SITE-SPECIFIC GENERATION OF NONSTABILIZED α-AMINOCARBANIONS. A NOVEL METHODOLOGY TOWARD THE SYNTHESIS OF 2-AMINOALCOHOLS

Article abstract of DOI:10.1016/0040-4039(91)85017-Y

A method for site-specific generation of "nonstabilized" α-aminocarbanions with no activating or stabilizing group was developed, utilizing a metal-sulfur exchange reaction.The method provides a novel methodology for the synthesis of 2-aminoalcohols.

Full text of DOI:10.1016/0040-4039(91)85017-Y

Tetrahedron L~tt~rs, Vol.32. No.17, pp 1975 1978, 1991
Printed ha Great Britain
0040-4039/91 $3.00 t .00
Pergamon Press plc
SITE-SPECIFIC GENERATION OF NONSTABILIZED cc-AMINOCARBANIONS
A NOVEL METHODOLOGY TOWARD THE SYNTHESIS OF 2-AMINOALCOHOLS
Tetsuto Tsuneda,.1) Kenshu Fujiwara, Ye-ichi Yamamote and Sh6 It6.1)
Department of Chemistry, Tohoku University, Sendal 980, Japan
Abstract: A method for site-specific generation of "nonstabilized" m-amino-
carbanione with no activating or stabilizing group was developed, utilizing a
metal-sulfur exchange reaction. The method provides a novel methodology for
the synthesis of 2-aminoalcohols.
Despite their potential utility, the "non-stabilized" carbanions carrying a nitrogen
atom on the anionic carbon (~-aminocarbanions) unlike other c~-heterocarbanions have
received little attention in organic synthesis.2,3, 4) This seems to be due mainly to
the fact that an appropriate method for their generation at the desired position has
not yet been developed.5, 6) Of the two presently available methods,6b,c) the
metatlation of aminoalkanes which produces the thermodynamically most stable
anions lacks site-specificity.
The only site-sDecific method, transmetatlation of
organetin compounds, conveniently starts from {z-amino-suLfides. 5) We present
herein reductive lithiation, another and perhaps more convenient and versatile
reaction, for site-specific generation of the "non-stabilized" cz-aminocarbanions 1,
also starting with c~-amino-sulfides 2.7 )
NR22
LiDBB
NR22
NR2
OH
2
1
3
a : Rl=,/-Pr, R2=-(CH2)4-, R3= p-MeOCsH4-
b : Rl=t-Bu, R2=-CH2CGH5, RS=/-pr
c : R~=t-Bu, R2=-(CH2)4-, R3=t-Bu
The reaction is simple, as the following typical experimental procedure
exemplifies. Two molar equivalents of Li di-t-butylbiphenyl radical anion (LiDBB) 8)
in THF (0.12M) were added quickly at-95°C under argon atmosphere to a THF solution
of the c~-amino-sulfides 2a. The greenish-blue color of LiDBB changed instantly to
1975

1976
the dark red of the desired anion la. The formation of l a was verified by adding
p-anisaldehyde at -95°C to the dark red solution after 90 sec. to give the 2-amino-
alcohol 3a in 79% yield after easy purification by SiO2 chromatography. Absence of
a radical species in the dark red solution was established when no 3a was formed
with the use of only 1 tool equivalent of LiDBB.
The anion la is fairly stable below -30°C (a few percent decomposition after 5
min), but decomposes slowly at -10°C (ca. 40% after 10 min and ca. 90% after 30 rain)
and rapidly at 0°C (80% after 10 mir~) 9,10)
The anion formation is quite site-specific, as was shown by the fact that even 2b
with more acidic hydrogens afforded lb with the negative charge at the position
where the PhS group originally was. This was again verified by the formation of 3b
in 71% yield.
The method is quite versatile. The bulkiness of R1 does not affect the anion
formation; t-Butyl (2¢) gives the anion le in approximately the same yield (73%,
quenched with pivalaidehyde) as ethyl (71%) and isopropyl (72%). Not only secondary
but also tertiary carbanion can be generated by the present method; 1-phenylthio-
1
-(1-pyrrolidinyl)cycfohexane gave the corresponding anion which was trapped by
p-anisaldehyde in 65% yietd.
In order to establish the stability of 1 compared with the "stabilized"6) cc-amino-
carbanions and examine the feasibility of their synthetic application, a few reactions
were carried out. Some of the examples are shown below. The reactions of 1 with
aliphatic aldehydes give anti(erythro)-2-aminoalcohols
3 predominantly in contrast
to the syn preference of c~-alkoxycarbanions, 11) but diastereoselectivity (anti/syn
ratio) in the reaction of lc is highest (6.1) with pivalaldelyde and drops to 3.5 and
o
-
@
+
- . M
4
3%
OH
Bu3SnCl
R1
SnBu 3
4
8%
RI=Et or /-Pr.
CsH13Br ~ ""T~c6H13~r~
+
/
~
2
3
5%
24%

1
977
4
.0 with isobutyl- and propionaldehydes, respectively. The amino group with bulkier
dibenzyl (lb) gives larger ratios (32, 10 and 9 in the same order), p-Anisaldehyde
gives very low (0.5~1.5) selectivity with both anions, lb and lc.12,13) Enolizable
ketcnes (e.g. cyciohexanone) give the pretonation product, 1-alkylpyrrolidine,
lowering the yield of 3. Halogenoalkyls afford alkylation products and bis(1-
pyrrolidinyl)alkanes resulted frcm single electron transfer and subsequent radical
coupling.
Trialkyltin chlorides yield the corresponding aminostannanes.
Thus, the present study disclosed a convenient method of generation of "non-
stabilized" c~-aminocarbanions, their thermal stability, and reactions with some
electrophiles. Since the "stannane route" is not applicable to the generation of
secondary carbanions, 14) the method may be useful in diastereoselective synthesis
of 2-aminoalcohols.
REFERENCES AND NOTES
1
2
) Present address; Faculty of Pharmaceutical Sciences, Tokushima Bunri University,
Tckushima 770, Japan.
) The "stabilized" c~-aminocarbanions are defined as those with a stabilizing group
on the anionic carbon (e.g. ~-system) or on the nitrogen (e.g. acyl group). (cf. D.
Seebach, I. M. P. Huber, and M. A. Syfrig, Helv. Chem. Acta, 70, 1357 (1987), A.
I. Meyers, D. B. Miller, and F. H. White, J. Am. Chem. Soc., 110, 4778 (1988), and P.
Beak and W.-K. Lee, Tetrahedron Lett., 30, 1197 (1989)). ~-Aminocarbanions only
with alkyl or aralkyl group are consequently "non-stabilized".
3
4
)
)
References on the carbanions stabilized by the 2nd-row elements are numerous.
For reviews, e.g.W.-K. Li, R.H. Nobes, D. Poppinger, and L. Radom in Comprehensive
Carbanion Chemistry, Part C. E. Buncel and T. Durst ed., Elsevier, 1987, p25.
For the "non-stabilized" carbanions with c~-oxygen, see J.S. Sawyer, A. Kucerovy,
T.L. Macdonald, and G.J. McGarvey, J. Am. Chem. Soc., 110, 842 (1988) and the
references cited therein.
5
6
) a. D.J. Peterson, J. OrganometaL Chem., 21, p63 (1970). b. Idem, J. Am. Chem.
Soc., 93, 4027 (1971).
)
The "stabilized" ~-aminocarbanions are well known and utilized in organic
syntheses. For reviews, see a. P. Beak, W.J. Zajdel, and D.B. Reitz, Chem. Rev., 84,
471 (1984). b.A. Kfief, Tetrahedron, 36, 2531 (1980). c. D. J. Peterson,
OrganometaL Chem. Rev., A, 7, 295 (1972).
7
8
) Cohen et al. have successfully applied a radical anion to generate "non-stabilized"
c~-oxycarbanions from ~-aikoxy-sulfide. T. Cohen and J.R. Matz, J. Am. Chem. Soc.,
1
02, 6900 (1980). T. Cohen and M~-T. Lin, ibid., 106, 1130 (1984).
)
P.K. Freeman and L.L. Hutchinson, J. Org. Chem., 46, 1924 (1980).

1978
9
) All percent yields in this paper are isolated yields of the 2-aminoalcohols 3,
the trapping products, based on the c~-amino-sulfides 2, the precursors of the
anions 1.
1
0) The only isolable product besides 3a is N-isobutylpyrrolidine, the product of
proton abstraction from the reaction mixture.
1
1
1) G.J. McGarvey and M Kimura, J. Org. Chem., 47, 5420 (1982).
2) While the anti/syn ratio was determined by GLC, relative stereochemistry of the
products was established after separation of diastereoisomers in most cases by
the chemical correlation shown below.
Bn2N
NH2
~ . / R 3
R I ~ R 3 H2/PdC
Br ~ B r
=
OH
MeOH
R
.
i.Pr2N Et
EtOH, ref.
R3
OH
R1
{
CICOOCC;3
i-Pr2N Et/CH2CI2
OH
O
O
H N "IJ'O
HN'JJ"O 3
H
H
R1
H
Stereochemistry of the cyctic carbamates was established by NOE
measurements and/or coupling constants (Jtrans: 3.3~4.6 Hz, Jcis: 8.2~7.5Hz) of
two protons depicted. Our assignments are generally in accord with those based
on J values in 2-aminoalcohols (Janti < Jsyn. D. Seebach, A.K. Beck, T.
Mukhopadhlyay, and E. Thomas, Helv. Chem. Acta., 65, 1101 (1982)), but the
reverse is true, or both Js are comparable, in several cases where R2 is -(CH2)4-
and R1 is bulkier than isopropyl group.
1
3) Mechanism of the stereoseleetion will be discussed in the forthcoming full paper.
4) None of these stannanes reacted with BuLi in contrast to the aminomethyl-
stannanes (ref. 4a), probably due to steric reasons.
1
(
Received in UK 14 February 1991)