyf
a
Basis set: B3LYP/6-31G(d,p); 1 kcal mol21 ¼ 4.186 kJ mol21
.
b
c
d
e
f
Ref. [16].
Ref. [17].
rHðMeÞ–HðC4Þ
:
Relative energy of 2eq-4eq (IV) is 0 kcal mol21
Methylcyclohexane.
.
5. Discussion
carbons. Averages of these results are presented in
Table 4.
These energies were also calculated by DFT/6-
31G** for the axial and equatorial conformers of I
and II and for 2,4-dimethyl-1,3-dithiane in three
different conformations: 2eq,4eq-dimethyl-1,3-
dithiane, 2eq,4ax-dimethyl-1,3-dithiane and 2ax,4eq-
dimethyl-1,3-dithiane. The calculated conformational
energies of the 2-methyl and of the 4-methyl group in
the dithiane ring are unchanged when going from the
monosubstituted compounds to the 2,4-disubstituted
compounds (Table 4). This finding suggests that there
is no buttressing effect in the compounds studied by
Eliel and Hutchins [16] and by Pihlaja [17], despite
the observation of shorter hydrogen–hydrogen dis-
tances in the disubstituted compounds. Consequently,
the values obtained experimentally for the disubsti-
tuted compounds should agree with ours.
The buttressing effect in a 1,3-dialkylcyclohexane
is expected to shorten the distances between the
methyl group and one syn-axial hydrogen, because of
the greater difficulty of this hydrogen to bend out of
the ring.
The calculated distances between syn-axial ring
hydrogen atoms with the closest axial methyl
hydrogen are presented in Table 4. There is a
˚
shortening of 0.014 A in the distance for the methyl
hydrogen of 2ax-4eq (IV), compared to the corre-
sponding distance in I (axial). The comparison
between II (axial) and 2eq-4ax(IV) indicates a
shortening in the distance between the proton of 4-
˚
axial-methyl and H-axial-C-2 by 0.021 A. This, in
itself, would indicate a buttressing effect, but
the analysis of conformational energies of I and II
indicates that this effect is irrelevant for the studied
compounds (see below).
In fact, there is an excellent agreement between
our experimental values and the literature values,
but the agreement is far better with the results of
Eliel and Hutchins [16]. Also, there is a very good
agreement between DFT energies and experimental
energies.
These distances are definitively larger in I than they
are in methylcyclohexane, so whatever the distortions
of the dithiane ring, compared to the cyclohexane ring,
they do not promote a proximity of these hydrogen
atoms comparable to those of methylcyclohexane, as
suggested by Eliel et al. [13,16].
It should be pointed out that the energy value
obtained by Pihlaja [17] for I was based on the energy
value of II, because he studied the isomerization of
2,4-dimethyl-1,3-dithiane. For II, this author per-
formed the same experiment as Eliel and Hutchins
[16], but obtained a lower value for the conformation-
al energy. Had he used the value obtained by Eliel and
Hutchins [16] in his calculations, the outcome for I
The conformational energies (DH) of 2-methyl-
and 4-methyl-1,3-dithianes have been measured by
the curve fitting of 13C chemical shifts as a function of
temperature into Eq. (5). This procedure was carried
out for two distinct carbon atoms for each equilibrium
and consistent DH values were obtained for both

D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
91
would have been 1.72 kcal mol21, much closer to our
experimental and theoretical results and to those of
Eliel and Hutchins [16].
D.S.R.), and CNPq for a fellowship (to R.R.).
CENAPAD-SP is also gratefully acknowledged for
the computer facilities (Gaussian98) as is Professor
C. H. Collins’ assistance in revising this
manuscript.
The conformational energies of I and II are
indeed very similar to that of methylcyclohexane,
despite the very different distances between the
closest methyl hydrogen and the ring syn-axial
hydrogens. Thus, the present study has shown that,
although there is some reduction in the considered
distances between the mono- and disubstituted
dithianes, the buttressing effect does not lead to
any difference in the conformational energies.
The qualitative expectation that the methyl deriva-
tive (I and II) should have a lower conformational
energy, in comparison to methylcyclohexane, has not
been observed.
References
[1] E.L. Eliel, S.H. Wilen, L.N. Mander, Stereochemistry of
Organic Compounds, Wiley, New York, 1994, pp. 665–
834.
[2] K.B. Wiberg, J.D. Hammer, H. Castejon, W.F. Bailey,
E.L. DeLeon, R.M. Jarret, J. Org. Chem. 64 (1999)
2085–2095.
[3] E. Juaristi, Conformational Behavior of Six-membered Rings,
VCH Publishers, New York, 1995.
[4] J.A. Hirsch, Top. Stereochem. 1 (1967) 199–222.
[5] J. March, Advanced Organic Chemistry, fourth ed., Wiley,
New York, 1992, pp. 144.
Recently Wiberg et al. [2] have reported some
evidence that the 1,3-syn-axial interaction is not so
important in determining the conformational energy
in methylcyclohexane. The present work leads to the
same conclusion. However, because this 1,3-syn-axial
interaction is such a long established rationale it
obviously needs further studies to settle this question.
[6] G. Solomons, C. Fryhle, Organic Chemistry, vol. 1, Wiley,
New York, 2000.
[7] V.M. Gittins, E. Wyn-Jones, R.F.M. White, in: Internal
Rotation in Molecules, Wiley, London, 1974, pp. 433
[8] A. Streitwieser, C.H. Heathcock, E.M. Kosower, Introduction
to Organic Chemistry, fourth ed., Macmillan, New York,
1992, (Chapter 7).
[9] E.L. Eliel, Angew. Chem., Int. Ed. Engl. 11 (1972) 739–860.
[10] F. Freeman, M.L. Kasner, W.J. Hehre, J. Mol. Struct.
(Theochem) 574 (2001) 19–26.
6. Conclusion
[11] N. Allinger, L.A. Freiberg, J. Am. Chem. Soc. 84 (1962)
2201–2203. W.D. Cotterill, M.J.T. Robinson, Tetrahedron 20
(1964) 777–790.
The experimental conformational energies for 2-
methyl- (1.76 kcal mol21) and 4-methyl-1,3-dithiane
(1.75 kcal mol21) are in excellent agreement with both
calculated and literature values. Consequently the
possibility of a buttressing effect on the compounds
studied by Eliel and Hutchins [16] and by Pihlaja [17]
can be ruled out.
[12] E.L. Eliel, K.D. Hargrave, K.M. Pietrusiewicz, M. Mano-
haran, J. Am. Chem. Soc. 104 (1982) 3635–3643. D. Kim, T.
Baer, J. Phys. Chem. A 104 (2000) 509–513.
[13] R.L. Willer, E.L. Eliel, J. Am. Chem. Soc. 99 (1977)
1925–1936.
[14] E.L. Eliel, D. Kandasamy, C. Yen, K.D. Hargrave, J. Am.
Chem. Soc. 102 (1980) 3698–3707.
The calculated distances between 1,3-syn-axial ring
hydrogens and the closest axial methyl hydrogen in
axial methylcyclohexane and in axial 2-methyl- and 4-
methyl-1,3-dithiane are considerably different, being
larger in the dithiane ring. Thus, the qualitative
explanation of steric effects in the axial conformation
being due to repulsive 1,3-interactions should be
further investigated.
[15] E.L. Eliel, M. Knoeber, J. Am. Chem. Soc. 90 (1968)
3444–3458. F.G. Riddell, M.J.T. Robinson, Tetrahedron 23
(1967) 3417–3425.
[16] E.L. Eliel, R.O. Hutchins, J. Am. Chem. Soc. 91 (1969)
2703–2715.
[17] K. Pihlaja, J. Chem. Soc., Perkin Trans. 2 (1974) 890–896.
[18] D. Danneels, M. Anteunis, Tetrahedron 31 (1975) 1689–1694.
[19] R.J. Abraham, D.S. Ribeiro, J. Chem. Soc., Perkin Trans. 2
(2001) 302–307.
[20] D.S. Ribeiro, R.J. Abraham, Magn. Reson. Chem. 40 (2002)
49–56.
Acknowledgements
[21] E.L. Eliel, V.S. Rao, F.G. Riddell, J. Am. Chem. Soc. 98
(1976) 3583–3590.
We acknowledge FAPESP for financial support
(grant no. 2000/07692-5) and for a fellowship (to
[22] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery,

92
D.S. Ribeiro, R. Rittner / Journal of Molecular Structure 657 (2003) 85–92
R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham,
C.Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P.M.W. Gill, B.G. Johnson, W. Chen, M.W. Wong, J.L.
Andres, M. Head-Gordon, E.S. Replogle, J.A. Pople, GAUS-
SIAN 98, Revision A.7, Gaussian, Pittsburgh, 1998.
Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomeli, C.
Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala,
Q. Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B.