Application of Baldwin's rules for the preparation of stable, β-leaving group-bearing organolithium compounds [19]

Full text of DOI:10.1021/ja005745j

J. Am. Chem. Soc. 2001, 123, 2095-2096
Application of Baldwin’s Rules for the Preparation
2095
of Stable, â-Leaving Group-Bearing Organolithium
Compounds
M. Isabel Calaza, M. Rita Paleo, and F. Javier Sardina*
Departamento de Qu´ımica Orga´nica
Facultad de Qu´ımica
UniVersidad de Santiago de Compostela
15782 Santiago de Compostela, Spain
Figure 1.
ReceiVed October 31, 2000
Scheme 1
A great deal of research has been devoted to the development
of highly functionalized organometallic reagents.1 Despite this
interest, polar organometallic compounds possessing a leaving
group (RO-, R2N-, halogen) â to the anionic center remain
notoriously elusive, due to their tendency to undergo elimination
reactions to give alkenes.2 Thus, C-lithioaziridines (1)3 and
R-acyloxy-â-aminoalkyl-lithiums (2, 3)4 are the only stable
examples known of the potentially useful â-amino-alkyl-lithium
reagents. The factors that prevent these â-amino-substituted
lithiated species from undergoing elimination reactions have not
been definitely established, but lithioaziridines 1 (and the
analogous lithiooxiranes)3b may owe their stability to their cyclic
nature, since electrocyclic ring-opening pathways should be
inhibited by an enforced syn arrangement of the C-Li bond and
the lone pair of the heteroatom,3d while the stability of 2 and 3
may arise from their dipole-stabilized nature.4d
In this communication we present our studies on the preparation
and synthetic applications of the cyclic R-alkoxy-â-aminoalkyl-
lithium compounds 4, 5 (Figure 1). These species are of great
interest from a mechanistic point of view, as well as for their
great potential in organic synthesis.
We hypothesized that the relative stability of cyclic, â-func-
tionalized organolithium compounds such as 4 and 5 could be
predicted by using the principle of microscopic reversibility along
with Baldwin’s rules,5 since their â-eliminative decompositions
are in fact the reverse of n-endo-trig cyclizations.6 If this were
the case, the stability of these species would decrease with
increasing ring size.7
To test this hypothesis we first prepared R-alkoxy-â-amino-
stannanes 7, as precursors of the corresponding organolithium
reagents 4. Cyclic â-aminoketones 6a-c (n ) 1 to 3) were reacted
with Bu3SnLi/CeCl3 to provide the corresponding stannyl-
alcohols,8 which were straightforwardly protected (MOMCl/
DIPEA) to provide the desired aminostannyl-acetals 7a-c in good
overall yields (50-60%). Reaction of the azepinone 6d with the
tin nucleophile gave only low yields of the desired alcohol, the
main product being the acyclic ketone 10.
Treatment of stannanes 7a-c with 115 mol % of n-butyllithium
in THF at -78 °C (for 2-80 min) resulted in tin-lithium
exchange,9 as evidenced by quenching experiments with the
electrophiles shown in Scheme 1. The cyclic amines 8a-c, which
arise from the reaction of the organolithium intermediates with
the corresponding electrophiles (entries 1, 2, 4, and 6), were
isolated as the exclusive or very major reaction products. Only
minute amounts of enol ethers 9b,c, formed via â-elimination of
the organolithium intermediates, were detected in the crude
products of the reactions of the five- and six-membered ring
stannanes 7b,c. Azetidine 7a afforded only substitution products
(8a). The outcome of these experiments clearly showed that the
cyclic R-alkoxy-â-aminoalkyl-lithium compounds 4a-c are stable
at -78 °C for extended periods.10
This paper is dedicated to Professor Henry Rapoport. Financial support
from the CICYT (Spain, Grant SAF99-0127) and the Xunta de Galicia (Grant
PGIDT00PXI20909PR and a fellowship to M.I.C.) is gratefully acknowledged.
We also thank Professor Rafael Suau (Universidad de Ma´laga, Spain) for the
elemental analyses.
(1) Rottla¨nder, M.; Boymond, L.; Be´rillon, L.; Lepreˆtre, A.; Varchi, G.;
Avolio, S.; Laaziri, H.; Que´guiner, G.; Ricci, A.; Cahiez, G.; Knochel, P.
Chem. Eur. J. 2000, 6, 767 and references therein.
(2) Foubelo, F.; Gutie´rrez, A.; Yus, M. Synthesis 1999, 503. (b) Foubelo,
F.; Gutie´rrez, A.; Yus, M. Tetrahedron Lett. 1997, 38, 4837 and references
therein.
(3) Bisseret, P.; Bouix-Peter, C.; Jacques, O.; Henriot, S.; Eustache, J. Org.
Lett. 1999, 1, 1181. (b) Satoh, T. Chem. ReV. 1996, 96, 3303. (c) Vedejs, E.;
Kendall, J. T. J. Am. Chem. Soc. 1997, 119, 6941. (d) Vedejs, E.; Moss, W.
O. J. Am. Chem. Soc. 1993, 115, 1607.
(4) Schwerdtfeger, J.; Kolczewski, S.; Weber, B.; Fro¨hlich, R.; Hoppe, D.
Synthesis 1999, 1573. (b) Weber, B.; Kolczewski, S.; Fro¨hlich, R.; Hoppe,
D. Synthesis 1999, 1593. (c) Weber, B.; Schwerdtfeger, J.; Fro¨hlich, R.; Go¨hrt,
A.; Hoppe, D. Synthesis 1999, 1915. (d) Schwerdtfeger, J.; Hoppe, D. Angew.
Chem., Int. Ed. Engl. 1992, 31, 1505. (e) Beak, P.; Carter, L. G. J. Org. Chem.
1981, 46, 2363.
(5) Johnson, C. D. Acc. Chem. Res. 1993, 26, 476. (b) Baldwin, J. E.;
Thomas, R. C.; Kruse, L. I.; Silberman, L. J. Org. Chem. 1977, 42, 3846. (c)
Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734. (d) Baldwin, J. E.;
Cutting, J.; Dupont, W.; Kruse, L.; Silberman, L.; Thomas, R. C. J. Chem.
Soc., Chem. Commun. 1976, 736.
(6) An analogous argument has already been put forward to explain the
stability of â-alkoxy-substituted, four- and five-membered-ring cyclic eno-
lates: Seebach, D.; Hungerbu¨hler, E. Modern Synthetic Methods; Otto Salle
Verlag: Frankfurt a. Main, 1980; Vol 2, p 91.
(7) 4- and 5-endo-trig cyclizations are disallowed, while 6- and 7-endo-
trig processes are allowed. See ref 5c.
(8) In the absence of Ce(III) a large proportion of starting ketone was
recovered after the reaction, presumably due to an enolization process. For a
preparation of tributylstannyllithium, see ref 9c.
(9) Sawyer, J. S.; Kucerovy, A.; Macdonald, T. L.; McGarvey, G. J. J.
Am. Chem. Soc. 1988, 110, 842 (b) Still, W. C.; Sreekumar, C. J. Am. Chem.
Soc. 1980, 102, 1201. (c) Still, W. C. J. Am. Chem. Soc. 1978, 100, 1481.
10.1021/ja005745j CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/10/2001

2096 J. Am. Chem. Soc., Vol. 123, No. 9, 2001
Communications to the Editor
Scheme 2
Scheme 3
These results show the synthetic potential of this kind of
â-amino-organolithium reagents, but they are somewhat surprising
from the mechanistic standpoint, since Baldwin’s rules predict
that the decomposition of lithiopiperidine 4c should be a facile
process, contrary to its observed stability in THF at -78 °C.7
The confirmation that Baldwin’s rules correctly predict the relatiVe
stabilities of 4a-c was obtained when the tin-lithium exchange
experiments on 7a-c were carried out in DME at -50 °C for 80
min (entries 3, 5, and 7). Lithioazetidine 4a was found to be stable
toward elimination under these conditions, but lithiopirrolidine
4b and lithiopiperidine 4c afforded noticeable amounts of
elimination products 9b,c, the six-membered ring compound
undergoing elimination at a higher rate than the pyrrolidine
analogue.
This trend of decreasing stability of 4a-c with increasing ring
size might explain first, the well-known stability of lithioaziridines
1 and the analogous lithioepoxides,3 and second, why acyclic
ketone 10 was the main product of the attempted stannylation of
azepinone 6d. The intermediate stannyl-alkoxide 11 could undergo
the tin counterpart of a Brook rearrangement11 to afford C-lithio
azepine 12 which should rapidly undergo elimination to provide
a tin enol ether that, on work up, should give rise to ketone 10.
We have observed that N-disubstituted acyclic R-amino aldehydes
14a,b experience the same process when reacted with anionic
tin nucleophiles (Scheme 2).
Once we had established the validity of our hypothesis, we
studied its extension to the preparation of other types of cyclic
â-amino-organolithium reagents that could show even greater
synthetic potential than 4a-c. We chose lithio-oxazolidines 5 as
targets, as their reactions with electrophiles should give rise to a
wide variety of synthetically and pharmacologically interesting
â-amino alcohols. Scheme 3 summarizes the synthesis of 5a,b
and the results of their reactions with several representative
electrophiles. Stannyl-oxazolidines 17a,b, precursors of 5a,b, were
obtained from N-Pf-alaninal 16.12 Addition of tributylstannyl-
lithium to 16 proceeded stereoselectively to give one sensitive
stannyl-alcohol which was immediately cyclized to give oxa-
zolidine 17a (69% overall yield). NOE experiments performed
on 17a allowed us to establish its stereochemistry and showed
that the addition to the aldehyde group of 16 had taken place
under chelation control.
The epimeric oxazolidine 17b was obtained by oxidation of
the alkoxide intermediate of the tin-lithium addition to 16 with
1,1′-(azodicarbonyl)dipiperazine (0 °C, 50% yield),13 followed
by reduction of the resulting stannyl-ketone with LiAlH4. The
mixture of alcohols obtained was cyclized to provide a mixture
of 17a and 17b (2:1 ratio, 65% overall yield) which was resolved
by HPLC.
Treatment of stannyl-oxazolidines 17a,b with 115 mol % of
BuLi (THF, -78 °C, 2-15 min) followed by addition of a variety
of electrophiles (Scheme 3) afforded clean addition/substitution
reactions to yield oxazolidines 18a,b; no traces of byproducts
coming from â-elimination reactions at the stage of the organo-
lithium intermediates were detected.14 Analysis of the products
of the reactions of 5a,b with D2O and acetone (Scheme 3, entries
1, 4, 8, and 9) showed that the tin-lithium exchanges and the
addition to the electrophiles had occurred with retention of
configuration.15
We believe that the methodology presented herein can have
wide application for the development of synthetically useful,
highly functionalized organolithium reagents, previously in-
accessible by other routes.
(10) This kind of tertiary â-amino-organolithium compound cannot be
prepared by deprotonation of the corresponding carbamates, since only tertiary
hydrogens that are activated by an extra carbanion-stabilizing group can be
abstracted using this methodology: (a) Derwing, C.; Hoppe, D. Synthesis 1996,
149. (b) Paetow, M.; Kotthaus, M.; Grehl, M.; Fro¨hlich, R.; Hoppe, D. Synlett
1994, 1034. (c) Carstens, A.; Hoppe, D. Tetrahedron 1994, 50, 6097. (d)
Hoppe, D.; Paetow, M.; Hintze, F. Angew. Chem., Int. Ed. Engl. 1993, 32,
394. (e) Hoppe, D.; Carstens, A.; Kra¨mer, T. Angew. Chem., Int. Ed. Engl.
1990, 29, 1424.
JA005745J
(13) Marshall, J. A.; Welmaker, G. S.; Gung, B. W. J. Am. Chem. Soc.
1991, 113, 647 and references therein.
(11) Naganuma, K.; Kawashima, T.; Okazaki, R. Phosphorus, Sulfur Silicon
Relat. Elem. 1997, 124 and 125, 513. (b) Schiesser, C. H.; Styles, M. L. J.
Chem. Soc., Perkin Trans. 2 1997, 2335.
(14) Pf-NH2 would be formed. The rest of the material was composed
exclusively of protonated oxazolidine 18, R1 ) R2 ) H.
(15) For removal of protecting groups in 18a,b see: Ferna´ndez-Meg´ıa, E.;
Paz, M. M.; Sardina, F. J. J. Org. Chem. 1994, 59, 7643.
(12) Lubell, W. D.; Rapoport, H. J. Am. Chem. Soc. 1987, 109, 236.