Synthesis and reactivity of conformationally locked alpha-aminoorganostannanes and alpha-aminoorganolithiums. Discovery of a surprising configurational requirement for transmetalation.

Article abstract of DOI:10.1021/ol005770u

[equation--see text] 2-Tributylstannyl-N-methylpiperidines that are conformationally locked by a 4-tert-butyl substituent were evaluated in transmetalations (Sn-Li exchange) and reactions with electrophiles. When the tin is equatorial, transmetalation occurs smoothly as does reaction with carbonyl electrophiles. Alkyl halides seem to undergo single electron transfer reactions, affording nonselective alkylation products, along with products of radical disproportionation. In a surprise, an axially oriented tin failed to transmetalate, suggesting that a synclinal relationship between the nitrogen lone pair and the carbon-tin bond is a requirement for transmetalation.

Full text of DOI:10.1021/ol005770u

ORGANIC
LETTERS
2000
Vol. 2, No. 11
1561-1564
Synthesis and Reactivity of
Conformationally Locked
r-Aminoorganostannanes and
r-Aminoorganolithiums. Discovery of a
Surprising Configurational Requirement
for Transmetalation
Gilles Chambournier and Robert E. Gawley*
Department of Chemistry, UniVersity of Miami, P.O. Box 249118,
Coral Gables, Florida 33124
rgawley@miami.edu
Received March 7, 2000
ABSTRACT
2-Tributylstannyl-N-methylpiperidines that are conformationally locked by a 4-tert-butyl substituent were evaluated in transmetalations (Sn−Li
exchange) and reactions with electrophiles. When the tin is equatorial, transmetalation occurs smoothly as does reaction with carbonyl
electrophiles. Alkyl halides seem to undergo single electron transfer reactions, affording nonselective alkylation products, along with products
of radical disproportionation. In a surprise, an axially oriented tin failed to transmetalate, suggesting that a synclinal relationship between the
nitrogen lone pair and the carbon−tin bond is a requirement for transmetalation.
R-Amino organolithiums have received a significant amount
of interest over the past two decades because of their
configurational stability and synthetic versatility.1 In 1993,
we reported the synthesis of N-methyl-2-lithiopyrrolidine and
-piperidine from the corresponding N-methyl-2-tri-n-butyl-
stannylpyrrolidine and -piperidine through tin-lithium ex-
change with n-butyllithium, to produce organolithium species
having extraordinary configurational stability.2 The lithiated
species were then reacted with different electrophiles to
provide 2-substituted N-methyl pyrrolidines and piperidines
in good to excellent yields. The steric course3 of these
reactions was found to depend on the type of electrophile:2b
ketones, aldehydes, and esters react with retention while alkyl
halides react with inversion. Reducible electrophiles such
as benzyl bromide, ethyl bromoacetate, and benzophenone
give complete racemization.
All of our previous studies evaluated the steric course of
reactions where the only stereocenter was the metal-bearing
carbon of enantiopure organolithiums. In this report, we
disclose the results of an investigation into the transmetalation
and electrophilic quench of diastereomeric, conformationally
locked, racemic piperidines. During the course of these
studies, we were surprised to find a severe stereoelectronic
requirement for tin-lithium exchange.
(1) Reviews: (a) Beak, P.; Reitz, B. D. Chem. ReV. 1978, 78, 275. (b)
Beak, P.; Zajdel, W. J.; Reitz, B. D. Chem. ReV. 1984, 84, 471. (c) Gawley,
R. E.; Rein, K. S. In ComprehensiVe Organic Synthesis; Trost B. M.,
Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 3, Chapter 1.2. (d) Gawley,
R. E. Curr. Org. Chem. 1997, 1, 71.
(2) (a) Gawley, R. E.; Zhang, Q. J. Am. Chem. Soc. 1993, 115, 7515.
(b) Gawley, R. E.; Zhang, Q. J. Org. Chem. 1995, 60, 5763. (c) Gawley,
R. E.; Zhang, Q. Z. Tetrahedron 1994, 50, 6077.
(3) Gawley, R. E. Tetrahedron Lett. 1999, 40, 4297.
(4) Go¨ssinger, E. Monatsh. Chem. 1982, 113, 339.
(5) (a) Beak, P.; Lee, W.-K. J. Org. Chem. 1990, 55, 2578. (b) Beak,
P.; Lee, W.-K. J. Org. Chem. 1993, 58, 1109.
10.1021/ol005770u CCC: $19.00 © 2000 American Chemical Society
Published on Web 05/05/2000

Because of the variability of the steric course of SE2
substitution in unsubstituted heterocyclic rings, our plan was
to evaluate reactivities of conformationally restricted 4-tert-
butyl-N-methylpiperidines having a 2-tri-n-butylstannyl group
in both the axial and equatorial positions. To this end,
piperidines 3 (Scheme 1) with an equatorial stannyl group
also has an equatorial methyl in the 6-position. Carbamate
2 was deprotonated and methylated with dimethyl sulfate to
obtain carbamate 4 in 76% yield. Stannylation of 4 with tri-
n-butylstannyl chloride afforded stannyl carbamate 5 in 82%
yield. Reduction of the carbamate moiety with excess
diisobutylaluminum hydride afforded N-methyl piperidine 6
in 60% yield. The relative configuration was assigned on
the basis of literature precedent5 and 1H NMR spectroscopy.6
Piperidine 3 was transmetalated by treatment with n-
butyllithium for 20 min at -78 °C in THF (Scheme 3), and
Scheme 1
Scheme 3
and 6 (Scheme 2) with an axial stannyl group were prepared
for this study. As shown in Scheme 1, 4-tert-butylpyridine
1 was hydrogenated, in the presence of a catalytic amount
of platinum oxide,4 in an ethanol/chloroform solvent mixture
Scheme 2
the resulting 2-lithiopiperidine 7 was allowed to react with
various carbonyl electrophiles to afford N-methyl piperidines
8a-c in excellent yields. As we found previously in the
unsubstituted case,2b deprotonation of cyclohexanone or
acetone does not compete effectively with addition to the
carbonyl. The percentage yields of educts 8a-c are similar
to the yields obtained with unsubstituted R-lithiopiperidine,
and the steric course at the carbanionic carbon was similar
to that observed in the unsubstituted series: retention
(SE2ret).
With benzyl bromide as electrophile (Scheme 4), the yield
of substitution product was lower, and the steric course was
random. With enantiopure organolithiums unsubstituted in
the 4-position, we also found stereorandom substitution2 and
believe that the reason is a radical mechanism involving
initial single electron transfer (SET).8 In this case, SET from
organolithium 7 to benzyl bromide would produce radical
11 after loss of bromide. This reaction mixture also yielded
a mixture of 12 and a compound tentatively identified as
13. The former was prepared independently (vide infra), the
latter appeared as a nearby peak in the GC-MS trace, with
a molecular weight and fragmentation pattern consistent with
this structure. These two compounds would result from
disproportionation of 11, again consistent with an SET
mechanism with this electrophile.
to afford 4-tert-butylpiperidine in 81% yield. Protection of
the amine functional group was done with di-tert-butyl
dicarbonate in dichloromethane to give carbamate 2 in 97%
yield. Deprotonation of 2 with sec-butyllithium in the
presence of TMEDA in diethyl ether at -78 °C, as reported
by Beak,5 and alkylation with tri-n-butylstannyl chloride (in
61% yield) was followed by reduction of the BOC group
with diisobutylaluminum hydride to give 3 in 75% yield
(36% from 1). The equatorial position of the tri-n-butylstan-
nyl group was assigned on the basis of NMR analysis.6
For comparison with (cis) 3, we would have liked to
examine the trans stereoisomer, having an axial tin, but
several attempts at introduction of an axial tin using iminium
ion chemistry7a,b or reduction of enamides7c were unsuccess-
ful. Knowing that 2,6-dimethyls could be introduced in a
trans orientation by a lithiation strategy,5 we followed the
route shown in Scheme 2, to obtain axial stannane 6, which
(6) Gawley, R. E.; Low, E.; Chambournier, G. Org. Lett. 1999, 1, 653.
(7) (a) Meyers, A. I.; Shawe, T. T.; Gottlieb, L. Tetrahedron Lett. 1992,
33, 867. (b) Gottlieb, L.; Meyers, A. I. Tetrahedron Lett. 1990, 33, 4723.
(c) Wilkinson, T. J.; Stehle, N. W.; Beak, P. Org. Lett. 2000, 2, 155.
(8) Gawley, R. E.; Low, E.; Zhang, Q.; Harris, R. J. Am. Chem. Soc.
2000, 122, 3344.
1562
Org. Lett., Vol. 2, No. 11, 2000

ments in which the transmetalations of piperidines 3 and 6
were followed by methanol quench. The crude mixtures were
analyzed by GC-MS after workup, and the traces were
compared to those of piperidines 12 and 17 which were
synthesized independently from 2 and 4, as shown in Scheme
6. The trace of the crude mixture from the transmetalation
of 3 compared favorably with independently prepared 12
while the trace coming from the transmetalation of 6 showed
no trace of 17! Seemingly, an equatorial tri-n-butylstannyl
group is required for the tin-lithium exchange to occur in
these species. This failure may provide a clue to the
explanation of other tin-lithium exchange failures reported
in the literature.9
Scheme 4
3
Using the Karplus relationship between JSn-C coupling
constants and torsion angles,10 we recently showed that
equatorial stannane 3 and its analogue lacking the tert-butyl
exist as half-chairs in solution.6,11 Axial stannane 6 shows
couplings to C-4 of 17.4 Hz and to C-6 of 21.8 Hz; however,
the Karplus relationship allows two solutions: 50 ( 5° or
115 ( 5° to C-4 and 49 ( 5° or 120 ( 5° to C-6. The ∼50°
solution is a slightly distorted chair; the ∼115-120° implies
a half-chair, since the anticlinal angle is approaching the
eclipsing value of 120°, as observed with the equatorial
stannanes.6,11 Although the solution conformation of 6 cannot
be determined from the present data with certainty, we are
working under the hypothesis that it is in a relatively
undistorted chair. If it was a half-chair, it would be difficult
to explain the differences in reactivity between 3 and 6.
In conclusion, we have shown that conformationally rigid
4-tert-butyl-2-lithio-N-methylpiperidines having an equatorial
lithium react stereoselectively with carbonyl electrophiles (SE-
2ret) in excellent yields. Stereorandom alkylation in modest
yield was observed with benzyl bromide, consistent with a
change in mechanism, most likely SET. However, unlike
unsubstituted 2-lithiopiperidines, equatorially lithiated 7 does
not react well with unactivated alkyl halides, apparently due
to the intervention of SET processes. This could be due to
the fact that the preferred mode of alkylation is SE2inv,
Finally, the same reaction was repeated with phenylpropyl
bromide (Scheme 5) as the electrophile. With enantiopure
Scheme 5
4-unsubstituted piperidines, this electrophile afforded com-
plete inversion (SE2inv) of configuration in 75-76% yield.2b
Since the lithium in 7 is formally equatorial (it is probably
bridged to the nitrogen, distorting a perfect chair), invertive
substitution should be severely inhibited. In the event, the
product of electrophilic substitution (14) was isolated in only
5% yield as a mixture of diastereomers. In this case, we
found a dimer (15) of unknown relative configuration, and
probable disproportionation products 12 and 13. All of these
products are consistent with the intervention of an SET
mechanism. Dimers were observed in the reaction of
lithiopyrrolidine with benzyl bromide, which is thought to
proceed by an SET mechanism.8 Only polar pathways are
followed for this electrophile in unsubstituted piperidines,
so it appears that when the preferred pathway for electrophilic
substitution, here SE2inv, is unlikely or impossible, SET can
intervene as a competitive pathway. The reasons for mecha-
nistic variations between retentive or invertive polar substitu-
tions and SET reactions have not been completely clarified
but are the subject of ongoing investigation.8
(9) (a) Tsunoda, T.; Fujiwara, K.; Yamamoto, Y.; Itoˆ, S. Tetrahedron
Lett. 1991, 32, 1975. (b) Burchat, A. F.; Chong, J. M.; Park, S. B.
Tetrahedron Lett. 1993, 34, 51. (c) Gawley, R. E.; Evanseck, J. D.; Pearson,
W. H.; Stevens, E. P. In ECHET96. Electronic Conference on Heterocyclic
Chemistry; Rzepa, H., Snyder, J., Eds.; Royal Society of Society of
Chemistry: London, 1997 (CD-ROM).
(10) (a) Kitching, W.; Olszowy, H.; Waugh, J.; Doddrell, D. J. Org.
Chem. 1978, 43, 898. (b) Kitching, W.; Olszowy, H. A.; Harvey, K. J.
Org. Chem. 1982, 47, 1893.
(11) Stannane conformers 18/19 could be observed as well-resolved,
unique conformers below -60 °C.6 For the equatorial conformer 18, the
3JSn-C coupling to carbon-6 is 24.6 Hz, corresponding to possible torsions
of 45° or 123°, according to the two possible solutions to Kitching’s Karplus-
like relationship.10 For tert-butylpiperidine 3, J values suggest that the
corresponding torsion is either 42° or 126°. Because of the 4-tert-butyl
group, a 42° torsion requires a boat conformation with the Bu3Sn group in
a flagpole orientation. Since this orientation would be of extraordinarily
high energy, the 123°/126° torsions are more likely. This angle is close to
an eclipsing angle of 120°, suggesting a flattened ring (half-chair). For 19,
the solutions to the Kitching/Karplus equation correspond to possible
torsions of 64° or 107°. Previously, we interpreted these data in terms of a
relatively undistorted chair with an axial tin, but the 107° value cannot be
completely ruled out.
It was of interest to investigate the corresponding axial
organolithium, since we anticipated that SE2inv would be
more easily accommodated with a formally axial lithium.
Much to our surprise, treatment of 6 with n-butyllithium did
not trigger any tin-lithium exchange to 16: only starting
material (6) was detected after workup (Scheme 6). This
surprising observation was confirmed in side-by-side experi-
Org. Lett., Vol. 2, No. 11, 2000
1563

Scheme 6
Acknowledgment. This work was generously supported
by the NIH (GM 56271). We are grateful to Mrs. J.
Esterbrook for technical assistance.
difficult when the lithium is equatorial. In a surprise, we
found that when the tin is trans to the tert-butyl group,
transmetalation fails completely. In these piperidines, a
requirement for transmetalation appears to be a synclinal
relationship between the C-Sn bond and the nitrogen lone
pair. It is possible that this effect is relevant to the failure or
reluctance of certain other (acyclic) R-aminoorganostannanes
to transmetalate.9
Supporting Information Available: Experimental details
for the preparation and characterization of all new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL005770U
1564
Org. Lett., Vol. 2, No. 11, 2000