Influence of intramolecular hydrogen bonding on the conformational equilibrium of cis-3-N,N-dimethylaminocyclohexanol compared with trans-3-N,N-dimethylaminocyclohexanol and cis- and trans-3-N,N-dimethylamino-1- methoxycyclohexane

Article abstract of DOI:10.1002/poc.896

¹H NMR data show that concentration increase shifts the conformational equilibrium of cis-3-N,N-dimethylaminocyclohexanol (1) (cis-3-DACH) from the 1aa conformer, stabilized by an intramolecular hydrogen bond (IAHB), to the lee conformer [43% (

Full text of DOI:10.1002/poc.896

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2005; 18: 513–521
Published online 5 January 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.896
Influence of intramolecular hydrogen bonding on the
conformational equilibrium of cis-3-N,N-dimethylamino-
cyclohexanol compared with trans-3-N,N-dimethylamino-
cyclohexanol and cis- and trans-3-N,N-dimethylamino-1-
methoxycyclohexane
P. R. de Oliveira, D. S. Ribeiro and R. Rittner*
Physical Organic Chemistry Laboratory, Chemistry Institute, State University of Campinas, Caixa Postal 6154, 13084-971 Campinas, SP, Brazil
Received 28 July 2004; revised 1 November 2004; accepted 8 November 2004
1
ABSTRACT: H NMR data show that concentration increase shifts the conformational equilibrium of cis-3-N,N-
dimethylaminocyclohexanol (1) (cis-3-DACH) from the 1aa conformer, stabilized by an intramolecular hydrogen
bond (IAHB), to the 1ee conformer [43% (0.01 M) to 70% (0.40 M), in CCl4], which can form intermolecular hydrogen
bonds (IEHB). The percentage of 1ee conformer also increases with the solvent basicity from 36% in C6D12 to 89% in
DMSO. The conformational equilibrium of the trans isomer (trans-3-DACH) is also dependent on concentration,
since 1ae increases from 77% (0.05 M) to 84% (0.40 M) in CCl4 but not with the solvent polarity. The occurrence of an
IAHB in cis-3-DACH was confirmed by the study of a model compound, cis-3-N,N-dimethylamino-1-methoxycy-
clohexane (2) (cis-3-DAMCH), lacking an OH group and presenting a single conformer 2ee (ꢀ95%). The
corresponding trans isomer (trans-3-DAMCH) behaves similarly to trans-3-DACH, since the 2ae conformer occurs
as ꢀ83%, in the studied solvents. The PCMODEL program gave very good coupling constant values for the
qualitative analysis of energy changes in the study of concentration and solvent effects, since the energy values
obtained for cis and trans isomers of 3-DACH were in good agreement with the theoretically calculated [B3LYP/6–
311þg(d,p) level] values. The averaged calculated energy of the IAHB, for conformer 1aa of cis-3-DACH, with
optimization of the reference structure and at several levels of theory, is 5.74 kcal molꢁ1 (1 kcal ¼ 4.184 kJ). Copyright
# 2005 John Wiley & Sons, Ltd.
Supplementary electronic material for this paper is available in Wiley Interscience at http://www.interscience.
wiley.com/jpages/0894-3230/suppmat/
KEYWORDS: conformational analysis; theoretical calculations; intramolecular hydrogen bond; NMR spectroscopy
INTRODUCTION
cyclohexane-1,2-diol and cis-cyclohexane-1,3-diol are
shifted towards the diequatorial and diaxial conformers,
respectively, which have the OH groups close to each
other, allowing the formation of hydrogen bonds.
The presence of a hydrogen bond has been detected by
experimental techniques such as H—O and C—O
stretching vibration frequency shifts and hydroxyl proton
chemical shifts.1–5 However, it was after the classical
book The Hydrogen Bond by Pimentel and McClelland6
that a new era of intense research on this subject was
launched.
Previous workers have shown that the OH groups of a
dihydroxy compound, if sufficiently close to each other,
may form an intramolecular hydrogen bond.7 Kuhn8
showed that the conformational equilibria of trans-
1
Abraham et al.9 have demonstrated, through H NMR
and theoretical data, that the OHꢂ ꢂ ꢂF hydrogen bonding,
in trans-2-fluorocyclohexanol, is responsible for the pre-
dominance of eq–eq conformation of this molecule.9
Substituted six-membered rings are useful models for
studies of conformational analysis. Mono- and 1,2-disub-
stituted cyclohexanes have been thoroughly studied,10–14
but publications on 1,3-disubstituted cyclohexanes are
rarely found in the literature.15,16
This paper describes a study of the behavior of the cis
and trans isomers of the previously unknown 3-N,N-
dimethylaminocyclohexanol (3-DACH, Figs 1 and 2), at
different concentrations and in different solvents, giving
emphasis on how their conformational equilibria are
affected by inter- and intramolecular hydrogen bonds
*Correspondence to: R. Rittner, Physical Organic Chemistry Labora-
tory, Chemistry Institute, State University of Campinas, Caixa Postal
6154, 13084-971 Campinas, SP, Brazil.
E-mail: rittner@iqm.unicamp.br
Contract/grant sponsor: FAPESP; Contract/grant numbers: 00/07692-
5; 03/10591-4.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

514
P. R. DE OLIVEIRA, D. S. RIBEIRO AND R. RITTNER
which is clearly due to a change in the conformer
populations. For dilute solutions the 1aa conformer
(Fig. 1) is favored, but an increase in concentration makes
1ee more stable than the 1aa conformer.
The molar fraction (X) and free energy difference
(ÁGee–aa) for 1aa and 1ee conformers of cis-3-DACH
were determined through Eqns (1) and (2), taking the H-1
coupling constant values of the 1aa and 1ee conformers
Figure 1. Conformational equilibrium for the cis isomer of
3-DACH (1, R ¼ H) and 3-DAMCH (2, R ¼ CH3)
individually (3JH1e/H2e or H6e ¼ 4.41 and 3JH1a/H2a or H6a
¼
11.15 Hz), obtained through the PCMODEL program,17
from the Haasnoot–Altona equations,18 since the experi-
mental data from observed vicinal coupling constants
(3Jobs) are averaged values (Table 1).
The mole fraction of the diequatorial conformer (Xee)
is given directly from observed vicinal coupling constants
(3Jobs) as follows:
Figure 2. Conformational equilibrium for the trans isomer
of 3-DACH (1, R ¼ H) and 3-DAMCH (2, R ¼ CH3)
3Jobs ꢁ JH1e;H2e
3
Xee
¼
ð1Þ
3JH1a;H2a ꢁ JH1e;H2e
3
(IEHB and IAHB, respectively). The compounds cis- and
trans-3-N,N-dimethylamino-1-methoxycycloexane (3-
DAMCH, Figs 1 and 2), also unknown, were included
in this study for comparison. The geometries of the more
stable conformers of 3-DACH and 3-DAMCH and the
strength of IAHB for the cis-3-DACH in the gas phase
were also determined.
since Xee þ Xaa ¼ 1.
The free energy difference (ÁG ꢃ) is readily obtained
from Eqn (2), where R ¼ 1.99 cal mol Kꢁ1 (1 cal ¼ 4184J),
T ¼ 298 K, K ¼ Xee/Xaa and K ¼ Xae/Xea for the cis and
trans isomer, respectively, of 3-DACH and of 3-
DAMCH:
ÁGꢃ ¼ ꢁRT ln K
ð2Þ
RESULTS AND DISCUSSION
Concentration effect
The molar fraction of the diequatorial conformer (1ee)
and the free energy (ÁGee–aa) between the 1aa and 1ee
conformers, for a concentration range of 0.01–0.40 M, in
CCl4 and in CDCl3 are given in Table 1. The positive
values of ÁGee–aa show that the 1aa conformer predo-
minates for low concentrations, i.e. up to 0.05 M in CCl4
(57%). Its stability can be attributed to an IAHB, while an
increase in concentration makes ÁGee–aa negative, favor-
ing the 1ee conformer (70%, for a 0.40 M solution) owing
to the prevalence of an IEHB over the IAHB.
1H NMR spectra of nine samples in CCl4 (with 10%
C6D6) and nine samples in CDCl3 as solvent with con-
centrations of 0.01, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35
and 0.40 M for cis- and trans-3-DACH were recorded and
the results are presented in Table 1.
The coupling constant (3J) values, from Table 1, for the
H-1 hydrogen of cis-3-DACH, increase significantly on
increasing the concentration, in both CCl4 and CDCl3,
b
Table 1. Hydrogen H-1a coupling constants (3J), equatorial–equatorial molar fractions (Xee)b and energy differences (ÁGee–aa
for cis-3-DACH at different concentrations in CCl4c and CDCl3 solutions
)
CCl4
3JH1/H2e
CDCl3
3JH1/H2e
Concentration
(M)
3JH1/H2a
3JH1/H2a
Xee
ÁGee–aa
Xee
ÁGee–aa
or H6a
or H6e
or H6a
or H6e
0.01
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
7.30
7.50
7.85
8.27
8.36
8.61
8.82
8.94
9.13
3.62
3.68
3.85
3.91
4.00
4.08
3.22
3.18
3.56
0.43
0.46
0.51
0.57
0.59
0.62
0.65
0.67
0.70
0.17
0.10
—
—
—
—
0.00
7.77
7.96
8.12
8.16
8.22
8.26
8.45
8.59
3.83
3.81
3.99
3.88
3.80
4.01
3.98
3.98
0.50
0.53
0.55
0.56
0.57
0.57
0.60
0.62
ꢁ0.03
ꢁ0.17
ꢁ0.21
ꢁ0.30
ꢁ0.38
ꢁ0.43
ꢁ0.50
ꢁ0.06
ꢁ0.12
ꢁ0.13
ꢁ0.16
ꢁ0.17
ꢁ0.24
ꢁ0.29
a The H-3 hydrogen signal was overlapped by the N(CH3)2 hydrogens.
b Molar fraction and ÁGee–aa (kcal molꢁ1), obtained from experimental coupling constants (3JH1/H2a or H6a) and calculated values by the PCMODEL program.
c For the mixture CCl4–C6D6 (9:1).
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

CONFORMATIONAL EQUILIBRIUM OF cis-3-N,N-DIMETHYLAMINOCYCLOHEXANOL
515
Table 2. Hydrogen H-3 coupling constants (3J) axial-equatorial molar fractions (Xae)a and energy differences (ÁGae–ea)a for
trans-3-DACH at different concentrations in CCl4b and CDCl3 solutions
CCl4
CDCl3
Concentration (M)
3JH3/H2a or H4a
3JH3/H2e or H4e
3JH3/H2a or H4a
3JH3/H2e or H4e
Xae
ÁGae–ea
0.01
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
9.91
10.14
10.29
10.42
10.42
10.49
10.51
10.48
10.54
3.40
3.32
3.31
3.28
3.23
3.27
3.26
3.25
3.27
—
—
0.77
0.79
0.81
0.83
0.83
0.84
0.84
0.84
0.84
ꢁ0.70
ꢁ0.80
ꢁ0.87
ꢁ0.93
ꢁ0.93
ꢁ0.97
ꢁ0.98
ꢁ0.96
ꢁ1.00
10.57
10.61
10.55
10.70
10.61
10.58
10.60
10.77
3.42
3.43
3.43
3.39
3.45
3.44
3.40
3.42
a Molar fraction and ÁGae–ea (kcal molꢁ1) obtained from experimental coupling constants, in CCl4 (3JH3/H2a or H4a), and calculated by the PCMODEL program.
b For the mixture CCl4–C6D6 (9:1).
The IAHB effect in the cis isomer can be proved by a
comparison of changes in the H-1 coupling constants for
this cis isomer (Table 1) with those for the trans isomer
(Table 2), whose conformational equilibrium is presented
in Fig. 2, since the trans isomer cannot form IAHB.
Actually, the results displayed in Table 2 show that there
is a very small change in the coupling constants for the
trans isomer with increase in concentration, due to an
increase in the population of the 1ae conformer (Fig. 2)
from 77% (at 0.01 M, in CCl4) to 84% (at 0.40 M, in CCl4),
taking the calculated values obtained through the
PCMODEL program for the 1ae and 1ea conformers
Solvent effect
Tables 3 and 4 present the H-1 and H-3 coupling
constants for cis- and trans-3-DACH, in different sol-
vents, at a low concentration (0.05 M) to reduce the
interference of IEHB. H-1 coupling constants for the
cis isomer (Table 3) increase with the increase of solvent
polarity, showing that 1aa is more stable than the 1ee
conformer in less polar solvents (C6D12) owing to the
favorable formation of an IAHB (Fig. 1) in the absence of
solvation effects. The experimental coupling constant of
6.82 Hz in C6D12 leads to a value of 64% for the 1aa
conformer (Xee ¼ 0.36, Table 3) for the conformational
equilibrium at room temperature. Therefore, the steric
effect is less important or smaller than the effect of
stabilization provided by IAHB, for less polar solvents,
whereas the opposite is observed in more polar solvents,
when the 1ee conformer becomes more stable. This
behavior can be explained by taking into account that a
polar solvent has a larger affinity for the OH and N(CH3)2
groups than a non-polar solvent,20 and it can approach
those groups more easily when they are equatorial.
Therefore, the 1ee conformer can be more efficiently
3
(3JH3/H2a or H4a ¼ 11.81 Hz and JH3/H2e or H4e ¼ 3.72 Hz,
respectively). The large stability of the 1ae conformer
can be attributed to the 1,3-diaxial steric effect, since an
axial dimethylamino group presents a larger steric effect
than an axial hydroxy group. It seems that the increase in
the population of the 1ae conformer, which occurs on
increasing the concentration, may be due to their smaller
bulk, suggesting that the steric effect in the case of the
trans isomer is more important than the stabilization
provoked by intermolecular hydrogen bonds (IEHB),19
as observed for the cis isomer.
b
Table 3. Hydrogen H-1a coupling constants (3J), equatorial–equatorial molar fractions (Xee)b and energy differences (ÁGee–aa
for cis-3-DACH in solventsc of different dielectric constants (") and different basicities (SB)
)
Solvent
SBd
"
3JH1/H2a or H6a
3JH1/H2e or H6e
Xee
ÁGee–aa
C6D12
CCl4
CDCl3
CD2Cl2
CD3CN
Acetone-d6
Pyridine-d5
DMSO-d6
0.07
0.04
0.07
0.17
0.29
0.48
0.58
0.65
2.05
2.24
4.81
9.08
37.50
20.70
12.40
46.70
6.82
7.50
7.77
7.96
2.99
3.68
3.83
3.84
3.83
4.01
4.02
3.48
0.36
0.46
0.50
0.53
0.81
0.84
0.90
0.89
0.35
0.10
0.00
ꢁ0.06
ꢁ0.87
ꢁ1.00
ꢁ1.33
ꢁ1.26
9.89
10.09
10.50
10.43
a The H-3 hydrogen signal was overlapped by the N(CH3)2 hydrogens.
b Molar fraction and ÁGee–aa (kcal molꢁ1) obtained from experimental coupling constants (3JH1/H2a or H6a) and calculated values by the PCMODEL program.
c Concentration: 0.05 M.
d Data from Ref. 21.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

516
P. R. DE OLIVEIRA, D. S. RIBEIRO AND R. RITTNER
Table 4. Hydrogen H-1 and H-3 coupling constants (3J), axial–equatorial molar fractions (Xae)a and energy differences
(ÁGae–ea)a for trans-3-DACH in solventsb of different dielectric constants (")
Solvent
"
3JH1/H2a or H6a
3JH1/H2e or H6e
3JH3/H2a or H4a
3JH3/H2e or H4e
Xae
ÁGae–ea
C6D12
CCl4
CDCl3
CD2Cl2
Pyridine-d5
Acetone-d6
CD3CN
DMSO-d6
2.05
2.24
4.81
9.08
12.40
20.70
37.50
46.70
5.06
4.82
—
4.78
—
5.25
5.05
—
2.79
2.61
—
2.65
—
2.78
2.74
—
9.85
10.14
10.57
10.16
9.88
9.88
9.69
9.84
3.28
3.32
3.42
3.43
3.34
3.34
3.33
3.36
0.76
0.79
0.85
0.80
0.76
0.76
0.74
0.76
ꢁ0.68
ꢁ0.80
ꢁ1.01
ꢁ0.81
ꢁ0.69
ꢁ0.69
ꢁ0.61
ꢁ0.67
a Molar fraction and ÁGae–ea (kcal molꢁ1) obtained from experimental coupling constants (3JH3/H2a or H4a) and calculated values by the PCMODEL program.
b Concentration: 0.05 M.
Table 5. Hydrogen H-1a coupling constants (3J) for cis-3-
solvated. The increase in stabilization of the 1ee con-
former by a solvation effect is confirmed by the data in
DAMCH, equatorial–equatorial molar fractions (Xee)b and
energy differences (ÁGee–aa)b for cis-3-DAMCH in solventsc
of different dielectric constants (")
3
DMSO, a very polar solvent. Thus, JH1/H2a or H6a is
10.43 Hz in DMSO, leading to a 1ee conformer popula-
tion of 89%. However, data from Table 3 show that the
3JH1/H2a
3JH1/H2e
Solvent
"
Xee ÁGee–aa
1ee proportion is also large when the solvent is pyridine
(90%), which is less polar (smaller ", Table 3) than
acetone, acetonitrile and dimethyl sulfoxide. Therefore,
these results show that the conformational equilibrium is
shifted by the solvent basicity (SB, Table 3) and not by
the solvent polarity, estimated by the relative permittivity
or dielectric constant.
For the trans isomer (Fig. 2), the coupling constants are
very close to each other in solvents of very different
dielectric constants (Table 4), indicating that the confor-
mational equilibrium of this isomer is not affected by
solvent, giving an average value of ꢀ78% for the 1ae
conformer. The preference for this conformer has been
already discussed in the previous section.
A further proof that an IAHB occurs in cis-3-DACH
can be obtained from a similar study with cis-3-N,N-
dimethylamino-1-methoxycyclohexane (cis-3-DAMCH,
2), chosen as a model compound since it occurs as a
single conformer, exhibiting the dimethylamino and
methoxyl groups in the equatorial orientation, and it is
very similar to the target compound but lacks an OH
group. Table 5 presents the coupling constants for the H-1
hydrogen, which do not change with solvent polarity, in
contrast to cis-3-DACH, and indicate that the ee con-
or H6a
or H6e
CCl4
CDCl3
Pure liquid
2.24
4.81
—
10.88
10.96
10.92
10.63
10.86
10.92
10.96
4.04
4.04
4.03
4.14
4.08
4.12
4.10
0.95 ꢁ1.78
0.96 ꢁ1.95
0.96 ꢁ1.86
0.92 ꢁ1.43
0.95 ꢁ1.74
0.96 ꢁ1.86
0.96 ꢁ1.95
Pyridine-d5 12.40
Acetone-d6
CD3CN
DMSO-d6
20.70
37.50
46.70
a The H-3 hydrogen signal was overlapped by the N(CH3)2 hydrogens.
b Molar fraction and ÁGee–aa (kcal molꢁ1) obtained from experimental
coupling constants (3JH1/H2a or H6a) and calculated values by the PCMODEL
program; 3JH1/H2a or H6a (11.22 Hz) and 3JH1/H2e or H6e (4.04 Hz).
c Concentration: 0.05 M.
former (Fig. 1) is predominant in the conformational
equilibrium of cis-3-DAMCH (ꢀ95%). trans-3-DAMCH
presents a conformational equilibrium very similar to that
of trans-3-DACH, since the 2ae conformer (Fig. 2)
occurs as ꢀ83% and does not change with increase in
solvent polarity (Table 5).
Theoretical calculations
The geometry for the stable conformers of cis and trans
isomers of 3-DACH (1) and of 3-DAMCH (2) were
Table 6. Hydrogen H-1 and H-3 coupling constants (3J), axial–equatorial molar fractions (Xae)a and energy differences
(ÁGae–ea)a for trans-3-DAMCH in solventsb of different dielectric constants (")
Solvent
"
3JH1/H2a or H6a
3JH1/H2e or H6e
3JH3/H2a or H4a
3JH3/H2e or H4e
Xae
ÁGae–ea
CCl4
CDCl3
Pure liquid
Pyridine-d5
Acetone-d6
CD3CN
2.24
4.81
—
12.40
20.70
37.50
46.70
4.66
4.66
4.62
4.86
4.92
4.85
4.70
2.46
2.38
2.54
2.51
2.59
2.56
2.59
10.54
10.89
10.37
—
10.25
10.27
10.43
3.41
3.45
3.38
—
3.35
3.36
3.44
0.84 ꢁ1.00
0.89 ꢁ1.22
0.82 ꢁ0.91
0.81 ꢁ0.85
0.81 ꢁ0.86
0.83 ꢁ0.94
DMSO-d6
a Molar fraction and ÁGae–ea (kcal molꢁ1) obtained from experimental coupling constants (3JH3/H2a
H4a) and calculated values by the PCMODEL
or
program; 3JH3/H2a or H4a (11.81 Hz) and JH3/H2e or H4e (3.72 Hz).
3
b Concentration: 0.05 M.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

CONFORMATIONAL EQUILIBRIUM OF cis-3-N,N-DIMETHYLAMINOCYCLOHEXANOL
517
Table 7. Conformer relative energies (ÁE) and dipole moments (ꢀ) for cis and trans isomers of 3-DACH and 3-DAMCH at the
B3LYP/6–311þG** level
cis-3-DACH
trans-3-DACH
cis-3-DAMCH
trans-3-DAMCH
ÁE
ÁE
ÁE
ÁE
Rotamer ꢀ (D) kcal molꢁ1 Rotamer ꢀ (D) kcal molꢁ1 Rotamer ꢀ (D)
kcal molꢁ1 Rotamer ꢀ (D) kcal molꢁ1
a
1aa1
1ee1
1ee2
1ee3
1ee4
1ee5
1ee6
2.90
1.77
2.05
2.04
1.52
1.44
1.46
0.00
1.12
1.03
1.09
0.58
0.43
0.43
1ae1
1ae2
1ae3
1ae4
1ae5
1ae6
1ae7
1ae8
1ae9
1ea1
1ea2
1ea3
1.99
1.19
1.40
1.18
2.42
1.79
1.87
1.79
1.57
2.19
1.61
2.16
0.21
0.00
1.01
1.07
1.16
1.67
0.25
0.34
1.29
1.16
0.93
1.24
2aa1
2aa2
—
1.23
—
1.74
1.70
1.26
1.14
1.14
—
4.33
—
0.69
0.73
2.46
0.00
0.00
2ae1
2ae2
2ae3
2ae4
2ae5
2ae6
2ea1
2ea2
2ea3
1.55
0.81
0.87
1.57
1.13
1.66
—
0.16
0.00
0.99
0.19
0.87
0.13
—
b
a
2ee1
2ee2
2ee3
2ee4
2ee5
2ee6
a
1.25
1.66
0.94
1.27
a ÁE > 3 kcal molꢁ1
b The most stable diaxial conformer.
.
with the value of 0.35 kcal molꢁ1 obtained from experi-
mental data in C6D12 (Table 3), since this solvent has the
dieletric constant value (" ¼ 2.05) nearest that of the
considered value in the theoretical calculations (" ¼ 1.0).
The other possible rotamers for 1aa conformer
(ÁE > 4.8 kcal molꢁ1) are not stable owing to the large
1,3-diaxial interactions, which are decisive in the con-
formational equilibrium of the 1,3-disubstituted cyclo-
hexanes cis isomer. It can also be observed that most of
the other stable rotamers of the 1ee conformer present a
stability (ÁE ꢄ 0.58–1.12 kcal molꢁ1) near to 1ee5 and
1ee6 (ÁE ꢄ 0.43 kcal molꢁ1), which indicates that the
rotation of the C—O bond does not introduce large
changes in their energies.
The data in Table 7 also show that the dipole moment
for 1aa1 rotamer (2.90 D) is larger than for 1ee5 and 1ee6
rotamers (1.44 and 1.46 D, respectively). Therefore, an
increase in solvent polarity should shift the equilibrium
towards the 1aa conformer. However, an opposite beha-
vior was observed, 1ee being favored in polar solvents
(Table 3), owing to the interaction of the basic solvent
with the hydroxyl group (IEHB) as discussed above.
Figure 4 presents the stable rotamers for trans-3-
DACH. The corresponding energies are given in Table 7.
The most stable rotamers for the 1ae and 1ea conformers
are 1ae2 and 1ea2, respectively, and the 1ae2 rotamer is
0.93 kcal molꢁ1 (83%) more stable than 1ea2. This is in
very good agreement with the ÁGae–ea values in Table 4,
obtained from experimental coupling constants and cal-
culated by the PCMODEL program.
Figure 3 shows the stable rotamers for 2aa and 2ee
conformers, of cis-3-DAMCH (2), and the corresponding
energies are given in Table 7. The cis isomer of 2, which
cannot form an IAHB, is expected to occur mostly as a
single conformer (2ee), since the diaxial conformer (2aa)
energy is very large. The most stable rotamers for 2ee
and 2aa are 2ee5 and 2ee6, and 2aa2, respectively, but the
Figure 3. Stable rotamers for the cis isomer of 3-DACH (1,
R ¼ H) and 3-DAMCH (2, R ¼ CH3), obtained at the B3LYP/
6–311þg** level
obtained through theoretical calculations using Gaussian
9822 at the 6–311þG** basis set from Becke’s three-
parameter functional using the Lee–Yang–Parr correla-
tion functional (B3LYP)23 level of theory. The relative
energies and dipole moments for all stable rotamers of
compounds 1 and 2 with ÁE < 3 kcal molꢁ1 are given in
Table 7. The rotamers with ÁE > 3 kcal molꢁ1 were not
considered because they represent a very small propor-
tion in the equilibrium. The corresponding geometries for
the stable rotamers are presented in Fig. 3.
Data from Table 7 show that the rotamers 1aa1, for 1aa,
and 1ee5 and 1ee6, for 1ee, respectively, are the most stable
ones, and 1aa1 is more stable than 1ee5 and 1ee6 by
0.43 kcal molꢁ1, reinforcing the important role of an
IAHB in the conformational equilibrium of cis-3-
DACH. This relative energy is in very good agreement
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

518
P. R. DE OLIVEIRA, D. S. RIBEIRO AND R. RITTNER
Figure 4. Stable rotamers for the trans isomer of 3-DACH
(1, R ¼ H) and 3-DAMCH (2, R ¼ CH3), obtained at the
B3LYP/6–311þg** level
Figure 5. PES for the 1aa1 rotamer: (a) at the B3LYP/6–
311þg** level and (b) at the HF/6–311þg** level
2
ee5 and 2ee6 rotamers are 4.33 kcal molꢁ1 (99.9%) more
IAHB. Both theory levels [Fig. 5(a) and (b)] show two
minima, one corresponding to the most stable rotamer
(1aa1), which forms an IAHB, and the other for a less
stable (1aa2) rotamer with the hydrogen far from the
N(CH3)2 group.
stable than 2aa2. These values are in good agreement with
data in Table 5, obtained from experimental coupling
constants and calculated by the PCMODEL program,
showing that the steric effect is predominant in this
equilibrium.
The hydrogen bond energies can be calculated as the
difference between the energies of a non-bonded and a
bonded species (ÁE ¼ Enon-bonded ꢁ Ebonded). There are
two established ways to estimate the strength of IAHB,
either by optimizing,24–26 or not,27,28 the reference struc-
ture, which in the present case is the 1aa2 rotamer (Fig. 4).
The calculated energy of IAHB for conformer 1aa,
with (ÁE1) and without (ÁE2) optimization of the
reference structure, through the HF and B3LYP levels
with 6–31G** and 6–311þG** basis sets, and also at the
MP2/6–31g** level, are presented in Table 8. The values
obtained, for the IAHB, with optimization (ÁE1) are in
agreement for different theory levels, but the values
without optimization (ÁE2) are higher than those with
optimization (ÁE1), probably owing to the small Nꢂ ꢂ ꢂO
distance for 1aa2 not optimized in comparison with the
Figure 4 shows the stable conformers for trans-3-
DAMCH, and the corresponding energies of the 2ae
and 2ea conformers are given Table 7. The most stable
rotamers for the 2ae and 2ea conformers are 2ae2 and
2
ea2, respectively, and the 2ae2 rotamer is 0.94 kcal molꢁ1
(83%) more stable than 2ea2. This is also in agreement
with the ÁGae–ea values in Table 6, obtained from
experimental coupling constants and calculated by the
PCMODEL program.
Hydrogen bonding
Studies of the potential energy surface (PES) through the
HF and B3LYP levels with the 6–311þG** basis set for
the 1aa conformer of cis-3-DACH, were performed to
observe how changes in the position of the OH hydrogen,
near or far from the N(CH3)2 group, which modifies the
C6—C1—O—H dihedral angle from the 1aa1 geometry,
would affect the rotamer energy and the formation of an
1
aa2 optimized (Table 8). This distance is very close to the
˚
sum of the appropriate van der Waals radii (ꢀ3.0 A).
Therefore, the total energy of 1aa2 would contain a
considerable amount of non-bonded Pauli repulsion en-
ergy, when not optimized,29 since the ÁE2 values were
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

CONFORMATIONAL EQUILIBRIUM OF cis-3-N,N-DIMETHYLAMINOCYCLOHEXANOL
519
Table 8. Relative energies for the cis-3-DACH conformer 1aa1, in comparison with 1aa2, with (ÁE1) and without (ÁE2)
geometry optimization of conformer 1aa2, and Nꢂ ꢂ ꢂO distance for conformer 1aa2, with and without geometry optimization
Theory level
ÁE1 (kcal molꢁ1
)
ÁE2 (kcal molꢁ1
)
d(Nꢂ ꢂ ꢂO)a
d(Nꢂ ꢂ ꢂO)b
HF/6–31g**
5.28
4.70
6.25
5.57
6.89
5.74
8.71
7.89
10.51
9.58
10.56
9.45
3.126
3.153
3.141
3.173
3.036
3.123
2.892
2.924
2.819
2.851
2.793
2.856
HF/6–311þg**
B3LYP/6–31g**
B3LYP/6–311þg**
MP2/6–31þg**
Average values
a
˚
In A for the optimized conformer.
b
˚
In A for the conformer not optimized.
larger for B3LYP/6–311þg** and MP2/6–31þg**
because, at these levels, the d(Nꢂ ꢂ ꢂO) values were the
smallest. Averaged values, presented in Table 8, suggest
that the best value for the strength of IAHB, for cis-3-
in the conformational equilibrium of the studied com-
pounds, which are in very good agreement with the values
calculated at the B3LYP/6–311þg(d,p) level.
The agreement for the hydrogen bond energy, at
several theory levels, using an optimized geometry, for
cis-3-DACH (average value 5.74 kcal molꢁ1) suggests
that optimization of the reference structure is very im-
portant for the estimation of IAHB for 1,3-disubstituted
cyclohexanes.
DACH, is near 5.74 kcal molꢁ1
.
CONCLUSION
The results describing the concentration effects underline
the importance of using dilute solutions in studies of
conformational equilibria for compounds that exhibit
substituents that can form intermolecular hydrogen
bonds, to allow the discrimination from intramolecular
hydrogen bonds.
Studies of the solvent effects showed that the solvent
basicity may be more important than the solvent polarity,
as measured by the dielectric constant, and that the
increase in the solvent basicity can shift the conforma-
tional equilibrium from a diaxial conformer (stabilized by
IAHB) towards the diequatorial conformer (stabilized by
IEHB), which is more efficiently solvated.
The conformational equilibrium for cis-3-DACH is
shifted from the 1aa conformer (64% in C6D12), stabi-
lized by IAHB, to the 1ee conformer (89% in DMSO),
owing to the solvent basicity, which leads to more
solvated species. A similar effect is observed on changing
the concentration, since the 1aa conformer is more
stable at low concentration (57% in 0.01 M CCl4), but
the increase in concentration leads to predominance of
the 1ee conformer (70% in 0.40 M CCl4), owing to
prevalence of the IEHB over the IAHB effect.
The coupling constant values, for cis-3-DAMCH,
showed that it occurs mostly as the diequatorial confor-
mer (2ee), in all solvents studied, which must be true for
any 1,3-disubstituted cyclohexane, when no other effects
are present, such as the formation of an IAHB, owing to
the 1,3-diaxial steric effect. The coupling constant values
for the trans isomer of 3-DACH and of 3-DAMCH
showed that the corresponding conformational equilibria
are not affected by the solvent, but are also determined by
the steric effect.
EXPERIMENTAL
The 1H NMR spectra for the study of solvent and
concentration effects were recorded on a Varian INOVA
500 spectrometer, with a probe temperature of 20 ꢃC,
operating at 499.88 (1H) and 125.70 MHz (13C). Spectra
were recorded at concentrations of 0.05 M for the solvent
effect and at 0.01–0.40 M in CDCl3 and in CCl4
(CCl4:C6D6 ¼ 9:1, the latter for the deuterium lock)
solutions. In all cases, SiMe4 (TMS) was used as internal
reference. The spectral windows ensured a digital resolu-
tion of at least 0.04 Hz per point, and zero-filling helped
to define lineshapes further. Most FIDs were processed
with Gaussian multiplication, typically of gf ¼ 0.25 and
0.35 for spectral resolution improvement. The typical
1
conditions for H spectra were 128 transients, 32 K data
points, pulse width 37ꢃ, sweep width ꢀ3000 Hz and
acquisition time (AT) ꢀ2.7 s; and for 13C NMR spectra
1024 transients, 32 K data points, pulse width 45 ꢃ, sweep
width ꢀ10 000 Hz and AT 1 s. Assignments of the signals
in the 1H and 13C NMR spectra of 1 and 2 were
performed through gCOSY and HSQC experiments.
Optimized geometries were computed at the B3LYP
levels of theory, using the 6–311þG** basis sets. The
strengths of IAHB were computed at the Hartree–Fock
(HF) and B3LYP levels of theory, using the 6–31G** and
6–311þG** basis sets, and the MP2/6–31þg** level.
The PES were obtained at the HF and B3LYP levels with
6–311þG** basis sets, by changing the C6—C1—O—H
dihedral angle for the 1aa1 rotamer by 10ꢃ until complet-
ing 360ꢃ. For each 10ꢃ the structure was optimized.
Cis- and trans-3-DACH (1). A 70 ml volume of 40% di-
methylamine (0.60 mol) was placed in a round-bottomed
The PCMODEL program gave very good coupling
constants for the qualitative analysis of energy changes
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

520
P. R. DE OLIVEIRA, D. S. RIBEIRO AND R. RITTNER
flask fitted with magnetic stirrer, 20 g (0.20 mol) of 2-
cyclohexen-1-one were added dropwise and the reaction
mixture was stirred at room temperature for 4 h. The
organic layer was extracted with diethyl ether, dried over
MgSO4, filtered and the solvent was evaporated. The
product obtained (3-N,N-dimethylaminocyclohexanone)
was added dropwise to a three-necked 250 ml round-
bottomed flask, containing a suspension of lithium alu-
minum hydride (4.0 g, 0.10 mol) in tetrahydrofuran
(60 ml), with stirring, at ꢁ 10 ꢃC and in a nitrogen
atmosphere. The mixture was allowed to warm to room
temperature and stirred for more 1.5 h. Water was added,
carefully, to destroy excess of lithium aluminum hydride.
The organic layer was separated with diethyl ether, dried
over MgSO4, filtered and the solvent was evaporated. The
product was distilled to give cis- and trans-3-N,N-di-
methylaminocyclohexanol (1), in a ratio of 76:24 (15.4 g,
52%); b.p. 71–73 ꢃC/1.0 mmHg. cis-DACH was purified
by column chromatography using hexane–acetone (1:2)
as eluent and 230–400 mesh silica gel.
in a ratio of 93:7 (1.7 g, 82%). cis-DAMCH was purified
by column chromatography, using hexane as eluent and
230–400 mesh silica gel.
1
cis: H NMR (500 MHz, CDCl3), ꢁ 3.13 (tt, 10.89,
3.45, 1H), 3.36 (s, 3H), 2.27 (s, 6H), 2.26 (m, 1H), 2.06
(m, 1H), 1.85 (m, 1H), 1.82 (m, 1H), 1.21 (m, 1H), 1.15
(m, 1H), 1.12 (m, 1H), 1.06 (m, 1H). 13C NMR
(500 MHz, CDCl3), ꢁ 79.0, 61.9, 55.7, 41.3, 34.4, 31.7,
27.6, 22.2.
1
trans: H NMR (500 MHz, CDCl3), ꢁ 3.62 (m, 1H),
3.31 (s, 3H), 2.52 (tt, 10.80, 4.06, 1H), 2.26 (s, 6H); the
remaining signals could not be attributed, because the
trans isomer was present in only a very small amount in
the mixture.
Catalysts. Rhodium oxide catalyst, Rh(Ox)Li, was
prepared by lithium nitrate fusion with rhodium chloride
trihydrate, as described by Nishimura et al.30
cis: 1H NMR (500 MHz, CDCl3), ꢁ 3.69 (tt, 7.77, 3.83,
1H), 2.28 (m, 1H), 2.28 (s, 6H), 1.92 (m, 1H), 1.83 (m, 1H),
1.81 (m, 1H), 1.68 (m, 1H), 1.52 (m, 1H), 1.40 (m, 1H),
1.35 (m, 1H), 1.28 (m, 1H). 13C NMR (500 MHz,
CDCl3), ꢁ 69.3, 61.6, 42.1, 36.4, 35.0, 27.9, 20.0.
Acknowledgments
The authors thank FAPESP for financial support of this
research and for a scholarship (to P.R.O.), CNPq for a
fellowship (to R.R.) and CENAPAD-SP for computer
facilities (Gaussian 98).
1
trans: H NMR (500 MHz, CDCl3), ꢁ 4.18 (m, 1H),
2.64 (tt, 10.57, 3.42, 1H), 2.25 (s, 6H), the remaining
signals could not be assigned, since it was not possible
obtain the pure trans isomer. 13C NMR (500 MHz,
CDCl3), ꢁ 66.7, 58.1, 41.4, 36.0, 33.0, 27.6, 19.6.
REFERENCES
1. Allan EA, Reeves LW. J. Phys. Chem. 1962; 66: 613–617.
2. Hay RW, Williams PP. J. Chem. Soc. 1964; 2270–2272.
3. Koelle U, Forsen S. Acta Chem. Scand., Ser. A 1974; 28: 531–540.
4. Fateley WG, Carlson GL, Bentley FF. J. Phys. Chem. 1975; 79:
199–204.
5. Schaefer T. J. Phys. Chem. 1975; 79: 1888–1890.
6. Pimentel GC, McClelland AL. The Hydrogen Bond. W. H. Free-
man: San Francisco, 1960.
7. Finegold H, Kwart H. J. Org. Chem. 1962; 27: 2361–2365.
8. Kuhn LP. J. Am. Chem. Soc. 1952; 74: 2492–2499.
9. Abraham RJ, Smith TAD, Thomas WA. J. Chem. Soc., Perkin
Trans. 2 1996; 1949–1955.
10. Eliel EL, Wilen SH, Mander LN. Stereochemistry of Organic
Compounds. Wiley: New York, 1994; 690–707.
11. Wiberg KB, Hammer JD, Castejon H, Bailey WF, Deleron EL,
Jarret RM. J. Org. Chem. 1999; 64: 2085–2095.
12. Basso EA, Oliveira PR, Caetano J, Schuquel ITA. J. Braz. Chem.
Soc. 2001; 12: 215–222.
13. Kost D, Peor N, Moriah GS, Sharabi Y, Durocher DT, Raban M.
J. Org. Chem. 2002; 67: 6938–6943.
14. Freitas MP, Tormena CF, Oliveira PR, Rittner R. J. Mol. Struct.
(Theochem) 2002; 589: 147–151.
15. Corey EJ, Feiner NF. J. Org. Chem. 1980; 45: 765–780.
16. Franzus B, Hudson BE. J. Org. Chem. 1963; 28: 2238–2244.
17. PCModel, Version 75, Serena Software: Bloomington, IN, 2001.
18. Haasnoot CAG, Leeuw FAAM, Altona C. Tetrahedron 1980; 36:
2783–2792.
19. Eliel EL, Schroeter SH. J. Am. Chem. Soc. 1965; 87: 5031–5038.
20. Abraham RJ, Chambers EJ, Thomas WA. J. Chem. Soc., Perkin
Trans. 2 1993; 1061–1066.
21. Wypych G. Handbook of Solvents. ChemTec Publishing: Toronto,
2001; 598.
22. Frisch MJ, Trucks GW, Schlegel HB, Cheeseman JR, Gill PMW,
Wong MW, Foresman JB, Johnson BG, Robb MA, Replogle ES,
Gomperts R, Andres JL, Raghavachavi K, Blinkley JS,
3-N,N-Dimethylaminoanisole. A 5.0 g (46 mmol) amount
of 3-aminophenol and 70 ml of dry THF were placed in a
two-necked 125 ml round-bottomed flask, fitted with a
calcium chloride protected reflux condenser, a dropping
funnel and magnetic stirrer. Then 4.4 g (0.18 mol) of
sodium hydride were added and the reaction mixture
was stirred at room temperature for 1.5 h. The reaction
mixture was cooled to 0 ꢃC and 19.6 g (0.14 mol) of
methyl iodide, in 25 ml of dry THF, were gradually
added. The ice-bath was removed and stirring continued
for 1.5 h, under reflux. The solution was cooled to 20 ꢃC
and water was gradually added to the reaction flask to
destroy the excess of NaH. The organic layer was
separated with diethyl ether, dried over MgSO4, filtered
and the solvent was evaporated. The product was distilled
to give 3-N,N-dimethylaminoanisol (5.9 g, 85%); b.p. 82–
84 ꢃC/5 mmHg.
cis- and trans-3-DAMCH (2). A 2.0 g amount of 3-N,N-
dimethylaminoanisole in 15 ml of tert-butyl alcohol was
hydrogenated, in a 100 ml autoclave, in the presence of
0.5 g of rhodium oxide catalyst, Rh(Ox)Li, at 60 ꢃC,
under a hydrogen pressure of 500–700 psi. The reduction
was allowed to proceed for 6 h. The catalyst was filtered
and the clear solution was concentrated to give cis- and
trans-3-N,N-dimethylamino-1-methoxycyclohexanol (2)
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521

CONFORMATIONAL EQUILIBRIUM OF cis-3-N,N-DIMETHYLAMINOCYCLOHEXANOL
521
Gonzalez C, Martin RL, Fox DJ, Defrees DJ, Baker J, Stewart JP,
Keith T, Peterson GA, Montgomery JA, Al-Laham MA, Zakr-
zewski VG, Ortiz JV, Cioslowski J, Stefanov BB, Nanayakkara A,
Challacombe M, Peng CY, Ayala PY, Chen W, Head-Gordon M,
Pople JA. Gaussian 98, Revision A.7. Gaussian: Pittsburgh, PA,
1998.
25. Craw JS, Bacskay GB. J. Chem. Soc., Faraday Trans. 1992; 88:
2315–2321.
26. Schaefer T, Sebastian R, McKinnon DM, Spevack PW, Cox KJ,
Takeucki CS. Can. J. Chem. 1993; 71: 960–967.
27. Simperler A, Mikenda W. Monatsh. Chem. 1997; 128; 969–980.
28. Lampert H, Mikenda W, Karpfen A. J. Phys. Chem. 1996; 100:
7418–7425.
29. Millefiori S, Di Bella S. J. Chem. Soc., Faraday Trans. 1991; 87:
1297–1302.
30. Nishimura S, Uchino H, Yoshino H. Bull. Chem. Soc. Jpn. 1968;
41: 2194.
23. Foresman JB, Frish A. Exploring Chemistry with Electronic
Structure Methods. Gaussian: Pittsburgh, PA, 1996.
24. George P, Bock CW, Trachtman M. J. Mol. Struct. (Theochem)
1985; 133: 11–24.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 513–521