Chiral bases as useful probes of lithium amide reactivity

Article abstract of DOI:10.1002/anie.201206558

Experiments involving competition between chiral and achiral lithium amides provide a qualitative impression of the kinetics of different amide motifs in enolization reactions. The level of enantioselectivity was not diminished by the inclusion of lithium

Full text of DOI:10.1002/anie.201206558

.
Angewandte
Communications
DOI: 10.1002/anie.201206558
Kinetic Probes
Chiral Bases as Useful Probes of Lithium Amide Reactivity**
Mark R. Prestly and Nigel S. Simpkins*
Lithium amides are indispensable as strong bases in organic
synthesis, particularly for the generation of metal enolates
[1]
=
from compounds that possess C O functions of all types.
These reagents exhibit complex structural behavior, partic-
ularly in solution, in which the formation of aggregates and
mixed aggregates is commonplace, leading to a mechanisti-
cally rich repertoire of enolization and metalation chemis-
try.[2]
Our own interest in this area has focused on the use of
chiral lithium amides in asymmetric lithiation processes.[3]
One key finding in this regard is the exceptional levels of
enantioselectivity that are frequently provided by lithium
amide 1 and bislithium amide 2 (both derived from diamine
3), in diverse asymmetric metalation reactions (Scheme 1,
Eq. (1)–(3)).[4]
In the asymmetric alkylation of piperidine diester 4, the
use of lithium bisamide 2 proved essential for attaining both
optimal yield and asymmetric induction.[5] The explanation
for the much cleaner reactions observed using this base
compared to LDA and simpler chiral lithium amides was not
clear. In the case of systems that are prone to overreaction,
such as ring-fused imide 6, we found the use of the diamine
motif as a monolithium amide preferable.[6] In reactions of
certain chromium arenes, for example, 8, the diamine motif
again provided excellent results.[7] In their studies in this area,
Gibson and co-workers found identical ee values with 1 and 2
using a benzyl methyl ether complex, and speculated that the
bislithium amide could be the more reactive base. Thus,
reactions involving diamine 3 with one equivalent or less of
BuLi might still involve the more reactive bislithiated base 2,
formed by lithium exchange of 1.[8]
It occurred to us that experiments involving in situ
competition between a chiral and an achiral base would
directly provide the relative reaction rates for the two bases
through the ee values of the formed products. This concept
appeared attractive to us, because, despite extensive mecha-
nistic and structural research on lithium amides, it is difficult
to extract meaningful comparisons of rates of (typical
Scheme 1. Enantioselective deprotonation reactions using diamine-
derived bases 1 and 2.
substrate) deprotonation by different bases under the usual
types of conditions used for synthesis.[9]
Previously, OꢀBrien and co-workers used competition
experiments to assess the relative reactivity of sBuLi/diamine
complexes, enabling them to identify ligand combinations
suitable for catalytic deprotonation of N-Boc pyrrolidine.[10]
Herein we describe our own preliminary competition
results in this area, which provide some new insights into
lithium amide reactivity, and enable some clarification of the
apparent superior reactivity of diamine-derived bases. The
competition results can also be extended to provide data on
lithium exchange between different amines (see below).
Bearing in mind the ubiquitous nature of ketone enoliza-
tions, we chose 4-tert-butyl-cyclohexanone (10) as our test
[*] M. R. Prestly, Prof. N. S. Simpkins
School of Chemistry, University of Birmingham
Edgbaston, Birmingham, B15 2TT (UK)
E-mail: N.simpkins@bham.ac.uk
system to use with both chiral and achiral lithium amides.[11]
selection of lithium amide bases was tested for the formation
of enol silane 11 (Table 1).
A
[**] We acknowledge the University of Birmingham for support of M.R.P.
The NMR and GC instruments used in this research were obtained
through Birmingham Science City: Innovative Uses for Advanced
Materials in the Modern World (West Midlands Centre for Advanced
Materials Project 2), with support from Advantage West Midlands
(AWM) and part funded by the European Regional Development
Fund (ERDF). We would also like to thank Graham Burns and Chi
Tsang of the School of Chemistry for assistance with HPLC and GC.
All enolizations were carried out in the presence of LiCl,
which has been demonstrated to improve the level of
asymmetric induction in enolizations with chiral bases.[12]
Base 13 (Table 1, entry 2) was included as an achiral analogue
of 1 (entry 3), and gave reproducibly modest chemical yields,
even if reaction times were extended.[13] The simple chiral
bases 14, which was used by us in many applications, and 16,
one of a range of fluorinated bases developed by Koga and co-
workers,[14] gave very similar results, the sense of induction
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201206558.
12068
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 12068 –12071

Angewandte
Chemie
Table 1: Benchmark enol silane synthesis.
system that “wins out” (Table 2, entry 4). Thus, despite the
fact that the simple base 14 leads to a significantly higher
enantioselectivity (81% (S)) than base 1 (58% (R)), the
overall selectivity emerging from the competition is 35% (R).
Finally, when chiral base 14 (81% ee) competes with the
achiral diamine-derived lithium amide 13, the asymmetric
induction is almost completely eroded (Table 2, entry 5).
Although the diamine-derived base dominates the outcome
of the reaction, the involvement of 13 again leads to modest
overall conversion.
Confirmation of the trends observed in enolization and
enol silane formation were sought in a dissimilar system. We
chose to conduct asymmetric carboxymethylation of imide 6
using Manderꢀs reagent (Table 3).
Entry[a]
Base
Conv. [%]
ee [%]
Configuration
1
2
3
4
5
LDA
13
1
14
16
99
39
99
99
99
–
–
58
81
78
–
–
(R)
(S)
(S)
[a] Both the conversion and the ee value of the product were determined
by GC analysis of the crude product, using a Chiralsil DEX-CB capillary
column. All values are averages from duplicate runs. LDA=lithium
diisopropylamide.
Table 3: Carboxymethylation of 6 using single or mixed base systems.
Entry
Base A
Base B
Yield[a] [%]
ee[b] [%]
being opposite to that observed with base 1 (Table 1, entries 4
and 5).
Competition experiments, in which ketone 10 was reacted
with a 1:1 mixture of two bases, were then run, again in the
presence of LiCl (Table 2).
1
2
3
LDA
1
LDA
–
–
1
43
93
89
–
97
99
[a] Yield of isolated product. [b] Determined by HPLC analysis using
a Chiralpak OD column.
Table 2: Competition experiments using lithium amide base mixtures.
For this system, we achieved low levels of conversion
using LDA as base (43%), and the yields obtained in the
absence of LiCl were even worse (around 10–20%). As
expected from our previous work,[3,4c,6a] including that sum-
marized in Eq. (2), base 1 provided excellent results, with
substituted imide (À)-17 being generated in high yield and
with exemplary levels of selectivity. As observed for the
ketone system, the enantioselectivity that was achieved in the
synthesis of (À)-17 using base 1 was not affected by the
presence of LDA.
The results that show diamine base 1 effectively compet-
ing with LDA for enolizable substrates provide impetus for
probing the possible use of base 1 (or analogues) as a catalytic
chiral base, in the presence of LDA or a similar lithium amide
as bulk recycling base. Although the chemistry of such
catalytic chiral lithium amide bases is reasonably well-
established for epoxide rearrangement to allylic alcohols,[15]
the corresponding systems for enolization, explored by Koga
and co-workers, are not very effective.[16]
In order for a base such as 1 to act as a catalyst, it must be
more reactive toward a substrate than the bulk recycling base,
and there must be an effective lithium transfer between the
two types of base. To probe this aspect, we devised experi-
ments in which the enolization event is preceded by competi-
tion between the secondary amines for lithium, that is,
a secondary amine and a lithium amide (derived from
a different amine) are premixed and allowed to “equilibrate”
before the addition of ketone 10 (Table 4).[17]
Entry
Base A
Base B
Conv. [%]
ee [%]/Configuration
1
2
3
4
5
LDA
LDA
LDA
14
1
98
99
99
99
56
57 (R)
36 (S)
50 (S)
35 (R)
11 (S)
14
16
1
14
13
The most striking result for this substrate is that reaction
of 10 with a 1:1 mixture of base 1 and LDA gives exactly the
same product outcome as reaction with base 1 alone (Table 1,
entry 1). Whatever the myriad possibilities for base aggre-
gates and mixed aggregates, it is clear that the presence of an
equimolar amount of LDA does nothing to diminish the
(albeit moderate) selectivity afforded by the diamine motif.
It can be estimated that both (monolithium) bases react
with the ketone at similar rates (Table 2, entry 2), with LDA
decreasing the ee value that was obtained with 14 alone from
81% (Table 1, entry 4) to only 36%. The same situation
prevails for base 16 (Table 2, entry 3). When two chiral bases
are in competition with each other, it is the diamine-derived
Angew. Chem. Int. Ed. 2012, 51, 12068 –12071
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12069

.
Angewandte
Communications
Table 4: Competition experiments with lithium equilibration.
Scheme 2. Possible mode of enolization by bislithium amide 2 (sub-
stituents omitted from second and third structures for clarity).
Entry
Base A
Base B
Conv. [%]
ee [%]/Configuration[a]
1
2
3
4
5
6
7
LDA
16
LDA
13
16
1
3
iPr2NH
15
15
12
77
81
98
86
37
46
25
6 (R)
69 (S)
83 (S)
56 (S)
61 (S)
rac.
be facilitated between two diamine molecules? The results
summarized in entries 6 and 7 of Table 4 suggest that this is
indeed the case, that is, in a mixed system of chiral and achiral
diamine components, lithium equilibration favors achiral
lithium amide 13, thus leading to racemic enol silane 11.
This intriguing result, although unexpected, opens up new
possibilities for catalysis and requires further study in other
systems.
12
3
13
rac.
[a] rac. signifies less than 2% ee.
Having established that base 1 outpaces LDA in the
enolization of 10 (Table 2), it can be seen that very little of
active 1 is being generated during the low-temperature pre-
equilibration of 3 and LDA (Table 4, entry 1). In other words,
LDA does not deprotonate 3. In contrast, lithium exchange
occurs in the reactions summarized in entries 2 and 3
(Table 4), with highly effective transfer of lithium from
LDA to fluorinated amine 15, generating lithium amide 16
(entry 3). In both cases, the level of asymmetric induction
approximates that achieved with 16 alone. Analogous lithium
transfer is also observed from achiral diamine-derived lithium
amide 13 to fluorinated base 15, although the level of
induction appears somewhat decreased (Table 4, entry 4).
The reciprocal base pair 12 and 16 shows a similar level of
induction (Table 4, entry 5).
In conclusion, we have demonstrated that competition
experiments involving mixtures of achiral and chiral (non-
racemic) lithium amides can provide insights into the kinetic
competency of these bases with two different types of
substrates (Tables 2 and 3). With knowledge of the relative
reaction rates of these experiments, pre-equilibration studies
can also provide information about lithium transfer between
different types of amines (Table 4). Our results show excep-
tional reactivity of lithium bases derived from 1,2-diamines
(especially 3), possibly because of the intermediacy of
a bislithium amide base. Further studies are underway, with
the aim of establishing “league tables” of lithium amide
reactivity with key types of substrates, and new and efficient
catalytic systems for reactions with chiral bases.
As previously established by Koga and co-workers,[16] the
acidifying influence of a b-trifluoroethyl substituent is highly
effective in promoting lithium transfer. While the diamine-
derived lithium amide 1 appears exceptional in some cases in
terms of reactivity and enantioselectivity, there is little lithium
transfer to diamine 3 using LDA as bulk base. Therefore, in
order to design an exceptional base with the potential as
a catalytic lithium amide, we probably need to incorporate
CF3 groups into a diamine motif such as 3.
At present, we can only speculate as to the reasons for the
high kinetic reactivity of diamine-derived base 1. In epoxide
rearrangements, lithium amides with an internally coordi-
nated tertiary amine appear more reactive than LDA.[15]
Amines and chelating ligands can also accelerate enolizations
under certain conditions.[2a,c] For our vicinal secondary amine
systems, we support the suggestion of Gibson et al. that
reactions involving 1 or 2 proceed via a reactive bislithium
amide structure.[8] The doubly bridged N2Li2 arrangement has
been observed for this type of structure,[18] which we would
expect to be highly activating toward a coordinated carbonyl
moiety (Scheme 2).[19,20]
Received: August 14, 2012
Published online: October 25, 2012
Keywords: basicity · enantioselectivity · enolization · lithiation ·
.
lithium amide
[1] a) D. B. Collum, A. J. McNeil, A. Ramirez, Angew. Chem. 2007,
119, 3060 – 3077; Angew. Chem. Int. Ed. 2007, 46, 3002 – 3017,
and references therein; b) J. Eames in Science of Synthesis,
Vol 8a (Ed.: V. Snieckus), Georg Thieme, New York, 2005,
pp. 173 – 241.
[2] a) P. F. Godenschwager, D. B. Collum, J. Am. Chem. Soc. 2007,
129, 12023 – 12031; b) A. Ramirez, X. Sun, D. B. Collum, J. Am.
Chem. Soc. 2006, 128, 10326 – 10336; c) P. Zhao, D. B. Collum, J.
Am. Chem. Soc. 2003, 125, 14411 – 14424; d) X. Sun, D. B.
Collum, J. Am. Chem. Soc. 2000, 122, 2459 – 2463.
[3] a) N. S. Simpkins, M. D. Weller, Top. Stereochem. 2010, 26, 1 –
52; b) P. OꢀBrien, J. Chem. Soc. Perkin Trans. 1 1998, 1439 – 1457;
c) P. J. Cox, N. S. Simpkins, Tetrahedron: Asymmetry 1991, 2, 1 –
26; d) N. S. Simpkins, Chem. Soc. Rev. 1990, 19, 335 – 354; e) N. S.
Simpkins, Chem. Ind. 1988, 387 – 389.
[4] a) Original diamine synthesis: K. Bambridge, M. J. Begley, N. S.
Simpkins, Tetrahedron Lett. 1994, 35, 3391 – 3394; b) initial use in
ketone enolization toward anatoxin A synthesis: N. J. New-
combe, N. S. Simpkins, J. Chem. Soc. Chem. Commun. 1995,
831 – 832; c) use in imide desymmetrization: V. Rodeschini, N. S.
Simpkins, F. Zhang, Org. Synth. 2007, 84, 306 – 316; d) use in
kinetic resolution by bridgehead lithiation: V. Rodeschini, N. S.
This proposal, like Gibsonꢀs observations, mentioned
above, requires “a ready exchange of lithium cations between
all the nitrogen sites in the system”. However, this appears
contrary to the lack of lithium exchange seen when diamine 3
is mixed with LDA (Table 4 entry 1). Could lithium exchange
12070 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 12068 –12071

Angewandte
Chemie
Simpkins, C. Wilson, J. Org. Chem. 2007, 72, 4265 – 4267; For
related work on epoxides, see: e) J. P. Tierney, A. Alexakis, P.
Mangeney, Tetrahedron: Asymmetry 1997, 8, 1019 – 1022; f) O.
Equey, A. Alexakis, Tetrahedron: Asymmetry 2004, 15, 1069 –
1072.
ing”) of enolizations because of the formation of mixed lithium
amide/lithium enolate aggregates (but not usually in THF), see
references [2b] and [2d].
[14] a) K. Aoki, K. Koga, Tetrahedron Lett. 1997, 38, 2505 – 2506; for
recent use in total synthesis, see: J. S. Clark, D. Vignard, A.
Parkin, Org. Lett. 2011, 13, 3980 – 3983.
[5] N. J. Goldspink, N. S. Simpkins, M. Beckmann, Synlett 1999,
1292 – 1294.
[6] a) C. D. Gill, D. A. Greenhalgh, N. S. Simpkins, Tetrahedron
2003, 59, 9213 – 9230; b) D. J. Bennett, P. L. Pickering, N. S.
Simpkins, Chem. Commun. 2004, 1392 – 1393.
[7] a) R. A. Ewin, N. S. Simpkins, Synlett 1996, 317 – 318; b) R. A.
Ewin, D. A. Price, N. S. Simpkins, A. M. MacLeod, A. P. Watt, J.
Chem. Soc. Perkin Trans. 1 1997, 401 – 415.
[8] See, S. E. Gibson (nꢁe Thomas), P. OꢀBrien, E. Rahimian, M. H.
Smith, J. Chem. Soc. Perkin Trans. 1 1999, 909 – 912. Reactions
involving 1 or 2 give very similar ee values with all of the
substrates we have examined, including 6 and 10.
[9] See for example: a) M. Majewski, P. Nowak, Tetrahedron Lett.
1998, 39, 1661 – 1664; b) X. Sun, S. L. Kenkre, J. F. Remenar,
J. H. Gilchrist, D. B. Collum, J. Am. Chem. Soc. 1997, 119, 4765 –
4766.
[15] For a recent report, see: a) P. Dinꢁr, Tetrahedron: Asymmetry
2010, 21, 2733 – 2739; reviews: b) J. Eames, Eur. J. Org. Chem.
2002, 393 – 401; c) A. Magnus, S. K. Bertilsson, P. G. Andersson,
Chem. Soc. Rev. 2002, 31, 223 – 229; for pioneering reports
involving lithium amide competition in epoxide rearrangement,
see: d) M. Asami, T. Ishizaki, S. Inoue, Tetrahedron: Asymmetry
1994, 5, 793 – 796; e) M. J. Sçdergren, P. G. Andersson, J. Am.
Chem. Soc. 1998, 120, 10760 – 10761; f) D. Pettersen, M. Amedj-
kouh, S. O. Nilsson Lill, P. Ahlberg, J. Chem. Soc. Perkin Trans. 2
2002, 1397 – 1405.
[16] T. Yamashita, D. Sato, T. Kiyoto, A. Kumar, K. Koga,
Tetrahedron 1997, 53, 16987 – 16998.
[17] For related exchange processes, see: a) R. R. Fraser, A. Baignꢁe,
M. Bresse, K. Hata, Tetrahedron Lett. 1982, 23, 4195 – 4198; b) N.
Duguet, A. Harrison-Marchand, J. Maddaluno, K. Tomioka,
Org. Lett. 2006, 8, 5745 – 5748. See also reference [16].
[18] a) M. G. Gardiner, C. L. Raston, Inorg. Chem. 1996, 35, 4047 –
4059; b) D. R. Armstrong, D. Barr, A. T. Brooker, W. Clegg, K.
Gregory, S. M. Hodgson, R. Snaith, D. S. Wright, Angew. Chem.
1990, 102, 443 – 445; Angew. Chem. Int. Ed. Engl. 1990, 29, 410 –
411.
[19] For a structure incorporating acetone, doubly coordinated to
lithium, see: D. A. Grossie, W. A. Feld, L. Scanlon, G. Sandi, Z.
Wawrzak, Acta Crystallogr. Sect. E 2006, 62, m827 – m829.
[20] Note that LiCl, which is present in our reactions, may facilitate
formation of a mixed aggregate with 1 or 2, as has been shown
for 14. See: K. Sugasawa, M. Shindo, H. Noguchi, K. Koga,
Tetrahedron Lett. 1996, 37, 7377 – 7380. It is less clear if LiCl
should be included as an active component in Scheme 2.
[10] M. J. McGrath, J. L. Bilke, P. OꢀBrien, Chem. Commun. 2006,
2607 – 2609.
[11] a) Original report: R. Shirai, M. Tanaka, K. Koga, J. Am. Chem.
Soc. 1986, 108, 543 – 545; b) our preliminary findings: R. P. C.
Cousins, N. S. Simpkins, Tetrahedron Lett. 1989, 30, 7241 – 7244;
c) Reinvestigation of selectivity levels: K. Aoki, M. Nakajima,
K. Tomioka, K. Koga, Chem. Pharm. Bull. 1993, 41, 994 – 996;
d) Use in a-arylations: V. K. Aggarwal, B. Olofsson, Angew.
Chem. 2005, 117, 5652 – 5655; Angew. Chem. Int. Ed. 2005, 44,
5516 – 5519.
[12] a) B. J. Bunn, N. S. Simpkins, J. Org. Chem. 1993, 58, 533 – 534;
b) M. Majewski, R. Lazny, P. Nowak, Tetrahedron Lett. 1995, 36,
5465 – 5468.
[13] The reasons for this result are unclear at present, although
Collum and co-workers have described autoinhibition (or “stall-
Angew. Chem. Int. Ed. 2012, 51, 12068 –12071
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12071