Catalytic asymmetric chiral lithium amide-promoted epoxide rearrangement: A NMR spectroscopic and kinetic investigation

Article abstract of DOI:10.1016/j.tetasy.2010.10.034

The lithium amide derived from the chiral diamine (1R,3S,4S)-3-(1- pyrrolidinyl)methyl-2-azabicyclo[2.2.1]heptane, has been reported to catalytically deprotonate cyclohexene oxide and other epoxides, yielding chiral allylic alcohols in excellent enantiomeric excess. In this work, ⁶Li, ¹H and ¹³C NMR spectroscopy have been used to study the aggregation of the chiral lithium amide in THF and the influence on the aggregation by the addition of additives, such as 1,8-diazabicyclo-[5.4.0]undec- 7-ene (DBU). The activated complex under catalytic deprotonation of cyclohexene oxide, that is, with excess Li-DBU and free DBU, is built from one monomer of the chiral lithium amide, one molecule of epoxide and one additional molecule of DBU. The reaction order (-0.97) obtained for the bulk base Li-DBU shows an inverse dependence on the concentration, suggesting a deaggregation of the initial mixed dimer to a monomer-based transition state containing a monomer of the lithium amide.

Full text of DOI:10.1016/j.tetasy.2010.10.034

Tetrahedron: Asymmetry 21 (2010) 2733–2739
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Catalytic asymmetric chiral lithium amide-promoted epoxide rearrangement:
a NMR spectroscopic and kinetic investigation
⇑
Peter Dinér
Department of Biochemistry and Organic Chemistry, Uppsala University, PO Box 576, S-751 23 Uppsala, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 October 2010
Accepted 28 October 2010
The lithium amide derived from the chiral diamine (1R,3S,4S)-3-(1-pyrrolidinyl)methyl-2-azabicy-
clo[2.2.1]heptane, has been reported to catalytically deprotonate cyclohexene oxide and other epoxides,
yielding chiral allylic alcohols in excellent enantiomeric excess. In this work, 6Li, 1H and 13C NMR spec-
troscopy have been used to study the aggregation of the chiral lithium amide in THF and the influence on
the aggregation by the addition of additives, such as 1,8-diazabicyclo-[5.4.0]undec-7-ene (DBU). The acti-
vated complex under catalytic deprotonation of cyclohexene oxide, that is, with excess Li-DBU and free
DBU, is built from one monomer of the chiral lithium amide, one molecule of epoxide and one additional
molecule of DBU. The reaction order (ꢀ0.97) obtained for the bulk base Li-DBU shows an inverse depen-
dence on the concentration, suggesting a deaggregation of the initial mixed dimer to a monomer-based
transition state containing a monomer of the lithium amide.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
nyl)methyl-2-azabicyclo-[2.2.1]-heptane 1 in the presence of LDA
as the bulk base and DBU as an additive (Scheme 1).7,10,19,20 In or-
Stereoselective deprotonations by chiral lithium amides of
epoxides, ketones, and so on, have found increasing use in syn-
thetic chemistry, for example, to make biologically interesting
compounds.1 Since the report by Whitesell and Fellman2 on the
first asymmetric deprotonation of cyclohexene oxide yielding
2-cyclohexen-1-ol, significant progress has been made, a large
number of structurally diverse chiral lithium amides have been
synthesized and their stereoselective capabilities evaluated.3–13
In the deprotonation reaction, the chiral lithium amide is con-
sumed and the corresponding chiral amine is formed.14–17 There-
fore, it is desirable to be able to regenerate the chiral base during
the reaction, and thereby use catalytic amounts of the chiral lith-
ium amide. Lithium diisopropylamide (LDA) has been used as a
bulk base to regenerate the chiral lithium amide from the chiral
amine. However, this procedure resulted in lower stereoselectivi-
ties compared to deprotonation under stoichiometric conditions.
It has been assumed that this is due to competing (racemic) reac-
tion by LDA. However, improvements in both stereoselectivity and
reactivity have been achieved by the addition of 1,8-diazabicyclo-
[5.4.0]undec-7-ene (DBU) to the catalytic mixtures as first reported
by Asami.18
der to explain the change in the reactivity and enantioselectivity,
DBU was suggested to solvate the lithium amide and hence decom-
pose lithium amide aggregates into smaller aggregates.19–21
N
N
O
OH
Li
Li-1 (5 mol %)
LDA (1.5eq), DBU (10eq)
THF, 0 °C
2
(S)-3
97% ee
Scheme 1. Catalytic enantioselective deprotonation of cyclohexene oxide.
Ahlberg et al. showed that LDA also deprotonates DBU yielding
lithiated DBU (Li-DBU) under these conditions. An equilibrium is
set up between LDA and DBU on one side and Li-DBU and diisopro-
pylamine (DIPA) on the other.15 It was found that Li-DBU together
with the chiral lithium amide Li-5 forms a new complex, namely a
mixed dimer Li-5ꢁLi-DBU (Scheme 2).
Andersson et al. reported excellent stereoselectivity and reac-
tivity in the deprotonation of epoxides with the chiral lithium
amide Li-1 derived from diamine (1R,3S,4S)-3-(1-pyrrolidi-
Herein we report a detailed NMR spectroscopic investigation of
the initial states and a kinetic study of the reaction orders of the
system. The study shows the nature of the reagents and the com-
position of the rate limiting activated complexes, which show the
intricate role of DBU and Li-DBU. The results are interpreted in
the light of findings with some previously used bulk bases.
⇑
Tel.: +46 18 471 3755; fax: +46 18 471 3818.
E-mail address: peter.diner@biorg.uu.se
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.10.034

2734
P. Dinér / Tetrahedron: Asymmetry 21 (2010) 2733–2739
N
N
NLi
Li-5
N
Li
Li
LDA
DIPA
+
+
N
DBU
Li-DBU
N
Li-DBU•Li-5
Scheme 2. LDA deprotonates DBU and forms mixed dimers with the chiral lithium
amide 5.
2. Results and discussion
2.1. NMR-spectroscopic study
Organolithium chemistry is a complex chemistry since lithium
organic compounds aggregate in solution and interconvert be-
tween different types of aggregates. Therefore, studies of these
aggregates of lithium amides in solution, and how solvents and
additives influence these aggregates, become essential. The com-
position of the reagents in a solution of Li-1 in deuterated THF at
ꢀ90 °C has been investigated using 6Li, 1H and 13C NMR spectros-
copy (Scheme 3). The 6Li-labelling of reagents was carried out
using n-Bu-[6Li] as the lithiating agent.
N
N
N
Li
N
H
1
Li-1
N
N
N
N
Li
DBU
Li
Li-DBU
N
N
Li
N
Figure 1. 6Li NMR spectra obtained at ꢀ90 °C of a THF-d8 solution of 0.3 M Li-1 and
added DBU and Bu[6Li]: (a) 0.3 M n-Bu[6Li] and 1.0 equiv 1; (b) 1.0 equiv DBU; (c)
1.0 equiv n-Bu[6Li]; (d) 1.0 equiv DBU.
N
N
Li
Li
N
N
N
improve the reactivity and enantioselectivity of lithium amides.
It has been assumed that additives, such as DBU, deaggregate the
lithium amides, and hence increase the reactivity of the lithium
amide. In order to investigate the possible solvation effects on
(Li-1)2, DBU (1 equiv, 0.2 mmol) was added to a solution of
(Li-1)2. The addition caused only minor changes in the chemical
shifts of the signals (Fig. 1b). However, the intensity of the minor
signals (@ 1.73, @ 1.47) had increased slightly relative to the major
singlet (@ 1.54, @ 0.84), which is probably due to the DBU solvation.
However, the 6Li spectrum showed no new signals originating from
aggregates other than (Li-1)2. Furthermore, the 6Li or 13C NMR
spectra showed no evidence of the presence of lithiated DBU,
which shows that Li-1 is not a strong enough base to deprotonate
DBU. The addition of a second equivalent of n-Bu[6Li] (1 equiv,
0.2 mmol) to the DBU containing mixture gave rise to three new
signals in the 6Li spectrum at @ 0.18, @ 0.42 and @ 1.15 (the latter
two of equal intensity) in addition to the signals from (Li-1)2. Com-
parison with earlier results obtained by Ahlberg et al. indicates that
(Li-1)2
Li-1• Li-DBU
Scheme 3. Structures of precursors and their lithiated counterparts and dimeric
complexes.
2.1.1. Deprotonation of 1 by n-BuLi
The addition of diamine 1 (1 equiv, 0.2 mmol) to a solution of n-
Bu[6Li] (0.2 mmol, 0.3 M) in THF-d8 at ꢀ90 °C in a NMR tube (see
Section 4) gave a clear solution. The 6Li NMR spectrum showed
two major signals (@ 1.54, @ 0.78) in a 1:1 ratio and also two minor
signals (@ 1.72 and @ 1.44) with about equal intensity (Fig. 1a). The
ratio between the pairs of signals was 6:1. The pair of signals with
the higher intensity is suggested to come from a dimer (Li-1)2 with
two non-equivalent lithiums. This result is in line with previous re-
ports that lithium amides exist as dimers with equivalent or non-
equivalent lithiums.9,22 Additives such as DBU have been used to

P. Dinér / Tetrahedron: Asymmetry 21 (2010) 2733–2739
2735
the signal at @ 0.18 corresponds to lithiated DBU (Li-DBU).15 This is
further supported by the appearance of one signal at @ 164.4 in the
13C NMR spectrum. Previously, it has been shown that the carbon
at the 7-position in DBU shifts from @ 160.1 to @ 164.4 upon depro-
tonation to yield Li-DBU. The other two new signals (@ 0.42 and
@ 1.15) are suggested to come from a novel complex—a mixed
dimer (Li-1ꢁLi-DBU)—built from a monomer of Li-1 and a monomer
of Li-DBU (Scheme 4).
N
N
N
N
N
N
N
Li
Li
Li
N
Li
Li
Li
N
N
N
N
(Li-1)2
Li-DBU
Li-1• Li-DBU
Scheme 4. (Li-1)2 and (Li-DBU)2 is in equilibrium with the mixed dimer Li-1ꢁLi-
DBU.
The two lithiums in these mixed dimers are non-equivalent. This
is in analogy with the mixed dimer that has been shown to form be-
tween the Li-5 and Li-DBU (Scheme 2). This conclusion is further
supported by the 13C NMR data. A closer inspection of the 13C NMR
spectrum of the mixture reveals that there is also a signal slightly
downfield (@ 164.9) of the signal belonging to lithiated DBU (@
164.4). It has previously been found that the amidine carbon upon
mixed dimer formation is shifted slightly downfield (@ 165.4).15
It is noteworthy that the 6Li NMR spectrum (Fig. 1c) shows that
not all Li-DBU (presumably present as a dimer) and (Li-1)2 have re-
acted with each other to yield the mixed dimer Li-1ꢁLi-DBU. In-
stead an equilibrium has been established in which there is
about equimolar amounts of (Li-1)2 and Li-1ꢁLi-DBU and about half
as much of (Li-DBU)2 in the solution. Further addition of DBU
(1 equiv, 0.2 mmol) gave only minor changes of the 6Li NMR chem-
ical shifts of the signals. Further addition of DBU (3 equiv,
0.6 mmol; in total 4 equiv) resulted in formation of needle-like
crystals in the brownish solution. A X-ray crystallographic study
revealed that these were composed of Li-DBU dimers [(Li-DBU)2]
with each of the 2 equiv lithium coordinated to a DBU molecule
(Fig. 2, unpublished data).
Figure 2. X-ray structure of crystals formed in the NMR-tube at ꢀ90 °C in THF
(unpublished data).
and @ 164.6, corresponding to Li-DBU and Li-1ꢁLi-DBU, respec-
tively. The addition of DBU (1 equiv, 0.2 mmol) to this mixture only
caused minor changes as shown by the spectra in Figure 3e (or
Fig. 1d). The addition of another equivalent of n-Bu[6Li] (1 equiv,
0.2 mmol) generated another equivalent of Li-DBU and this
(Fig. 3f) changes the composition of the equilibrium favoring the
mixed dimer Li-1ꢁLi-DBU over the dimer with two non-equivalent
lithium (Li-1)2.
Increasing the concentration of Li-DBU shifts the equilibrium to-
wards the mixed dimer. Under the catalytic conditions, the lithiated
DBU concentration is 40 times higher than the concentration of
Li-1, which means that the equilibrium is completely shifted to-
wards the mixed dimer. With these experimental results as a basis,
we turned our attention to an investigation of the reagent compo-
sition when using LDA as a deprotonating agent rather than of
n-Bu[6Li].
2.1.3. Deprotonation of 1 using LDA
The addition of n-Bu[6Li] (1 equiv, 0.2 mmol) to diisopropyl-
amine (0.3 M, 0.2 mmol) in THF-d8, gave a clear solution. The 6Li
NMR spectrum at ꢀ90 °C (Fig. 4a) showed one major singlet
(@ 1.91) originating from LDA (a dimer in THF23).
2.1.2. Deprotonation of 1 using Li-DBU
The 6Li-NMR spectrum at ꢀ90 °C of a solution of Li-DBU (0.3 M,
0.2 mmol) in THF displays only a singlet as previously reported
(Fig. 3a).15 The addition of diamine 1 (0.5 equiv, 0.1 mmol) gave a
yellow clear solution with the spectrum shown in Figure 3b. The
chemical shifts and relative intensities were similar to those in
the spectrum in Figure 1c of a solution containing (Li-DBU)2,
(Li-1)2 and mixed dimer Li-1ꢁLi-DBU. It appears that (Li-DBU)2 is
a strong enough base to deprotonate the chiral diamine. The fur-
ther addition of chiral diamine (0.5 equiv, 0.1 mmol) gave the spec-
trum shown in Figure 3c, which is closely similar to the one
displayed in Figure 1b. This shows that all of the added Li-DBU
had been consumed in the deprotonation of the diamine. The spec-
trum in Figure 3d shows that Li-DBU is regenerated by the addition
of n-Bu[6Li] (1 equiv, 0.2 mmol) yielding an equilibrium mixture of
the dimers (Li-1)2 and (Li-DBU)2, and the mixed dimer Li-1ꢁLi-DBU
with about the same composition as that of the solution in
Figure 1c. The formation of mixed dimer Li-1ꢁLi-DBU was con-
firmed by the 13C NMR spectrum showing two signals at @ 164.4
The addition of chiral diamine 1 (0.5 equiv, 0.1 mmol) to the
LDA-solution, gave the 6Li-spectrum shown in Figure 4b. It shows
three major peaks, one originating from LDA (@ 1.92) and two
new signals of equal intensity at @ 1.54 and @ 0.77 corresponding
to (Li-1)2. Obviously, LDA is a strong enough base to deprotonate
the diamine 1. The spectrum gives no indication of the formation
of any mixed dimer between LDA and Li-1.
The further addition of diamine (0.5 equiv, 0.1 mmol) was
found to consume the remaining LDA as indicated by Figure 4c.
The LDA signal disappeared and the remaining signals originate
from the lithiated diamine (Li-1)2. This shows that LDA is capable
of fully deprotonating 1 and (Li-1)2 is strongly favored in the equi-
librium. After the addition of another equivalent of n-Bu[6Li]
(1 equiv, 0.2 mmol), LDA was regenerated as shown by the reap-
pearance of the singlet at @ 1.92 in the 6Li spectrum (Fig. 4d). Addi-
tion of DBU (1 equiv, 0.2 mmol) to this solution gave the spectrum

2736
P. Dinér / Tetrahedron: Asymmetry 21 (2010) 2733–2739
Figure 4. 6Li NMR spectra obtained at ꢀ90 °C of a THF-d8 solution of 0.3 M LDA and
added 1, n-Bu[6Li] and DBU: (a) 0.3 M LDA; (b) 0.5 equiv 1; (c) 1.0 equiv 1; (d)
1 equiv n-Bu[6Li]; (e) 1 equiv DBU; (f) 2 equiv DBU; 3 equiv DBU.
Figure 3. 6Li NMR spectra obtained at ꢀ90 °C of a THF-d8 solution of 0.3 M Li-DBU
and added 1 and n-Bu[6Li]: (a) 0.3 M Li-DBU; (b) 0.5 equiv 1; (c) 1.0 equiv 1; (d)
1.0 equiv n-Bu[6Li]; (e) 1.0 equiv DBU; (f) 1.0 equiv n-Bu[6Li].

P. Dinér / Tetrahedron: Asymmetry 21 (2010) 2733–2739
2737
shown in Figure 4e. Apparently, LDA has deprotonated DBU to
yield lithiated DBU. In this mixture the molar ratio between LDA
and (Li-DBU)2 is about 1:1. The signal from lithiated DBU (@ 0.72)
shifts downfield indicating a solvation by DIPA, which is in accor-
dance with previous observations.15 The two small peaks at @ 0.97
and @ 0.47 indicate the presence of the mixed dimer Li-1ꢁLi-DBU.
Further addition of DBU (1.3 equiv, 0.26 mmol) led to the forma-
tion of more lithiated DBU and the accompanying formation of
DIPA. This resulted in a shift of the Li-DBU signal to lower field
(@ 1.15). The two mixed dimer signals at @ 1.11 and @ 0.44 in-
creased in the 6Li spectrum (Fig. 4f). Addition of more DBU
(1 equiv, 0.2 mmol) shifts the Li-DBU signal even more downfield
(@ 1.32) (Fig. 3g). The 13C NMR spectrum of the solution shows
two signals at @ 164.9 and @ 165.3 consistent with the presence
of lithiated DBU and the mixed dimer Li-1ꢁLi-DBU, respectively.
Under the catalytic deprotonations, the DBU concentration is typ-
ically 5–7 times higher than the LDA concentration, which suggests
lithiated DBU is the major bulk base in solution.
DBU as an additive in THF at 0 °C (Table 1). In these initial rate
measurements, the concentrations of Li-1, epoxide 2, DBU and
Li-DBU have been varied independently in order to acquire infor-
mation about how the initial rate is influenced by a change in con-
centration of each component in the catalytic system. The
composition of the standard catalytic deprotonation system is
Li-1 (0.0050 M, 5 mol % of epoxide concentration) and a large ex-
cess of Li-DBU (0.20 M, 40 times Li-1) and DBU (0.805 M). Under
these conditions, the chiral lithium amide is present as a mixed di-
mer, Li-1ꢁLi-DBU, which is the reference state that relates to the
reaction orders obtained. The logarithms of the initial rates for for-
mation of 3 were plotted against the logarithms of the concentra-
tion for each component of the catalytic system.
Li-1 ꢁ Li-DBU þ 2 þ DBUꢀ½Li-1 ꢁ 2 ꢁ DBUꢂz þ Li-DBU ! 3
ð1Þ
ð2Þ
½Li-1 ꢁ 2 ꢁ DBUꢂz½Li-DBUꢂ
Kz ¼
½Li-1 ꢁ Li-DBUꢂ½2ꢂ½DBUꢂ
½Li-1 ꢁ Li-DBUꢂ½2ꢂ½DBUꢂ
½Li-1 ꢁ 2 ꢁ DBUꢂz ¼ Kz
ð3Þ
ð4Þ
½Li-DBUꢂ
2.1.4. The addition of DIPA to (Li-1)2
1
1
1
d½3ꢂ=dt ¼ kobs½Li-1 ꢁ Li-DBUꢂ ½2ꢂ ½DBUꢂ ½Li-DBUꢂꢀ1
The addition of diamine 1 (1 equiv, 0.2 mmol) to a solution of
n-Bu[6Li] (0.2 mmol, 0.3 M) in THF-d8 at ꢀ90 °C yielded a clear, yel-
low solution. The 6Li spectrum showed the same features as found
previously (spectra not shown). The addition of DIPA (0.1 equiv,
0.2 mmol) did not lead to any observable changes in the 6Li spec-
trum, showing that (Li-1)2 is not a strong enough base to deproto-
nate diisopropylamine.
The reaction orders for the lithium amide Li-1 and epoxide 2
were determined to 0.94 and 1.04, respectively (Figs. 5 and 6). In
addition, the reaction order of free DBU was determined to 0.98
(Fig. 7). This suggests that the composition of the activated com-
plex would be built from a mixed dimer Li-1ꢁLi-DBU and one mol-
ecule of cyclohexene oxide 2 with one additional molecule of DBU
solvating to lithium in the activated complex.
However, upon variation of the concentration of the bulk base
Li-DBU, it was found that the initial rate of the reaction increased
at lower concentrations (and decreased at higher concentrations)
of Li-DBU, relating to a reaction order of ꢀ0.97 (Fig. 8). This sug-
gests that the activated complex is not composed of a mixed dimer,
Addition of n-Bu[6Li] (1 equiv, 0.2 mmol) deprotonated the
formed diisopropylamine yielding a 6Li spectrum with mainly sig-
nals from LDA (@ 1.60) and (Li-1)2 (@ 1.20 and @ 0.40). Addition of
another equivalent of DIPA (0.1 equiv, 0.2 mmol) showed no signif-
icant changes in the 6Li spectrum. Finally, the 6Li spectra after addi-
tion in portions of DBU (0.5 equiv, 1 equiv, 2 equiv, 3 equiv) show
the formation of Li-DBU and mixed dimer at the expense of LDA.
2.2. Kinetic study
In order to obtain information about the composition of the rate
limiting activated complexes involved in the stoichiometric and
the catalytic deprotonation of cyclohexene oxide, the reaction or-
ders of the deprotonation reaction were determined via measure-
ment of the initial rates at different concentrations of each
reactants. The deprotonations of 2 yielding 3 were carried out in
THF at 0.0 0.1 °C.
Table 1
Initial rates obtained from runs with different concentrations of Li-1, 2, DBU and Li-
DBU in THF at 0.0 °C yielding 2-cyclohexen-1-ol 3
N
N
Li
The initial rates for the formation of 3 were determined with a
quench-extraction-GC procedure and the conversions of the reac-
tion were typically lower by a few percent. The logarithm of the
initial rates was plotted against the logarithm of the concentra-
tions of the varying component and the reaction orders were ob-
tained from the gradient. It should be remembered that the
reaction orders obtained are related to the change in composition
on going from the reactants to the activated complex, and there-
fore must be related to the composition of the initial state. Knowl-
edge of the major reagent species does not imply that the
composition of the activated complex is also known. However,
kinetics may yield reaction orders with respect to the reagents,
and these inform us about the composition of the activated com-
plexes, as long as the composition of the reagent is known. Several
previous studies have shown the complexity of the composition of
the activated complexes related to the organolithium compounds
in the solution.24–26
Li-1 (2.5-10 mol%)
DBU (5-10eq)
O
OH
THF, 0 °C
2
3
Li-1 (M)
2 (M)
DBU (M)
Li-DBU (M)
Initial rates (107 M)
0.0025
0.0025
0.0050
0.0050
0.0050
0.0050
0.010
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.050
0.050
0.20
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.805
0.805
0.805
0.805
0.805
0.805
0.805
0.805
0.805
0.805
0.805
1.14
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.10
0.10
0.15
0.30
13.6
12.1
23.5
23.4
24.0
23.3
49.6
45.7
10.5
11.7
46.0
34.0
35.1
51.1
51.5
46.0
37.7
18.2
0.010
0.0050
0.0050
0.0050
0.0050
0.0050
0.0050
0.0050
0.0050
0.0050
0.0050
1.14
1.815
0.805
0.805
0.805
0.805
2.2.1. Catalytic deprotonation of 2 with Li-1, excess Li-DBU and
DBU in THF
The initial reaction rates have been determined for the deproto-
nation of cyclohexene oxide 2 with Li-DBU as a bulk base and free

2738
P. Dinér / Tetrahedron: Asymmetry 21 (2010) 2733–2739
Figure 5. Plot of log[d[3]/dt] versus log[Li-1] in THF for the catalytic deprotonation
of cyclohexene oxide (2, 0.10 M) in the presence of Li-DBU (0.20 M) and free DBU
(0.805 M) at different concentrations of Li-1 at 0 °C. The reaction order was
obtained as the slope (0.94) of the line.
Figure 8. Plot of log[d[3]/dt] versus log[Li-DBU] in THF for the catalytic deproto-
nation of cyclohexene oxide (2, 0.10 M) with Li-1 (0.0050 M) in the presence free
DBU (0.805 M) at different concentrations of Li-DBU at 0 °C. The reaction order was
obtained as the slope (ꢀ0.97) of the line.
complex of the mixed dimer. This is in contrast to DBU which deag-
gregates the dimer into monomers of the initial state in the
solution.
3. Conclusion
The solution structure of Li-1 in THF was determined to be
mainly composed of a dimer (Li-1)2 with two non-equivalent lith-
iums. It has also been shown that the addition of DBU does not
cause any major changes to the composition of the lithium amide
dimer (Li-1)2, that is, the lithium amide does not deaggregate into a
monomeric structure or deprotonate DBU to form a mixed dimer.
Bulk bases such as LDA have previously been shown to deproto-
nate DBU, and thus form lithiated DBU. Li-DBU deprotonates 1 and
forms Li-1. Li-1 in the presence of Li-DBU forms a equilibrium of
the dimeric lithium amide (see above, (Li-1)2) Li-DBU, and the
mixed dimer Li-1ꢁLi-DBU. Under catalytic conditions, that is,
0.005 M Li-1 and 0.2 M Li-DBU, the mixed dimer Li-1ꢁLi-DBU is
the main lithiated species together with Li-DBU. The composition
of the activated complex for the catalytic deprotonation of cyclo-
hexene oxide with Li-1 in THF, excess Li-DBU and free DBU is built
from a monomer of Li-1, one molecule of cyclohexene oxide, and
one additional molecule of DBU. This shows that the mixed dimer
deaggregates from Li-1ꢁLi-DBU to the monomer Li-1 solvated by
DBU on going to the transition state.
Figure 6. Plot of log[d[3]/dt] versus log[2] in THF for the catalytic deprotonation of
cyclohexene oxide at different concentrations with Li-1 (0.0050 M) in the presence
of Li-DBU (0.20 M) and free DBU (0.805 M) at 0 °C. The reaction order was obtained
as the slope (1.04) of the line.
Previous interpretations have suggested that additives of DBU
deaggregate the lithium amide dimer into monomers in solution,
and this would be the reason for the acceleration of the deprotona-
tion reaction upon addition of DBU. Herein, it has been shown that
the chiral lithium amide Li-1 in the presence of excess Li-DBU
forms a mixed dimer Li-1ꢁLi-DBU in solution. However, in the acti-
vated complex for the deprotonation of 2, Li-1ꢁLi-DBU decomposes
to an activated complex composed of Li-1ꢁ2ꢁDBU.
4. Experimental
4.1. General
Figure 7. Plot of log[d[3]/dt] versus log[DBU] in THF for the catalytic deprotonation
of cyclohexene oxide (2, 0.10 M) with Li-1 (0.0050 M) in the presence of Li-DBU
(0.20 M) at different concentrations of free DBU at 0 °C. The reaction order was
obtained as the slope (0.98) of the line.
All of the syringes and NMR tubes used were dried overnight in
an oven. All handling of the compounds were carried out with gas-
tight syringes. THF-d8 was stored over molecular sieves (4 Å) in the
glove box. DBU was purified by distillation from CaH2 over nitro-
gen and diisopropylamine was distilled from NaOH over nitrogen.
n-Bu[6Li] were prepared as previously described.27 The chiral dia-
mine 1 was synthesized as previously described and all spectro-
but rather of a monomer of the lithium amide Li-1, the epoxide 2
and one additional molecule of DBU (Eq. 1–4). The reason for the
observed rate acceleration, with DBU as an additive/bulk base,
can be explained by DBU coordinating into the monomeric acti-
vated complex, which is stabilized compared to the activated

P. Dinér / Tetrahedron: Asymmetry 21 (2010) 2733–2739
2739
scopic data were in accordance with that previously
published.7,19,28
DBWX-30W column (30 m, 0.25
l
m) from Varian Inc. was used
with hydrogen as the carrier gas (2 ml minꢀ1). Reaction samples
(1.0 ll) were introduced onto the column via a split injector (split
4.2. NMR spectroscopy
flow 15 ml minꢀ1) and the components were separated using a
temperature programme. Initially the temperature was held at
80 °C for 2 min and then during 2 min it was increased to 120 °C.
The injector temperature was 225 °C and the detector was held
at 250 °C. Gas chromatography response factors for 2 and 3 were
determined, using 1-hexanol as a reference, to 1.01 and 0.85,
respectively.
All NMR experiments were performed in Wilmad tubes (5 mm)
fitted with a Wilmad/Omnifit Teflon valve assembly (OFV) and a
Teflon/Silicon septum. All NMR spectra were recorded with a Var-
ian Unity 500 spectrometer equipped with a 5-mm triple-reso-
nance probe head, built by Nalorac. Measuring frequencies were
499.9 MHz (1H), 125.7 MHz (13C) and 73.57 MHz (6Li). The 1H
and 13C spectra were referenced to signals from residual protons
at C2 (d 1.73) and from C2 carbon (d 67.57), respectively, in the sol-
vent THF-d8. Lithium resonances were referenced to external [6Li]
in a 0.3 M [6Li]Cl in MeOH-d4 (d 0.0) in a separate NMR tube. The
probe temperature was measured using a calibrated methanol
thermometer from Varian Inc.
Acknowledgement
P.D. thanks Vetenskapsrådet (621-2009-4018) for financial
support.
References
4.3. Typical NMR experiment
1. Asami, M.; Suga, T.; Honda, K.; Inoue, S. Tetrahedron Lett. 1997, 37, 6425–6428.
2. Whitesell, J. K.; Felman, S. W. J. Org. Chem. 1980, 45, 755–756.
3. Asami, M. Chem. Lett. 1984, 829–832.
4. Bhuniya, D.; DattaGupta, A.; Singh, V. K. J. Org. Chem. 1996, 61, 6108–6113.
5. Asami, M.; Suga, T.; Honda, K.; Inoue, S. Tetrahedron Lett. 1997, 38, 6425–6428.
6. Tierney, J. P.; Alexakis, A.; Mangeney, P. Tetrahedron: Asymmetry 1997, 8, 1019–
1022.
Diamine 1 (38
taining THF-d8 (650
n-Bu[6Li]. DBU (30
l
l, 0.2 mmol) was added to an NMR tube con-
l). [6Li]-1 was prepared by titration of 1 with
l, 0.2 mmol, 1 equiv) was added and the solu-
l
l
tion was allowed to equilibrate for 15 min before spectra were re-
7. Södergren, M. J.; Andersson, P. G. J. Am. Chem. Soc. 1998, 120, 10760–10761.
8. Colman, B.; de Sousa, S. E.; O’Brien, P.; Towers, T. D.; Watson, W. Tetrahedron:
Asymmetry 1999, 10, 4175–4182.
9. Pettersen, D.; Amedjkouh, M.; Nilsson Lill, S. O.; Dahlén, K.; Ahlberg, P. J. Chem.
Soc., Perkin Trans. 2 2001, 1654–1661.
corded at ꢀ90 °C.
4.4. Typical kinetic procedure
10. Bertilsson, S. K.; Södergren, M. J.; Andersson, P. G. J. Org. Chem. 2002, 67, 1567–
1573.
11. Cox, P. J.; Simpkins, N. S. Tetrahedron: Asymmetry 1991, 2, 1–26.
12. O’Brien, P. J. Chem. Soc., Perkin Trans. 1 1998, 1439–1458.
13. Eames, J. Eur. J. Org. Chem. 2002, 393–401.
14. Pettersen, D.; Amedjkouh, M.; Ahlberg, P. Tetrahedron 2002, 58, 4669–4673.
15. Pettersen, D.; Amedjkouh, M.; Nilsson Lill, S. O.; Ahlberg, P. J. Chem. Soc., Perkin
Trans. 2 2002, 1397–1405.
16. Amedjkouh, M.; Pettersen, D.; Nilsson Lill, S. O.; Davidsson, Ö.; Ahlberg, P.
Chem.-A Eur. J. 2001, 7, 4368–4377.
17. Nilsson Lill, S. O.; Pettersen, D.; Amedjkouh, M.; Ahlberg, P. J. Chem. Soc., Perkin
Trans. 1 2001, 3054–3063.
18. Asami, M.; Ishizaki, T.; Inoue, S. Tetrahedron: Asymmetry 1994, 5, 793–796.
19. Södergren, M. J.; Bertilsson, S. K.; Andersson, P. G. J. Am. Chem. Soc. 2000, 122,
6610–6618.
20. Bertilsson, S. K.; Andersson, P. G. Tetrahedron 2002, 58, 4665–4668.
21. Brandt, P.; Norrby, P.-O.; Andersson, P. G. Tetrahedron 2003, 59, 9695–9700.
22. Arvidsson, P. I.; Davidsson, Ö. Angew. Chem., Int. Ed. 2000, 39, 1467–1469.
23. Williard, P. G.; Salvino, J. M. J. Org. Chem. 1993, 58, 1–3.
24. Olsson, R. I.; Ahlberg, P. Tetrahedron: Asymmetry 1999, 10, 3991–3998.
25. Wiedemann, S. H.; Ramirez, A.; Collum, D. B. J. Am. Chem. Soc. 2003, 125,
15893–15901.
26. Ramírez, A.; Collum, D. B. J. Am. Chem. Soc. 1999, 121, 11114–11121.
27. Hilmersson, G.; Davidsson, Ö. Organometallics 1995, 14, 912–918.
28. Modin, S. A.; Andersson, P. G. J. Org. Chem. 2000, 65, 6736–6738.
Amine (1R,3S,4S)-3-(1-pyrrolidinyl)methyl-2-azabicyclo-[2.2.1]-
heptane 1 (2.5
THF, was added in a reaction vessel under nitrogen, followed by
DBU (150 l, 1.0 mmol). Next, THF (717 l) was added, followed
by the addition of n-butyllithium (80 l, 0.2 mmol, 2.48 M in hex-
ll, 2.0 M 0.0050 mmol), from a stock solution in
l
l
l
anes) under a nitrogen atmosphere. The yellow reaction solution
was allowed to equilibrate at 0.00 0.05 °C for 15 min in a thermo-
stat. The reaction was started by the addition of cyclohexene oxide
2 (50
drawn from the reaction vessel at approximately 1–2 min intervals
and quenched in hydrochloric acid solution (100 l, 0.6 M satu-
rated with sodium chloride). Compounds 2 and 3 were extracted
with carbon tetrachloride (500 l) containing the standard 1-hex-
anol (3.08 mM) and were then placed on ice. The liquid phases
were separated by centrifugation and 250 l of the organic phase
ll, 2.0 M in THF, 0.1 mmol) and samples (50 ll) were with-
l
l
l
was transferred to a vial and analyzed by capillary gas chromatog-
raphy. Gas chromatography analyses were performed on a Varian
3400 chromatograph equipped with an 8200 Cx auto sampler
and a flame ionisation detector (FID). For the separation an achiral