close

Вход

Забыли?

вход по аккаунту

?

Enantioselective Synthesis with Lithium()-Sparteine Carbanion Pairs.

код для вставкиСкачать
[(q5-C,H,)Re(NO)(PPh3)]+ and analogons unsaturated chiral complexes (see element symbols along
the fingertips) can form two diastereomeric adducts
with prochiral alkenes, aldehydes, and ketones. The
ratio of these diastereomers is a measure of the chiral recognition, symbolized by the hands, of the
reaction partners. The rhenium complex mentioned is
depicted above the hands (Re: light gray, P: yellow,
N: light blue, 0: dark blue, Ph, Cp: dark gray). The
selectivity of alkene complexation can be analyzed
(and predicted) by considering the sizes of the substituents a 4 on the alkene and correlating them with the
steric conditions in the chiral complexes, which can
be represented with three-dimensional bar graphs.
REVIEWS
Enantioselective Synthesis with Lithium/( -)-Sparteine Carbanion Pairs
Dieter Hoppe* and Thomas Hense
Dedicated to Professor Dieter Seebach on the occasion of his 60th birthday
“Chiral carbanions”-that
is, enantiomerically enriched lithium-carbanion pairs in which the carbanionic center
carries the chiral information-were regarded until recently as “exotic species.”
In the past ten years it has become clear
that they can, in fact, play a meaningful
role in enantioselective synthesis, since
substitution for lithium occurs here
stereospecifically, usually with retention
of configuration. They are also more
readily and commonly accessible than
was originally assumed. The trick lies in
the use of lithium cations with chiral ligands, whether in the form of alkyllithium species used as bases in kinetically
controlled, enantiotopically discriminating deprotonation, or in thermodynamically controlled equilibration in
configurationally labile epimeric ion
pairs. The lupine alkaloid (-)-sparteine
has shown itself admirably suited as a
1. Introduction and Delineation of the Problem
In a series of reviews that also provide insight into the fundamental contributions of his research group, D. Seebach has
since 1969 popularized the concept of “reactivity umpolung.”[’l
As a consequence, the synthetic chemist has acquired not only
a simple tool for the rational planning of syntheses, but has also
been stimulated in the search for new synthetic building blocks.
A higher standard has been achieved particularly in the case of
reagents for carbanionic synthons of the d’ and d3 type, which
now leaves little to be desired with respect to controlling reactivity and various levels of selectivity. Seebach has repeatedly noted that the reactivity and selectivity of carbanionic reagents is
heavily dependent on the nature of the counterion and the structure of the relevant aggregatesc2]It will become clear in what
follows that it is precisely the apparent disadvantage of most d’
and d3 reagents--a lack or at least a weakening of resonance
stabilization, and thus restricted accessibility-that lends new
qualities to these species.
The carbanionic center of a lithium-carbanion pair in the
vicinity of M substituents seeks to achieve a pyramidal and
therefore potentially chiral configuration; in this way it is able,
under appropriate conditions, to act as a carrier of chiral information, thereby opening the way to new strategies for enantiose-
+
[*] Prof. Dr. D. Hoppe, Dr. T. Hense
Organisch-chemisches Institut der Universitat
Corrensstrasse 40, 11.48149 Miinster (Germany)
Fax: Int. code +(251)833-9772
Angeu. Chem. In!. E d Engl. 1997, 36,2282-2316
chiral bidentate ligand, and its efficiency
and breadth of application are so far unsurpassed. This contribution constitutes
an overview of the preparation of chiral
reagents, covering primarily “umpoled”
synthons such as homoenolates, 1oxyalkanides with a broad pattern of
substitution, and a-aminobenzyl anions.
Keywords: asymmetric synthesis * chiral
building blocks lithium (-)-sparteine
-
lective synthesis. Chiral ligands attached to the cation-here
primarily the alkaloid (- )-sparteine (1) (Scheme 1)-have
shown themselves to be astonishingly efficient aids.
1A
Scheme 1. Conformations of (-)-sparteine (1).
1B
In nonpolar solvents lithium salts of carbanions form tight
ion pairs, usually present as dimers, tetramers, and occasionally
also higher aggregates.I2.31 The lithium cation usually prefers an
approximately tetrahedral coordination geometry saturated by
four donor ligands. It is thus tempting to provide the cation with
chiral, enantiomerically pure ligands and to hope that stereoselectivity will be introduced in reactions with achiral, prostereogenic substrates. For purposes of clarity our first “thought experiment” will involve a bidentate, C,-symmetric ligandC4.’]
and a ql-bound anion; in addition, we will consider a monomeric ion pair. The ion pair A bears an achiral alkanide residue; that
means that any chiral induction must be due to the chiral cation
(Scheme2). If both components of the ion pair (the cation
as well as the anion) are chiral, the result is the pair of
epimers B and epi-B. The consequences will be explored in more
detail.
0 WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1997
0570-083319713621.2283 S 17.50+ 5010
2283
D. Hoppe and T. Hense
REVIEWS
r
A
B
1
epiS
Scheme 2. Types of chiral lithium-carbanion pairs. X is, for example, OR, NR,;
L is, for example, OEt,, THF; A,B are C- or heterosubstituents.
A is already in a position to “recognize” enantiotopic faces
and groups in prochiral[61reaction partners, since it will proceed
through diastereomorphic transition states, which will in principle be energetically nonequivalent (Scheme 3 ) . This is apparent in complex D, comprising a combination of A and C: the
previously enantiotopic faces Re and Si are now diastereotopic
in the intermediate D, and the alkanide residue ACHY will be
transferred at different rates to form the adducts E and epi-E.
Hydrolysis leads finally to the enantiomers F and ent-F, one of
which will be present in excess.[71Scheme 4 depicts several examples from the early stages of the development of this technique.
A similar problem is posed when A interacts as a chiral base
with the prostereogenic substrate A-CH,-B:
in this case the
enantiotopic groups pro-S-H atom H, and pro-R-H atom H,
must display different reactivities. Insofar as the resulting ion
pairs B and epi-B are configurationally stable under the reaction
conditions and substitution occurs stereospecifically (that is,
with either retention or inversion), the enantiomeric relationship G:ent-G reflects directly the result of enantiotopic differentiation between H, and H, (Scheme 5 ) . Thus, the newly created
stereocenter at the carbanionic C atom determines the stereochemistry of substitution.
However, with few exceptions (which will be discussed later),
configurational stability of the carbanionic intermediate cannot
be assumed;[’51the ion pairs B and epi-B are subject to equili-
H
E
epi-E
OH
OH
F
enf-F
Scheme 3. Enantiofacial selection in a prochiral carbonyl compound on use of an
ion pair containing an achiral residue.
bration. Scheme 6 summarizes the possible situations, where for
simplicity only stereoretentative substitution is illustrated. Under the simplifying assumption that k , z k,> keDi,the ratio
G :ent-G corresponds to the position of equilibrium B :epi-B.
Especially fortunate circumstances obtain when one of the
diastereomers B or epi-B crystallizes, so that in the reaction
mixture essentially only one diastereomer arises. This special
case of a dynamic kinetic resolution through crystallization is
also referred to as “asymmetric transformation of the second
order.”“ 61 An additional limiting case-kinetically controlled
Dieter Hoppe, born in Berlin in 1941, worked as a chemical technician in Hannover before beginning his chemistry studies at the University of Gottingen in 1965. He received his doctorate in 1970 with
U. Schollkopx submitting a dissertation on metalated isonitriles
and completed his habilitation there in 1977 with a topic drawn from
the realm of p-lactam chemistry. From 1977 to 1978 he was a
postdoctoralfellow with R. B. Woodward at Harvard University in
Cambridge (Massachusettsj .After serving as a Privatdozent at the
University of Gottingen he accepted a call to a C-4 professorship at
the University of Kiel in 1985. Thenfolloweda call to the University
of Miinster (1992), after a call he had declined to Hamburg
(1991). His work was recognized in 1993 with an Otto-Bayer
Award. He is a coeditor of “Synthesis.” His areas ofresearch relate
to the development of stereoselective and especially enantioselective
synthetic methods based on carbanion chemistry.
D. Hoppe
T. Heme
Thomas Hense, born in 1964 in Oldenburg, began his studies in 1987 in Miinster, and joined the research group of D. Hoppe
in 1992. He completed his dissertation in 1996 on the subject of kinetic resolution of racemates through (-)-sparteine-induced
deprotonation. In f 997 he entered upon a postdoctoral appointment with J. Otera in Okayama (Japan) in the field ofenantioselective catalysis.
2284
Angew. Chem. Int. Ed. Engi. 1997, 36, 2282-2316
REVIEWS
Enantioselective Synthesis
d
+
H
rrC,H,Li
chiral additive
-
L;Li H
H OH
PhX(CH&CH,
-
El H
entG
EIX
Mq
CH,
H3C
kR
B
A
y0
G
Me,N
enantiotopedifferentiatinq
deprotonation
NMe,
OMe
OMe
e x [a] = 53 : 47 (6%ee)[b]
e.r. = 67 : 33 (34%ee)[c]
[el
[91
[lo, 111
PhPh
e.r. = 93 : 7 (86%ee) [b]
e.r. z 99 : 1 (> 98%ee) [d]
Vbl
[I21
stereoselective
substitution
Scheme 6. Limiting cases in the substitution of chiral ion pairs
stereoselective substitution with dynamic kinetic resolution" 6a* dl-occurs when the rate constants k, and k, differ
o - ~ greatly,
g
and kcpimuch larger than these. In this limiting case
o- (e.g., k, >> k, > kepi),the enantiomeric ratio G/mt-G is determined solely by the quotient k,/k,."71
It should also be recalled that achiral carbanions, which in the
e.r. = 94 : 6 (88%ee)
ground state are planar and which bear three different sub1131
stituents at the carbanionic center, become chiral in the course
of interacting with a cation (Scheme 7). Since exchange of the
'9
x
\
ex. = 56 : 44 (6%ee)
equilibration
H
entH
Scheme 7. Chiral tight ion pairs and their racemization via achiral separated ion
pairs.
e.r. = 95 : 5 (90% ee)
~ 4 1
Scheme 4. Addition of complexes of n-butyllithium and chiral ligands to benzaldehyde with differentiation between enantiotopic faces. [a] e.r. = enantiomeric ratio;
[b] 2.0 equivalents of ligand utilized; [c] with 9 equivalents, 86% ee was achieved;
[dl With butylmdgneslum chloride.
L A
-R*-cH,
R
-H,
R
cation between the enantiotopic faces on the "top" and "bottom" of the benzylmetal compounds H and ent-H normally
occurs with exceptional ease, enantiomerically enriched ion
pairs are subject to rapid racemization, as demonstrated by D. J.
However, if the cation is provided with a configuratively uniform ligand, the result is epimeric, configurationally
stable ion pairs, to which the above argument applies.
Thus, the areas of application for chiral lithium carbanions in
enantioselective synthesis can be subdivided according to the
effective mechanisms :
1. Enantiofacial selection with achiral carbanions bearing a chiral cation.
2. Kinetically determined enantiotopic selection through deprotonation by a chiral base.
3. Thermodynamically controlled selection between epimeric
ion pairs.
4. Kinetically controlled selection between diastereomeric ion
pairs during a substitution step.
We will also discuss and then utilize the complications that
arise out of interaction with other stereocenters that may be
present.
2. (-)-Sparteine-Modified Ion PairsEarly Experiments and Disappointments
G
entG
Scheme Deprotonation of a prochlral CH acid ACH,B by a chiral base and
stereoselective substitution of the epimeric ion pairs.
Angeu. Chem. Int. Ed. Engl. 1997, 36. 2282-2316
The alkaloid ( -)-sparteineI' 91 (1) is extremely well-suited to
the chiral modification of cations in carbanion pairs,'''] because
in the only slightly more energy-rich conformation 1A (see
2285
REVIEWS
D. Hoppe and T. Hense
Scheme I)["] it can function as a bidentate ligand. (-)Sparteine is available in large quantity and is obtained through
the extraction of certain papilionaceous plants such as Scotch
broom.["] In contrast to its d i a s t e r e o m e r ~ [ ~( ~ ])-a-isosparteine (2) and (-)-p-isosparteine (3) (Scheme 8) it deviates
_.
-.
J~ N\
( -)-a-iscsparteine
2
tempts at rearranging cyclopropyl car be no id^;['^".^] the best
results achieved in each case are illustrated without comment in
Scheme 10.
J
58%; 3% ee [27d] [a]
(-)-p-isosparteine
1. s B u L ~ / ~ , - ~ O ~ C
3
WCH3
2.co2
15%; 30% ee [27a]
H
H
70%;
20
[a,
= +1.4 [27a]
ca.l%ee
4
5
6
Scheme 8. Various sparteine diastereomers, along with starting material 6 and intermediates 4 and 5 for semisynthetic transformations.
Scheme 10. Early applications of (-)-sparteine-mediated lithiation. [a] The configuration is unknown.
Only very small enantiomeric excesses have been observed in
slightly from C,-symmetry; it is true that 2 and 3 also occur
naturally, but they are best obtained by isomerization of (-)sparteine (1) via the dehydro derivatives 4 and 5.rz41 (+)Sparteine (ent-1) has been described as a constituent of the
It is most conveniently
shrub Sophorapachycarpa C. A . Mey.[251
obtained in the form of the racemic lactam rac-lupanine (6)by
extraction of seeds of the bitter lupine Lupinus albus, followed
by racemate resolution through the salt of (-)-camphor-10-sulfonic acid and finally by deoxygenation to give ent-1.[261All the
sparteine diastereomers can be easily recovered from alkaline
suspensions as a result of their low water solubility.
Nozaki, Aratani, Toraya, and Noyori[". 271 first studied the
suitability of (-)-sparteine (1) as a chiral additive in carbanion
reactions in the years 1968-70 (Scheme 9). The greatest enan-
PhCHO + C,H,MgBr/l
15%
ation utilizing palladium catalysis,r311methyltitanium trichloride addition to alkanals,[321and the acylation of lithiomethyl
p-tolyl s ~ l f o x i d e . [ ~ ~ ]
An astonishingly high chiral induction of 98% ee was
achieved in the case of 7 by M. Guette, J.-P. Guette, and J.
Capillon in the Reformatzky reaction of ethyl bromoacetate
and benzaldehyde in the presence of ( - ) - ~ p a r t e i n e [ ~ ~ I
(Scheme 11). However, the authors themselves and others as
well[35*361 established, for unknown reasons, much lower ee
values with substrates that differ only slightly.
EtO
HO H
L C H ,
Ph
61
Scheme 9. Early attempts at (-)-sparteine-mediated
benzaldehyde. [lo, 27a]
:
+
HO H
L C H ,
Ph S
39 (22%ee)
nucleophilic alkylation of
tiomeric excess (22 'YOee) was achieved by addition of ethylmagnesium bromide/l to benzaldehyde."n, 27a1 The procedure they
since today efficient
utilized is only of historical
catalytically asymmetric methods are known for nucleophilic
alkyl tran~fer.[~"*~.'*
Other early applications relate to the asymmetric lithiation of
isopropylferr~cene['~~*~]
and ethylben~ene,['~"'as well as at-
2286
( -)-sparteine-mediated conjugate addition to en one^,[^'^ allyl-
Y Z n B r I l
0
PhCHO
0
H OH
7 ; 46%; 98% ee
Scheme 11 (-)-Sparteine-catalyzed Reformatzky reaction [34a].
( -)-Sparteine-modified alkyllithium and alkylmagnesium
complexes found early application in stereoselective anionic
in this context is the reacp ~ l y m e r i z a t i o n .24b1
~ ~ ~Remarkable
.
tion of racemic I-phenylethyl methacrylate (rac-8) with cyclohexylmagnesium bromide/(l) , as well as with other Grignard
complexes (Scheme 12). Under kinetic resolution, (S)-8is largely transformed into isotactic poly(methacry1ate), whereas (R)-8
remains behind.[24b1.
Angew. Chem. Int. Ed. Engl. 1997, 36, 2282-2316
REVIEWS
Enantioselective Synthesis
POlY-I(W1
MgBrfl
52%:
up to 92% isotactic
toluene, -78 “c
0
CH,
83% ee
(W-8
Scheme 12 Anionic polymerization of a methacrylic acid ester with kinetic resolution via (-)-sparteine [24b].
Kinetic ’H NMR studies by Fraenkel et al. document relatively slow ligand exchange in dialkylmagnesium/( - )-sparteine
compIexe~.[~~]
The metal complexes of (-)-sparteine and its isomers display
a strong tendency toward crystallization. This has permitted
corresponding X-ray crystal structure analyses of various Grignard
an allylpalladium complex,[401and several
transition-metal salts.[411
White, Raston, et al. solved the structure of the (-)-sparteine
complex with (methylphenylphosphinyl)methyllithium[421(9),
which crystallized as a dimer, and that with l-pyrid-2-yl-l(trimethylsilyl)methyllithium[431
( l o ) , which can be regarded as
an azaallyllithium derivative. The structure of a-isocyanodiphenylmethyllithium/1/2THF (11) was characterized by
Boche et al. (Scheme 13).[441It is interesting that here lithium
9 [a1
Li-N, (cis)
221.3 (218.3)
Li-N, (trans) 216.2 (210.2)
10
11
206.0
202.0
206.5
210.3
Scheme 13. Li-N bond lengths [pm] in selected lithium/( -)-sparteine complexes.
[a] Two different aggregates are present in the unit cell.
seeks contact with the isonitrile C atom, but not with the carbanionic center. The bond lengths between lithium and the
(nonequivalent) N atoms of (-)-sparteine are quite variable:
202-216 pm and 206-221 pm. The structures of additional
lithium complexes that possess stereogenic carbanionic centers
are discussed in Section 3.
3. Enantioselective Syntheses with Configurationally
Labile Lithium/Sparteine-Carbanion Pairs
3.1. Lithiated Ally1 Carbamates and
Enantioselective Homoaldol Reactions
We established in 1985 that enantiomerically enriched secondary 2-alkenyl diisopropylcarbamates can be deprotonated
with retention of configuration using b u t y l l i t h i ~ m . [ This
~~]
Angew Chem. Int. E d Engl. 1997, 36, 2282-2316
formed the basis for an enantioselective variant of the homoaldo1
Not aware at the time of the previously described sparteine efforts, 0. Zschage began in 1988 a diplom a
thesisr4’] with the task of deprotonating the (E)-2-butenyl carbamate 12 in the presence of (-)-sparteine instead of the ordinarily employed achiral complexing agent N,N,N‘,N‘-tetramethylethylenediamine (TMEDA), and subsequently investigating the resulting capture products with respect to possible
enantiomeric excesses. We anticipated that the chiral base secbutyllithium/( - )-sparteine would distinguish between the
enantiotopic protons pro-R-H and pro-S-H in the methylene
groups. In that case, one of the epimeric ion pairs (R)-13.1 or
(S)-13.1 should be generated in excess, which would be discernible from a corresponding enantiomeric excess in the capture product 14 (Scheme 14, Table 1).
It quickly became apparent that the deprotonation here is not
kineticalIy controlled, but the process is nevertheless extremely
e f f i ~ i e n t . [ ~ ~In. ~the
* ] very first experiment deprotonation with
sec-butyllithium/( - )-sparteine (1) in isopentane/cyclohexane
and addition of tetra(i~opropoxy)titanium[~~~
to the suspension
of the organolithium intermediate, followed by 2-methylpropanal, led to the (2)-anti-configured homoaldol adduct 17 a
[R’ = CH(CH,),] with 83% ee. Because of the complete 1,3chirality transfer[s41during carbonyl addition, the titanated intermediate (R)-14 must have been present here with the same
enantiomeric excess, corresponding to the diastereomeric excess
in the (-)-sparteine complex 13.1. Nevertheless, the reaction at
first proved not to be reproducible; only a nearly racemic
product was subsequently isolated. In contrast to the successful
experiment, crystallization failed to occur after deprotonation.
Our suspicion was confirmed that dynamic kinetic racemate
resolution[551occurred in the course of crystallization; the
epimeric (R)- and (S)-13.1 are in equilibrium in solution, and
one diastereomer crystallizes preferentially. A certain amount of
cyclohexane is required for the crystallization, however, and this
had been introduced in sufficient quantity into the first experiment only due to the use of a sec-butyllithium solution in
isopentane/cyclopentane with the unusually low concentration
of 0.5 M. An optimized procedure, in which a defined quantity of
cyclohexane was added to the hexane solution and n-butyllithium was employed as the base, resulted unambiguously in
substitution products with 90-94% ee.[551
Metal exchange occurs under the refined conditions[551with
complete inversion of configuration, and the titanium intermediate (R)-14 is configurationally stable up to at least -40°C.
Thus the epimeric ratio of ( S ) - and (R)-13.1 determines the
enantiomeric ratio of the homoaldol adducts 17/ent-17. The
stereoselectivity is predominantly reagent-controlled; that is,
equally good results are achieved for the “mismatched
pair”[16c*d*
s61 with chiral aldehyde components.[57.581
The crystalline suspension of (S)-13.1 can also be trapped
with retention of optical activity using trialkyltin chloride
(Scheme 15).[”] This results, with low regioselectivity and
about 80% ee, in a mixture of the a-adduct (S)-20 (inversion)
and the y adduct (S)-19 from an anti-S,‘ reaction. However,
stannylation of the titanium intermediate (R)-14 leads with high
efficiency to the (IZ,3R)-configured stannane 19, from which
the tributylstannyl derivative (R)-19b is obtained with about
95% ee.
2287
D. Hoppe and T. Hense
REVIEWS
Ti(ORr),
____________
NPr,
(S)-14
crystallization
12
-
NRr,
Ti(ORr),
+
= OCb
H2.Cq
l
(R)-14
(R)-14
+
R'I?CO
-[
]-
RZ
H 3 C e 2Ti(ORr)a
-7O'C
R1$Zb
-
!
!
NPr,
1. Hg(OAc),
9' OM
_2._ BF3-OEtz
_ _ _ _ _ _I _mCPBA+
___________--
R1%
CH,
(R)-15
r
"z0L
~
16
17
OCb
M = Ti(OPr),
Scheme 14. Dynamic kinetic racemate
resolution in the (-)-sparteine-induced
deprotonation of 12; lithium-titanium
exchange and enantioselective homoaldo1 reaction [48.51,52,55].
-
18
M=H'
Table 1. Various optically active homoaldol products that were prepared.
~
~
~
~
17
R'
R2
Yield [%]
ee[%]
Yield(l8)[%]
Ref
a
(CH,),CH
CH,
CH,C(CHd,
CH,CCH,
(CH,),C=CH-CH,
H,C=CMe
H,C=rPrC
CH,
H
H
H
H
H
H
H
CH,
90
95
93
90
90[a]
SO[aI
84[a]
[bl
92
WaI
90
82
89
[bl
90
[481
[481
[481
1491
~501
b
c
d
e
f
g
h
62
I8
81
92
7O[c]
[bl
[bl
[bl
I
nBuLi I (-)-sparteine
1511
1511
1481
[a] Nonoptimized preparation of 13.1. [b] Not determined. [c] A different oxidation method was utilized; see text.
The stannanes 19 are storable homoenolate reagents that are
activated by the addition of titanium tetrachloride (Scheme 16,
Table 2) .[55,5 9 , 601 All indications suggest a stereospecific tin-titanium exchange (similar to that shown by Marshall for tin-tin
to give a trichlorotitanate, and pericyclic syn-S,'
addition of the latter to the aldehyde. Starting with stannane
(R)-19 one obtains the enantiomerically enriched homoaldol
adduct 22, whereas the antipode (S)-19 leads to ent-22. Thus,
both series of enantiomers are accessible via this Lewis acid
mediated variant from the same lithium intermediate (S)-13-1.
The strength of the variant lies in a smooth reaction with sterically demanding ketones.
The extension of thermodynamically controlled asymmetric
lithiation with a subsequent homoaldol reaction to the diisopropylcarbamates of (E)-2-hexen-l-o1, 3-methyl-2-buteno1, or
(E)-3-trimethylsilyl-2-butenol provided only modest enantiomeric excesses (41 - 7 6 % e e ) , because optimized conditions
for the crystallization had not yet been established.[481
2288
(inversion)
4
"30n
R,Sn
OCb
(S)-l9a R = Me: 48%; 82% ee
I
(S)-19b R = Bu: 22%; 82%' ee
R,SnCI
(anfi-S;)
(S)-2Oa R = Me: 21%; 82% ee
(S)-2Ob R = BU: 58%; 82% 88
R,Sn
OCb
(R)-19a R = Me: 73%; 86% ee
(R)-19b R = Bu: 80%; 95% ee
Scheme 15. Synthesis of enantiomerically enriched alkenylstannanes from the lithium complex ( S ) - W l [55].
Angew. Chem Int Ed. Engl. 1997,36,2282-2316
REVIEWS
Enantioselective Synthesis
TiCI,
H3c-l
R3Sn
OCb
1
H 3 C q T i C I ,
RLRsC=O
OCb
L
J
(R)-19
(R)-21
H , C p ,
R,Sn
I
TiCI,
OCb
(antis;)
I
H
3
C
9 TiCI,
1
RLRSC=O
OCb
(S)-19
Figure 1. X-Ray crystal structure analysis of ~‘-[(lS,2E)-1-(N,N-diisopropylcarbamoyloxy)-3-trimethylsilyl-2-propen-l-yl]lithium~(-)-sparteine(23.1)[63]. A11
hydrogen atoms have been omitted, with the exception of H-l and H-2. (C Atoms
gray, 0 atoms dark gray, N atoms black).
L
OCb
_]
enf-22
VC
Scheme 16. Lewis acid induced enantioselective homoaldol reaction with 3-stannyl1-alkenyl carbamates [55]. RS, RL: see Table 2; OCb = O,CON(iF’r),.
OCb
24
Table 2. Enantioselective Lewis acid catalyzed homoaldol reactions [55]
_____~
22
Starting
material
RL
RS
Yield[%]
eel%]
Ref.
22a
enr-22a
22a
22b
22c
22d
22e
(R)-19a
(S)-19a
(R)-19b
(R)-19b
(R)-19b
(R)-19b
(R)-19b
(CH,),CH
(CH,),CH
(CH,),CH
(CHd3C
EtOCO(CH,),
EtOCO(CH,),
CzH,
H
H
H
CH3
CH,
CH,
PhCO(CH,),
91
82
96
80
84
91[a]
67[a1
88
82
96
74
94
94
94
[55]
I551
[55]
[55]
[55]
[SS]
[621
[a] See text.
In collaboration with Boche et al. it proved possible to obtain
an X-ray crystal structure analysis of sparteine complex 23.1
(Figure 1) .I631 This revealed the (IS)-configuration at the carbanionic center and verified our earlier assumptions regarding
the structure of lithioallyl carbamates: they are monomeric even
in the crystalline state, because the lithium cation is offered
optimal preconditions for four-fold coordination in the a-position. The allylic system is I?’-bonded, and the electron-donating
carbamoyl 0x0 group and the nitrogen atom of the tertiary
diamine function as residual ligands. Firm attachment of the
cation at the a-position is the key to high y-selectivity in the
course of carbonyl addition. This also causes the observed torsional stability in most cases of the j?,y-double bondr641and
presumably contributes much to increasing the barrier to racemization in chiral lithium derivatives.
The lithiated secondary 2-alkenyl and 2,4-alkadienyl carbamates 24 and 25 (Scheme 17), readily accessible in enantiomerically enriched form, are configurationally stable below
- 70 “ C ;[451 to our knowledge they represent the first known
examples of allyllithiurn derivatives with configurationally
Angeu. Chem. h i t . Ed. Engl. 1997,36, 2282-2316
25
Scheme 17. Configurationally stable alkenyl carbamates 24 and 25. L,
______~
= TMEDA.
stable stereogenic, metal-bearing C atoms. As we reported in
1990, the enantiomerically enriched lithium derivative 24 can be
obtained not only from the optically active carbamate, but also
with sec-butyllithium/( -)-sparteine under kinetic resolution of
the corresponding racemic allyl carbamate 24 (H for Li/L, in
ra~-24).[~~7
601 The rich chemistry of the compound[45*
60*6 5 1
and 25,r661as well as their analogues, is governed by their torsional behavior; further discussion of this point would exceed
the bounds of this review, however.
The value of the enantiomerically enriched homoaldol adducts in stereoselective synthesis extends far beyond formal hydrolysis to carbonyl compounds; numerous stereoselective
transformations of the (Z)-vinyl carbamate group have been
developed by P. Kocienski as well as by us.[673Oxidative deblocking1711of adduct 17c or 17e (Table 1) leads to completion
of the syntheses of (+)-quercus lactone A[52,481
(18c) and (+)eldanolide (18e) .Iso1 This procedure is readily transferable to the
synthesis of diverse y-lactones (Scheme 18).
3.2. Benzyllithium Compounds
Primary and secondary benzyl N,N-dialkylcarbamates are deprotonated with the same ease as the corresponding allyl esters.[”’ Whereas the lithium derivatives of chiral secondary benzyl carbamates are configurationally stable at - 70 “ C , this is
not true for lithiated primary benzyl carbamates, although the
“Hoffmann test”[571for the complex 27.TMEDA demonstrates
2289
D. Hoppe and T. Heme
REVIEWS
18c
82%ee
18e
92%ee
H 3 C q %..Li
. I1
H+
o
0H\3:
R
c
HO
OCb
epimer leads to a dynamic resolution. The configuration of the
favored ion pair has not been established with certainty; since
we have found inversion in carboxylation of a similar complex
(29, CH, for l-H),[74a*b1
there are strong arguments for the ( S )
c~nfiguration."~*
761
P. Beak et al. succeeded in carrying out stereoselective lithiation of a series of benzyl systems; the results are collected along
with an extensive discussion of the mechanistic aspects in a
recent review.["]
N-Methyl-3-phenylpropionamide(30) reacts with two equivalents of butyllithium/( - )-sparteine to give the dilithium
derivative 31 .1,[771
which leads with various electrophiles to
amides substituted at the benzyl position with 60-94% ee
(Scheme 20). Similar results were achieved when 30 was pre-
Scheme 18. Synthesis of optically active fl-methyl-y-lactones by enantioselective
homoaldol reactions [SO]
increased stability at the microscopic
This can also be
concluded from the following series of experiments; Benzyl
N,N-diisopropylcarbamate (26) was deprotonated with secbutyllithium/( -)-sparteine in ether at 78 "C, and after 4 h the
epimeric mixture of ion pairs 27.1 was trapped by introduction
of carbon dioxide into the homogeneous reaction mixture
(Scheme 19). After esterification with diazomethane one iso~
0
so!vent,-70 "C
PhXOKNPr,
EIX
332b
32 c
32d
32e
32 f
sBuLi/l
m
Li
31-1
MeSiCI
CH,I
I
Yield ["A] ee [%]
94
86
84
70
-
1. EIX
2. HzO
26
Bu,Sc
\ /
(5)-27-1
_____)
(R)-27-1
MeO,C
H
(a-28
Et,O
hexane
hexane (solid)
hexane (solution)
Ms0,C H
*
A N , C H 3
HSC6
H
32a
(60% ee)
+
0
PhLOKNPrz
(W-28
(91%) 57:43 (14%ee)
(77%) 91 :9 (82% ee)
(50%) 95 : 5 (90% ee)
(18%) 69 :31 (38% ee)
Scheme 19. Investigations into the configurational stability of lithiated benzyl carbamates.
lates the mandelic acid derivatives (+)-(S)-28 and (-)-(R)-28
with only 14% ee in favor of (S)-28.Under identical experimental conditions but with hexane as the solvent crystallization
occurred, and (+)-(S)-28 was formed with 82 % ee. In a third
experiment the solution was separated from the crystalline mass,
and the two were treated separately; the enantiomeric excesses
amounted to 38% and 90% ee, respectively.
It is thus very probable that the lithium complexes equilibrate
even at - 70 "C, and that here also the crystallization of one
2290
Me,Sj
1. sBuLi
2. Me,SiCI
Scheme 20. Enantioselective substitution of N-alkyl-3-phenylpropionic acid
amides [77]
0
PhAOKNPr,
0
a N , C H 3
HSCB
I
H
321
(60%ee)
pared first in the absence of (-)-sparteine (1) and the latter was
added only subsequently. This argues in favor of equilibration
between the epimers 31.1, a suspicion that has recently been
verified.['*] On the other hand, racemization-free lithiodestannylation/silylation proved successful with the enantiomerically
enriched stannane 32f to give silane 32a; under these reaction
conditions the carbanionic intermediate was configurationally
stable. It is noteworthy that reaction in this case can be carried
out in THF without displacement of ( - ) ) - ~ p a r t e i n e . [Equally
~~~l
efficient are transformations of the 2-methoxyphenyl derivative; adducts of the type 32 (2-MeOC6H, for C6H5)offer the
possibility of transformation into enantiomerically enriched
co~marins.[~'"l
As in the case of 31.1, configurational stability at - 78 "C was
demonstrated for a series of additional benzyllithium-( -)sparteine complexes. Warming of the reaction mixture is required to cause epimerization. This opens up interesting pathways for controlling stereoselectivity, as will be demonstrated.
Angew. Chem. Int. Ed. Engl. 1991,36,2282-2316
REVIEWS
Enantloselective Synthesis
M. Schlosser and D. Limat["] discovered that N-Boc-Nmethylbenzylamine (33) is smoothly deprotonated by sec-butyllithium/( - )-sparteine (Scheme 21). The resulting enantiomeric
excess and the direction of chiral induction depend not only on
the solvent and electrophile, but also on the length of time the
H YS 0
d - N A 0 B
I u
sBuLV 1
solvent
*
Me
\
34-1 + epi34-1
33
H El
d,,KOfBu
2
1.. lEIX
h
Figure 2. X-Ray crystal structure analysis of z-(N-methyl-,V-piva1oylamino)benzyllithium.( -)-sparteine (36.1)[82]. All hydrogen atoms have been omitted
with the exception of the benzylic hydrogen atoms. (C atoms gray, 0 atoms dark
gray, N atoms black).
0
+
I
\
Me
35
ent-35
37
EIX
35 : en135 ee [%l
solvene
90
THF
95.0 : 5.0
7.5: 92.5
hexane
THF
hexane
THF
87.5: 12.5
10.0: 90.0
91.5 : 9.5
7.5: 92.5
75
DCaCPh hexaneIa1
CH,,
co*
85
80
81
85
Scheme 21 Enantloselective lithiation and substitution of N-Boc-N-methylbenzylamine (33) [79]. [a] The same excess enantiomer is formed in ether.
reagent was kept before quenching it. Only after about 2 h
at - 75 "C is maximum enantiomeric excess achieved; exchange
of hexane for T H F causes the configuration of the product to
be reversed, and precipitation of a solid is observed in T H F
after 1 h.
Through deprotonation and subsequent carboxylation of
(S)-[I 0133 with sec-butyllithium/TMEDA~solit was established
that the ion pair was configurationally labile in all the solvents employed at 75°C. The ratio of the ion pairs 34.1
and epi-34. I is therefore thermodynamically determined.[*'l
Boche et al. conducted an X-ray crystal structure analysis on
the N-pivaloyl derivative 36.1 and found it to possess the
(IS)-configuration (Figure 2).18'] What is the cause of the
solvent-dependent stereoselectivity of the substitution step?
The authors suspect that in both cases an ion pair with the
(S)-configuration is present, but the stereoselectivity in the
substitution step is reversed under the influence of the solvent. However, all the results are also consistent with the following interpretation: The predominant ion pairs in hexane and
ether solution possess a reversed configuration relative to the
aggregate obtained through THF crystallization, and the substitution step follows the same configurational pathway in each
case.
Beak et al. describe a (-)-sparteine-mediated lithiation of
N-Boc-N-(pmethoxypheny1)benzylamine (37), followed by
enantioselective alkylation or addition to carbonyl compounds
and imines (Scheme 22).IS3]In a formal sense the p r o 3 H atom
Angeu Chem Int Ed EngI
1997,36,2282-2316
3&1
- 0";:""
EIX
39
EIX
8 MeOTf
b EtOTf
c AllOTf
d BnOTf
Scheme 22. Enantioselective
rnethoxypheny1)benzylamine
All = CH,=CHCH,.
81%; 94% ee
78%;94%ee
69%; 93%73%;96%ee
lithiation and substitution of N-Boc-N-(p(37) [83]. Ar = 4-MeO-C6H,; Tf = F,CSO,;
in 37 is substituted, and the resulting enantiomeric excesses in
most cases range around 95% ee.
The configuration of the dominant intermediate 38.1 is unknown.[841As we have previously demonstrated for a lithiated
benzyl arba am ate,'^^^] it can be reversed by stannylationdelithiostannylation (Scheme 23) .Is3] It is important to note
here that the intermediates complexed with ( - )-sparteine are
configurationally stable. It follows that one can obtain the oppositely configured tertiary amines ( S ) -and (R)-41 by the same
sequence of methylation, deprotonation, and alkylation, with
subsequent oxidative d e b l o ~ k i n g . [ ~ ~ " ]
In the case of the N-Boc-N-(3-chloropropyl)benzylamine42a
it proved possible to show with the aid of the (@-deuterium
derivative 42 b that the (- )-sparteine reagent preferentially abstracts the pro-S-proton in 42 a (Scheme 24) .[8s1 Intramolecular
cycloalkylation of the (S)-configured lithium derivative 42 a - 1
to pyrrolidine (S)-44a occurs very rapidly, and therefore virtually free of racemization. Alkylation with stereoretention is an
astonishing
The reaction is transferable with comparable efficiency to a whole series of arylmethyl- and heteroarylmethylamines of the type 42.
N,N-Diisopropyl-2-ethylbenzamide
(45) was transformed by
P. Beak et al. into the (-)-sparteine-lithium complex 46.1[s61
2291
D. Hoppe and T. Hense
REVIEWS
-
phAN,~c
nBuLil1
I
Ar
37
Li I 1
,!,,~oc
CH,OV
Ph
.
943
phAN,Bo
nBuLiITMEDA
~
I
I
38 1
1.nBuliIl
2. Me,SnCI
.
nBuLi I 1
Ar
40-TMEDA
(S)-39a
Lill
Ph
H,C Li I TMEDA 1 , AllOTf
CH
Ai,&c
I
nBuLilTMEDA,
Ar
enf-38-1
2.CAN
I
Ar
enf40-TMEDA
BuLdl
Ar
.
I
Ar
(W-41
98% ee
90% ee (99% ee)
Scheme 23. Enantiodivergent synthesis of tertiary benzylamines [83a]. CAN = (NH,),Ce(NO&;
= p-MeOC,H,.
0
Ha Ldl
&BOC
toluene, -78 “C
____t
CI
xNOBoc
Ph
I
Ar
39e
97%; 90% ee
WH
I
90% ee (99% ee)
-
,Ay,&c
.
Ar
Ar
4
SnMe,
Ph
I
k
Ar
H,C
HaC Li I TMEDA 1, AllOTf
&3oc
2 CAN
Ph
CI
42
(S)434
(944
a H:=H
72%; 96% ee
bH=D
Scheme 24. Synthesis of (S)-N-Boc-2-phenylpyrrolidineby intramolecular substi42-44
48
tution [85].
and then alkylated. All experimental indications are that the
rare case of dynamic kinetic resolution (Scheme 6) has been
realized.[871The epimers (R)and (S)-46.1 are in equilibrium,
and one [presumably (S)-46.1]reacts preferentially. In the process, substitutions with alkyl halides and alkyl tosylates take
opposite courses (Scheme 25). No such strong dependence on
the leaving
had previously been observed for alkylating agents.
49-1
epi-49 -1
50
83:17 to 955
en150
Scheme 26. Enantioselective lithiation and substitution of 2-ethyl-N-pivaloylaniline [88].El = alkyl, Me,Si, and others.
45
(S)-46.1
,
I
(R)46.1
1
1
47
enf-47
Scheme 25. Enantioselective lithiation and substitution of 2-alkyl-N,N-diisopropylbenzamides [86.87].
The (-)-sparteine complex of the dilithium salt of 2-ethyl-Npivaloylaniline (49.1) is configurationally stable at - 78 “C. Only after equilibration between the epimers 49.1 and epi-49.1,
which occurs upon raising the temperature, does electrophilic
substitution lead to good enantiomeric excessesr881
(Scheme 26).
The two diastereomeric complexes 49.1 and epi-49.1 are
present before equilibration in roughly equal amounts. Their re2292
activities toward electrophiles differ; from subsequent experiments a difference AGGi+49.1- AG&.l = 3.4 kJmol-’ was established for trimethylsilylation at - 78 oC.[88b1
This can be exploited as follows (Scheme 27): If one adds to this mixture only
half an equivalent of trimethylsilyl chloride, primarily 49.1 is
trapped, with formation of the (R)-configured silane 50a. Excess
ally1 bromide transforms the residual epi-49.1 into the oppositely configured product ent-50b.
The configurational lability of lithiated intermediates at
higher temperature permits recycling of the less reactive epimer
into the other (“diastereomeric recycling”[88b1)
through a warming and cooling sequence (“warm and cool protocol”[88b1)as a
means of increasing the enantiomeric excess. As a result of
warming to - 25 “C, a 49.1:epi-49.1 ratio of 92: 8 is established,
which is then frozen by cooling the reaction mixture to - 78 “C.
A trapping experiment with excess trimethylsilyl chloride produces the enantiomers 50a and ent-50a in a ratio of 92:s (84%
ee). However, if the electrophile is added in two portions of 0.45
equivalents each, with warming to -25 “C in between, the result
is 50a with a 98 % ee (e.r. = 99: 1).
Angen. Chem. Int. Ed. Engl. 19!37,36,2282-2316
Enantioselective Synthesis
Piv,NOLi
W C
Lill
H ,
REVIEWS
3.3. Chiral Lithium Indenides
1. 0.5equiv
Piv,
,Li
N
+ &CH,?
li/l
MesSiCI
2.AllBr
The epimeric ( - )-sparteine complexes of I-lithioindenyl
N,N-dii~opropylcarbamate[~’]
(55a. 1 and epi-55a. 1) rapidly
equilibrate at temperatures around 0 “C, and produce upon
silylation the optically active silane ( +)-56a’921with only 16%
ee (Scheme 29). In a similar way, one obtains from the 2methylindenyl derivative 54b via the corresponding indenides
\
\
49-1
epiQ9 4
Piv,
,H
50a
ent-5Ob
39%; 52% ee
Piv,
N
/Li
P\ C
49.1
32%; 44% ee
H
,
+
+
-25%
1
+ Me,SiCI
50a
~\
\
Piv,N,Li
Ldl
NPr,
C
H
54
3
=OCb
a:R=H
b: R = CH,
I
SiMe,
epiQ9.1
R
i
\
ent-50a
Scheme 27. Differing reactivities of the diastereomeric ion pairs 49.1 and epi-49.1:
“diastereomeric recycling.”[88b]
a-Monosubstituted benzyllithium compounds thus assume
an intermediate position with respect to their configurational
stability, and in favorable cases their behavior can be prearranged by the selected reaction temperature. Configurational
lability permits efficient dynamic kinetic resolution at higher
temperature, whereas freezing the equilibrium between the
epimers opens the possibility of thermodynamically governed
reaction control.
V. Snieckus et al. established a high enantiomeric excess by
deprotonation of the 2-ethylphenyl carbamate 51 with subsequent silylation (Scheme 28) .[901 The configuration of the dominant products 53 and intermediates 52.1, as well as the extent
of the configurational stability, are so far unknown.
OCb
O Y
O
I
NPr,
56
55.1 + epi65.1
a: 80%; 16% ee
b: 97%; 6% ee
C: 63%; 51% ee with tj-a-isosparteine (2)
Scheme 29. Lithiation and silylation of indenyl carbamates 191. 961
55b.l and epi-55b.1 the silane ( +)-56b‘92] with only 6 % ee.
‘H NMR spectra in [DJtoluene could be obtained from both of
the indenide complexes 55a. 1 and 55b 1 (Figures 3 and 4), and
dkNEt2
2 equiv
sBuLil1
,Pr20,-78 OCC
x
_ -
51
I
7.6
a X=OMe
b X = SiMe,
7.0
6.4
-6
Figure 3. Excerpt from a ‘H NMR spectrum (300 M H r ) of [l-(~~”.N-diisopropylcarbamoyloxyj-lH-inden-l-yl)lithium.(
- j-sparteine (55a- I and epi-SSa 1) in
[D,]toluene at -15°C: [91]. The 5pectrum indlcales the presence of tuo
diastereomers in the ratio 60:40; these interconvert only slowly. if at ali.
53
a X = OMe; 50%; 92% ee
b X = SiMe,; 58%; 69% ee
Scheme 28. Enantioselective lithiation and substitution of 2-ethylphenyi N,N-diethylcarbamates [90].
Angrw. Chrm. Int. Ed. Engl. 1997, 36. 2282-2316
these verify the presence in each case of two diastcreomers in a
ratio of roughly 60:40.[91- 9 4 1 This in turn indicates an energy
difference AG,,, of 0.9 kJ inolSemiempirical calculations (MOPAC, PM3)[951predict for
the ground state energies of the complexes 55b.2 and epi-55b.2,
’.
2293
D. Hoppe and T. Hense
REVIEWS
vided by the I-butyl derivative 57b, although here the regioselectivity is reversed even sooner to the benefit of the y-adduct
(S)-60h.
Table 3 . Lithiation and stereoselective substitution of I-alkyl-3H-indenes [97].
I
I
I
7.5
7.0
6.5
-6
Figure 4. Excerpt from a ‘HNMR spectrum (300 MHz) of[l-(NJ-diisopropylcarbamoyloxy)-2-methyl-lH-inden-l-yl]lithium~(
-)-sparteine (55b. 1 and epi-55b. 1)
in [DJtoluene at 10 “C[91].The sharp signals of the “cyclopentadienyl protons”
H-3 and H-3’ (in the diastereomers) are A 6 ~ 0 . 1apart.
~
which bear the C,-symmetric (-)-a-isosparteine as ligand, a
higher difference of 2.5 kJmol- I , and the silylation experiment
in fact led to (+)-56b with 51 YOee.[961
To our great surprise, the ion pairs 58.1 derived from l-alkyl3H-indenes 57 proved to be the most efficient representatives
(Scheme30).[971As I. Hoppe observed as early as 1990, the
a
Starting EIX
material
Products [a]
El
59:60
57 a
57 a
57 a
57 a
57 a
57 a
57 b
57 b
(R)-59a
(R)-59 b
(R)-59c
(R)-59 d [bl
(R)-59e[b]
(R)-59f + (S)-60f
(R)-59g
(R)-59h[b] (S)-60h[b,c]
MeOCO
MeCO
PhCO
PhCHOH
tHuCHOH
Me,COH
PhCO
PhCHOH
>91:3
>91:3
>91:3
>91:3
Me0,CI
MeCOCl
PhCOCl
PhCHO
tBuCHO
MeCOMe
PhCOCl
PhCHO
+
Yield[%]
64
63
14
67
<3:97 55
31169 36
>95:5
79
35:65 52
[a] Unless indicated otherwise. the enantiomeric excess is greater than 95 % ee.
[b] Epimers with respect to the 1’-stereocenter. [c] The ee value could not be determined.
These results are best interpreted with an assumption of equilibrating dissolved ion pairs (1s)- and (1R)-58.1, from which
one epimer crystallizes and reacts stereoselectively as a solid at
low temperature with the carbonyl compound added as electrophile. It was long unknown which configuration applied to
the crystallizing ion pair. An X-ray structural analysisr971
was
ultimately successful for the dominant complex 58b. 1 (Figure 5). This established a (1s)-configuration along with y3-c0ordination for the indenide to the lithium cation. The C1 -Li
and C3-Li distances of 243.2 and 233.4 pm are unusually long.
1
57
nBuLi, EbO, 1
k
a: R = CH,
b: R =n-C,H,
El
q+
6
4
El
A
(IS)-58.1
crystallization
li
(1R)S8.1
K3-59
(RPO
Scheme 30. Lithiation, dynamic kinetic resolution of (1S)-58.1 by crystallization
and enantioselective substitution of I-alkyl-3H-indenes [97].
1-methyl derivative 57a is rapidly deprotonated by n-butyllithium/( -)-sparteine in ether, and a yellow solid crystallizes
upon warming the reaction mixture. After addition of an acid
chloride or an
the 1-substitution product (R)-59
is usually isolated with an enantiomeric purity greater than
95% ee. Only with bulky carbonyl compounds such as 2,2dimethylpropanal or acetone does a 3-substitution product of
the type (S)-60 predominate (Table 3). Similar results are pro2294
Figure 5. X-Ray crystal structure analysis of q3-[(1S)-l-butylindenyl]lithium.(-)sparteine 58b’l[47]. All hydrogen atoms have been omitted.
Carbonyl addition thus proceeds with retention. Apparently
the attacking carbonyl group reversibly displaces one coordination site of the ally1 anion with the formation of intermediates
61 or 62, which react further by transfer of the carbonyl compound with allylic inversion to give adducts 63 or 64
(Scheme 31).[991Route A should be favored because of the
weaker C1-Li interaction. Only when the C-C bonding step
leading to 63 is heavily burdened sterically should route B with
formation of the y-adduct 60 gain the upper hand.
The (-)-sparteine complexes 58.1 are quite labile. Addition
of THF causes one coordination site of the indenide and also
Angew. Chem. Int. Ed. Engl. 1997,36,2282-2316
Enantioselective Synthesis
@$
-
[
q
o
L
i
R’
route^
R
61
X R ’
REWEWS
/I-
(Wd9
63
k
.
)
an intermediate, which can be trapped with various electrophiles
to give products with high enantiomeric enrichments. Conclusive control experiments prove the configurational stability of
67.1 under the selected reaction conditions. The configuration
of 67.1 was derived under the assumption that transformations
with alkylating and silylating agents proceed as anti-E reactions,
and hydroxyalkylations with ketones as syn-E reactions.[’041
- -4.’
(S)-60
nBuLll
toluene. -78
Ph
R
R
62
Ph
OC
64
Scheme 31. Formation of the b- and y-adducts 59 and 60.
0
10
EIX
66
67.1
sparteine on lithium to be displaced; (1S)-58a-1 is transformed
into the ql-(THF), complex rac-65,which has also been characterized by crystal structure analysis (Scheme 32, Figure 6) .I9’, “]’
Reaction of the latter with benzaldehyde produces the completely racemic y-adduct me-60a. Quite generally,
the regioselectivity of substitution is strongly influenced by the
ligands on lithium, since in the complex 58-TMEDA there is a
considerable tendency toward y-substitution.“’”
HO
D
L
i
,
,
LiflHF),
+
Table 4. Enantioselective deprotonation and substitution of 66.
Major product
Yield[%]
ee [“lo]
68 a
74
73
92
95
Y4
96
96
YO
98
EN
CH3
(lS)-58a.l
C,H,
CH3
CH3
rac-65
raC-608
Scheme 32. Formation of the complex ruc-65 and its addition to benzaldehyde.
H,COTf
H,CI
H,C=CHCH,Br
PhCH,Br
68a
68b
12
70
46
49
77
68C
68dIaI
en1-68e[b]
enr-68 f [c]
Me,SnCI
(CH,),C=O
_____________
[a] Together with 24% of the psilane 69d (94% ee) [b] Together wlth 24% of the
7-stannane enr-69e. [c] El = 1-hydroxycyclohexyl.
In order to gain access to the opposite series of enantiomers,
the authors again utilized the sequence of stannylation of 67-1
followed by lithium-tin exchange,[”51 which led to ent-68
(Scheme 34). This lithiostannylation was carried out in the pres1. nBuLl1
N
Ar’
Me3Sn
‘Boc
66
I
Ph+YSn
N
N
‘Boc
Ar’
ent68e
49%; 90% ee
I
Figure 6. X-Ray crystal structure analysis of rac-(3-methylinden-l-yl)lithium.
(3THF) (ruc-65) [97]. All hydrogen atoms with the exception of those on C-l and
C-2 have been omitted.
Ar’
‘Boc
ent69e
24%; 90% ee
1 . nBuLi/l
1. nBuLl 1
2. AllylBr
2.AllylBr
3.4 Lithiated Cinnamylamides
Ph
As P. Beak and G . A. Weisenburger recently
one
of the enantiotopic methylene protons in (E)-N-(p-methoxypheny1)cinnamylamide (66) is released with outstanding differentiation by n-but)..llithiuml(-))-sparteine
(Scheme
4).
The q3-allyllithium compoundr103J67.1 has been formulated as
337
Angew Chem h i Ed Engl 1997, 36,
Ar’
N
’Boc
68
72%; 94% ee
Ph
Ar /“Boc
entab
60%; 74% ee
from enf68e
77%; 80% ee
from ent69e
Scheme 34. Configurational inversion through stannylation and lithiodestannyla
tion [102].
2282-2316
2295
D. Hoppe and T. Hense
REVIEWS
ence of (- )-sparteine, because uncomplexed 67 is configurationally labile.
The high y-selectivity and enantioxlectivity of the methylation remains intact in corresponding transformations of the
cyclohexylallyl derivative 70 as well, although the result is the
oppositely configured products 71 and 72, with ( Z ) - and ( E ) double bonds['061(Scheme 35). Since the enamides of the corresponding aldehydes are subject to hydrolysis, these experiments
demonstrate an additional route to enantiomerically enriched
homoenolate equivalents.
1. nBuLi/l, toluene
2. CH,I
Ar'
N
'Boc
70
71 43%; 84% ee
72 27%; 92% w
Scheme 35. Lithiation and methylation of the cyclohexyl derivative 70 [102].
Ar = 4-MeOC6H,.
- naphthalene
76
I-
1
RITOj
+ Li+ NaphthLR'
OR'
-naphthalene
)FL Li
R2
77 + ent-77
78
Scheme 37. Preparation of racemic a-oxyalkyllithium derivatives by reductive
desnlfenylation.
it proceeds via a configurationally labile radical intermediate 77
due to the single-electron transition involved. This reaction thus
leads to the racemic products 78.['13]
Formation via deprotonation is ruled out by the extraordinarily low CH-acidity of dialkyl ethers. The situation is more
favorable for esters because of the possibility of efficient complexation of the lithium reagent[114]and the associated increase
in kinetic acidity, as well as dipole stabilization of the ion pair,
although a competitive nucleophilic attack is a potential threat.
Thus, only lithiomethyl derivatives of a tert-butyl methyl
ether"' 51 (79) and sterically blocked benzoic acid esters (80" 16]
and 81" "I) could be obtained (Scheme 38); persistent homologues with primary alkyl chains cannot be generated in this
4. Configurationally Stable Chiral Ion Pairs through
Deprotonation of Achiral and Racemic Precursors
4.1. Preparations of Non-Resonance-Stabilized
1-HydroxyalkyllithiumDerivatives
Enantiomerically enriched 1-(a1koxymethoxy)alkyllithium
derivatives of the type 74 were recognized to be configurationally stable by W. C. Still and C. Streekumar in 1980; the
compounds fail to racemize below - 40 "C in ethereal solvent~.["~* They were prepared by lithium-tin exchange
starting from the precursor 73. Transmetalation is accomplished, as is the subsequent methylation to 75,with stereoretention (Scheme 36).[107,109,1101
73
Li-0
R
,CH3
CH3
R
74
79
80 R = iPr
81 R = fBu
Scheme 38. I-Oxymethyllithium
[I 15,116a, 1171.
derivatives
accessible
by
deprotonation
As expected, the methyl N,N-diisopropylcarbamate" 19] 82
is easily deprotonated by sec-butyllithiumlTMEDA in diethyl
ether or pentane as solvent (Scheme 39); addition of the lithium
intermediate 83.TMEDA to aldehydes and ketones to give diol
derivatives of type 84 is straightforward.['201
This reaction is applicable to the ethyl carbamate, although
"the faithful servants" that one called upon in the case of the
0/\OMe
75
Scheme 36. Preparation of enantiomerically enriched 1-oxyalkyllithium derivatives
by lithiodestannylation [107].
The tin derivative 73 was obtained through a laborious racemate resolution by way of diastereomeric esters, although simpler pathways are now known, most of which involve asymmetric reduction of acylstannanes." 'I
In contrast to the above-described lithiodestannylation, reductive cleavage of chiral monothioacetals 76 is not suitable for
stereoselective carbanion generation (Scheme 37) ,[I 'I because
2296
83
84
Scheme 39. Preparation and carhonyl addition of I-(N,N-diisopropylcarbamoyloxy)methyllithinrn (83) [120b].
Angew. Chem. I n f . Ed. Engl. 1997,36,2282-2316
REVIEWS
Enantioselective Synthesis
bulky carbamoyl group proved hard to dismiss; subsequent hydrolytic cleavage of N,N-dialkylcarbamates of saturated alcohols cause considerable difficulties, and only reductive methods
with large excesses of diisobutylaluminum hydride[”’] under
drastic conditions have so far led to success.
A general solution to the problem was achieved with IJ-oxazolidine-3-carboxylic esters of the types 89[1221and 90[1231
(Scheme 40) .[’221 2,2,4,4-Tetrasubstitution accomplishes extremely efficient shielding of the carbonyl group, but incorpora-
91
92
94
93
Scheme 41. Stepwise cleavage of the 1,3-oxazolidine-3-carbonyl
residue.
0
H
O
MeSO H
3
-
4.2. ( - )-Sparteine-Induced Deprotonation
of Achiral Alkyl Carbamates
clco(occ13)
,N-H
X
The deprotonation of “ordinary” achiral primary alkyl carbamates of types 89 or 90 with see-butyllithium/( -)-sparteine in
diethyl ether or hydrocarbons proceeds with a reIiable preference for the pro-S-proton to give the lithium derivatives 95.1 or
96.1, which can be substituted with various electrophiles under
stereoretention to give 97 or 98, respectively (Scheme 42,
Table 5 ) . The resulting enantiomeric enrichment is normally
85 R’ = CH,
06 2R’ = (CHJ,
07 R’ = CH,
00 2R‘ = (CH,),
09 Cb= Cby
90 Cb = Cbx
E
09
R’ = CH,
90 2R‘ = (CH,),
95 Cb= Cby
96 Cb= Cbx
97 C b = Cby
90 Cb = Cbx
2
Scheme 40. General synthesis of 1,3-oxazolidine-3-carboxylic
acid esters.
tion of an aminoketal group also ensures the presence of an
Scheme 42. Enantioselective deprotonation and substitution of “oxazolidine
carbamates” 89 and 90.
acid-labile potential cleavage site. The acid chlorides 87 and 88
required for introduction of an activating 0-protective group
are obtained by
of 2-amino-2-methyl- Table 5. Selected examples of the ( - )-sparteme-mediated lithsation and substitution of
propanol with acetone or cyclohexanone to give the distillable alkyl carbamates.
oxazolidines 85 and 86, followed by chlorocarbonylation with Product R
El
EIX
Yield[%] Config.[a] Ref
diphosgene.[’22.
The carbamates 89 and 90, generally obCH,
C0,Me
CO,[b]
75
R
[123,130]
tained by the acylation of alcohols with acid chlorides 87 and 88, 98aa
Me,%
Me,SnCI
76
S
[123,130]
98ab
CH,
are present as E / Z mixtures, which in the time frame of 97ac
CH,
PhCHOH
PhCHO
76
R
11301
60
R(42)
[130]
CH,
H,C=CHCH, AllBr
‘H NMR spectroscopy interconvert only slowly. The conse- 97ad
97aa
CH,
C0,Me
MeOCOCl 73
R
~1301
quence is a doubling of the signals when ‘H NMR spectra are 97ae
PhC=O
PhCOCl
38
CH,
R
~301
recorded without special precautions.[’251Spectroscopic evalua- 97af
CH,
Me,Si
Me,SiCI
86
S
(1 301
CH,
Me,%
Me,SnCI
72
S
[1301
tion is less hampered for Cby-esters of the type 89 than for the 97ag
97ah
CH,
Me,Pb
Me,PbBr
61
S(93)
11301
originally utilized spirocyclic Cbx-esters 90.
R
[123,130]
(CH,),CH
CO,
52
97ba
CO,Me[b]
[123,130]
62
S
Me,SnCI
(CH,),CH
Me,Sn
The 2,2,4,4-tetramethyl-1,3-oxazolidinylgroup corresponds 97bg
S
[123,130]
87
CH,
CH,I
H,C(CH,),
approximately in terms of spatial demand to a di-(tert-butyl- 97ci
R
[123,130]
H,C(CH,),
co,
79
97ca
C0,Me
amino) group, so it is hardly surprising that secondary alkyl 97da
88
PhCH,CH, C0,Me
co,
R
[1311
Ph,PCI
70
FcCH,[e]
Ph,P
S
11321
esters 91 (R and El = alkyl) usually are not attacked even by 97ek
S(98)
[122,130]
CH,
9
7
f
i
H,C(CH,),,
CH,I
60[d]
lithium aluminum hydride[’261 in boiling T H E Stirring in
64
S(192)
[130]
CH3
CH,I
If1
979i
methanol containing methanesulfonic acid releases the N-@-hydroxyalkylurethane) 92 from the N,O-acetonide 91. Thereafter, [a] The enantiomeric excess is greater than 95% ee. Deviations are indicated in parentheses. [b] Subsequent methylation of the carboxylic acid with diazomethane. [c] Mixture
the hydroxy group plays an active role in base- or acid-catalyzed of epimers [d] Use of 3 equivalents of sec-butyllithium/l. [el Fc = ferroceny!.
deblocking to alcohol 93 (Scheme 41).
I f I W,C),C=CH-(CHdz.
Rngew Chem. in?. Ed. Engf. 1997, 36, 2282-2316
2297
D. Hoppe and T. Hense
REVIEWS
more than 95% ee. Suitable reaction partners include methyl
iodide, CO,, alkyl chloroformates, aldehydes, ketones, carboxylic acid chlorides and esters, and trialkylsilyl and trialkyltin
chlorides. Partial racemization is observed in reactions with
benzylic and allylic halides, which speaks for the involvement
here of single-electron transfer
in the substitution
step, which is favored by resonance stabilization of the benzylic
or allylic radical. Primary alkyl iodides can be caused to react[’281 by the addition of HMPTA-substitutes such as 1,3dimethyltetrahydro-2( 1H)-pyrimidinone (DMPU) .[1291 It is especially noteworthy that acylation occurs without racemization.
This points to weak kinetic basicity of the lithium compounds 95
and 96, which presumably has its origin in the severe steric
burden of the lithium cation, that in turn hinders “docking” at
the ketone on the carbonyl group and thus activation of the
neighboring CH-bond.
Methylation product 97fi led, with 98 % eerl”] after deblocking and acetylation, to (S)-( )-Ztridecylacetate 99,a pheromone
of the fruit fly Drosophila muelleri,[’331 whereas (S)-( -)-sulcato1 (100),1’34a1a pheromone of the beetle Gnathotricus sulcat ~ s , [ ’ (92%
~ ~ ~ ee)
] was obtained from 97gi (Scheme 43).11301
+
0
n
U
I
Scheme 45. General synthetic strategy.
displacement of (-)-sparteine by Lewis-basic substituents in
the substrate; opposite stereochemical preferences in chiral substrates through the formation of “mismatched pairs.”[56]
First, however, we discuss insights gained to date into the
mechanistic pathway of (-)-sparteine-mediated deprotonation.
4.3. On the Mechanism of Kinetically Controlled
Carbarnate Deprotonation
A series of simple experiments[’ 361 with the deuterium-labeled substrate [lD197a supports the validity of the previously
described interpretation (Scheme 46):“ 371 Through double application of the sequence (- )-sparteine-induced deprotonation
’2
CH,
Cbfl
1. SBULill
2.MeOD
97a
100
99
- ‘2
CH,
CbyO
I
[1Dl978 >99% D
Scheme 43. Synthetic insect pheromones *[I221 and 100[130].
Unlike in the case of the lithium ion pairs of resonance-stabilized a-carbamoyloxyalkanides (see Section 3.1), substitution
reactions of the saturated analogues 95.1 and 96.1 always occur
with retention of configuration. Thus, the synthetic ester 97aa is
identical with a sample prepared from (R)-lactic a ~ i d . 1 ”(R)~~
97aa was also obtained via the detour of stannylation of 101.1
to (S)-97ag and destannylation with n-butyllithiumlTMEDA by
way of the sparteine-freecomplex 101 .TMEDA (Scheme44).“ 231
1. sBuLi/l
2. Me,SiCI
97af
sBuLil
TMEDA
n SM+.Me,SiCl
CbyO
2
97at
CH,
ent-lola-TMEDA
[lD}lOl8-TMEDA
enf-[lD]97a?
>96% ee;
98.7% D
Scheme 46. Determination of the kinetic isotope effect k,lk, in the deprotonation
of an alkyl carbamate [136].
SBULil
H,C,SnMe,
J
Me,SnCI
-
H3C)
sBuLill H,C,.Ln
i
OCby
97a
OCby
101.1
OCby
97ag
1. CO,
2. CH,N2
TMEDA_H,C
and deuterolysis of the organolithium intermediate it was posYLflMEDA
sible to isolate the a-deuterioethyl carbamate [lD197a with a
OCby
101-TMEDA
1. co2
12.CH,N,
- H3cY.C0fle
OCby
97aa
Scheme 44. Stereochemical correlation of the metal derivatives with (R)-lactic acid
[123].
’
Since according to all previous experience[’09* 351 the
lithiodestannylation step is always accompanied by retention,
stannylation must also have taken this configurative course.
Thus, the process permits a broad range of electrophilic substitution reactions of pro-S-protons in alkanols; the lithiated
carbamates 95.1 or 96.1 correspond to the chiral synthon I
(Scheme 45).
Irregularities can arise if the starting material contains heterosubstituents. The following causes will be discussed in Sections
4.1 -4.6: 1,2- and 1,3-elimination of nucleofugic leaving groups;
2298
1-D,-content higher than 99 YO.An attempt at renewed deprotonation with sec-butyllithium/( - )-sparteine (1) and subsequent
trapping with chlorotrimethylsilane failed; the anticipated
silane 97af was detectable at most in trace amounts. Deprotonation of [1D]97a with the achiral base pair sec-butyllithiuml
TMEDA, on the other hand, is not subject to any stereochemical restriction: the ratio of rates according to which reaction
pathways A and B proceed is determined exclusively by the
kinetic isotope effect k,/k,. The reaction produced, after silylation, the silane ent-IlD197af with greater than 96% ee and a ‘H
content of less than 1.3%.[’381From this one can calculate a
k,/k, ratio of > 70.[1391
The following conclusions can be drawn from these results: 1)
Lithiated alkyl carbamates are completely stable configurationally in the form of TMEDA or (-)-sparteine complexes
under the reaction conditions. 2) The deprotonation step is kinetically controlled, and this step determines the stereochemical
course. 3) A pro-S-proton is abstracted with high selectivity
Angew. Chem. Inl. Ed. Engl. 1997,36,2282-2316
REVIEWS
Enantioselective Synthesis
89
+ sBuLi/ 1
configurationally stable
Scheme 47. Diastereomorphic transition states in the (-)-sparteine-mediated deprotonation of ethyl carbamate 97a with sec-butyllithium
under the influence of (-)-sparteine, and the incoming electrophile occupies its topochemical position. 4) The observed
kinetic isotope effect k,/k, of at least 70 points to the intervention of a quantum-mechanical tunneling
I4O1 Powerful H/D tunneling effects are observed when the reactants and
products are separated by high, very “thin” potential barriers.
This applies to proton transfer when the base and the resulting
carbanion have similar basicities, and when the reacting centers
are heavily shielded sterically. To our knowledge, the largest
H/D kinetic isotope effect for a deprotonation reaction (kH/
k , = 24.3) was determined in the deprotonation of 2-nitropropane with 2,4,6-trimethylpyridine.“40b‘
Additional experimental facts are also important if one wishes to propose a solid mechanistic model: 1) Without addition of
a bidentate ligand such as TMEDA or (-)-sparteine ( l ) , deprotonation of “simple” alkyl carbamates is excluded. 2) A reactive
alkyl carbamate like 89 or 90 added to an equimolar mixture of
sec-butyllithium, (-)-sparteine ( l ) , and the (not deprotonateable) isopropyl carbamate iprOCby at - 78 “C is also not attacked.[’41] This means that prior to the deprotonation step a
complex consisting of alkyllithium, the bidentate ligand, and the
carbamate forms virtually irreversibly. Beak et al.r’4z1verified
by NMR spectroscopy the presence in ethereal solution of an
unsymmetrical complex of the composition [(RLi),(Et,O); l ) ,
which is, however, under the conditions applicable here for the
kinetics of the deprotonation step of no significance.
Proton transfer in the aggregate 102.1 is turned in this way
into an intramolecular process with differentiation between
diastereotopic protons (Scheme 47). Abstraction of the pro-S-H
from conformation A occurs roughly 50 times more rapidly than
that of the pro-R-H from conformation B, which leads to the
minor diastereomer epi-103.1. From this result a free energy
difference AC* of about 6.3 kJmol-’ is calculated.
The limits of the deprotonation of very weak CH acids have
nearly been reached; it is apparent here that the combination
sec-alkyllithium/( -)-sparteine was a fortunate choice, and with
Angen. Chem. h
i .
Ed. Engl. 1997, 36, 2282-2316
respect to the basicity and steric demands of base and ligand an
optimum has almost surely been reached: Deprotonation of
nonactivated alkyl carbamates does not occur with use of n- or
tert-butyllithium/( -)-sparteine (1) or sec-butyllithium/( -)aisosparteine (2) .[1311
J. Haller and E.-U. Wiirthwein simulated the energy minima
for transition states along the diastereomorphic reaction pathways A and B with the aid of semiempirical calculations
(MOPAC, PM3) .I9’] For this process, sec-butyllithium was reand lithium parameplaced by the achiral is~propyllithium,[’~~~
ters from A n d e r ~ ‘ ~were
’ ~ ] utilized. In both transition structures
104.1A and 104.1B (Figure 7) the reaction center inserts into a
“niche” that is left free by a “wing” of the ligand (-)-sparteine.
Figure 7. Models of the calculated diastereomorphic transition states (MOPAC,
PM3) in the intramolecular removal of thepro-S-proton (left) and thepro-R-proton
(right) in carbamate 97a by isopropyllithium/( -)-sparteine[l41]. In the case of
sparteine (top of both illustrations) all hydrogen atoms have been omitted except for
those at the bridgeheads; the same applies to the 2,2,4,4-tetrarnethyloxazolidin-3carbonyl residue (bottom left). The view into the reaction center [the “triangle”
consisting of the lithium cation (violet), C-1 of the substrate (left), the proton to be
transferred (white), and C-2 of the base (right)] has been held constant. In both
transition states, the bulky base (bottom right) extends into the niche left free by the
cis-annelated outer six-membered ring of the sparteine. The pro-Stransition state
(left) corresponds to the experimentally observed stereochemical course; here the
methyl group (bottom left) extends forward. In the pro-R-transition state (right) it
points downward; here the distance to the isopropyl group of the base IS smaller,
and therefore the steric interaction greater.
2299
D. Hoppe and T. Hense
REVIEWS
The structures differ from each other in that the urethane alkyl group
in structure 104.1Aextends into the
free space, whereas that in 104'1B
collides with the isopropyl group.
The calculated energy difference
AAG* of 0.8 kJmol-' is much too
small in comparison to the experimental value of 26.3 kJmol-'; we
Figure 8. Models of the calculated diastereomorphic transition states (MOPAC, PM3) for removal of the pro-R-proton
suspect that the calculation method
in a complex consisting of methyllithium/ethylenediamineand (S)-1-phenylethyl formate (left) or (R)-1-phenylethyl
selected for the exact solution of
formate (right)[l45]. In the transition state a triangle is formed consisting of the now pentacoordinated lithium cation
this complex problem is, like other
(violet), the (also pentacoordinated) carbon atom of the base, and the C-H acid, on the basis of which the attacked
proton wanders. The calculated energy difference AAHin favor of the ul-proton (left) relative to the lk-process (right)
semiempirical methods, inadequate
amounts to 2.9 kJmol-'. In the most favorable conformations in each case the plane of the phenyl residue is parallel to
for the task, which is complicated
the C-H bond at the stereogenic center in order to avoid 1,3-allyl strain. In the favorable ul-transition state (left) the
further by the appearance of tunnelC1 -C2 bond can assume a completely eclipsed conformation at the expense of an interaction between an ortho-H atom
and the remaining I-H atom, whereas that conformation in the Ik-transition state apparently represents a poor comproing
mise. The two experiments are possible only by computer, since in a reaction flask the system would discover that removal
Better agreement between experiof an N-H proton and subsequent attack of lithium amide on the formyl group would be the favored reaction pathway.
mental and calculated results is obtained when diastereotopic differentiation is established through a
neighboring stereogenic center in
the substrate:['451 (R)-2-Phenylpropyl ~ a r b a m a t e t ' ~105
~ ] leads after deprotonation and methoxycarbonylation to the diastereomeric
esters 108 and 109 in a ratio of 95: 5
(Scheme 48); that is, the pro-S-proton is removed roughly 20 times
Figure 9. Models of the calculated diastereomorphic transition states (MOPAC, PM3) for removal of the pro-R-proton
more rapidly than the pro-R-proton.
-yl)methyl
in a complex consisting of methyllithium/ethyIenediamine and @)-(left) or (S)-(1,2,3,4-tetrahydronaphth-l
formate (right)[145J. In this case calculations (AAK = 2.9 kJmol-') and experiment (AG = = 2.9 kJmol-') show the
However, in (l-tetrahydronaphIk-transition state (left) to be the more favorable. Since the phenyl residue and the attached alkyl group are constrained
thy1)methyl carbamate 110, which
in a plane by the ring, the conformational energies are reversed relative to the txJ-(l,l')-C-C bond. In the ul-transition
has an equivalent configuration,['461 state (right) the interaction of the C-2-methylene group of the tetrahydronaphthyl residue with the reaction center la
apparently stronger than that of the Ik-transition state (left).
a reverse preference for 113:114 is
registered with a ratio of 13:87.r'471
Although the PM3 c a I c ~ 1 a t i o n are
s ~ ~carried
~ ~ out with simreagents (MeLi for sec-BuLi and ethylenediamine for TMEDA),
plified substrates (formyl in place of oxazolidinecarbonyl) and
they correctly reflect the observed trend (Figures 8, 9).11481
Calculated energy differences A(AH2 - AH:) correspond
well with the experimentally determined values A(AG2 - AG:
in parentheses): 2.5 (5.0) kJmol-I for the system 105 and -2.5
( - 2.5) kJmol-' for 110.
105
110
It is apparent that significant differences in the energies of
diasteromorphic reaction pathways for a kinetic racemate resolution under the influence of the chiral base pair sec-butyllithium/( -)-sparteine are susceptible to exploitation.1145*
1491 The
enantiomer that reacts most rapidly is the one in which the pro-S
preference of substrate matches that of the reagent, namely
(R)-105 and (S)-l10.c145.
1491
I I
106-TMEDA
107;TMEDA
111-TMEDA
1. co,
112-TMEDA
1. co,
2.CH2N,
108
955
109
113
ma7
314
Scheme 48. Diastereotope-differentiating deprotonation of chiral alkyl carbamates
[145].
2300
4.4. Deprotonation of Heterosubstituted Alkyl Carbamates
As described in the preceding section, the formation of a
complex consisting of alkyl carbamate, sec-butyllithium, and
( - )-sparteine is essential for achieving highly stereoselective
deprotonation. If the alkyl group bears a strong donor substituent, this may displace the sparteine.
Thus, deprotonation of the 3-(N,N-dimethylamino)propyl
carbamate 117a, followed by reaction with various electrophiles, leads to the nearly racemic product 118a
(Scheme 49) .[150. 15'] The bulkier N,N-dibenzylamino group on
Angew. Chem. In(. Ed. Engl. 1997,36, 2282-2316
REVIEWS
Enantioselective Synthesis
1. sBuL/l
2. co,
3.CH,N,
-RiR2Nq0cby
R’R‘N-OCby
117
cb@&ocby
R’
R2
Yield [%] ee [%I
a
b
CH,
CH,
CH,
PhCH,
92bl
70
c
PhCH,
PhCH,
94
A
w
122-1
CIO
80
80%; 95% ee
-
Cb@*ocby
-
Cb@&Ldl
H OCby
Li
I;;r
4
122-1A
or
”‘.H
-y - Y o c b y
OCby
125
I
El
122-1B
I
4
tEuMe$iiTf or
BF3-OEt,
" -OCby
H
ent-125
I
1. sEuLirrMEDA
2. CIC0,Me
1-20
Scheme 50 Enantioselective deprotonation and substitution of o-heterosubstituted alkyl carhamates.
Table 6. Enantioselectibe deprotonation and substitution of achiral w-heterosubstituted alkyl carbamates 119. TBDMS = ierf-butyldimethylsilyl, MEM = methoxyethoxymethyl.
Starting Product
material
n
Y
EK
Yield[%]
ee [YO] Ref.
119a
119b
119c
119d
119e
119f
119g
1
2
2
2
2
3
0
OCby
OCby
OMe
OTBDMS
OMEM
OCby
NBn,
Me1
Me1
Me,SnCI
CO,[a]
CO,[a]
Me,SiCA
CO,[a]
83
92
70
77
69
70
56
> 91
97
99
> 95
62
96
> 95
124
Scheme 51. Synthesis of (R)-pantolactone [153a].
1 . sBuLiI1
2. ED(
[153a]
[153a]
[153a]
[153a]
11551
[153b]
[is41
[a] Isolated as the methyl ester after treatment with diazomethane
Compound 120g is a protected derivative of D-isoserine.[1541
Carboxylation of lithiated 1,3-dicarbamates leads in two steps
to essentially enantiomerically pure y-lactones, as illustrated in
an exemplary way by the synthesis of D-pantolactone (124)
(Scheme 51).[153a,
1,3-Dicarbamates like 122.1 (Scheme 52) bear potential leaving groups that are activated by Lewis acids. In fact, we observed upon treatment with chlorotrimethylsilane the formation
of the cyclopropyl carbamate 125 with 1,3-elimination of the
lithium carbamate salt.[136.
15’] In the process the ( S ) enantiomer 125 results with greater than 95% ee, as could be
demonstrated by transformation into the carboxylic acid ester
126 and crystalline ketone 127 via the configurationally
stable“
intermediate lithium cyclopropanide. Investigations
with the deuterium labeled, enantiomerically enriched starting
material (S)-[lD]121 proved without question that ring closure
Angew. Chem. Int. Ed. Engl. 1991,36,2282-2316
U
123
97
the other hand does not interfere, and the product 118c of
electrophilic substitution is obtained with the customary enanIn this respect
tiomeric enrichment of greater than 95 % ee.[1501
the (N-benzyl-N-methylamino) group in 117b (80 YOee) occupies
an intermediate position.[”’]
The 3-, 4-, and 5-~arbamoyloxy,[’~~~
4-methoxy or 5-silylOX^,[^^^^^ and 2-dibenzylamino groupsr’541 do not disrupt the
enantioselectivity of deprotonation in the presence of ( -)sparteine (Scheme 50, Table 6).
120a
120b
l20c
120d
l2Oe
120f
12og
Cb@&OCby
a,Me
Scheme 49 ( - )-Sparteine-induced deprotonation of 3-(dialkylamino)alkyl carbamates [152]. [a] CH, instead of C0,Me.
119
c
121
118
Y 4 0 C b y
SBULdl
1. sBuLirrMEDA
2. CICO,Me
4
4
126
ent-126
“‘-CO,Me
OCby
““OCby
C0,Me
&CH3
OCby NEn,
127
Scheme 52. Stereochemistry of the 1,3-elimination of a 1,3-dicarbamate [157,160].
here occurs stereospecifically, with retention at C-I as well as
inversion at C-3, hence out of conformation 122.1a.[’361
To our
great surprise, stronger Lewis acids like tert-butyldimethylsilyl
triflate or boron trifluoride caused formation of the opposite
enantiomerically enriched cyclopropane, ent-125 ( > 95 YO ee
and 74% ee); stereoinversion at both reactive centers was verified.[’60 - 1631
Despite many possible complications, the carbamate group is
distinguished by its unsurpassed directing effects. The following
“acid test” (Scheme 53) underscores this assertion: Carbamate
1. sBuLdl
2. PhCOCl
OCby
OCby
0P
‘h
128
129 85%; >95% ee
Scheme 53. Competing acidic positions in (-)-sparteine-induced lithiation [165]
2301
D. Hoppe and T. Hense
REVIEWS
128 bears three extraordinarily easily activable protons in H,,
H,, and HB.;[1641nevertheless, under the usual conditions of
( - )-sparteine-induced deprotonation, exclusively the pro&
proton HA is released.f1651
epi-134 with trimethyltin chloride produces mainly the (1S,3S)diastereomer 135 (Scheme 55, Table 7). From this one can conclude that there is highly selective formation of the anti-annelated tricyclic chelate complex 134.f16831701
In the presence of
( -)-sparteine (1) the tendency toward release of the p r o 4 pro-
4.4.1. Competition with Stereogenic Centers in the Substrate;
Kinetic Resolution
0q
Provided that a good donor substituent is attached to a
stereogenic C atom in the y- or &position, it can ensure a significant amount of chirai induction in the deprotonation step. It is
useful when this chelate effect is subordinate in the competition
with (- )-sparteine (1). In favorable cases-as with 1,3-dicarbamate 130a-lithiation can then be directed preferentially in
either of the two directions.[1661Application of the sparteine
variant with subsequent methylation led to the (S,S)-2,4pentanediol derivative 132a with high selectivity, whereas in the
presence of TMEDA the meso-compound epi-132a dominated
(Scheme 54).[’661We suspect that a bicyclic chelate complex of
O
PO
-
C
b
y
133
HsHR
1
134
epi-134
o-n(ocby
OCby
El
El
-
OH
-0Cby
Me0
H
1
V
131-1
I
+ CH,I
132[a]
a n = 1; 56%; 945 [b]
b n = 2; 51%; >98:2
Scheme 55. Diastereoselective lithiation of the chiral acetonide 133 [169]
Table7. Deprotonation of the acetonide 133 and reaction with various electrophiles 11691.
--
--
CH,
CH,
epi-132
a n = 1; 44%; 98:2 [c]
b n = 2; 60%; 12:88
Scheme 54. (-)-Sparteine- vs. substrate-directed diastereoselective lithiation of
1,3- and 1,Cdicarbamates [166]. [a] Conducted with the carbamate residue Cbx.
[b] Ratio 132:epi-132 [ = (S,S):(S,R)]. [c] Ratio epi-132:132 [ = (S,R):(S,S)l.
the type 131[1671
forms, and the more favorable exo-position of
the stationary methyl group determines the transition
In the homologous (S)-2-methyl-l,4-butanediyldicarbamate
130b, reagent- and substrate-controlled deprotonations take the
identical stereochemical course, and lead with differing efficiency to the (S,S)-diol derivative 132b.
An effective substrate-inherent chiral induction can also be
exploited with the acetonide of the (S)-3,4-dihydroxybutyl carbamate 133.[16’]Deprotonation in diethyl ether (without any
other additive) and reaction of the resulting ion pairs 134 and
2302
136
+ CH,I
Cb@&OCby
cb@ +OCby
CH, CH,
1
OMe
K
131
1
epi-135
Additive L
Products
Em
Yield[%]
135:epi-135
Et,O
TMEDA
1
ent-1
Et,O
Et,O
Et,O
Et,O
Et,O
Et,O
Et,O
Et,O
135a
135a
135a
135a
135b
135b
135c
135e
135f
135g
135h
135i
Me,SnCI
Me,SnCI
Me,SnCI
Me,SnCI
Me1
MeOCO(0Me)
HCOOEt
iPrCOCl
Ph,CO
[a1
[bl
[CI
63
71
61
15
98:2
15: 35
>99:1
28:12
>95:5
98:2
96:4
>95:5
>95:5
10
35
69
57
16
39
51
62
>95:5
>95:5
>95:5
[a] (E)-Crotonyl chloride. [b] (S)-2-(N,N-Dibenzyl)alanine benzyl ester. [c] 6Valerolactone.
ton is further increased, because in this case epi-135a is no longer
identifiable in the reaction mixture. The 2,2-dimethyl-substituted 1,3-dioxolane ring is only a weak ligand for lithium, for in the
deprotonation step, as with sparteine, it is displaced by
TMEDA. As Table 7 indicates, diastereoselectivity decreases
Angew. Chem. Int. Ed. Engl. 1997,36,2282-2316
Enantioselective Synthesis
REVIEWS
under these conditions and is reversed by means of (+)sparteine (ent-1)" 'I with its preference for the pro-R-proton. In
other words, in the presence of a bidentate complexing ligand,
intramolecular complexation is not involved, and the normal
deprotonation pathway applies. Conformational rigidity in the
form of the acetonide is essential for achieving high substratecontrolled stereoselectivity, for the 3,4-dimethoxy derivative
136 reacts unselectively under the above-described conditi0ns.['~~1
As Table 7 shows, the ion pair 134 reacts smoothly with a
wide variety of electrophiles, and acylation also poses no problems. For this reason the substrate constitutes a valuable synthetic equivalent to the (S)-I,3,4-trihydroxybutanide (K) .I'
The method should be generalizable to the transfer of longer
chains and more highly functionalized analogues.
How might one now gain entrance to the epimeric series
epi-135 without the need for the awkwardly available (+)sparteine? This is easy so long as a deuteration of the product
can be tolerated (Scheme 56).[65e1Acetonide 133 is lithiated and
137
R,bi
138
IElX
NB",
C
b
$
y
OCby
El
139
Scheme 57. Substrate-directed regio- and diastereoselective deprotonation of dicarbamate 137 [173b].
Table 8. Substrate-directed deprotonation and substitution of the dicarbamate 137
[173b].
ElX
Yield ["A]
qocby
*
DOMe
Me1
LO
coz
95
93
92
El
Product [a]
~~
1. sBuLfib0
2.MeOD
0
HSHR
- 0
sBuLi/
D H
TMEDA
D
Me
CO,Me[b]
Me,COH
iPr,COH
EtC=O
[cl
Me$
Me&
PhS
PPh,
139a
139b
139c
139d
139e
139f
139g
139h
139i
139k[e]
1391[e]
Me,C=O
1Pr,C=O
EtCOCl
[dl
MeJiCI
Me,SnCI
PhSSPh
Ph,PCl
85
75
80
71
73
75
73
56
[a] In no case could a second diastereomer be verified by 'H NMR, so the diastereomeric ratio was at least 97:3. [b] After esterification with diazomethane.
[c] (E)-MeCH=CH-C=O. [d] (E)-MeCH=CHCOCI. [el Ref. [175].
[l DIepi-134-TMEDA
[l Dlepi-135
El = Me,%: 45%; d.r. = 98:2
Scheme 56. Reversal of diastereoselectiv~tythrough deuteration of 133 [65e].
deuterated under substrate control. Because of the large kinetic
H/D isotopic effect (see Section 4.2), TMEDA-assisted lithiation leads from ( S ) - [101133 to the diastereomerically pure ion
pair [IDIepi-lM, which is configurationally stable and can be
substituted to give [ID]epi-135 under stereoretention.
High substrate-controlled diastereoselectivities are also
achievable if the substituent capable of chelating is not itself on
a stereogenic center, but rather adjacent to one. Upon deprotonation according to either the TMEDA or the (-)-sparteine
variant, followed by reaction with electrophiles, dicarbamate
137 of (S)-2-(dibenzylamino)-1,4-butanediol,['73a~ derived
from (S)-asparaginic acid, leads to mixtures of regio- and
diastereomers." 73b1 However, if the transformation is carried
out in diethyl ether (without addition of a diamine), the pure
1-substitution product 139 results (Scheme 57, Table 8). The
breadth of the applicable electrophiles and the high reactivity
are noteworthy. We therefore conclude that formation of the
chelate complex 138 is kinetically controlled;[' 74,
the cause
of the highly stereoselective course of the reaction is thought to
be the strong tendency of the 2-dibenzylamino and 4-carbamoyloxy groups to occupy an equatorial position in the developing
six-membered ring.
Angew Chem. Inl. Ed. Engl. 1997,36, 2282-2316
The 4-methyl ether 142 shows very similar selectivity,['73b.1 7 7 1 whereas the 4-0-TBDMS ether 143 is unreactive," "I which also supports the important role played by the
complexing substituent in the 4-position.
If the 1-pro-S-position of substrate 137 is blocked, whether by
an alkyl group or deuterium,['78*' 791the (-)-sparteine-mediated deprotonation "at the other end" takes its usual course, with
the abstraction ofpro-S-H-4 (Scheme 58). Obviously the l-posi-
H H R NBn*
h
O
C
b
y
1. sBuWl
2.EIX
ybCO-,,
E'
NB".
CbYQ
Rk
139a
139b
R=D
R=CH,
R H
140 R = D; El = CH,;
86%; d.r. >95:5
141 R = CH,; El = C0,Me;
90X;d.r. z 955
142
143
R=CH,
R=SiMqiBu
144
145
R=Me
R=CPh,
Scheme 58. Lithiation of 2-amino-1,4-butanediol derivatives in the 4-position
[173b,177].
2303
REVIEWS
D. Hoppe and T. Hense
tion can also be blocked in a classical way after introduction of
a nonactivating protective group such as methyl or trityl at 0-1
(144 and 145, Scheme 5 8 ) , leading to regio- and stereoselective
deprotonation at C-4.[1771
In the corresponding 2-amino-1,s-pentanediol derivative 146,
made from (S)-glutamic acid, internally assisted deprotonation
is apparently retarded due to the unfavorable ring size of the
resulting lithium chelate complex (Scheme 59) .[I 73b1 Thus, in
(S)-2-(N,N-Dibenzylamino)alkylcarbamates, accessible in a
small number of steps from the natural amino acids,['801display
marked ul-induction[6a1in TMEDA-assisted deprotonation" 541
(L, = TMEDA). Therefore, the pro-R-proton is preferentially
abstracted, and one obtains after reaction, irrespective of the
electrophile, an excess of the diastereomer 152 (Scheme 61,
Table 9). The (S)-alaninol derivative 149a constitutes an excep-
146
1. sBuLd1. EbO.
2.co,
14'
1. sBuLi. EbO
1
sBuLi/L,
EbO,-78 "C
-1 -Hs
-54s
yB"2
7
OK
L,,L'-o
YBn,
Cbfl-OCby
cb@-Ocby
C02Me
147
i"
C02Me
148
61%; d.r. w 97:3
80%; d.r. w 97:3
Scheme 59. Regio- and diastereoselectivesubstitution of the 2-amino-1,S-pentanediol derivative 146 [173b].
the presence of (-)-sparteine, proton 5-H, is abstracted almost
exclusively, whereas in the absence of a diamine a slower deprotonation at C-1 is initiated; the intermediate lithium compounds
produce, after trapping with carbon dioxide and U-methylation,
the regioisomeric esters 147 and 148.
Therefore, the procedures described offer extremely facile access to equivalents for the stereoisomerically pure carbanionic
synthons L through 0 (Scheme 60).
r?Y
rtrH2
Ho*HOH
$ 2 -
8
n=l L
n=2 M
n=l N
n=2 0
Scheme 60. Carbanionic synthons accessible from 2-amino-1,w-alkanediols[173b].
L
151
iEiX
y".
yBn2
&OCby
R
NRz
i
El
R 7 0ElC b Y
1 52
153
Scheme 61. Deprotonation of (S)-2-(N,N-dibenzylamino)alkylcarbamates. For R,
see Table 9.
tion. One presumes that proton abstraction takes place in the
antiperiplanar conformation, and in the normal case the pro-RH is more accessible (Figure
As described above, we
have so far found no indication that the dibenzyiamino group
interacts as a complexing ligand.[18']
In the presence of (-)-sparteine (L, = 1) as complexing partner, the situation of a "mismatched pair" arises due to its
preference for the p r o - S - p r ~ t o n . [Strong
~ ~ ~ substrate-directed
stereoselection is directed toward the pro-R-H, and thus retards
Table 9. Substrate- and reagent-directed deprotonation of 2-(dibenzylamino)alkyl carbamates 149.
Starting
material
149a
149a
149a
149b
149b
149b
ent-149b
149c
149d
149e
149e
ruc-149f
L,= TMEDA
R
EIX
Products
CH3
CH3
CH3
H,CCH,
H,CCH,
H,CCH,
H,CCH,
BnzN(CHd3
(CH,),CHCH,
CH31
Bu3SnC1
CO,Me[a]
CH,I
Bu,SnCI
PhCOCl
CO,Me[a]
CIC0,Me
CO,Me[al
CU,Me[a]
PhCOOMe
CO,Me[a]
152aa, 153aa
152ab, 153ab
152ac, 153ac
152ba, 153ba
152bb, 153bb
152bd, 153bd
enr-152bq ent-153bc
1 5 2 q 153cc
152dc, 153dc
152ec, 153ec
152ed, 153ed
ent-l52fc, enf-153fc
C6H5CH2
C,H&Hz
c-C,H,CH,
L, = 1
Yield[%]
152: 153
Yield[%]
82
49
76
72
70
74
37: 63
36: 64
31:69
88:12
89:11
88: 12
[bl
-
-
57
[CI
[bl
43
65
[dl
78
56
84:16
>95:5
-
-
74
89
92:8
83: 17
152:153
Ref.
56
59
48
<3:97
<3:97
<3:97
PI
-
[183]
[183]
[183]
[I831
[I831
[I831
[183]
[I831
11541
[184]
[184]
[I851
PI
>97:3
-
11:89
7:93
Id1
[a] Isolated after esterification with diazomethane; CO, was used as electrophile. [b] No deprotonation. [c] Not carried out. [d] See text.
2304
Angew. Chem. Int. Ed. Engl. 1997,36,2282-2316
REVIEWS
Enantloselective Synthesis
v. P
Bn,N+OyNR,
5 i .
*/I Bu
HSHR
0
Figure l o Favored conformation for substrate-directed
deprotonatlOnof (S)-2-(N,Ndibenzy1arnino)alkyl carba-
formation of the diastereomeric ion
pair 151, which in the presence of
large residues R may be completely
excluded.
Also with the (S)-prolinol derivative 154 there is a strong preference
ruc-149f it was possible in this way to acquire the ester
ent-152fc ( > 9 5 % ee) and recover (S)-carbamate ( S ) - 1 4 9 f
(Scheme 64).['28,1851 The advantage relative to classical or enzymatic racemate resolution lies in the possibility of accomplishing a highly diastereoselective C-C coupling in addition.
qoc..
for abstraction of the pro-R-proton,
which leads via the ion pair 155 to
mates.
the diastereomerically pure substitution products 156 (Scheme 62).1'541
Quite surprisingly, when deprotonation is carried out in ether
solution in the presence of ( - ) - ~ p a r t e i n e [ ' ~or
~ Ientirely without
the addition of a diamine, the very same diastereomer res u l t ~ . [This
' ~ ~observation
~
permits only one conclusion: Here
the less shielded amino function intervenes in the deprotonation
step in an intra- or intermolecular way.
(N%OCby
in
+
<ocby
HSHR
HSHR
14s
enf-14%
I
q
;
1.5 equiv sBuL/ 1
EbO,- 78 "C
OCby
G
;
OCby
LV 1
Lit 1
-
ent-l5Of-1
1
sBuLi
151f.1
1 . CO,, 2. CH,N,
q GOC
H,c~,
i
OCby
154
C02Me
156
155
C0,Me
en?-152fc
40%; > 95% ee
a El = CH,; 72%; d.r. > 95:5
b El = BuSn; 71%; d.r. > 95:5
c El = (CH3),CHOH 33%; d.r. > 95:5
~
Scheme 62. Diastereoselective deprotonation of the prolinol carbamate 154 11541.
153fc
6%
GOC
42%;80% ee
In the ( -1-sparteine-induced deprotonation of (R)-2-(dibenzy1amino)alkyl carbamates enr-149, both the substrate- and the
reagent-controlled preferences point to the pro-S-proton. As
a consequence, the reaction is rapid and stereoselectivity virtually complete (compare with enr-149b in Table 9)
(Scheme 63).[183.l E 4 ] The sharply differing reactivities of the
Scheme 64. Kinetic resolution of the carbarnate rat-149f Il29.1851
As was demonstrated with 3-(piperidine-2-yl)ethyl carbamate
the method can also prove valuable when differentiation between the enantiomers is less efficient.['861One obtains
after methylation the epimeric (2S)-alkyl carbamates 157
( > 9 5 % ee) and 158 (89% ee), which after deblocking lead to
the alkaloids ( +)-~edridine['~'l(159) and (+ )-allosedridine
(160) (Scheme 65). The sequence possesses the character of a
ruc-156,
r c (yO
OCby
ent-15Ob-1
I
ent-152bc
rjn
El = CO,;65%;d.r. > 95:5
Q
rac-156
Scheme 63. Highly diastereoselective deprotonation and substitution of the (R)-2(N,N-dibenzy1arnino)butylcarbamate enr-149b [lS3].
,",i\o
CH3
158 49% (89%ee)
two enantiomers with respect to sec-butyllithium/( -)-sparteine
can be exploited for efficient kinetic racemate resolution. As
with all kinetic racemate resolutions with carbamates, at least
one equivalent of alkyllithium is required; apparently this is
irreversibly bound in a precomplex even with the less reactive
enantiomer. From the racemic P-cyclopropylalaninol derivative
Anxeu Chem lnr €0.Erin/ 1997, 36, 2282-2316
H
160
Scheme 65. Synthesis of (+)-sedridine and (+)-allosedridine [1X6]
highly enantioselective substitution on a racemic starting material, which simplifies racemate resolution of the starting material by turning it into a separation of product diastereomers.
2305
D. Hoppe and T. Hense
REVIEWS
An interesting stereochemical problem is presented by mesodicarbamate 161 (Scheme 66).['53b31881 The compound bears
two enantiotopic side groups, each with two diastereotopic protons. The chiral base sec-butyllithium/( -)-sparteine abstracts
that pro-S-proton whose removal is supported by the substrateinherent preference, namely R-H,.['891 The [lR,I(lR),2S]-configured lithium intermediate 162.1 arises in excess together with
the [IS,l(lR),2R]-diastereomerand leads to the substitution
products 164 with high stereoselectivity.
($ -
L/ 1
sBuLdl
+
q
Or! C
b
161
Ld 1
162-1
1
EIX
H \\
X
:g 122 RuClJNalO,
168
'*..Ln
EIX
mu
('***El
mu L O
169.1
170
170 a EIX = MeOS0,OMe; 88%; 94% ee
b EIX = CO,; 55%; 88% ee
c EIX = Ph,CO; 75%: 90% ee
d EIX = Bu,SnCI; 83%; 96% ee
OCby
H
C
A0
mu
OC
Scheme 67. (-)-Sparteine-controlled deprotonation of N-Boc-pyrrolidine 11931
OCby
HSHR
Hs
y
OCby toluene,-78 "C
H
Q
A0
sBuLdl
EbO. -78
Q+R
H
163-1
: EIX
(Scheme 68)
As expected, reagent-induced chiral induction dominates, as has also been shown to be the case with the
2-phenylpyrrolidine derivative 173.[' 51
v
H
Boc
17Oa
164
a EIX
165
Boc
171 >99%ee
90
172
10
COJCH,N,; 57%;
d.r. = 982; > 95% ee
b EIX = CH& 65%;
Q
d.r. = 955; > 95%ee
Scheme 66 Desymmetrization of a meso-1,Cdrcarbamate [153b].
I
BOC
173
Carboxylic acid ester 164a was transformed with 5~ hydrochloric acid into the tetrahydrofuran derivative 166, which
was oxidized to the lactone acid ester 167.1'90]At this stage the
expected configuration was verified by an X-ray crystal structure
All in all, desymmetrization of the meso-substrate was associated with highly diastereoselective C-C bond
formation.
4.5. Enantioselective Lithiation of N-Boc-Pyrrolidines
As Beak et al. were able to show as early as 1984 that N-(tevtbutoxycarbony1)pyrrolidines and -piperidines are easily deprotonated to the corresponding racemic dipole-stabilized lithiumcarbanion pairs.['921 Application of sec-butyllithium/( -)sparteine to N-Boc-pyrrolidine leads through enantiotopic differentiation, after removal of the pro-S-2-H atom, to the configurationally stable intermediate 169.1, which can be substituted
with retention by various electrophiles (Scheme 67) .I' 931
The (R)-diphenylprolinol derivative 170c can be enriched to
99.3 % ee by recrystallization and provides a valuable ligand for
enantioselective ketone reduction according to Corey-It~ u n o . ~Tin
' ~ ~compound
]
170d is the starting material for the
synthesis of enantiomerically enriched 2-lithio-N-methylpyrrolidines developed by R. E. G a ~ l e y . [ ' ~Double
~]
lithiation/
methylation offers easy access to the (S,S)-2,5-dimethylpyrrolidine derivative 171; a possibility has also been worked out for
separation of the roughly 10% content of meso-compound 172
2306
BOC
BOC
174
93
175
7
Scheme 68. Reagent-directed deprotonation of chiral 2-substituted N-Boc-pyrrolidines [SS].
From a mechanistic standpoint there are many parallels to the
deprotonation of 0-alkyl carbamates (Sections 4.1 -4.3);['961
kinetic and deuteration studies suggest rapid formation of a
precomplex of the reaction partner^.['^**'^'^ M. E. Kopach and
A. I. Meyers were able to show with a conformationally biased
pyrrolidine that both the deprotonation and substitution steps
proceed with retention.['981In a paper that is well worth reading, about two dozen ligands-mainly chiral diamines-were
investigated to see to what extent they could replace (-)sparteine (l), or even lead conveniently into the other series of
enantiomers."
The most effective among them have been
collected in Scheme 69.
Essential, sufficiently high configurational stability of the
lithiated intermediates seems to be limited to pyrrolidines like
168.[2001
The greater configurational lability of the corresponding piperidines can be compensated for to some extent through
a rapid, intramolecular subsequent reaction. Thus, Y. S. Park
and P. Beak obtained the l-azabicyclo[3.l.0]hexane 182 with
55 % ee after deprotonation of N-Boc-4-tosylpiperidine (179) in
the presence of (-)-sparteine by the addition of chlorotrimethylsilane; the configuration is still unknown
(Scheme 70).Iz011In this case, rapid 1,3-substitution of the enantiomerically enriched intermediate 180.1 preserves most of the
chiral information.
Angew. Chem. Int. Ed. Engl. 1997,36,2282-2316
REVIEWS
Enantioselective Synthesis
Q
1. sBuLiR'
EbO, -78 "C
2. Me3SiCI
QSiMe,
I
Q'***SiMe3
I
Boc
BOC
168
BOC
17Oe
(3-sparteine (1)
ent-17Oe
87%(96%ee)
98
63%(72%ee)
86
H
183
L
J
184-1
165
73%; 52% ee
2
:
14
H
186
187
86%;84% ee
Scheme 71. (-)-Sparteine-induced rearrangement of meso-epouides via 1 8 4 1
I
CH3
177
'OH
Scheme 69. Testing of various bidentate ligands in reagent-controlled deprotonation of N-Boc-pyrrolidine [199].
ing of the epoxide ring and insertion into the y-endo-CH bond.
cis-Epoxides of medium-sized cycloalkanes undergo similar reactions with transannular CH-insertion.[202a1
An interesting vinylogous nucleophilic ring opening of mesooxabicyclic system 188 by n-butyllithium/( - )-sparteine was discovered by M. Lautens et al. (Scheme 72).I2O3lThe best results
[60% yield of (-)-189 with 52% eel were obtained with five
equivalents of butyllithium and 15 mol% (-)-sparteine at
-40°C in pentanelhexane.
BOC
179
180-1
188
1. sBuLi
2.M+SiC;
Boc
181
I
BW
SiMe,
182
77%; 55% ee
Scheme 70. Enantioselective deprotonation and intramolecular alkylation of an
N-Boc-piperidine [201]. Ts = 4-MeC6H,SO,.
4.6. Enantiotopically Differentiating Lithiation at Other
Prochiral Groups; Planar- and Axial-Chiral Intermediates
In the preceding sections, chirality was in most cases introduced through differentiation by the chiral base between enantiotopic protons of a methylene group. This means that configurational stability of the lithiated intermediate under the
reaction conditions as well as a subsequent stereoselective substitution are necessary prerequisites. A few other cases in which
no stereogenic carbanionic C atom is formed will now be discussed.
Hodgson et al. investigated a rearrangement of meso-epoxides induced by chiral
Treatment of exo-norbornene
epoxide (183) with sec-butyllithium/( -)-sparteine in pentane
permitted the isolation of (-)-nortri~yclanol['~~~
in 73 % yield
and 52% ee (Scheme 71).[202b1
The base differentiates between
the (R)- and (S)-sites on the oxirane unit during proton abstraction, and the carbenoid stabilizes immediately through openAngew. Chem. Int. Ed. Engl. 1997,36,2282-2316
OH
H6
(-)-US9
76
.
(+)-189
:
24
Scheme 72. Nucleophilic ring opening ofmeso-oxabicycliccompound 188. R
=
Bu.
Both methyl groups in the dimethylphosphane derivatives
190- 192 (Scheme 73) are enantiotopic, and the attached hydrogen atoms are acidified. During deprotonation to 191.1- 193.1
the phosphorus atom becomes a stereogenic center. First experiments in deprotonation of dimethylphenylphosphine oxide
(190a, R = C,H,) with n-butyllithium/( -)-sparteine were described in very brief form by Raston, White, et al.,[421with an
observed diastereomeric ratio in the range of 60:40.
Better results were obtained by D. A. Evans et al. with
aryldimethylphosphine sulfides 190b and the borane complex
ee values of 74-94% were achieved in the capture
products with benzophenone (194b and 194c).
Oxidative coupling of the lithium derivative 193.1 with copper@) pivalate to give the bis(phosph0nium) complexes 195 and
196 is associated with an enhancement of the enantiomeric ratio
to 98:2 (96% ee) (Scheme 73), because the greatest part of the
epimer (PR)-193.1(present to the extent of about 12%) is removed as the meso-diastereomer (R,S)-196. Deblocking of the
2307
D. Hoppe and T. Hense
REVIEWS
19oax=o
19ObX=S
190~
X = BH,
191.1 x = o
192.1 X = S
193.1 X = BH,
194a X =0,R = Ph
x = s (79%. 79% ee)
194c X = BH, (84%, 87% ee)
104b.c: R =
6g,,,,
CH,
B e
Cu(O,CtRu), &. . - rA
H3C
67%
Good enantiomeric enrichments have
also been achieved through enantioselective
deprotonation of other ferrocene derivatives : (diphenylphosphany1)ferrocene(2Ola)
with lithium (R,R)-bis(1-phenylethyl)amide,
55 % ee,[2081and (dimethylaminomethyl)ferrocene (201b) with ( R , R ) - N , N , N ' , N tetramethyl-I ,2-~yclohexanediamine/nbuee~(Scheme 74).
t y l l i t h i ~ m ,80%
[~~~
A breakthrough was achieved by V.
Snieckus et al.,[2071when they subjected
N,N-diisopropylferrocene
carboxamide
€BH,BBH,
&CH,
Ar
w
+
*tolyl
€SH,BBH,
&.CH3
Ar
Ar$@
HC
,
w
(202) to deprotonation with n-butyllithium/( -)-sparteine in diethyl ether
96% ee 88
:
12
(Scheme 75). The expected products 203
were isolated with a host of electrophiles in
62-96% yield and 81-99% ee.
Related stereochemical problems are
Ar ...; j
jj..*CH3
Are.;.
ij.;.CH,
posed by monosubstituted arenetricarH,C'
LsiJ
.Ar
H 3 C v p U Ar
bonylchromium complexes; in this case
chiral
bases other than alkyllithium/
Me 'Me
(S,S)-l98
(-)-sparteine
prove to be most useful.
(R,R)-197
.Ar
.. = D
- tnlvl
Scheme 76 collects examples from the reScheme 73. Enantioselective, (-)-sparteine-induced deprotonation of aryldimethylphosphane derivatives
search groups of Kiindig,[2101Uernura,l2l
[204].
Simpkins,[212s2131 and S c h m a l ~ . [ ~ * ~ ]
borane complex (S,S)-195with diethylamine leads to the free
Axial chirality can also be created through enantiotopically
diphosphane (S,S)-198. Just as in the case of the bis(phosdiscriminating deprotonation. The first examples, with 22'-dimethyl-1,l'-binaphthyl and 2,2',6,6'-tetramethyl-l,l'-biphenyl
phiny1)-substituted silane (R,R)-197, C,-symmetric diphosphanes obtained in this way represent valuable potential ligands
for enantioselective catalyst systems.
Based on the seminal work by I. Ugi and H. B. Kagan,[20s1
ferrocenes are extraordinarily versatile chiral auxiliaries and
L[Ezoc
catalyst ligands. When one succeeds in using an enantiomerical2.EIX
ly pure base to distinguish between the enantiotopic sites of the
achirai monosubstituted ferrocene 199 (Scheme 74), racemiza193.1
(S,S)-195
+
(R,S)-l%
I-
&(A
&A
aN4
202
203
El
X
X
a Me,Si; 96%; 98% ee
b PkCOH; 91%; 99% ee
c PhS; 90%; 93% ee
d I; 85%; 96% ee
X
Scheme 75. Lithiation and substitution of monosubstituted ferrocenes with enantiotopic differentiation.
199
200
ent-200
Scheme 74. (-)-Sparteine-induced ortho-lithiation of a ferrocene derivative 12071.
bases:
tion of the enantiomerically enriched planar-chiral intermediates 200 and ent-200 remains possible only through a proton
transfer reaction or dissociation- recombination of the ferrOcene ; that is, configurational stability can be presupposed.1zo6]
2308
&
yh
'
sBuLd
MqNhNMe2
Ph
Li
Li
PhyNvPh
PhyNyPh
Ph
Scheme 76. Enantioselective deprotonation of arenetricarbonylchromium complexes [210-21 2,2141.
Angen Chem Int. Ed Engl. 1997, 36, 2282-2316
REVIEWS
Enantioselective Synthesis
from Raston et al., are unfortunately not well documented with
respect to quantitative results and configurational assignrnent.I2l5]P. Beak and D. P. Curran recently described a (-)sparteine-induced deprotonation and methylation of the 1naphthoic acid amide 204 with 50% ee (Scheme 77).['16] The
yield at - 78 "C amounted to only 53 YO,
and from that one can
conclude that the chiral base apparently accomplishes a kinetic
racemate resolution between conformers that are interconverted
only very slowly at - 78 0C.[2171
&
-
/
(R)-204
\
2104
Li
I
H*,k
f
f
/
(S)-204
CH,]
1. sBuLi11
EGO. -78 'C
2. CH,I
Ph
OH
213
63%; 82%ee
pb
R S..'
lLi,
OH
ti
211-1
212
82%; 80% ee
Scheme 79. Enantio- and diastereoselective carbolithiation of lithium (E)-3phenyl-2-propen-I-olate [221]
/
(W-205
30%. e.r. = 75 : 25 (50% ee)
6)
-205
R = CH,(CH,),
Scheme 77 Induction o f axial chirality through (-)-sparteine-induced deprotonation [216]
5. Sparteine-Induced Carbolithiation
So far we have discussed only the most popular route to
"carbanion generation", the deprotonation of CH-acidic precursors. The addition of organolithium to C=C double bonds
(carbolithiation)[21*I is an attractive alternative, because it
yields an additional C-C bond "gratis" (Scheme 78). This reac-
206
1 LP
i 3 - f
OLi
03""' (3y"'
/
hexane
epirnerization
/
I
.
nButiIl
(€)-209
H
1
J.-F. Normant, and co-workers succeeded in carrying out an
enantioselective (-)-sparteine-controlled addition of alkyllithium to cinnamyl alcoholates and cinnamylamines
(Scheme 79) .[2z1] Reaction of the (E)-alcoholate (E)-209 with
207.1
208
poly-208
Scheme 78. Sparteine-induced asymmetric polymerization.
tion represents the first step in anionic polymerization;[2181its
use in organic synthesis depends upon retarding (relative to the
first step) addition of the intermediate 208 to another molecule
of alkene 206. This can be accomplished, for example, by conducting carbolithiation with a 5-alkenyllithium derivative in an
intramolecular way according to W. F. Bailey,~Z'9~2z01
or in a
case where the adduct 208 is specially stabilized.
As a benefit, control is extended over two or even three (if the
residue R3 is chiral) stereogenic centers in the product. I. Marek,
Angebr. Chem. Int. Ed. EnxI. 1997, 36, 2282-2316
butyllithium/( - )-sparteine provides good insight into the
course of the reaction. In a syn-addition, typical for carbometalation,[2221
butyllithium/( -)-sparteine [presumably bonded in a
precomplex to the alcoholate (E)-2091 attacks the C=C double
bond from the Si-face. The resulting five-membered ring chelate
complex 210.1 bears phenyl and butyl residues in cis-orientation, hence the configurationally labile benzyllithium derivative
epimerizes with formation of the more stable trans-substituted
chelate 211-1. Protonation produces the (S)-configured alcohol
212 with 80% ee, whereas methylation with inversion[2231
leads
to the doubly branched (S,S)-alkanol213. Locking the intermediate into the chelate and the formation of reactive, presumably
h e x a m e r i ~ [lithium
~ ~ ~ I alcoholate clusters with diminished reactivity protects adducts 210.1 and 211.1 from polymerization.
Quite remarkably, 5 mol YOof (-)-sparteine is sufficient to produce enantioselective reaction."' s]
Enantiofacial differentiation on the part of the ( -)-sparteine
reagent at C-2 of the double bond in the conformationally restricted complex determines the stereochemical course of the
reaction, since a configurationally stable stereocenter is created
at C-2. Consequently, the (Z)-alcoholate (2)-209 leads to
the opposite series of enantiomers.[2z1a1Similarly successful
additions have 'been realized with (E)-cinnamylamines and
-acetals.[zz bl
Kiindig et al. have recently reported on enantioselective carbolithiation of the arenetricarbonylchromium complex 214
from 4,4-dimethyl-2-phenyl-l,3-oxazolidine(Scheme 80)
Here the (-)-sparteine reagent attacks one of the enantiotopic
sites preferentially from the "upper side," and treatment of the
carbanionic intermediaterzz7]with propargyl bromide leads to
a 5,6-trans-disubstituted cyclohexadiene 215. Greater enantiomeric excesses (65-93% ee) with somewhat lower yields
(51 -67 YO)were achieved on using (S,S)-1,2-dimethoxy-1,2diphenylethane (216) in place of ( -))-~parteine.[~'~]
2309
D. Hoppe and T. Hense
REVIEWS
1. sBuWl
Et20, - 78 'C
2. PhMe,SiCI
c
OCby
Ph
SiMe,Ph
221
q-P
Me0
a C,H,
72%
b CH,=CH, 87%
48%; d.r. > 98:2; >95% ee
%%ee
3 4 % ~
CH,
70% 47%ee
d CH3(CHJ365% 36% ee
C
OMe
216
Scheme 80. Diastereo- and enantioselective reaction of the tricarbonylchromium
arene complex 214 [226]. HMPA = hexamethylphosphoramide.
Orientational investigations with a-styryl-N,N-diisopropyl
carbamate (217) reveal a pathway to enantiomerically enriched
secondary benzyl alcohols (Scheme 81) .[228] Attack by alkyllithium/( -)-sparteine leads-albeit with reduced enantiofacial
selectivity-to
the configurationally stable adducts 218.1,
which are carboxylated with inversion174b1
and protonated with
retention of configuration.L2291
Scheme 82. Stereohomogeneous cyclopentyl carbamates through cyclocarbolithiation [231]
have turned out in all the relevant studies to be especially configurationally labile. Experimental results from Reich et al. point
to the fact that the formation of solvent-separated ion pairs
tends to increase the barrier to inversion rather than lower
it.[232b1
Nevertheless, we thought it might be possible that incorporation into a rigid chelate complex might lead to an increase
in configurational stability. Alky1r2331
and vinyl N,N-dimethylmonothiocarbarnates[2341
are distinguished by high kinetic acidity. B. Kaiser thus investigated the ( - )-sparteine-induced deprotonation of S-alkyl thiocarbamates of the type 224;[235*
2361
the reactions with S-butyl esters (Scheme 83) will be discussed as
sBuLil1
217
a R=nBu;73%;32%ae
b R = iPr; 80%; 45% ee
C R = fBu; 77%; 25% ee
222
R
1
EiX
M~O,C*'*
'R
219
220
Scheme 81. Enantioselective carbolithiation of or-styryl carbamate 217 [228].
(R)-226
Intramolecular carbolithiations of (E)-alkenyllithium compounds proceed rapidly, as documented by numerous studies by
Bailey et al., by 5-exo-trig-ring closure.[219.2301 However, this
pathway has so far been used only to synthesize racemic or
achiral cyclopentane derivatives. The gap has been closed by a
highly stereoselective carbolithiation developed by us, which
also assures control over the configuration of the attacking Cnucleophile; it produces in the case of the substituted cyclopentane 222, for example, three consecutive, uniformly configured
stereogenic centers (Scheme 82) .[2311
6. Ion Pairs from a-Thiocarbanions
In contrast to a-alkoxyalkyllithium derivatives, a-thio-substituted a l k y l l i t h i ~ mand
[ ~ ~benzyllithium
~~
corn pound^^^^^^ 223-1
2310
a El = SiMq; 92%; 63 : 37 (46% M)
b €1 = C0,Me; 91%; 63.5 :36.5 (47% W )
Scheme 83. (- )-Sparteine-mediated deprotonation of S-alkyl thiocarbamates
[235].
an example of the results of this study. Deprotonation of 224a
with sec-butyllithium/( -)-sparteine in diethyl ether and trapping of the carbanionic intermediate with chlorotrimethylsilane
led to the silane (+)-(S)-226a with 46% ee. A nearly identical
result was recorded in carboxylation and esterification to (-)(S)-226b.I2 61
Is the cause here insufficient configurational stability of the
diastereomeric intermediates (R)-225.1 and (S)-225.1, which
enter into equilibrium under the reaction conditions, or inadequate enantiotopic differentiation in the deprotonation step? As
demonstrated by the following series of experiments with the
Angeu,. Chem Int. Ed. Engl. 1997,36, 2282-2316
REVIEWS
Enantioselective Synthesis
deuterated substrate ruc-[D]224, in this case both factors are
responsible (Scheme 84) .[2351 During deprotonation with secbutyllithium/TMEDA (a) and capture of the intermediate ion
me[D]224
rao(D1226a
93%, w 99% D
1. 0.5 equiv sBuLil1
2.Me,SiCl
m~[D]224
(S>[Dl226~1
+
Why do S- and 0-lithioalkyl derivatives behave so differently? Presumably the relatively long C( = 0)-S and C-S bonds
(about 180 pm compared to 140 pm for C - 0 ) establish a less
compact transition state for deprotonation, coupled with diminished enantiotopic differentiation. Moreover, the mechanisms
of enantiomer interconversion differ. Evidence is increasing that
rotation of the attached C-S bond from the antiperiplanar to
the synperiplanar conformation is the rate-determining step in
the enantiomer interconversion, whereas for a-hydroxyalkyllithium derivatives inversion at the carbanionic center appears
to be the slowest step (Scheme 85) .[2401
ra~IDJ224
34% ee
Scheme 84 Reactions with deuterium-labeled thiocarbamates
pair rac-[D]225 to give the silane ruc-[D]226a, all the deuterium
remains in the molecule; that is, the kinetic H/D isotope effect
is on the order of about 100. In the next experiment (b), ruc[Dl224 was treated with one equivalent of the chiral base system
see-butyllithium/( -)-sparteine, and silylation was initiated; the
result was a greater than 99% deuterated product (S)-[D]226a
in 93% yield and 34% ee.[2371
If there were a great preference
for one of the enantiotopic protons pro-S-H or pro-R-H, one of
the enantiomers would have been preferentially deprotonated,
thus introducing a kinetic resolution like that observed with the
corresponding 0-alkyl carbarnates (Section 4.2). However, the
substrate was nearly completely deprotonated; nevertheless, an
overproportional amount of enantiomerically enriched product
was isolated. From this it can be concluded that the
diastereomeric ion pairs (1R)- and (1 S)-[D]225.1 equilibrate under the reaction conditions. Experiment (c) solidifies this conclusion; it differs from experiment (b) only in a reduced amount
of base (0.5 equivalents). Again the product was (S)-[D]224a
with 34% ee, and the recovered unchanged substrate 101224
took the form of a racemate. There was thus no kinetic resolution, and the product ratio was determined thermodynamically
at the stage of the ion pairs.
Also for the (-)-sparteine complex of the dilithium compound derived from N-methyl-3-phenylthiopropionamide,
which represents a valuable homoenolate reagent, Takei et al.
found evidence of configurational lability in THF at - 78 " C ,as
well as similarly low enantiomeric enrichment in the aldehyde
Scheme 85 The rate-determrning step in enantiomer interconversion of z-heterosubstituted alkyllithium compounds.
It can be concluded from the work of R. W. Hoffmann et al.
on racemic I-(arylse1enyl)alkyllithium derivatives that bulky
arylthio residues increase the barrier to enantiomer interconverion.^^^^' An increased crowding of groups at the carbanionic
center should for the same reason lead to increased configurational stability. Indeed, the lithium compound ( S ) 225a.TMEDA prepared from (S)-224a (46% ee) with secbutyllithium in diethyl ether proved to be configurationally
stable at - 78 "C, because after standing for 2.5 h, deuterolysis
produces the capture product (S)-[D]226a with 44% ee
(Scheme 86).[2351
This is, to the best of our knowledge, the first enantiomerically enriched a-thioalkyllithium derivative. The world record with
respect to configurational stability is held by benzyllithium
derivative (S)-227 (99% ee).[24232431 No racemization could be
demonstrated after standing for 24 h at -78 "C, and even
10 min of warming the reaction mixture to 0 "C caused less than
1 YOracemization.
EbO, - 70 '
C
Cbys
(5)-224a 46%-
ad duct^.^^^''
For an efficient utilization of chiral a-thiocarbanions, conditions and ligands must be found that lead to the largest possible
energy difference between the diastereomeric lithium complexes. Corresponding investigations with (similarly configurationally labile) lithium-I-(phenylseleny1)alkanides carried out
by R. W. Hoffmann et al. show through an NMR study that this
is not a simple
A n p w Chrm. f n t . Ed Engl. 1997, 36, 2282-2316
227
Scheme 86. Configurationally stable z-thioalkyllithium derivatives [235,241.242].
231 3
REVIEWS
D. Hoppe and T. Hense
7. Summary and Prospects
The interaction between a lithium ion bearing chiral ligands
and a developing or previously formed carbanion forces the
latter into a specific configuration, whether through a preferred
diastereomorphic transition state in the course of deprotonation
or because of the establishment of a state of equilibrium. In
combination with a stereoselective substitution step, this provides a new and effective strategy for enantioselective synthesis.
The synthetic value of the process is increased by the fact that
the carbanionic reagents are obtained by simple deprotonation
of the corresponding CH acids. The examples discussed above
were discovered through intuition, whereby spontaneous crystallization of one of the diastereomers occasionally played a
crucial role in increasing efficiency. One goal is to understand
the interaction between chiral ligand, cation, and substrate or
carbanion better, as well as to quantify it through quantum-mechanical calculations in order ultimately to succeed in predicting
favorable ligand -carbanion combinations. We are closer to
this goal with respect to estimating the relative energies of
diastereomeric ion pairs than for diastereomorphic transition
states.
The results described from our own research group have been
achieved over the past eight years by capable and enthusiastic
co-workers. Their names are included in the bibliography. Special
thanks are due to Prof. Dr. E.-U. Wurthwein, who convinced me
( D . H . ) of the value of quantum-mechanical computational methods for the solution of our problems, introduced several co-workers to the application of these methods, and stood at our side with
unseyish advice and assistance. A . Deiters and Dipl.-Chem I:
Heinl gave valuable support with the presentation of the
manuscript and the graphics. The work was supported for many
years by the Deutsche Forschungsgemeinschaft, the Fonds der
Chemischen Industrie, the Stiftung Volkswagenwerk, and Bayer
AG.
Received: March 5 , 1997 [A2151E]
German version. Angen. Chem. 1997, 109, 2376-2410
Translated by Dr. W E Russey, Huntingdon, PA (USA)
[l] a) D. Seebach, Angew. Chem. 1979,91,259; Angew. Chem. I n t . Ed. Engl. 1979,
18,239; b) D. Seebach, Synthesis 1969,17; c) D. Seebach, Angew. Chem. 1969,
81,690; Angew. Chem. In[. Ed. Engl. 1%9,8,639; d) D. Seebach, D. Enders,
ibid. 1975, 87, 1; and 1975, 14, 35; e) D. Seebach, K.-H. Geiss, J Orgunornet
Chem. Libr. 1976, 101, 1; 0 D. Seebach, A. R. Sting, M. Hoffmann, Angew.
Chem. 1996, 108,2881 ; Angew. Chem. I n t . Ed. Engl. 1996, 35, 2795
[2] a) D Seebach, Angew. Chem. 1990, 102, 1363; Angen. Chem. In!. Ed. Engl.
1990,29, 1320; b) D. Seebach, ihid. 1988, 100, 1685; and 1988,27, 1624.
[3] Reviews of the structure oforganolithium compounds: a) P. von R. Schleyer,
Pure Appl. Chem. 1984,56,151; b) G. Boche, F. Haller, K. Harms, D. Hoppe,
W. Koch, J. Lorenz, M. Marsch, A. Opel, C. Tummler, 0 . Zschage, “Crystal
Structure and Reactivity (Selectivity) of Oxygen (Aryloxy; Carbamoyloxy;
Sily1oxy)-SubstitutedLithium Compounds” in N e u Aspects ofOrganic Chemistry 11, Kodansha, Tokyo, 1992, pp. 159-179; c) E. Weiss, Angeu. Chem.
1993, 10S, 1565; Angew. Chem. Int. Ed. Engl. 1993, 32,1501; d) C. Lambert,
P. von R. Schleyer, ibid. 1994,106, 1187 and 1994,33, 1129; e) C. Lambert,
P. von R. Schleyer, “Carbanionen - Polare Organometall-Verbindungen,” in
Methoden Org. Chem. 1952- (Houhen- Weyl), Carbanionen, 4th ed., Vol.
E19d, 1993, pp. Z - 106,
[4] Regarding the advantages of C,-symmetric ligands, see the following reviews:
a) J. K. Whitesell, Chem. Reu. 1989,89,1581; b) H. Waldmann. Nuchr. Chem.
Tech. Lab. 1991, 39, 1142.
[S] In the preceding example, a nonsymmetric C , ligand would have had the
consequence that an additional stereogenic center would have been established at the lithium cation, and it would have been necessary for us to
consider twice as many diastereomers.
[6] The term “two-dimensional stereogen” has been considered better defined: a)
D. Seebach, V. Prelog, Angew. Chem. 1982, 94, 696; Angeu Chem. Int. Ed.
2312
Engl. 1982, 21, 654. Although not as well defined, we continue to use the
simpler designation “prochiral”: b) E. L. Eliel, S. H. Wilen, Stereochemistry
oforgunic Compounds, 1st ed., Wiley, New York, 1994.
[71 On applying the term “cation” rather generotisly and even including covalently bonded metal substituents (e.g., magnesium, boron, zinc. kind titanium),
this simple concept has been shown to be exceptionally successful. Magnesium: a) R. Noyon, M. Kitamura, Angew. Chem. 1991,103,34; Angeu. Chem.
Int. Ed. Engl. 1991, 31, 49; b) B. Weber, D. Seebach, Tetrahedron 1994,50,
6117; Boron: C) H. C. Brown, P. V. Ramachandran, J. Orgunomet. Chem.
1995,500, 1 ; d) D. S. Matteson, Chem. Rev. 1989,89, 1535; e) W. R. Roush,
“Formation of C-C Bonds by Addition of Allyl-Type Organometallic Compounds to Carbonyl Compounds. Allylboron Reagents,” in Methoden Org.
Chem 1952- (Houben- Weyl), Stereoselective Synthesis, 4th ed., 1952-,
E21b. 1995. pp. 1410-1486; f) R. h!Hoffmann, Angeu. Chem. 1987,99,503;
Angen. Chem Int. Ed. Engl. 1987,26,489;g) R. W Hoffmann, R. Stiirmer,
“Synthesis of Natural Products of Polyketide Origin, An Exemplary Case,”
in Stereoselectwe Synthesis (Eds.: E. Ottow, K. Schollkopf, B. G. Schulz),
Springer, Berlin, 1994. pp. 91-108; Zinc: h) P. Knochel, Synlett 1995, 393;
Titanium: i) R. 0. Duthaler, A. Hafner, Chem. Rev. 1992, 92, 807; k) D.
Seebach, A. K. Beck, R. Imwinkelried, S. Roggo, A. Wonnacott, Helv. Chim.
Actu 1987,70,954; 1) D. Seebach. L. Behrendt, D. Felix, Angen. Chem. 1991,
103,991 ; Angeu,. Chem. Int. Ed. Engl. 1991.30, 1006; m) D. Seebach, D. A.
Plattner, A. K. Beck, Y. M. Wang, D. Hunziker, W. Petter, Helv. Chim. Actu
1995, 7S, 2171; n) A L. Costa, M. G. Piazza. E. Tagliavini, C. Trombini, A.
Umani-Ronchi, J A m . Chem. Soc. 1993, 115, 7001; 0 ) D. Seebach, A. K.
Beck. B. Schmidt, Y. M. Wang, Tetrahedron 1994,50,4363
[XI H. L. Cohen, G. F. Wright, J. Org. Chem. 1953, 18, 432.
[9] D. Seebach, G. Grass, E.-M. Wilka, D. Hilvert, E. Brunner, Helu. Chim. Acta
1979.62.2695.
[lo] H. Nozaki. T. Aratani, T. Toraya, Tetrahedron Lett. 1968, 4097.
1111 D. Seebach, H. Dorr, B. Bastani, V. Ehrig, Angew. Chem. 1969, 81, 1002;
Angew. Chem. Int. Ed. Engl. 1969,8, 982.
[12] T. Mukaiyama, K. Soai, T. Sato. H. Shimuzu, K. Suzuki, J Am. Chem. Soc.
1979, 101.1455.
[13] R. Noyori, S. Suga, K. Kawai, S. Okada, M. Kitamura, Pure Appl. Chem.
1988,60,1597
[14] J.-P. Mazaleyrat, D. J. Cram. J Am. Chem. SOC.1981, 103,4585.
[I 51 For this reason, the first successful examples involving this strategy of enantiotopically differentiated deprotonation relate to the formation of chiral
enolates from 4-substituted cyclohexanones; here the stereogenic (C-4) and
the carbanionic (C-2) centers are separated from each other; reviews: a) K.
Koga, Pure Appl. Chem. 1994,66,1487; b) K. Koga, M. Shindo, J. Synth. Org.
Chem. Jpn. 1995,53, 1021;c) N. S. Simpkins, Pure Appl. Chem. 1996,68,691;
d) P. J. Cox, N. S. Simpkins, Tetruhedron. Asymmetry 1991, 2, 1.
[16] a) R. Kuhn, Ber. Dtsch. Chem. Ges. 1932, 6S, 49; b) M. M. Harris, Progr.
Stereorhem. 1958, 2, 157; overviews of dynamic racemate resolution: c) S.
Caddick, K. Jenkins, Chem. SOC.Rev. 1996, 447; d) M. Reggelin, Nuchr.
Chem. Tech. Lab. 1997,45, 789.
[17] A somewhat more penetrating discussion is provided in P. Beak, A. Basu,
D. J. Gallagher, Y S . Park, S. Thayumanavan, Acc. Chem. Res. 1996, 29,
552.
(181 D. J. Cram. I. L. Mateos, F. Hauck, A. Langemann, K. R. Kopecky, W. D.
Nielsen, J. Allinger, J Am. Chem. Soc. 1959,81, 5774; b) D. J. Cram, C. A.
Kingsbury, B. Rickborn, h i d . 1961, 83, 3688.
[19] a) J. Stenhouse, Ann. Chem. Pharm. 1851, 78, 1 ; b) E. J. Mills, h i d . 1863, 125,
71; c) W. Boczon, Heterocycles 1992, 33, 1101.
[20] “( -)-Sparteine”: D. Hoppe in Encyclopedia of Reugents for Organic Synthesis, Vol. 7 (Ed: L. A. Paquette), 1st ed., Wiley, Chichester, 1995, p. 4662.
[21] a) F. Bohlmann, D. Schumann, C. Arndt, Tetrahedron Lett. 1965,2705; b) M.
Wiewiorowowski, 0 . E. Edwards, M. D. Bratek-Wiewiorowska, Cun. J
Chem. 1967, 45, 1447-1457.
[22] (-1-Sparteine is commercially available as the free amine (SIGMA, about 450
US dollars per mol-I) or as a sulfate pentahydrate (about 280 US dollars per
mol-I).
1231 M. Carmack, B. Douglas, E. W. Martin, H. Suss, J. Am. Chem. SOC.1955,87,
443s.
[24] a) F. Bohlmann, Chem. Ber. 1959, 92, 1798; b) Y. Okamoto, K. Suzuki, T.
Kitayama, H. Yuki, H. Kageyama, K. Miki, N. Tanaka, N. Kasai, J. Am.
Chem. Soc. 1982,104,4618;c) F. Galinovsky, P. Knoth, W. Fischer, Monutsh.
Chem. 1955, 86, 1014.
[25] A Orechoff, M. Rabinowitch, R. Kolowanowa, Ber. Dtsch. Chem. Ges. 1933,
66,621.
[261 a) T. Ebner, M. Eichelbaum, P. Fischer, C. 0.Meese, Arch. Phurm. 1989,322,
399; b) G. R. Clemo, R. Raper, W. S. Short, J Chem. SOC.1949, 663.
[27] a) H. Nozaki, T. Aratani, T. Toraya, R. Noyori, Tetrahedron 1971,27,905; b)
H. Nozaki, T. Aratani, R. Noyori, Tetrahedron Lett. 1968, 26, 2087; cf T.
Aratani. T. Gonda, H Nozaki, ibid. 1969, 27,2265; d) T. Aratani. T. Gonda,
H. Nozaki, Tetrahedron 1970, 26, 5453.
[28] In a recent paper by Denmark et al. it was shown, however, that 1 is a very
efficient catalyst for the nucleophilic alkylation of imines: S . E. Denmark, N
Nakajima, 0. JLC. Nicaise, 1 Am. Chem. SOC.1994, 116, 8797.
Angew. Chem. Int. Ed. Engl. 1997,36,2282-2316
Enantioselective
Synthesis
[29] a ) P. Knochel. Chemtracts- Org. Chern. 1995, 8, 205; b) K. Soai, S. Niwa.
Chem. Rev. 1992, 92, 833.
[30] R. A. Kretchmer, J Org. Chem. 1972, 37, 2744.
[31] B. M . Trost. T. J. Dietsche, J Am. Cheni. SOC.1973, 95, 8200.
[32] M. T Reek. J. Westermann, Syn. Commun. 1981, 11. 647.
[33] N. Kunieda. M. Kinoshita, Phosphorous Sulfur Relat. Elem. 1981, 10, 383.
[34] a) M. Guette. J.-P. Guette, J. Capillon, Tetrahedron Lett. 1971, 2863; b) M.
Guette. J. Capillon, J.-P. Guette, Tetrahedron 1973, 29, 3659.
[35] M. M. Hansen, P. A. Bartlett, C. H. Heathcock, Organometallics 1987, 6,
2069.
[36] K. Soai. A. Oshio, T. Saito, J Chem. SOC.Chem. Commun. 1993, 811
[37] a ) Y. Okamoto, K. Ohta, H. Yuki, Chem. Lett. 1977, 617; b) G. Wulff, R.
Sczepan, A Steigel. Tetrahedron Lett. 1986,27,1991;c) Review: Y Okamoto,
T. Nakano, Chem. Rev. 1994, 94. 349.
[38] a ) G Fraenkel. C. Cottrell, J. Ray, J. Russell, Chem. Commun. 1971, 273, b)
G. k‘rdenkel. B. Appleman, J. G. Ray, J Am. Chem Soc. 1974, 96, 5113.
[39] a) EtMgBr.1 and EtMgBr.l.(24b]; b) EtMgBr.1: H. Kageyama, K. Miki, N.
Tandka, N. Kasai, Y Okamoto, H. Yuki, Bull. Soc. Chem. Jpn. 1983,56,1319;
cj EtMgBr-2: H Kageyama, K. Miki, Y. Kai, N. Kasai, Y. Okamoto, H Yuki,
Acto Crysrullogr.B 1982,38,2264; d) rBuMgCl.1 and MgC1;l: H. Kageyama, K. Miki. Y Kai. N. Kasai, Y Okamoro, H. Yuk!, BUN. Chem. Sor. Jpn.
1983. 56. 241 1 .
[40] [(q,-Cyclopentenyl)Pd/l]+PF;: A. Togni, G. Rihs, P. S . Pregosin, C. Ammann. Hell’. Chim. Acra 1990, 73, 723.
[41] a ) CoC1,-I- R. Kuroda, S. F. Mason, J Chem. SOC.Dalton Trans. 1977. 371;
b) COCI,.~:R. Kuroda, S . F. Mason, ibid. 1979, 727; c) CuCI,.3: L. S .
Childers. K. Folting, L. L. Merritt Jr. W. E. Streib, Acra Crystallogr. B 1975,
31. 924.
[42] L. T. Byrne. L. M. Engelhardt, G. E. Jacobsen, W.-P. Leung, R. 1. Papasergio,
C. L. Raston, B. W. Skelton, P. Twiss. A. H. White. J Chem. SOC.Datton
Trans. 1989, 105.
1431 R. I. Papasergio, B. W. Skelton, P. Tw!ss,A. H. White, C. L. Raston, J Chem.
Sor. Dalton Truns. 1990, 1161
[44] B Ledig, M. Marsch, K. Harms, G. Boche, Angew. Chem. 1992, 104, 80;
Angen. Chem. Int. Ed. Engl. 1992, 31, 79.
[45] D. Hoppe, T. Kramer, Angew. Chem 1986, 98, 171 ; Angew. Chem. Int. Ed.
Engl. 1986. 25, 160.
[46] Reviews- a) D. Hoppe, Angew Chem. 1984,Y6, 930; Angen.. Chem. Inr. Ed.
Engl. 1984.23,932;b) D. Hoppe, T. Kramer, J.-R. Schwark, 0.
Zschage. Pure
Appl. Chem. 1990, 62, 1999; c) D. Hoppe, 0. Zschage, “Chiral Metalated
Cdrbamates: Tools for New Strategies in Asymmetric Synthesis” in Organic
Synthesis viu Oganometalhcs (Ed.: K. H. Dotz, R. W. Hoffmann), 1st ed..
Vieweg. Brdunschweig 1990, pp. 267; d) H. Kunz, H. Waldmann, Chemrractst
Org. Chen7. 1990, 3, 421; e) “Formation of C-C Bonds by Addition of
Allyl-Type Organometallic Compounds to Carbonyl Compounds”: D.
Hoppe, Methoden Org. Chem. (Houben-Weyl) ,41h ed., 1952-, E21b, 1995, p.
1551 ; f) “Generation and Reactions of Chiral. Mesomerically Stabilized
Lithium Carbanions”: D. Hoppe in New Horizons in Organic Synthesis (Eds.:
V. Nair. S. Kumar). New Age, New Delhi, 1996, p. 130.
[47] 0. Zschage. Diplomarbeit, Universitit Gottingen, 1989.
[481 Correction of the configuration for 13.1: 0. Zschage, D. Hoppe, Tetrahedron
1992, 48, 5657.
[49] a ) P. Le Menez. V. Fargeas, J. Poisson, J. Ardisson, J.-Y. Lallemand, A. Pancrazi, TerrahedronLerr. 1994,35,7767;b) P. Le Menez, N. Firmo, V. Vargeas,
J. Ardisson, A. Pancrazi, Synktt 1994, 995.
[50] H. Paulsen, D. Hoppe. Tetrahedron 1992, 48, 5667.
[St] B. Peschke. J LiiRmann, M. Dyrbusch, D. Hoppe, Chem. Ber. 1992, 117,
1421
[52] D. Hoppe. 0. Zschage. Angew. Chem. 1989,101.67; Angen! Chem. Int. Ed.
Engl. 1989, 28. 67.
[53] Reviews of lithium-titanium exchange: a) “Titanium in Organic Synthesis”:
M. T Reetz in Organometallics rn Synihesis (Ed.: M. Schlosser), Wiley,
Chichester. 1994, p. 1994;b) ref[7i]; C)M. T. Reetz, Organotitanium Reagents
111 Organic Synthesis, Springer, Berlin, 1986; d) “Titanium and Zirconium
Derivatives in Organic Synthesis”: D. Seebach, B. Weidmann, L. Widler in
Modern Synfheric Methods 1983 (Ed.: R. Scheffold), Salle und Sauerlander,
Frankfurt und Aarau, 1983, p. 217; e) B. Weidmann, D. Seebach, Angew.
Chern. 1983. 95. 12; Angew. Chem. In!. Ed. Engl. 1983, 22, 32.
[54] The configurational stability of the titanium intermediate, together with complete 1.3-chirdhty transfer, must be concluded from the distinctively different
diastereomeric ratios achieved on addition of enantiomerically pure and
racemic aldehydes [571.
[55] H. Paulsen. C. Graeve, D. Hoppe. Synthesis 1996, 141.
[56] Review S. Masamune, W. Choy, J. S . Petersen, L. R. Sita, Angew. Chem.
1985, Y7. 1 : Angew. Chem. Int. Ed. Engl. 1985, 24, 1
[57] R. W Hoffmann, J. Lanz, R. Metternich,G. Tarara, D. Hoppe, Angew. Chem.
1987. Y9. 1196; Angew. Ckem. Int. Ed. Engl. 1987,26, 1145.
[58] D. Hoppe. G. Tarara. M. Wilckens. Synthesis 1989, 83.
[59] 0. Zschage. J:R. Schwark, D. Hoppe, Angew. Chem. 1990,102,336; Angew.
Chem. Int Ed. Engl. 1990 29, 296.
[60] 0. Zschage. D. Hoppe, Tetrahedron 1992, 48. 8389.
Angew. Chem Int Ed. Engl 1997,36, 2282-2316
[61] Reviewen: a) J. A. Marshall, Chemtracts. Org. Chem. 1992. 5. 75; b) Chem.
Rev. 1996, 96, 31
[62] H. Paulsen, C. Graeve, R. Frohlich, D. Hoppe, Sjnthesrs 1996, 145.
1631 M. Marsch, K. Harms, 0.Zschage, D. Hoppe, G. Boche, Angew. Chem. 1991,
103, 338; Angew Chem. Int. Ed. Engl. 1991, 30, 321.
[64] D. Hoppe, R. Hanko, A. Bronneke, F. Lichtenberg, E. van Hiilsen, Chem.
Ber. 1985, 118, 2822.
[65] a) D. Hoppe. I Kramer, J.-R. Schwark, 0. Zschage, Pure Appl. Chem. 1990,
62. 1999; b) T. Kramer, D. Hoppe, Tetrahedron Lett. 1987. 28, 5149; c) T.
Kramer, J.-R. Schwark, D. Hoppe, &id. 1989,30,7037; d j 0. Zschage, J.-R.
Schwark, T Kramer, D. Hoppe, Tetrahedron 1992.48. 8377; e) H. Helmke,
Dissertation. Universitat Miinster, 1995.
[66] J.-R. Schwark, Dissertation, Universitat Kiel, 1991.
[67] For reasons of space. a presentation in context must be deferred to a planned
separate review article. A few representative examples are provided in
refs. [68, 69. 701 as well as [46].
[68] Metalate rearrangements and related reactions: a) P. Kocienski, N. J. Dixon,
Synlett 1989, 52; b) A. Pimm, P. Kocienski, S. D. A. Street, ibid. 1992, 886;
c) P. Kocienski, C . Barber, Pure Appl. Chem. 1990, 62, 1933; d) [49a]; e) D.
Madec, S Pujol, V. Henryon, J. P. Ferezou, Synletr 1995. 435.
I691 Epoxjdation and furanosde synthesis: a) D. Hoppe, .I. LiiBmann, P. G. Jones,
D. Schmidt, G. M. Sheldrick, Tetrahedron Lett. 1986. 27. 3591; b) J.
LiiRmann, D. Hoppe, P. G. Jones, C. Fittschen, G. M. Sheldrick, ihid. 1986,
31,3595; c) ;[58] d) G. Tarara, D. Hoppe, Svnrhesis 1989,89;e) [SO];3-acyltetrahydrofurans: f ) D. Hoppe, T. Kramer, C Freire Erdbriigger, E. Egert,
Tetrahedron Lett. 1989,30, 1233; synthesis of dipeptide isosteres: g) R. Hanko, K. Rabe, R. Dally, D. Hoppe, Angew. Chem. 1991. 103. 1725; Angew.
Chem. Int. Ed. Engl. 1991,30,1690;h) F. Rehders. D. Hoppe. Synthesis 1992,
859; i) F. Rehders, D. Hoppe, ihid. 1992, 865.
[70] Applications in the synthesis ofcomplex natural products: Dehydroavermectin BI,: a) J. P FCrezou, M. Julia, R. Khourzom, Y. Li, A. Pancrazi, P Robert,
Synlerr 1991, 611; Desepoxy-Rosaramycin: b) ref.[49b], c) P. Le Menez. I .
Berque, V. Fargeas, J. Ardisson, A. Pancrazi, S.vnlerr 1994.998;d) Le Mtnez.
N. Firmo, V. Fargeas, J. Ardisson, A. Pancrazi, hid. 1994, 995; jaspamide:
e) P. Ashworth, B. Broadbelt, P. Jankowski, A. Pimm. P. Kocienski, Synthesi.s 1995, 199; herboxidiene A: f) N. D. Smith, P. J. Kocienski, S. D. A.
Street, ibid. 1996, 652; rue-tylosine: g) P. Le Menez, V. Fargeas, I. Berque, J.
Poisson, J. Ardisson, JY
: . Lallemand. A. Pancrazi. J Org. Chem. 1995,
60, 3592.
[71] D. Hoppe, A. Bronneke, Tetrahedron Lett. 1983. 24, 1687.
[72] D. Hoppe, A. Bronneke, Synthesis 1982, 1045.
[73] R. W. Hoffmann, T. Riihl, J. Harbach, Liebigs Ann. Chem. 1992,725.
[74] a) A. Carstens, D. Hoppe, Terrahedron 1994, 50, 6097; b) C Derwing, D.
Hoppe, Synthesi.? 1996,149; c) E Hammerschmidt. A. Hanninger, Chem. Ber.
1995, 128, 1069.
[75] Information on the stereochemical course of electrophilic substitution of chiral benzyllithium compounds is still quite fragmentary. Only in the case of
(I-lithio-1-phenylethy1)-NNdiisopropylcarbamate [74a.b] and 2,4,6-triisopropyl benzoate,[74c) which are accessible in enantiomerically enriched fom
with known configuration through deprotonation of optically active precursors, do there exist firm correlations between the configurations of starting
materials and products. The trend toward stereoinversion in reactions with
alkyl and acyl halides, carbon dioxide, carbon disulfide, and silyl and stannyl
chlorides is strong. Protonation with alcohols and carboxylic acids, hydroxyalkylation with aliphatic aldehydes, and acylation with esters occur with
retention; in these cases one can assume a preassociation of the reagent at the
cation. The experimental basis is too small for solid generalization, moreover,
a strong dependence on individual factors is to be anticipated.
[76] Zhang and Gawley carried out similar experiments in THF and isolated
racemic products. This once again verities that (-)-sparteine is easily displaced from the lithium cation by THF: P. Zhang, R E Gawley, J Org.
Chem. 1993.58. 3223.
[77] a) P. Beak, H. Du, J Am. Chem. Sac. 1993, 115,2516; bj G. P. Lutz, H. Du,
D. J. Gallagher. P. Beak, J Org. Chem. 1996, 61, 4542
1781 D. J. Gallagher, H. Du, S . A. Long, P. Beak, J Am. Chem. Soc. 1996, 118,
11391.
[79] M. Schlosser, D. Limat, J Am. Chem. SOC.1995, 117, 12342.
[801 As we have already established in related cases [I36, 2351, deprotonation is
subject to an unusually large kinetic H/D isotope effect.
[XI] Synthesis of (R)-phenylsarcosine: N. Voyer, J. Roby, Tetrahedron Lett. 1995,
36, 6627
[821 G. Boche, M. Marsch, J. Harbach, K. Harms, B Ledig. F Schubert. J. C. W.
Lohrenz, H. Ahlbrecht. Chem. Ber. 1993, 126, 1887.
[83] a) Y. S. Park, M. L. Boys, P. Beak, J Am. Chem. Soc. 1996. 118,3757; other
applications and examples: b) Y S . Park, P. Beak, J Org. Chem 1997, 62,
1574.
I841 We refrain here from provjding an interpretation m depth. If t h e individual
lithiation and substitution steps are subject to the same stereochemical course
as was proven to be the case for the lithium derivatives of secondary benzyl
carbamates [75], 38-1 must possess the (R)-configuration [83b].
[85] S. Wu, S. Lee, P. Beak, J Am. Chem. SOC.1996, 118. 715.
2313
REVIEWS
[86] S. Thayumanavan, S. Lee, C. Liao, P. Beak, L Am. Chem. SOC.1994, 116,
9755.
[87] The configuration at the benzylic center is unknown.
[SS] a) A. Basu, P. Beak, L An?. Chem. Soc. 1996, 118, 1575; b) A. Basu, D. J.
Gallagher, P. Beak, L Org. Chem. 1996, 61, 5718.
[89] Configurational assignment is based on the assumption of stereoinversion in
the course of silylation.
[90] V. Snieckus, lecture on December 16, 1996, in Miinster.
[91] S. Retzow, dissertation, Universitat Kiel, 1993.
[92] The absolute configuration of the predominant stereoisomer is unknown.
1931 For 55a.llepI-55a.1, a free energy of epimerization AGG, of about
67 kJmol- was estimated from the coalescence temperature, whereas for the
pair 55b. llepr-55b. 1 a value of > 78 kJ mol- results in the same way.
[94] An additional problem will be touched upon only briefly: Since (-)-sparteine
is not C,-symmetric, a stereogenic center is also created at the tetrahedrally
coordinated lithium cation. In this way four diastereomers compete. All experimental indications are consistent with the view that epimerization at lithium is much more rapid than that a t the carbanionic C atom.
I951 a) J. J. P. Stewart, L Comp. Chem. 1989, 10, 209; b) E. Anders, R. Koch, P.
Freunscht, ibid. 1993, 14, 1301.
[96] T. Heinl, Diplomarbeit, Universitat Miinster, 1996.
[97] I. Hoppe, M. Marsch, K. Harms, G. Boche, D. Hoppe, Angew. Chem. 1995,
107,2328; Angew. Chem. h i . Ed. Engl. 1995,34,2158
[98] The addition of tetra(isoprop0xy)titanium leads to increases in the yield and
product purity of the resulting carbinols, but does not change either the
enantio- or the diastereoselectivity. From this we conclude that metal exchange does not arise in the carbanionic intermediate. Rather, we suspect that
the lithium alcoholate is bound as a titanate, and hence possible subsequent
and retroreactions are prevented.
I991 In a general sense, and with the same consequences with respect to regioselectivity, intermediates can be formulated in which the carbonyl oxygen atom
has displaced one coordination site of (-)-sparteine.
[lo01 A lithium-sparteine contact has also been demonstrated by NMR spectroscopy in diethyl ether solution, whereas in T HF solution no such evidence
was discovered; S. Gessler, Diplomarbeit, Universitat Marburg, 1996.
[loll I. Hoppe, unpublished.
[lo21 G. A. Weisenburger, P. Beak, J Am. Chem. Sac. 1996, 118, 12218.
11031 The authors suggest a $structure 67.1 and not a q‘-z-lithio ion pair [102].
Presumably the marked y-selectivity of the substitution reaction serves as an
argument. We were able to show that related ~i-c-lithioallylion pairs (for
which the I-chelate structure was established) display a strong tendency
toward y-alkylation [46a, 64, 6SdI.
I1041 Only the (S)-configuration of 68d was rigorously proven 11021.
[lo51 The authors claim that the provisional assignments of absolute configuration
for the stannanes ent-68e and ent-69e are based on analogies to our results
[102]. On this basis we come to the reverse conclusion (68eand 69e), but then
encounter contradictions with respect to the configurational assignment of
the alkylation products 68b and enr-68b.
[lo61 The high enantioselectivity is only useful in a preparative sense if the ( E ) -and
(2)-enamides 71 and 72 are separated and they do not interconvert during
workup and separation.
[lo71 W.C. Still, C. Sreekumar, J Am. Chem. SOC.1980, 102, 1201. See also
ref. [lOS].
[I081 J. S. Sawyer, A. Kucerovy, T. L. Macdonald, G. J. McGarvey, L Am. Chem.
SOC.1988,110, 842.
[lo91 J. S. Carey, T. S. Coulter, D. J. Hallett, R. J. Maguire, A. H. McNeill,
S. J. Stanway, A. Teerawutgulrag, E. J. Thomas, Pure Appl. Chem. 1996,68,
707.
[110] As demonstrated by many examples, sp3-hybridized (thus non-resonancestabilized) alkyllithium derivatives react with retention of configuration,
apart from a few very special exceptions.
[1111 a) P. C.-M. Chang, J. M. Chong, J Org. Chem. 1988, 53, 5584; b) J. A.
Marshall, W. Y. Gung, Terrahedron 1989, 45, 1043; c) Ref. [135].
[112] a) T. Cohen, M.-T. Lin,J. Am. Chem. SOC.1984,106, 1130; b) C. G. Screttas,
M. Micha-Screttas, L Org. Chem. 1978, 43, 1064.
[113] For elegant synthetic applications based on diastereoselective “capture” of
the lithium cation as well as thermodynamically controlled equilibrations, see
a) S. D. Rychnovsky, D. E. Mickus, Tetrahedron Lert. 1989,30,3011; b) S. D.
Rychnovsky, J Org. Chem. 1989,54,4982.
[114] N. G. Rondan, K . N. Houk, P. Beak, W. J. Zajdel, J. Chandrasekhar,
P. von R. Schleyer, J Org. Chem. 1981, 46, 4108.
[115] E. J. Corey, T. M. Eckrich, Terruhedron Letr. 1983, 24, 3165.
[116] a) P. Beak, B. G. McKinnie, J. Am. Chem. Sac. 1977, 99, 5213; b) P. Beak,
L. G. Carter, J Org. Chem. 1981, 46, 2363.
[117] R. Schlecker, D. Seebach, W. Lubosch, Helv. Chim. Acra 1978,61, 512.
[118] The 2-(dimethylaminoethyl) ester of type 80 constitutes an exception, activation through the chelating dialkylamino group here certainly plays a significant role; cf. [116b].
[119] First application of N,N-dialkylcarbamoyl groups in the activation of alcohols: D. Hoppe, R. Hanko, A. Bronneke, Angew. Chem. 1980, 92, 637;
Angew. Chem. Inr. Ed. Engl. 1980, 19, 625.
2314
D. Hoppe and T. H e n s
[1201 a) P. Tebben, dissertation, Universitat Kiel, 1991; b) T. Raffel, Diplomarbeit,
UniversitPt Kiel, 1990.
[I211 T. Nakai, personal communication, 1996.
[122] F. Hintze, D. Hoppe, Synfhesis 1992, 1216.
[I231 D. Hoppe, F. Hintze, P. Tebben, Angew. Chem. 1990, 102, 1457; Angew.
Chem. Inr. Ed. Engl. 1990, 29, 1422.
[1241 P. Mickon, A. Rassat, Bull. Soc Chim Fr. 1971, 3561; see also Ref. [122].
[12S] Coalescence is achieved in CDCl, by warming to about 70°C; J. Haller,
Diplomarbeit, Universitat Kiel, 1992.
[1261 If there are additional hydroxy groups in the vicinity of the carbamoyloxy
group, reductive ester cleavage with lithium aluminum hydride occurs readily.
Apparently these bring the metal hydride into place.
1127) a) J. Tanaka, J Chem. Soc. Perkin Truns. 2 1989,1009; b) K. S. Rein, TerrahedronLrtr. 1991,32.194l;c)Gawley,L Org. Chem. 1989,54,175;d) J. J. Eisch,
Res. Chem. Inrermed. 1996, 22, 145.
[128] T. Hense, dissertation, Universitat Miinster, 1996.
[I291 a) D. Seebach, R. Henning, T. Mukhopadhyay, Chem. Ber. 1982, f15, 1705;
b) T. Mukhopadhyay, D. Seebach, Helv. Chim. Aefa 1982,65, 385.
11291 T. Hense, Dissertation, Universitat Miinster, 1996.
[130] F. Hintze, Dissertation, Universitat Kiel, 1993.
[131] K. Behrens, dissertation, UniversitPt Miinster, 1997.
[132] I. Hoppe, unpublished (1991).
[I331 R. H. Bartelt, A. M. Schaner, C. L. Jackson, L Chem. Ecol. 1989, 15, 399.
[1341 a) K. Mori. Terrahedron 1975.31,3011; b) J. H. Borden, E. Stokkink, Can. L
Zool. 1973, 51. 469.
[135] Y. Yamamoto, Chemrractst Org. Chem. 1991, 4, 255
[136] D. Hoppe, M. Paetow. F. Hintze, Angew. Chem. 1993, 105, 430; Angew.
Chem. inr. Ed. Engl. 1993, 32, 394.
[I371 We wish to thank Professor D. 0. Collum for stimulating us to carry out this
investigation through a comment at the Gordon Conference in Newport
(1992).
[I381 Starting from (S)-configured I-deuterocarbamates of the type 97 one is able
to enter the series of enantiomeric products enr-97. However, these suffer the
blemish of being deuterated. We are aware of no precedent in which a kinetic
isotope effect has been similarly utilized for the synthesis of highly enantiomerically enriched products.
[139] A high, though not quantified, kinetic HID isotope effect was observed in the
deprotonation of S-ethyl thiobenzoates; D. B. Reitz, P. Beak, R. F. Farney,
L. S. Helmick, L Am. Chem. SOC.1978,100, 5428.
[140] a) Review. “Isotope Effects in Hydrogen Atom Transfer Reactions,” E. S.
Lewis in Isotopes in Organic Chemistry, Vol. 2 (Eds. E. Buncel, C C. Lee),
Elsevier, Amsterdam 1976, p. 127; b) E. S. Lewis, L. H. Funderburg, J. Am.
Chem. SOC.1967,89,2322.
[I411 J. Haller, dissertation, Universitat Miinster, 1995.
[142] D. J. Gallagher, S. T. Kerrick, P. Beak, 1 Am. Chem. Sac. 1992, 114, 5872.
[143] In this way the useless stereogenic center in the base is avoided. It follows from
studies by Beak that this exchange has virtually no influence on the stereoselectivity [142].
[I441 We believe that despite the small calculated energy difference the calculated
structures portray the decisive distinction. It is very likely that the “true”
transition states are more compact than the calculated ones. For the ground
state of the lithium-sparteine-methylindenide complex 58a-1: Li-N distances measured in the crystal structure (209.5 pm and 214.1 pm) are noticeably shorter than the calculated distances (225.2 pm und 226.4 pm). T. Heinl,
Diplomarbeit, Universitat Miinster, 1996.
[145] J. Haller, T. Hense, D. Hoppe, Liebigs Ann. Chem. 1996, 489.
(1461 These investigations were carried out with racemic mixtures, for greater clarity, one enantiomer has been singled out.
[147] T. Hense, Diplomarbeit, Universitat Miinster, 1993.
(1481 Those with little practice in stereochemistry are asked to forgive us for the
confusion we may have perpetrated. For reasons of clarity we have held
constant the configurational arrangement at the reaction center in the calculations, and instead varied the configuration at the stationary center. This is
legitimate, since all such transformations that obey the relative topicity Ik
(like) [6a]-irrespective whether R,Re, S,Si or R*,Re*-must overcome the
same energy barrier. The same applies in principle to the ul (unlike)series.
[I491 J. Haller, T. Hense, D. Hoppe, Synlerr 1993, 726.
[l SO] P. Sommerfeld, D. Hoppe, Synlerr 1992, 764.
[lSl] The great tendency toward complexation for y-Me,N groups in organolithium compounds is well known: G . W. Klumpp, M. Vos, F. J. J. de Kanter, C.
1985, 107,8292.
Slob, H. Krabbendam, A. L. Spek, J. Am Chem. SOC.
[I 521 Because of the unfavorable solubility characteristics of the intermediate carboxylic acid, methyl iodide was used as trapping reagent. The observed enantiomeric excesses are independent of the electrophile [ISO].
[153] a) M. Paetow, H. Ahrens, D. Hoppe, Tetrahedron Lett. 1992,33, 5323; b) H.
Ahrens, dissertation, Universitat Miinster, 1994.
[I541 J. Schwerdtfeger, D. Hoppe, Angew. Chem. 1992, 104, 1547, Angew. Chem.
Inr. Ed. Engl. 1992, 31, 1505.
[155] S. Kleinfeld, Diplomarbeit, Universitat Miinster, 1996.
[156] D-Pantolactone (124) is produced by BASE The key step here is a sophisticated racemate resolution on a technical scale.
Angew. Chem. Int. Ed. Engl. 1997, 36, 2282-2316
Enantioselective Synthesis
[157] M Paetow. M. Kotthaus, M. Grehl, R. Frohlich, D. Hoppe, Synlett 1994,
1034.
11581 I t is unclear whether Me,SiC1 or LiCl formed in the process functions as Lewis
acid.
[I591 Cyclopropyl anions possess a very high barrier to inversion: H. M. Walborsky, C. Hamdouchi, J. Am. Chem. Soc. 1993, 115,6406 and earlier papers
by these authors.
[160] M. Kotthaus, dissertation, Universitat Miinster, 1996.
[161] This represents the only case in which stereoinversion of a lithiated alkyl
carbamate was observed during the reaction.
[162] The stereochemistry at the nucleophilic center during 1,3-eliminations was
accorded little attention until recently: a) B. Beruben, I. Marek, J.-F. Normant, N. Platzer, Tetrahedron Lett. 1993,34,7575; b) J. Org. Chem. 1995,60,
2488; c) J.-F. Normant, Chemtracrs. Org. Chem. 1994, 7, 59; d) A. Krief, M.
Hobe, Tetrahedron Lett. 1992,33,6529; e) A. Krief, M. Hobe, W. Dumont, E.
Badaoui, E. Guittet, G. Evard, ibid. 1992, 33, 3381; f) N. Isono, M. Mori, J.
.Org. Chem. 1996, 61, 7867.
[163] Similar findings are obtained with 2-monosubstituted 1,3-dicarbamates ofthe
type 121 [160]. Because of the complicated stereochemical relationships (the
CH,OCbv groups are enantiotopic, their protons diastereotopic), a presentation will be dispensed with in this context.
[164] Reviews ofdirectea lithiation: a) P. Beak, A I Meyers, Arc. Chem. Res. 1986,
IY, 356, b) D D. Clark, A. Jahangir, Organic Reactions 1995, 47, 1; c) V.
Snieckus, Chenr. Rev. 1990, 90, 879.
[165] a ) C. Boie, dissertation, Universitdt Munster, 1996; b) C. Boie, D. Hoppe,
Synthesis 1997, 176.
[166] H Ahrens, M Paetow, D. Hoppe, Tetrahedron Lett. 1992,33, 5327.
[167] We do not know the precise structure of bis-chelate complexes like 131. It is
also possible that the more effectively donating carbonyl group of the y- or
3-carbamoyloxy residue binds, but then a seven- or eight-membered ring
would need to form. We can also only speculate about the fourth ligand L;
Et,O or Me,NCHICH,NMe, (TMEDA) bound in monodentate fashion are
possible candidates.
[I681 Irrespective of the fact that the deprotonation step is kinetically directed, the
energetically more favorable bicyclic chelate complex is formed in excess.
[169] H. Helmke. D. Hoppe, Synlert 1995, 978.
[170] Based on MOPAC calculations [65e], there is an energy difference of
4.6 kJ mol- between the simplified complexes 134 and epi-134 (Me,NC=O
for Cby. L = Me,O).
[171] For the preparation of (+)-sparteine from rac-lupinane, which is obtained
from the seeds of Lupinus alha, see ref. [26a].
[172] Another elegant route for diastereoselective generation of protected 1,3-dihydroxyalkanides was developed by S . E. Rychnovsky: a) [113b]; b) T. I.
Richardson, S. D. Rychnovsky, J. Org. Chem. 1996,61,4219; review: c) S. D.
Rychnovsky. Chern. Rev. 1995, 95, 2021.
[173] a) W. Guarnieri. Diplomarbeit, Universitat Kiel, 1992; b) W. Guarnieri, M.
Grehl, D Hoppe, .4ngew. Chem. 1994,106,1815; Angew. Chem. Int. Ed. Engl.
1994.33. 1734; c) P. Gmeiner. A. Kirtner, Synrhesis 1995, 83.
[174] The energy difference in the ground state between the complexes (1S)-138 and
(1 R)-138 is apparently smaller than the relevant free energies of activation,
because the addition of lithium bromide initiates an epimerization that leads
to a 1 :2 mixture [175]. Nevertheless, the possibility that a further LiBr-containing aggregate enters into competition cannot be ruled out [175].
[175] M. Sendzik, Diplomarbeit, Universitat Kiel, 1995.
[176] We do not know the structure of 138; our suggestion offers one plausible
interpretation of the experimental findings. In particular, it remains unclear
in what way the 4-OCby group is bound to the lithium cation, and whether an
additional ligand L (here Et,O) is acquired.
[177] M. Sendzik, dissertation, Universitat Miinster, 1997.
11781 This experiment proves that there is a kinetically determined deprotonation.
[179] As demonstrated here, the high H/D kinetic isotope effect permits a very
general utilization of deuterium as protective group in the deprotonation of
alkyl cdrbamates.
[180] Preparation of 2-(dibenzy1amino)alkanols: a) L. Velluz, D. Amiard, R.
Heymes, BUN Sor. Chim. Fr. 1954. 1012; 6 ) M. T. Reetz, M. W. Dtewes, A.
Schmidt, Angeu. Chem. 1987,99,1186; Angew. Chem. Int. Ed. Engl. 1987,26,
1141: c) ref. [173cl.
[181] The following problem has been avoided in Figure 10: With the exception of
two energetically unfavorable conformational arrangements, the pyramidal
dibenzylamino group contributes to varying degree to the shielding of the
front and rear faces. It is quite conceivable that the influence of the “small”
methyl group is overcompensated in this way.
[182] For experiments with N-(diphenylmethylene) derivatives see ref. [165]
[183] S Kolczewski, dissertation, Universitat Miinster 1995.
11841 B. Weber, D. Hoppe, unpublished.
[185] T. Hense. D. Hoppe, Synthesis 1997, in press.
[186] P. Sommerfeld. dissertation, Universitat Miinster, 1995.
I1871 Enanrioselective total syntheses of ( + )-sedridine: a) S. Murahashi, Y.Imada,
M. Kohno. T. Kawakami, Synfett 1993,395;b) C Louis, C. Hootel&,Tetrahedron- AJymmerry 1995, 6, 2149, c) B. J. Littler, T. Gallagher, I. K. Boddy,
P. D. Riordan, Synlett 1997, 22.
’
Angew. Chem. I n / . Ed. Engl. 1997,36, 2282-2316
REVlEWS
[188] J. van Bebber, D. Hoppe, unpublished.
11891 No indications of participation by the 3-OCby-group were detected.
11901 P. H. J. Carlsen, T. Katsuki, V. S . Martin, K. B. Sharp1ess.J. Org. Chem. 1981,
46,3936
[I911 R. Frohlich, M. Grehl, personal communication (1994).
[192] P. Beak, W. J. Zajdel, J. Am. Chem. SOC.1984,106, 1010.
[193] a) S. T. Kerrick, P. Beak, J. Am. Chem. Soc. 1991, 113,9708; b) P. Beak, S T.
Kerrick, S. Wu, J. Chu, ibid. 1994, fl6,3231.
[194] a) E. J. Corey, R. K. Bakshi, S. Shibata, J Am. Chem. Soc. 1987,109,5551;
b) E. J. Corey, R. K. Bakshi, S. Shibata, C. P. Chen, K. K. Singh. (bid. 1987,
109,7925; c) review: S . Wallbaum, J. Martens, Tetrahedron Asymmetry 1992,
3, 1475.
[I951 a) R. E. Gawley, Q . Zhang, J. Am. Chem. SOC.1993, 115, 7515; b) T. R.
Elworthy, A. I. Meyers, Tetrahedron 1994, 50, 6089; c) K. E Gawley, Q.
Zhang, J. Org. Chein. 1995, 60, 5763; d) R. E. Gawley, Q. Zhang, S. Campagna, J. Am. Chem. Soc. 1995, I f 7,11817: e) I. Coldham. R. Hufton, D. J.
Snowden, ibid. 1996, 118, 5322.
[196] For a detailed discussion, see a very recent review by Beak: ref. [I71
[197] D. J Gallagher, P. Beak, J. Org. Chem. 1995, 60, 7092.
[I981 M. E. Kopach, A. I. Meyers, J Urg. Chem. 1996,61, 6764.
11991 D. J. Gallagher, S. Wu, N. A. Nikolic, P. Beak, J Org. Chrm. 199560, 8148.
[200] In orientational experiments with N-Boc-piperidine we obtained nearly
racemic products; I. Hoppe, unpublished (1990).
[20l] Y S. Park, P. Beak , Tetrahedron 1996, 52, 12333.
[202] a) D. M. Hodgson, G. P. Lee, Chem. Commun. 1996,1015; b) D M. Hodgson, R. Wisedale, Tetrahedron ’ Asymmetry 1996, 7, 1275: c) D. M. Hodgson,
G. P. Lee, [bid. 1997, 2303.
Chem. Conzmun. 1993, 1193.
[203] M. Lautens, C. Gajda, P. Chiu, J. Chem. SOC.
12041 R. A. Muci, K. R. Campos, D. A. Evans, J. Am. Chem. So<..1995,117,9075.
12051 Brief reviews: a) A. Togni, Angew. Chem. 1996,108,1581: Angew. Chem. Int.
Ed. Engl. 1996,35, 1475; b) S. Borman, Chem. Eng. News 1996, 74 (30), 38.
[206] With respect to relevant experiments with isopropylferrocene (199,
X = CHMe,) by Nozaki et a]. see Section 2 and refs. [27c,d]
[207] a) M. Tsukazaki, M. Tinkl, A. Roglans, B. J. Chapell. N. J. Taylor, V.
Snieckus, J. Am. Chem. SOC.
1996, 116.685; b) an application. H. Jendralla,
E. Paulus, Synlett 1997,471
[208] D. Price, N. S. Simpkins, Tetrahedron Let/. 1995, 36, 6135.
[209] Y.Nishibayashi, Y Arikawa, K. Ohe, S. Uemura, J. Org. Chem. 1996, 61,
1172.
[210] E. P. Kiindig, A. Quattropani, Tetrahedron Lett. 1994, 35, 3497.
I2111 M. Uemura, Y Hayashi, Y. Hayashi, Tetrahedron: Asymmetry 1994,5, 1427.
[212] D. A. Price, N. S. Simpkins, A. M. MacLeod, A. P. Watt, J. Org Chem. 1994,
59, 1961.
[213] J. Aube, Chemtracrs: Urg. Chem. 1994, 7,413.
[214] H -G. Schmalz, K. Schellhaus, Tetrahedron Lett. 1995, 36. 5515.
[215] L. M. Engelhardt, W.-P. Leung, C. L. Kaston, G. Salem, P. Twiss, A. H.
White, J. Chem. Soc. Dalton Trans. 1988, 2403.
[216] S. Thayumanavan, P. Beak, D. P. Curran. Tetrahedron Lett. 1996, 37, 2899.
[217] A very effective approach to ( - )-sparteine-assisted synthesis of enantiomerically enriched binaphthyls consists of the oxidative coupling of naphthols: a)
M. SmrEina, J. Polakova, S. Vyskoi.11, P. KoEovsky, J. Org. Chem. 1993, 58,
4534; b) T. Osa, Y Kasiwagi, Y Yanagisawa, J. M. Bobbitt, J. Chem. SOC.
Chem. Commun. 1994,2535.
[218] Reviews: a) P. Knochel, Comprehensive Organic Synthesis, Vol. 4 (Eds.: B. M.
Trost, I. Flemming, M F. Semmelhack), Pergamon, Oxford, 1991, p. 865;
b) R. L. Subramanian, Methoden Org. Chem. (Houben- Weyl) , 4th ed. 1952-,
E19d, 1993, p. 744.
12191 a) W. F. Bailey, K. V. Gavaskar, Tetrahedron 1994,50, 5957; b) V. N. Drozd,
Uy. A. Ustynyuk, M. A. Tsel’eva, L. B. J. Dmitriev, J. Gen. Chem. USSR
(Engl. Transl.) 1968, 38, 2144; Zh. Obsch. Khim. 1968, 38, 2144.
[220] A. Krief, B. Kenda, B. Remade, Tetrahedron Lett. 1995. 36, 7917.
[221] a) S. Klein, I. Marek, J-F. Poisson, J.-F. Normant, J. Am. Chem. Soc. 1995,
117, 8853; b) S. Norsilian, 1. Marek, J.-F. Poisson, J. F. Normant, J. Org.
Cheni. 1997, 62, 4898.
[222] Brief review: ref. [162c]
[223] Proof of stereoinversion in the alkylation of benzyllithium derivatives:
ref. [74b].
[224] Structure of lithium alcoholates in nondonor solvents: B. Goldfuss, P. von R.
Schleyer, F. Hampel, J. Am. Chem. SOC.1996, 118, 12183 and references
therein.
[225] To our knowledge, this discovery constitutes the first example of a catalytic
enantioselective lithiation induced by (-)-sparteine.
[226] D. Amurrio, K. Khan, E. P. Kiindig, J. Org. Chem. 1996, 61, 2258.
12271 Suggestions regarding the structure of the carbanionic intermediate were not
published [226]. The propargyl residue enters from the side complexed by
chromium; thus a bonding interaction between the metal and the alkynyl
residue is very probable.
I2281 J. G. Peters, Diplomarbeit, Universitat Miinster, 1996.
12291 Stereochemistry of the electrophilic substitution of lithiated secondary benzyl
carbamates: ref 1741.
Chem. Commun. 1976. 738; b) review. C. D.
[230] a) J E. Baldwin, J. Chem. SOC.
Johnson, Arc. Chem. Res. 1993, 26, 476.
2315
D. Hoppe and T. Heme
REVIEWS
[231] M. J. Woltering, R. Frohlich, D. Hoppe, Angew,. Chem. 1997, 109, 18041805; Anger*,.Chem. In!. Ed. Engl. 1997,336, 1764-1766.
[232] a) H. J. Reich, M. D. Bowe, J. Am. Chem. SOC.1990,112,8994; b) H. J. Reich,
R. R. Dykstra, Angen,. Chem. 1993, 105,1489; Angen. Chem. Int. Ed. Engl.
1993, 32, 1469; c) H. J. Reich, R. R. Dykstra, 1 Am. Chem. SOC.1993, 115.
7041; d) H. J. Reich, K. J. Kulicke, ibid. 1995, l f 7 , 6621; e) H. J. Reich, K. J.
Kulicke, ibid. 1996, f18,273; f) ref. [73]; g) R. W. Hoffmann, M. Julius, F.
Chemla, T. Ruhland, G. Frenzen, Tetrahedron 1994. 50, 6049; h) H.
Ahlbrecht, J. Harbach, R. W. Hoffmann, T. Ruhland, Liebigs Ann. Chem.
1995,211; i) R. W Hoffmann, W. Klute, Chem. Eur. J 1996.2,694; k) R. W.
Hoffmann, R. K. Dress, T. Ruhland, A. Wenzel, Chem. Ber. 1995. 128, 861.
[233] P. Beak, P. D. Becker, J Org. Chem. 1982, 47, 3855.
12341 D. Hoppe, L. Beckmann, R Follmann, Angew,. Chem. 1980, 92.305; Angew
Chem. Int. Ed. Engl. 1980, 19, 303.
[235] B. Kaiser, D. Hoppe, Angew. Chem. 1995, 107, 344: Angew. Chem. Int. Ed.
Engl. 1995, 34, 323.
12361 The (R)-configuration of the predominant lithiated intermediate 225 was
derived from the carboxylic ester (S)-226b under the assumption of stereore-
tention during carboxylation (in analogy to the corresponding reactions of
oxygen analogues). The configurational assignment is thus not completely
secured.
[237] We do not know how to account for the marked deviation of the ee value
(34%) from that obtained with the undeuterated substrate (46% ee). Is it
likely that equilibrium between the diastereomers would be upset by a thermodynamic isotope effect on the order of magnitude of AAG = 0.4 kJmol-I?
[238] T. Shinozuka. Y Kikori, M. Asaoka, H. Takei, J. Chem. Soc. Perkin Trans I ,
1996, 119.
[239] R. W Hoffmann, W Klute, R. K. Dress. A. Wenzel. J Chem. Soc. Perkin
Trans. 2. 1995, 1721.
[240] G. Boche, A. Opel, M. Marsch, K. Harms, F. Hailer, J. C. W. Lohrenz, C.
Thiimmler, W. Koch, Chem. Ber. 1992. 125, 2265.
[241] T. Ruhland, R. Dress, R. W. Hoffmann, Angen. Chem. 1993. 105, 1487,
Angew. Chem. Int. Ed. Engi. 1993,32, 1467.
[242] B. Kaiser, dissertation, Universitat Miinster, 19%.
[243] D. Hoppe, B. Kaiser, 0. Stratmann, R. Frohlich, Angen. Chem. 1997, 109,
No. 24; Angew. Chem. Intl. Ed. Engl. 1997, 36, No. 24.
Time is money...
Save both by starting a personal subscription
to Angewandte Chemie
At 9e per page you can't go wrong.
With your own private copy you can read the
"hot" articles before they cool, and waste
less time in front of the copier.
Order now!
An order form can be found on
the last page of this issue.
2316
Angew. Chem. Int. Ed. Engl. 1997,36,2282-2316
Документ
Категория
Без категории
Просмотров
2
Размер файла
4 865 Кб
Теги
synthesis, sparteine, pairs, enantioselectivity, carbanion, lithium
1/--страниц
Пожаловаться на содержимое документа