Relevance of torsional effects to the conformational equilibria of 1,5-diaza-cis-decalins: A theoretical and experimental study

Article abstract of DOI:10.1021/jo000580i

1,5-Diaza-cis-decalin populates two conformations in which the nitrogen atoms are either gauche (N-in) or anti (N-out) to one another. The equilibrium mixture of the two conformers depends on the substituents at the nitrogen atom, as well as the reaction conditions. Ab initio (HF/6-31G*, B3LYP/6-31+G*) and molecular mechanics (Amber*) calculations have been performed to examine the possible role of stereoelectronics and steric effects in controlling the equilibrium of substituted 1,5-diaza-cis-decalins. In the present study, N,N′-diethyl- and N,N′-bistrifluoroethyl-1,5-diaza-cis-decalins have been synthesized, and the equilibrium mixtures have been measured using ¹H and ¹³C NMR experiments. Steric effects appear to control the equilibria between the two conformational isomers of 1,5-diaza-cis-decalin while torsional effects appear to dominate the equilibria for the N,N′-dialkyl derivatives.

Full text of DOI:10.1021/jo000580i

J . Org. Chem. 2001, 66, 1103-1108
1103
Releva n ce of Tor sion a l Effects to th e Con for m a tion a l Equ ilibr ia of
1,5-Dia za -cis-d eca lin s: A Th eor etica l a n d Exp er im en ta l Stu d y
Bishwajit Ganguly, Dana A. Freed, and Marisa C. Kozlowski*
Department of Chemistry, Roy and Diana Vagelos Laboratories, University of Pennsylvania,
Philadelphia, Pennsylvania 19104
marisa@chem.upenn.edu
Received April 17, 2000
1,5-Diaza-cis-decalin populates two conformations in which the nitrogen atoms are either gauche
(N-in) or anti (N-out) to one another. The equilibrium mixture of the two conformers depends on
the substituents at the nitrogen atom, as well as the reaction conditions. Ab initio (HF/6-31G*,
B3LYP/6-31+G*) and molecular mechanics (Amber*) calculations have been performed to examine
the possible role of stereoelectronics and steric effects in controlling the equilibrium of substituted
1,5-diaza-cis-decalins. In the present study, N,N′-diethyl- and N,N′-bistrifluoroethyl-1,5-diaza-cis-
1
decalins have been synthesized, and the equilibrium mixtures have been measured using H and
13C NMR experiments. Steric effects appear to control the equilibria between the two conformational
isomers of 1,5-diaza-cis-decalin while torsional effects appear to dominate the equilibria for the
N,N′-dialkyl derivatives.
In tr od u ction
as on solvent and additives.4 These experimental results
have generally been rationalized on the basis of stabiliz-
ing C-H to C-N delocalization, hydrogen bonding, and
solvent effects.4 Since the N-in form of these compounds
is presumably the effective form, a thorough understand-
ing of the factors controlling the equilibrium of the
neutral species in eq 1 was required so that derivatives
which would favor the N-in form could be selected for
synthesis and experimental evaluation.
We report herein a detailed computational study of the
conformational isomers of 1,5-diaza-cis-decalin (1) and
its substituted derivatives using molecular mechanics
and ab initio analyses. In addition, 1H and 13C NMR
experiments have been used to measure the conforma-
tional isomer ratio for the N,N′-diethyl- and N,N′-
bistrifluoroethyl-1,5-diaza-cis-decalin derivatives. These
studies have allowed the specific interactions controlling
the conformational equilibria of substituted 1,5-diaza-
cis-decalins to be delineated.
An ongoing research interest in our group is the design
and development of new chiral auxiliaries and catalysts
for asymmetric synthesis. The development of effective
chiral auxiliaries (stoichiometric) and catalysts (substo-
ichiometric) for asymmetric reactions depends largely on
the identification of the appropriate chiral ligands. As
an aid toward this end, a series of computer-aided design
protocols for the identification of new chiral ligands for
reactions which proceed through well-defined transition
states has been developed.1 This process identifies ligand
families via computational screening of data incorporated
within large structural databases such as the Cambridge
Structural Database. The critical motifs from the result-
ant lead compounds are identified and are used as the
basis for potential ligands. The benefit of this method
lies in the large number of potential scaffolds which can
be rapidly and efficiently screened by the end user. Using
this method, novel ligands containing a cis-decalin ring
system, including 1,5-diaza-cis-decalin (1), have been
identified as potential chiral auxiliaries for the boron
allylation and aldol addition reactions.
Resu lts a n d Discu ssion
A prior combined study by Santos et al.4 examined the
conformational equilibria of the 1,5-diaza-cis-decalins in
CDCl3 (Scheme 1, entries 1-3). The NH compound 1 has
a clear tendency to populate conformer a , in which
heteroatoms are gauche (N-in). The conformer equilib-
rium shifts in the direction of the b-type anti isomer (N-
out) upon alkylation of the nitrogens. The conformational
preference is completely shifted in the case of the N,N′-
di-n-butyl derivative (Scheme 1). 1H and 13C NMR studies
have shown that the N-in and N-out isomers in each case
are symmetric in nature.
The 1,5-diaza-cis-decalins, which have also found util-
ity in asymmetric lithiation/substitution2 and oxidative
biaryl coupling chemistry,3 exist predominantly in two
conformations (eq 1). The position of the conformational
equilibrium depends on the nitrogen substituents, as well
To gain additional insight to this problem, N,N′-diethyl
and N,N′-bistrifluoroethyl derivatives 4 and 5 have been
synthesized in our laboratory as outlined in eq 2 and have
(1) Kozlowski, M. C.; Evans, C. A. Submitted.
(2) Li, X.; Schenkel, L. B.; Kozlowski, M. C. Org. Lett. 2000, 2, 875-
8.
(4) Santos, A. G.; Klute, W.; Torode, J .; Bo¨hm, V. P. W.; Cabrita,
(3) Li, X.; Yang, J .; Kozlowski, M. C. Submitted.
E.; Runsink, J .; Hoffmann, R. W. New J . Chem. 1998, 993-997, 1411.
10.1021/jo000580i CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/26/2001

1104 J . Org. Chem., Vol. 66, No. 4, 2001
Ganguly et al.
Ta ble 1. Rela tive En er gies (k ca l/m ol) Ca lcu la ted w ith Am ber * a s w ell a s a b In itio HF /6-31G* a n d B3LYP /6-31+G* for
th e Low est En er gy F or m s of Con for m er s a a n d b of 1,5-Dia za -cis-d eca lin 1 a n d N,N′-Dia lk yla ted Der iva tives 2, 4, a n d 5
method
1a
0.0
0.0 (0.0)
0.0
1b
3.4
2.0 (1.7)
1.4
2a
0.0
1.7 (4.5)
1.1
2b
0.4
0.0 (0.0)
0.0
4a
0.7
3.9 (2.9)
3.8
4b
0.0
0.0 (0.0)
0.0
5a
0.0
3.6 (3.7)
2.9
5b
0.8
0.0 (0.0)
0.0
Amber*
HF/6-31G*a
B3LYP/6-31+G*
a
Relative solvation energies calculated using the IPCM method for CHCl3 are given in parentheses.
Sch em e 1
Sch em e 2
been characterized by 1H and 13C NMR spectroscopy
(Scheme 1). Our motivation for examining these deriva-
tives was to determine what effects groups with different
steric sizes and different electronic characteristics would
have on the conformational equilibria of these systems.
The experimental data from these additional analogues
would allow dissection of the stereoelectronic, electro-
static, and steric components in the conformational
equilibria of the diaza-cis-decalin systems. Analysis of the
NMR structural results revealed that the N,N′-diethyl
and N,N′-bistrifluoroethyl derivatives populate N-out
conformer b almost exclusively (Scheme 1).
mented in MacroModel provided the energies illustrated
in Table 1 (the N,N′-di-n-butyl derivative 3 was not
examined due to its relatively large size). In a second set
of calculations, further optimization at the ab initio HF/
6-31G* and B3LYP/6-31+G* levels of theory was under-
taken (Table 1). In addition, solvent calculations have
been incorporated into the HF/6-31G* calculations using
the IPCM model.
Due to the conformational flexibility of the ethyl and
trifluoroethyl groups, three conformers were examined
for each N-in and N-out isomer in these cases (Scheme
2). The illustrated conformations represent the most
stable form of the cis-decalin system, a chair-chair
conformation, and include all of the possible symmetrical
staggered conformations arising from rotation of the
substituent ethyl or trifluoroethyl groups (the nonsym-
metrical possibilities were also considered, but constitute
intermediary energy points and are not presented in
order to simplify the analysis). The φ-angles (180°, 60°,
-60°) shown in Scheme 2 illustrate the orientation of
ethyl and trifluoroethyl groups with respect to the
nitrogen lone-pair.
Table 2 summarizes the relative stability of N,N′-
diethyl 4 and N,N′-bistrifluoroethyl 1,5-diaza-cis-decalin
5 conformational isomers outlined in Scheme 2. The anti
conformation a 1 (Scheme 2) is preferred over the gauche
conformations a 2 and a 3 for the N-in isomer, while the
gauche conformations b2 and b3 are preferred over anti
b1 for N-out isomers (Scheme 2). Solvent calculations
performed using the IPCM model are qualitatively in
agreement with the calculated gas-phase results for 4a
and 5. For 4b, the solvent calculations show that 4b2 and
4b3 are very close in energy similar to the gas phase and
solvent results obtained for 5b. Regardless, the relative
stability of 4a vs 4b and 5a vs 5b is unaffected upon
consideration of solvation as illustrated in Table 1, which
contains the lowest energy geometries of the N-in and
N-out conformers from Table 2.
The relative stability of the conformers of 1,5-diaza-
cis-decalin and their substituted derivatives has been
calculated using molecular mechanics5 and ab initio6
approaches. The reliability of these methods to examine
the relative stability of conformers of similar N-C-C-N
systems has previously been acknowledged.7 In this
study, the possible role of stereoelectronic interactions
(such as C-H to C-N* delocalization, negative hyper-
conjugation), steric, electrostatic, torsional, and solvent
effects in controlling the conformational equilibria of 1,5-
diaza-cis-decalins has been examined and compared with
experimental results. The possibility of conformational
variation in the ethyl and trifluoroethyl groups was
considered by using different initial geometries.
Minimization of 1,5-diaza-cis-decalin 1 and its substi-
tuted derivatives using the Amber* force field as imple-
(5) (a) Macromodel v4.0, v5.0, v6.0; Still, W. C.; Columbia University.
(b) Mohamdi, F.; Richards, N. G.; Guida, W. C.; Liskamp, R.; Lipton,
M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J . Comput.
Chem. 1990, 11, 440-467.
(6) Hehre, W. J .; Radom, L.; Schleyer, P. v. R.; Pople, J . A. Ab initio
Molecular Orbital Theory; Wiley: New York, 1988.
(7) Chang, Y.-P.; Su, Li, T.-W.; Chao, I. J . Phys. Chem. A 1997, 101,
6107-6117. (b) Reany, O.; Goldberg, I.; Abramson, S.; Golender, L.;
Ganguly, B.; Fuchs, B. J . Org. Chem. 1998, 63, 8850-8869. (c) Star,
A.; Goldberg, I.; Lemcoff, N. G.; Fuchs, B. Eur. J . Org. Chem. 1999,
2033, 3-2043.

Conformational Equilibria of 1,5-Diaza-cis-decalins
J . Org. Chem., Vol. 66, No. 4, 2001 1105
Ta ble 2. Rela tive En er giesa (k ca l/m ol) Ca lcu la ted w ith Am ber * a s w ell a s a b In itio HF /6-31G* a n d B3LYP /6-31+G* Levels
for th e N-in a n d N-ou t Con for m er s of N,N′-Dieth yl 4 a n d N,N′-Bistr iflu or oeth yl 5
method
4a 1
0.0
0.0 (0.0)
0.0
4a 2
1.1
1.2 (1.0)
0.0
4a 3
4b1
3.9
4.5 (4.1)
4.5
4b2
0.0
0.0 (0.3)
0.0
4b3
0.5
1.0 (0.0)
0.7
Amber*
2.6
HF/6-31G*b
1.2 (1.0)c
0.0c
B3LYP/6-31+G*
5a 1
5a 2
5a 3
5b1
5b2
5b3
Amber*
0.0
0.0 (0.0)
0.0
2.2
1.5 (3.0)
1.7
3.0
5.3
3.9 (6.2)
3.1
0.0
0.0 (0.0)
0.0
0.4
HF/6-31G*b
B3LYP/6-31+G*
1.5 (3.0)d
1.7d
0.0 (0.0)e
0.0e
a
Relative energies are reported with respect to the lowest energy isomer within each of the series 4a 1-3, 4b1-3, 5a 1-3, and 5b1-3
.
Relative solvation energies calculated using the IPCM method for CHCl3 are given in parentheses. c Converged to the same conformation
b
as 4a 2. Converged to the same conformation as 5a 2. e Converged to the same conformation as 5b2.
d
Ta ble 3. Rela tive En er gies (k ca l/m ol) Ca lcu la ted w ith
Am ber * Usin g AM1 Ch a r ges for th e Con for m er s of 1, 2, 4,
a n d 5
Generally, the calculated results for the N-in and N-out
isomers (Table 1) are in agreement with experimental
observations (Scheme 1). Molecular mechanics results
indicate that N-in 1a should be the more stable conformer
for 1,5-diaza-cis-decalin in agreement with experimental
data. With the N,N′-dimethyl derivative 2 and N,N′-
diethyl derivative 4, the energy differences calculated
using molecular mechanics also corroborate the experi-
mentally determined stabilities. In one case (5, N,N′-
bistrifluoroethyl), the molecular mechanics results indi-
cate that the N-in isomer is more stable contrary to
experimental observations. This result has been inves-
tigated further and is discussed later in this study.
Further refinement was undertaken with ab initio
calculations (Table 1). Solvent calculations using the
IPCM model for the conformers of 1, 2, 4, and 5
qualitatively agree with the gas phase ab initio results.8
As such, it appears that solvation is not playing a key
role in influencing the relative stabilities of the conform-
ers in this series. The HF/6-31G* calculation for the
parent system 1 agrees with the molecular mechanics
result that N-in conformer 1a is more stable. On the other
hand, the HF/6-31G* calculation for N,N′-dimethyl de-
rivative overestimates the stability of N-out conformer
2b. Calculations performed at B3LYP6-31+G* level9
reduce the energy differences, but do not change the
preferences (Table 1). HF/6-31G* and B3LYP/6-31+G*
calculations correctly reproduce the stability of the N-out
form over the N-in for N,N′-diethyl 4 and N,N′-bistrif-
luoroethyl 5, in agreement with experimental observa-
tions (Scheme 1). It is interesting to note that the energy
difference between the N-in and N-out isomers for N,N′-
bistrifluoroethyl derivative 5 changes at the B3LYP/6-
31+G* level of theory by 0.7 kcal/mol but only by 0.1 kcal/
mol for the N,N′-diethyl derivative 4 (Table 1). The
greater energy difference for 5 can be accounted for by
negative hyperconjugation. The low-lying C-C σ*-orbital
of the CH2CF3 group is appropriately oriented to interact
with the nitrogen lone-pair in conformer 5a 1, which
stabilizes the N-in isomer. Apparently, such negative
hyperconjugation is not as significant for N,N′-diethyl
derivative 4.
1a
1b
2a
2b
4a
4b
5a
5b
Amber*
0.0
4.5
0.0
0.4
0.7
0.0
2.9
0.0
for the N,N′-bistrifluoroethyl case might be due to the
poor charge description of CH2CF3 group in 5 by molec-
ular mechanics. Since standard force fields typically do
not perform well with polar functional groups,5 the
electrostatic charges for N,N′-bistrifluoroethyl derivative
5 were calculated separately using a semiempirical (AM1)
method.10 These electrostatic charges were then incor-
porated into the molecular mechanics (Amber*) calcula-
tions. As mentioned above, three different conformers
were examined for the N-in and N-out isomers of 4 and
5 (Scheme 2). As before, the anti conformation a 1 is
preferred for the N-in arrangement and the gauche
conformation b2 is preferred for the N-out arrangement
in 4 and 5 (Scheme 2). 1,5-Diaza-cis-decalins 1, 2, and 4
were also reanalyzed using this method, and the results
are summarized in Table 3. The calculated results with
the AM1 charges do not significantly change for 1, 2, and
4 compared to the results using the Amber* charges.
With N,N′-bistrifluoroethyl derivative 5 there is a dra-
matic change upon application of AM1 charges (Table 2
vs Table 3). Using this technique, the calculated results
are qualitatively in agreement with the experimental
results (Scheme 1 vs Table 3). While the poor treatment
of the CH2CF3 using the standard Amber* force field
might arise from other parameters (such as van der
Waals and torsional terms), agreement between experi-
ment and theory across the series indicates that applica-
tion of corrected charges is useful for a qualitative
assessment in this particular case.
The interactions that control the relative stability of
conformers of 1,5-diaza-cis-decalin and their substituted
derivatives could arise from a number of sources (steric,
electrostatic, stereoelectronic, and torsional). In the fol-
lowing paragraphs each of these interactions is analyzed
with the aim of identifying the critical interactions. The
stability of N-in isomer 1a over N-out isomer 1b can be
explained on the basis of 1,4-nonbonded axial hydrogen
interactions. In the case of the N-in isomer, 1,4-non-
bonded hydrogen interactions are absent, whereas the
N-out isomer 1b has three 1,4-nonbonded hydrogen
interactions (within 2.4 Å), which presumably destabilize
the latter isomer (Scheme 3).11 However, these interac-
The failure of molecular mechanics calculations (Am-
ber*) to predict 5b (N-out) as the more stable conformer
(8) For one case, 2a , a substantial energy change between gas phase
(1.7 kcal/mol) and IPCM CHCl3 (4.5 kcal/mol) calculations is observed.
While the increase may reflect stabilization of 2b by CHCl3, the
experimental ratio in chloroform (60:40 2a :2b) indicates that the
calculations overestimate the magnitude.
(9) For this system (1, 2, 4, 5), less costly single point B3LYP/6-
31+G*//HF/6-31G* calculations gave essentially the same energies
(within 0.1 kcal/mol) as the more extensive B3LYP/6-31+G* calcula-
tions.
(10) (a) SPARTAN version 5.1.2. Wavefunction, Inc.; 18401 Von
Karman Ave., Suite 370; Irvine, CA 92612 U.S.A. (b) Chirlian, L. E.;
Francl, M. M. J . Comput. Chem. 1987, 8, 894-905. (c) Breneman, C.
M.; Wiberg, K. B. J . Comput. Chem. 1990, 11, 361-373.

1106 J . Org. Chem., Vol. 66, No. 4, 2001
Ganguly et al.
Sch em e 3
between the nitrogen atoms and stabilize the N-in
isomer. Experimentally, the N-out conformers were found
to be increasingly preferred over the N-in conformers in
the series R ) H, Me, Et, CH2CF3 (Scheme 1). As such,
the differences in nitrogen lone pair and nitrogen elec-
trostatic repulsions between these compounds do not
appear to be the predominant factors governing the
conformational equilibria of the 1,5-diaza-cis-decalins.
Sch em e 4
Stereoelectronic interactions were also evaluated as a
potential source of the different stabilities of the N-in and
N-out forms of the 1,5-diaza-cis-decalins.15,16 The relevant
interactions include (i) negative hyperconjugation be-
tween the nitrogen lone-pair and a properly oriented σ*-
acceptor (C-H and C-C bonds) and (ii) donation from
the best σ-donor (C-H bond) to the best σ*-acceptor
(C-N bond). Comparing negative hyperconjugation in-
teractions, the N-in isomer has four σ* acceptor C-H
bonds available to interact with the nitrogen lone-pair
while the N-out isomer has two σ* acceptor C-H and two
σ* acceptor C-C bonds. Upon examination of the relative
C-H and C-C bond lengths of 1 calculated at HF/6-31G*
level, it appears that the C-H and C-C bonds anti to
nitrogen lone-pairs are relatively longer than the other
C-H and C-C bonds. Similar trends were observed for
the N,N′-dialkyl derivatives 2, 4, and 5. While this
observation supports the presence of negative hypercon-
jugation interactions,16 Wiberg et al. have proposed that
such behavior may also be explained by repulsive inter-
actions between the nitrogen lone-pair and the backside
of the C-H σ-bond orbital.17
Examination of the parent system 1 reveals that four
C-H σ-bonds are aligned to interact with σ* acceptor
C-N bonds in the N-in isomer while two C-H and two
C-C σ-bonds can interact in the N-out isomer. Assuming
that C-H bonds are stronger donors than C-C bonds,18
the stability of N-in isomer 1a over N-out isomer 1b can
be explained, but the stability of N-out isomers 2b, 4b,
and 5b cannot be accounted for using this analysis
(Scheme 1 and Table 1). Since the number of stereoelec-
tronic interactions is equivalent in the N-out and N-in
isomers of 1, the real issue centers on the magnitude of
these interactions; there is no definitive evidence to
indicate that these interactions differ enough to cause
the degree of preferential stabilization that is ob-
served.19,20,21 Moreover, the stabilities of the N-in isomers
tions are also present in the N,N′-dialkylated derivatives.
As such, an additional interaction(s) must be responsible
for the shift in the equilibria toward the N-out form upon
N-alkylation.
For parent 1, an N-in conformer with an axially
disposed hydrogen which can undergo a weak intra-
molecular hydrogen bond is also possible.12 Fleischhauer
et al. have examined the stability of this hydrogen bonded
conformer and found that the hydrogen bonded N-in
isomer is more stable than the non-hydrogen bonded N-in
isomer (1a ).13 Therefore, the hydrogen bonded conforma-
tional isomer would also contribute toward the stability
of the N-in conformer. While such an effect might serve
to explain the greater stability of the N-in conformer for
parent 1 relative to 2, 4, and 5, it does not account for
the observed effects within the N,N′-dialkylated series.
In addition, a similar preference for the N-in isomer has
been observed for 1-aza-cis-decalin which cannot undergo
such a hydrogen bond indicating that other factors are
critical (see above).14 Even though hydrogen bonded 1a
was not used explicitly in this study (due to the different
geometry as compared to the N,N′-dialkyl derivatives),
inclusion of hydrogen bonded 1a does not alter any of
the trends or conclusions drawn.
Another factor that can influence the relative stability
of the N-in isomers of 1,5-diaza-cis-decalins arises from
repulsion between the nitrogens due to lone pair interac-
tions or nitrogen electrostatic interactions. From Scheme
4 it is anticipated that both of these effects would
destabilize the N-in form relative to the N-out form across
the series. The addition of electron donating N-alkyl
groups such as methyl (2) and ethyl (4) should enhance
this repulsive interaction between the nitrogen lone pairs,
destabilizing the N-in isomers relative to the parent 1
(R ) H). On the other hand, the electron-withdrawing
N-trifluoroethyl group (5) should reduce the repulsions
(15) (a) Epiotis, N. D.; Yates, J . R.; Larson, C. R.; Kirmaier, C. R.;
Bernardi, F. J . Am. Chem. Soc. 1977, 99, 8379-8391. (b) Brunck, T.
K.; Weinhold, F. J . Am. Chem. Soc. 1979, 101, 1700-1703. (c) Dionne,
P.; St-J acques, M. J . Am. Chem. Soc. 1987, 109, 2616-2623. (d)
Senderowitz, H.; Golender, L.; Fuchs, B. Tetrahedron 1994, 50, 9707-
9728. (e) Allinger, N. L.; Yuh, Y. H.; Lii, J -H. J . Am. Chem. Soc. 1989,
111, 8551-8566.
(16) (a) Brockway, L. O. J . Phys. Chem. 1937, 41, 185. (b) Patrick,
C. R. Adv. Fluorine Chem. 1961, 2, 1. (c) Albright, T. A.; Burdett, J .
K.; Whangbo, M.-H. Orbital Interactions in Chemistry; Wiley: New
York, 1985; p 171.
(17) (a) Wiberg, K. B.; Breneman, C. M. J . Am. Chem. Soc. 1990,
112, 8765-8775. (b) Wiberg, K. B. J . Org. Chem. 1991, 56, 544-561.
(18) Rablen, P. R.; Hoffman, R. W.; Hrovat, D. A.; Borden, W. T. J .
Chem. Soc., Perkin Trans. 2 1999, 1719-1726.
(11) (a) Carey, F. A.; Sundberg, R. J . Advanced Organic Chemistry,
3rd ed.; Plenum: New York; p 138. (b) Booth, H.; Bostock, A. H. J .
Chem. Soc., Chem. Commun. 1967, 177-178.
(12) The N-H-N bond angle in this “hydrogen bond” is almost 90°.
As such, the term hydrogen bond is used only in a descriptive sense
since almost no orbital overlap is possible. This interaction is more
accurately considered as an electrostatic interaction between the axial
hydrogen and the nitrogen across the ring.
(13) Fleischhauer, J .; Raabe, G.; Santos, A. G.; Schiffer, J .; Wollmer,
A. Z. Naturforsch 1998, 53a, 896-902.
(14) (a) Booth, H.; Griffiths, D. V. J . Chem. Soc., Perkin Trans. 2
1975, 111-118. (b) Booth, H.; Griffiths, D. V. J . Chem. Soc., Chem.
Commun. 1973, 666-667.
(19) Whether the σ-orbital of a C-H bond is a better donor than
that of a C-C bond has been a subject of debate for many decades
(see ref 17). The hyperconjugative effect as first suggested by Baker-
Nathan ranks the C-H bond as a stronger donor than the C-C bond;
however, there is no compelling evidence to support such a statement.
Even though negative hyperconjugation is manifested in the confor-
mational behavior of many experimentally studied systems, its origin
and magnitude are a subject of continual debate (see ref 21).
(20) (a) March, J . Organic Reactions and Mechanism, VCH: New
York. (b) Cooney, B. T.; Happer, D. A. R. Aust. J . Chem. 1987, 40,
7-1555.

Conformational Equilibria of 1,5-Diaza-cis-decalins
J . Org. Chem., Vol. 66, No. 4, 2001 1107
Ta ble 4. Th e Degr ee of P yr a m id a liza tion (360° - Σ RNR
An gles) of th e Tw o Nitr ogen Atom s a n d th e Tor sion a l
An gle (R-N1-C1′-C8) for 1, 2, 4, a n d 5 a s Obser ved in
th e B3LYP /6-31+G* a n d (Am ber *) Ca lcu la ted Geom etr ies
at the nitrogen atoms for the 1,5-diaza-cis-decalin con-
formers and the N,N′-dialkyl derivatives. N,N′-Bistrif-
luoroethyl 5 proves to be the exception as molecular
mechanics (Amber*) calculations appear to underesti-
mate the degree of pyramidalization of nitrogen atoms
(Table 4). This result is most likely a consequence of the
electrostatic charges introduced from AM1 calculations;
however, the direction of the perturbations are consistent
with all the other calculated results. Overall, the results
from the molecular mechanics and ab initio calculations
suggest that N-alkylation flattens the nitrogen atom,
inducing unfavorable eclipsing torsional interactions in
the N-in isomers a and slightly relieving unfavorable
gauche interactions in the N-out isomers b.
compound
N-in (a )
N-out (b)
pyramidalization (deg)
1
2
4
5
1
2
4
5
27.0 (28.6) 23.7 (25.2)
24.8 (27.0) 18.7 (20.6)
18.8 (20.2) 19.0 (21.0)
13.4 (5.0)
15.5 (0.3)
R-N1-C1′-C8 (deg)
54.9 (55.9) 57.5 (52.7)
55.7 (56.7) 62.8 (59.4)
48.0 (50.1) 63.1 (59.6)
41.1 32.4)
67.5 (109.3)
Sch em e 5
Con clu sion
The conformational equilibria of 1,5-diaza-cis-decalin
and its substituted derivatives have been calculated
using molecular mechanics (Amber*) and ab initio cal-
culations. N,N′-Diethyl- and N,N′-bistrifluoroethyl-1,5-
diaza-cis-decalins have been synthesized and the equi-
1
librium mixtures have been measured by means of H
and 13C NMR experiments. The calculated results cor-
relate with the experimentally determined stabilities of
conformational isomers in almost all the cases. The
apparent failure of molecular mechanics calculations to
predict 5b as the more stable isomer for the N,N′-
bistrifluoroethyl derivative has been accounted for by
introducing electrostatic charges from AM1 calculations.
Molecular mechanics and ab initio calculations have
allowed an understanding of the relative importance of
stereoelectronic, steric, and torsional effects in controlling
the stability of 1,5-diaza-cis-decalin conformers. Overall,
the calculated results suggest that stereoelectronic in-
teractions are largely equivalent in both forms, and the
relative stability of conformers is governed by steric and
torsional effects.
relative to the N-out isomer for 1, 2, 4, and 5 are borne
out in the molecular mechanics calculations even though
the geometries do not show lengthening of C-H and C-C
bonds anti to nitrogen lone pairs expected from such
stereoelectronic interactions. Since the standard force
fields used do not appear to reproduce these stereoelec-
tronic effects, our initial hypothesis was that the stability
of isomers N-in vs N-out is primarily governed by
nonbonded steric effects.22
Analyzing B3LYP/6-31+G* and molecular mechanics
calculated N-in and N-out geometries of 1 and the N,N′-
dialkyl derivatives 2, 4, and 5 reveals a possible source
for such a steric effect. The degree of pyramidalization
of nitrogen atoms decreases with the increasing size of
the alkyl groups (Table 4).23 As a consequence, the
R-N1-C1′-C8 torsional interaction becomes more
eclipsed in the case of N-in isomers a , which presumably
destabilizes the N-in form as the alkyl groups increase
in size (Scheme 5 and Table 4).24 On the other hand, the
R-N1-C1′-C8 angle increases with substitution in the
N-out isomers b relieving torsional strain between C1′-
C8 and N1-R.
Exp er im en ta l Section
Com p u ta tion a l Meth od ology. Molecular mechanics (Am-
ber*) calculations have been performed using Macromodel.5
The AM1 electrostatic charges have been calculated using
Spartan.10 Ab initio calculations have been performed using
Gaussian94 program.25 Geometries were fully optimized using
both the RHF method with a 6-31G* basis set and the B3LYP
HF-DFT method with a 6-31+G* basis set.26 The calculated
geometries have been characterized as true minima via
harmonic vibrational frequency analyses. The C2-symmetry
has been maintained in all the calculations. Nitrogen inversion
equilibrium has not been considered in this study; only
equatorial conformations have been calculated at all levels of
theory. The solvation energies have been calculated using
isodensity polarized continuum model (IPCM), which is a
dielectric continuum type method, where the shape and size
of the cavity occupied by the solute molecules is defined by an
isodensity surface of the solute.27 The IPCM calculations have
In general, the ab initio and molecular mechanics
geometries predict similar degrees of pyramidalization
(21) Wiberg, K. B.; Rablen, P. R. J . Am. Chem. Soc. 1993, 115, 614-
625.
(22) The molecular mechanics calculations used in this study do
contain explicit parameters for the anomeric effect, but parameters
for other stereoelectronic effects such as the lp-N-C-X (X ) H, C)
interactions under discussion are absent. The inability of molecular
mechanics to reproduce the expected geometrical parameters indicates
that the force fields did not incorporate such effects from the stan-
dardization set into other parameters.
(23) There is precedent for decreasing pyramidalization with in-
creasing substitution on nitrogen. For example, in the following series
the degree of pyramidalization (360 - Σ RNR angles) decreases with
increasing substitution: NH3: 39.9°, CH3NH2: 36.2°, (CH3)2NH: 28.0°,
and (CH3)3N: 27.3°. These pyramidalization angles were calculated
from the experimental bond angles: Pozzoli, S. A.; Rastelli, A.;
Tedeschi, M. J . Chem. Soc., Faraday Trans. 2 1973, 69, 256-261.
(24) For a related proposal in the cis-decahydroisoquinolines, see
ref 14.
(25) Gaussian 94: Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill,
P. M. W.; J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.;
Petersson, G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M.
A.; Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defreees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
Inc., Pittsburgh, PA, 1995.
(26) (a) Becke, A. D. J . Chem. Phys. 1993, 98, 5648-5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.
(27) Foresman, J . B.; Keith, T. A.; Wiberg, K. B.; Snoonian, J .;
Frisch, M. J . J . Phys. Chem. 1996, 100, 16098-16104.

1108 J . Org. Chem., Vol. 66, No. 4, 2001
Ganguly et al.
been performed using HF/6-31G* basis set at the appropriate
(gas phase) optimized geometries in chloroform.
and N-in conformers of 2 indicated that only the N-out isomer
was populated at room temperature.
N,N′-Bis-tr iflu or oa cyl-1,5-d ia za -cis-d eca lin . Trifluoro-
acetic anhydride (2.16 mL, 15.3 mmol) was added to a solution
of cis-1 (0.535 g. 3.82 mmol), Et3N (3.20 mL, 23.0 mmol), and
DMAP (∼2 mg) in CH2Cl2 (10 mL) at 0 °C. After stirring 1d,
water was added and the mixture was extracted with CH2Cl2.
The combined extracts were dried with MgSO4, and the solvent
evaporated. The residue was purified by chromatography (CH2-
Cl2) to provide the product as a light yellow solid in 85% yield
(1.08 g, 3.25 mmol): 1H NMR (200 MHz, CDCl3, 298 K) δ 1.63-
2.08 (m, 6H), 2.13-2.37 (m, 2H), 2.83 (t, J ) 13.1 Hz, 2H),
3.15 (m, 2H), 3.50-3.66 (m, 2H), 3.84 (bd, J ) 13.9 Hz, 2H)
3.97-4.10 (m, 2H) 4.46 (d, J ) 15.1 Hz, 2H), 4.60-4.73 (m,
2H).
N,N′-Bis-tr iflu or oeth yl-1,5-d ia za -cis-d eca lin (5). Selec-
tive reduction was accomplished without deacylation.29 After
BH3‚DMS (1.32 mL, 13.9 mmol) was added dropwise to a
stirred solution of N,N′-bis-trifluoroacyl-1,5-diaza-cis-decalin
(1.03 g, 3.09 mmol) in THF (20 mL) at 0 °C, the mixture was
heated at reflux for 2 d. While cooling in an ice bath, MeOH
was added until evolution of gas stopped, and then a large
amount of water was added. The resulting mixture was
extracted with Et2O, and the extracts were dried over Na2-
SO4. Concentration afforded the product as a yellow solid in
94% yield (0.88 g, 2.89 mmol): 1H NMR (500 MHz, CDCl3,
260 K) δ 1.42-1.69 (m, 8H), 2.61-2.63 (m, 4H), 2.93-3.07 (m,
6H); 13C NMR (500 MHz, CDCl3, 260 K) δ 16.93, 24.55, 47.20,
56.00 (q, J ) 123 Hz)) 60.37, 126.2 (q, J ) 1105 Hz); MS (ES)
m/z calc′d for C12H19F6N2+ requires 305.1439, found 305.1452.
Comparison of the NMR data from the N-out and N-in
conformers of 2 indicated that only the N-out isomer was
populated at room temperature.
Gen er a l In for m a tion . Unless otherwise noted, all non-
aqueous reactions and distillations were carried out under an
atmosphere of dry N2 in dried glassware. When necessary,
solvents and reagents were dried prior to use. THF and CH2-
Cl2 were dried and deoxygenated using a solvent column
purification system.28 A 450 mL reactor (No. N4767 from Parr)
was used for the high-pressure reactions with stirring by an
external magnetic stir plate. For the chromatographic and
instrumentation methods used as well as the spectral formats,
see the Supporting Information.
1,5-Dia za -cis-d eca lin (1).2 To a solution of naphthyridine
(20.0 g, 0.154 mol) in HOAc (120 mL) was added 10% Rh/Al2O3
(2.0 g). The mixture was stirred for 36 h at room temperature
under H2 (1100 psi). The mixture was filtered through Celite
to give a 90:10 mixture of the cis-1 and trans-1. While cooling
to 0 °C, saturated NaOH was added until pH > 14. After
saturation with NaCl, the mixture was extracted extensively
with Et2O. The combined extracts were dried over Na2SO4, and
the solvent was removed in vacuo to leave a 94:6 mixture of
the cis-1 and trans-1 isomers (13.7 g, 63%). Crystallization
from Et2O (∼20 mL) furnished cis-1 containing <5% of trans-
1. Distillation (88-90 °C, 2 mmHg) afforded the pure product
as a waxy white solid.
N,N′-Dim eth yl-1,5-diaza-cis-decalin (2).4 Compound cis-1
(0.730 g, 5.21 mmol) was added to a mixture of formic acid (2
mL) and 37% formaldehyde solution (4 mL). After warming
for 1 d to 70 °C, water was added followed by 50% aqueous
NaOH to obtain pH > 14. The mixture was extracted with CH2-
Cl2. The combined extracts were dried with K2CO3, and the
solvent was evaporated. Purification by distillation (100 °C, 1
mmHg) afforded the product as a colorless liquid in 87% yield
(0.760 g, 4.53 mmol). The 1H NMR data obtained from this
material was identical to that previously reported.
Ack n ow led gm en t. Financial support was provided
by the University of Pennsylvania, the National Science
Foundation (CHE-9730576), Merck Research Labora-
tories, DuPont, and Pharmacia-Upjohn. Computational
support was provided by the NCSA in the form of a
supercomputing time allotment. Acknowledgment is
made to the donors of the Petroleum Research Fund,
administered by the American Chemical Society, for
partial support of this research. The invaluable as-
sistance of Dr. George Furst in collecting the NMR
spectroscopic data is gratefully acknowledged.
N,N′-Diet h yl-1,5-d ia za -cis-d eca lin (4). Aqueous NaOH
(50%, 2 mL) was added to a solution of cis-1 (0.280 g, 2.00
mmol) in THF (3 mL) followed by EtI (0.400 mL, 5.00 mmol).
After stirring for 1 d, water was added and the mixture was
extracted with CH2Cl2. The combined extracts were dried with
K2CO3, and the solvent was evaporated. Purification by
distillation (110 °C, 2 mmHg) afforded the product as a
colorless liquid in 67% yield (0.262 g, 1.34 mmol): 1H NMR
(500 MHz, CDCl3, 260 K) δ 1.10 (t, J ) 7.2 Hz, 6H), 1.51-
1.60 (m, 6H), 1.73 (d, J ) 8.3 Hz, 2H), 2.34 (t, J ) 10.5 Hz,
2H), 2.51-2.59 (m, 6H), 3.10 (d, J ) 7.8 Hz, 2H); 13C NMR
(500 MHz, CDCl3, 260 K) δ 12.70, 14.34, 24.17, 46.46, 47.71,
Su p p or tin g In for m a tion Ava ila ble: Spectral assign-
ments for compounds 4 and 5 are provided. This material is
available free of charge via the Internet at http://pubs.acs.org.
+
56.84; MS (ES) m/z calcd for C12H24N2 requires 197.2020,
found 197.2017. Comparison of the NMR data from the N-out
J O000580I
(28) Modified procedure of Pangborn, A. B.; Giardello, M. A.; Grubbs,
R. H.; Rosen, R. K.; Timmers, F. J . Organometallics 1996, 15, 1518-
1520.
(29) Fuchigami, T.; Ichikawa, S. J . Org. Chem. 1994, 59, 607-615.