Expeditious syntheses of (±)-allo-sedamine and (±)-allo- lobeline via a combination of aza-Sakurai-Hosomi and hydroformylation reactions

Article abstract of DOI:10.1055/s-0028-1083543

The expeditious preparation of allo-sedamine and allo-lobeline via 1,3-diastereoselective aza-Sakurai-Hosomi reaction followed by hydroformylation is reported. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0028-1083543

LETTER
2859
Expeditious Syntheses of ( )-allo-Sedamine and ( )-allo-Lobeline via a
Combination of Aza-Sakurai–Hosomi and Hydroformylation Reactions
Expeditious
S
yn
h
theses of
(
o
)-allo
-
Sedam
m
ine and( )-allo-L
o
a
beline s Spangenberg, Etienne Airiau, Mathieu Bui The Thuong, Morgan Donnard, Manuella Billet, André Mann*
Département de Pharmacochimie de la Communication Cellulaire, UMR 7175-LC-1 ULP/CNRS, Faculté de Pharmacie,
74 route du Rhin, BP 60024, 67401 Illkirch, France
E-mail: andre.mann@pharma.u-strasbg.fr
Received 23 July 2008
In memory of our late colleague Philippe Klotz (deceased 9 June 2005)
lectivity.5 For the construction of lobeline and sedamine
analogues, the hydroformylative end game required the
homoallylamines derived from the corresponding 1,3-
amino alcohols. Therefore at the onset of our work we had
to explore the diastereoselectivity of the aSH-R3C of 1,3-
O-protected aldehydes. For the assignment of the stereo-
chemistry of the diastereomers resulting from this ami-
noallylation step, the subsequent syntheses of lobeline or
sedamine analogues were helpful.
Abstract: The expeditious preparation of allo-sedamine and allo-
lobeline via 1,3-diastereoselective aza-Sakurai–Hosomi reaction
followed by hydroformylation is reported.
Key words: alkaloids, multicomponent reaction, diastereoselectiv-
ity, hydroformylation
A large body of work has been devoted to the preparation
of homoallylamines,1 the main substrates for the construc-
tion of piperidines which are privileged scaffolds for bio-
molecules. Indeed, homoallylamines have a great
chemical potential due to the presence of a terminal dou-
ble bond enabling a number of chemical manipulations,
such as metathesis, oxidative hydroboration or hydro-
formylation. Several syntheses of naturally occurring pi-
peridines are performed with functionalized hydroxy
homoallylamines including sedamine (1a) or allo-
sedamine (1b). Most of them are based on the following
classical strategy: a multistep preparation of the 1,3-hy-
droxy homoallylamine and a heterocyclization either by a
ring closure metathesis or an internal nucleophilic dis-
placement.2 Herein we report the synthesis of ( )-allo-
sedamine (1b) and of the unknown ( )-allo-lobeline (2)
based on a new strategy (Scheme 1): (i) an improved
method for the preparation of 1,3-O-protected homoallyl-
amines via an aza-Sakurai–Hosomi three-component re-
action (aSH-R3C), (ii) a linear hydroformylation of a
terminal alkene, yielding an aldehyde that collapses to an
internal enamine that is easily convertible into sedamine
analogues or which, in a one-pot sequence, can be con-
verted into an enone undergoing an intramolecular aza-
Michael reaction yielding lobeline analogues.
HO
OH
O
HO
Ph
N
Ph
N
Ph
Ph
N
H
H
H
H
Me
Me
Me
(+)-sedamine (1a)
(+)-allo-sedamine (1b)
(+)-allo-lobeline (2)
cyclization
sedamine
family
R
N
PG
H2/CO
RhI
TMS
NH2
CHO
Lewis
acid
R
NH
PG
R
NH
PG
R
O
PG
R = PhCH(OR3)CH2
O
Ph
Wittig
reaction
Michael
lobeline
family
R
NH
PG
addition
Scheme 1 General strategy
In our synthetic plan, the starting homoallylic alcohol 3
was prepared by allylation of benzaldehyde in 96% yield
(Scheme 2). Then the secondary alcohol was protected
with TBSCl in quantitative yield and subjected to ozono-
lysis leading to the corresponding racemic 1,3-O-protect-
In recent years multicomponent reactions have been re-
fined into powerful tools in organic synthesis,3 exempli-
fied by the preparation of homoallylamines from any kind
of aldehyde via the aSH-R3C.4 In this one-pot reaction the
acyliminium produced in situ from an aldehyde and a car-
bamate reacted with an allylsilane in the presence of a
Lewis acid, to afford homoallylamines.4 In the past, we
have reported that the aSH-R3C performed on 1,2-O-pro-
tected hydroxy aldehydes provided syn 1,2-diastereose-
OTBS
OH
i) TBSCl, imidazole
DMF, r.t.
O
ii) O3, Ph3P
CH2Cl2–MeOH, –78 °C
95% (over 2 steps)
4
3
SYNLETT 2008, No. 18, pp 2859–2863
1
4
.1
1
.2
0
0
8
Advanced online publication: 15.10.2008
DOI: 10.1055/s-0028-1083543; Art ID: G26108ST
Scheme 2 Preparation of aldehyde 4
© Georg Thieme Verlag Stuttgart · New York

2860
T. Spangenberg et al.
LETTER
ed aldehyde 4, the starting material for performing the action occurred with or without carbamate (entries 6 and
aSH-R3C.
11) yielding the corresponding homoallylalcohols with
the expected anti diastereoselectivity.7 The same scenario
was observed when the carbamate was omitted during the
reaction with allylsilane (entry 10); the Sakurai–Hosomi
adducts were obtained with a similar diastereomeric ratio.
The use of a more hindered allylsilane had no impact ei-
ther on the yield or on the diastereomeric ratio (entry 5).
When TiCl4 was used as Lewis acid compound 4 was con-
verted into cinnamaldehyde (data not shown) via a dehy-
dration pathway, an observation that excluded further
investigations with bidendate Lewis acids. The nature of
the nitrogen protecting group had little impact on the dia-
stereomeric ratio (entries 7–9). The Boc group showed
good diastereoselectivity (entry 7; 5b/6b = 16:84) al-
though with an eroded yield due to its sensitivity to acidic
conditions, even on lowering the amount of the Lewis acid
to 0.5 equivalent. Urethane gave good diastereoselectivi-
ties (entry 8; 5c/6c: 19:81) and excellent yields whereas
the tosyl group showed a modest yield and moderate dia-
Then, based on our previous work on the aSH-R3C, dif-
ferent solvents, carbamates and allylating agents were
screened. As Lewis acid, BF3·OEt2 was retained, only its
stoichiometry was modified. Orientating experiments
have revealed that the TBS group on the benzylic alcohol
function was the best choice as the protecting group.6 The
results of the chemical screening are depicted on Table 1.
According to the reaction conditions different O-silylated
(5a–d/6a–d) or O-desilylated homoallylamines (5e/6e)
were obtained. In dichloromethane, it appeared that the
aSH-R3C was operative with allylsilane, BF3·OEt2 and
several carbamates including TsNH2, and indeed homo-
allylamines 5a–d/6a–d were obtained in acceptable yields
as mixtures of diastereomers (Table 1, entries 1, 3–5, and
7–9). However conducting the same experiment in
MeCN, O-desilylated adducts 5e and 6e were obtained in
equal amount (entry 2). As the reaction produces a
stoichiometric amount of H2O, it is conceivable that in
acetonitrile, a basic solvent, fluorine is liberated from the
Lewis acid, being at the origin of a fast desilylation of the
benzylic hydroxyl. As a result, no facial discrimination
was obtained. Finally with allyltin a Sakurai–Hosomi re-
1
stereoselectivity (entry 9; 5d/6d = 30:70). From the H
NMR spectra of the crude mixtures, it was determined that
the diastereomeric ratio always favored the same diaste-
reomer (6a–d) with unknown relative stereochemistry.
Therefore the major diastereomer 6a8 (entry 1) was en-
Table 1 Study of the 1,3-Diastereoselective Aza-Sakurai–Hosomi Reaction
a: R2 = Cbz, R3 = TBS
b: R2 = Boc, R3 = TBS
c: R2 = CO2Et, R3 = TBS
d: R2 = Tos, R3 = TBS
e: R2 = Cbz, R3 = H
H2NR2
BF3⋅OEt2
R3
R3
R2
R2
TBS
Ph
O
HN
O
HN
O
O
R1
Ph
Ph
5a–e
6a–e
4
CH2Cl2, 0 °C, 1 h
syn
anti
R1
R2
BF3·OEt2
1 equivd
1 equivd,e
Products
5e/6e
syn/antib
18:82
50:50
17:83
15:85
16:84
ND
Yieldc
89%
93%
84%
89%
75%
90%
68%
95%
51%
76%
94%
a
Entry
1
2
TMS
TMS
TMS
TMS
Cbz
Cbz
Cbz
Cbz
5e/6e
3
1 equiv, 1 Mf
0.5 equiv, 1 Mf
1 equiv, 1 Mf
1 equiv, 1 Mf
0.5 equiv, 1 Me
1 equiv, 1 Mf
1 equiv, 1 Mf
1 equiv, 1 Mf
1 equiv, 1 Mf
5a/6a
5a/6a
5a/6a
4
5
SiMe2Ph
SnBu3
TMS
Cbz
g
6
Cbz
–
7
Boc
5b/6b
5c/6c
5d/6d
16:84
19:81
30:70
30:70
25:75
8
TMS
CO2Et
Tos
9
TMS
h
g
10
11
TMS
–
–
h
g
SnBu3
–
–
a Freshly distilled BF3·OEt2 was used.
b Determined by GC–MS, except for entries 10 and 11 that were determined by 1H NMR.
c Isolated after column chromatography.
d Added neat.
e Reaction was performed in MeCN.
f Diluted in CH2Cl2.
g Sakurai products.
h Reaction was performed without carbamate.
Synlett 2008, No. 18, 2859–2863 © Thieme Stuttgart · New York

LETTER
Expeditious Syntheses of ( )-allo-Sedamine and ( )-allo-Lobeline
2861
gaged in subsequent transformations towards the prepara- anti stereochemistry could be assigned to the major ad-
tion of sedamine or allo-sedamine depending on if 6a had duct 6a or its analogues 6b–d. Consequently the aSH-R3C
the syn or anti configuration. We expected that the corre- performed on 1,3-O-protected aldehyde followed an anti
lation with the reported physical data of sedamine (1a) bias. To account for the observed anti-preferred diastereo-
would not only solve the structural ambiguity but also selectivity, a nonchelated transition state is suggested.
provide some rationale about the observed 1,3-diastereo- Therefore, the model proposed by Evans and Reetz14 for
selectivity in the aSH-R3C.
the nucleophilic attack on aldehydes possessing a b
stereocenter (Figure 1) could be transposed to the corre-
sponding iminiums (A), intermediates in the aSH-R3C, to
explain the observed anti selectivity. From this prelimi-
nary study we propose an expeditious preparation of 1,3-
O-protected homoallylamine from the corresponding al-
dehyde using a simple R3C reaction and the synthesis of
( )-allo-sedamine (1b) in seven steps (from benzalde-
hyde) with an overall yield of 44%. Probably this is one of
the shortest syntheses reported to date.15
As mentioned before homoallyamine 6a was submitted to
a hydroformylation on the terminal alkene in THF as sol-
vent. This reaction is an attractive synthetic transforma-
tion as a new carbon–carbon bond is formed by addition
of H2 and CO to an alkene catalyzed by Rh(I). Further-
more an aldehyde function is introduced under generally
mild conditions allowing the direct use of this highly im-
portant group into a wide variety of reactions, making this
process a prototype of an atom economic transformation.9
In our case, the linear aldehyde was desired for the subse-
quent construction of a piperidine ring; we, thus, selected
the bulky diphosphite biphephos as ligand for hydro-
formylation (Scheme 3).10 As a nucleophilic nitrogen is
embedded in substrate 6a, the hydroformylation step is
followed by the desired cyclohydrocarbonylation to give
the six-membered heterocycle 711 in 93% isolated yield,
formed via a transient aminal and the assistance of catalyt-
ic amount of PPTS.12 The conditions for the conversion of
6a into 7 were very mild using low loading of Rh(I) and
reasonable syngas pressure, demonstrating that the cyclo-
hydrocarbonylation is a viable alternative to metathesis
for the transformation of homoallylamines to piper-
idines.13 Regarding the presence of functional groups on
the substrates, the catalytic mixture Rh(I)–biphephos used
for the hydroformylation reaction is less demanding than
the Ru(II)-based catalyst, used for metathesis. Then the fi-
nal steps were carried out in one pot under hydrogen re-
ductive conditions in the presence of Pearlman’s catalyst:
(i) the enamide of the tetrahydropyridine was reduced, (ii)
the Cbz group was deprotected, (iii) the presence of form-
aldehyde furnished by reductive amination yielded the
sedamine analogue 8 in 78% yield. Finally cleavage of the
TBS protecting group with TBAF gave the hydroxypiper-
idine in 88% yield after purification by column chroma-
tography.
Ph
H
TBSO
TBS
Ph
O
NHCbz
Cbz
H
+
N
H
H
H
TMS
1,3-anti product
A
Figure 1 Rationale for the 1,3-diastereoselectivity
Now that we had secured the relative stereochemistry of
allylamines 6a–d, we envisioned the preparation of ( )-
allo-lobeline (2) starting from 6b, a compound that has
never been prepared before. Therefore, we targeted inter-
mediate 12, an epimer of a late stage intermediate reported
by Lebreton, towards the synthesis of (–)-lobeline.
As the hydroformylation proved to be useful in the prepa-
ration of allo-sedamine, we decided to exploit the same
sequence towards the synthesis of allo-lobeline with some
new sophistication. The synthesis started with the Boc-
protected allylamine 6b (Scheme 4). To prevent the cy-
clohydrocarbonylation, that would inevitably follow the
hydroformylation of the terminal double bond, and to
meet the structural requirements, substrate 6b was reacted
with methyl iodide in the presence of LiHMDS to produce
the N-methylated compound 9. For the access to enone 12,
we first opted for a domino hydroformylation–Wittig ole-
fination, a sequence initially described by Breit et al.16 But
under the hydroformylative conditions the conjugated
double bound was reduced, precluding the possible aza-
Michael reaction. Therefore the reaction sequence was
modified in the following way: after hydroformylation of
9 the pressure was released, ylide 11 was added in the
same pot yielding enone 12 in 77% isolated yield, as a
mixture of E- and Z-isomers. The final steps were adapted
from the reported synthesis of lobeline. Heating 12 under
hydrolytic condition in isopropanol allowed N-Boc depro-
tection followed by the aza-Michael reaction and desilyla-
tion affording a cis-2,6-piperidine identified as ( )-allo-
lobeline (2). As the intramolecular Michael addition in 12
is reversible, the stereochemistry of the double bond in the
enone 12 has no influence on the 1,4-addition because the
thermodynamic more stable 2,6-cis-piperidine is the fa-
1
The H and 13C NMR data of this compound clearly
matched those of reported allo-sedamine (1b). This chem-
ical correlation (6a with allo-sedamine) revealed that the
Rh(CO)2acac (1 mol%)
biphephos (2 mol%)
H2, Pd(OH)2/C
MeOH, r.t.
OTBS
6a
N
H2/CO (1:1), 5 bar
PPTS (5 mol%)
THF, 65 °C, 18 h, 93%
then aq HCHO
78%
H
Cbz
one-pot
7
OH
OTBS
TBAF
THF, r.t., 88%
N
H
N
H
Me
Me
( )-1b
8
Scheme 3 Preparation of ( )-1b
Synlett 2008, No. 18, 2859–2863 © Thieme Stuttgart · New York

2862
T. Spangenberg et al.
LETTER
(5) (a) Ella-Menye, J.-R.; Dobbs, W.; Billet, M.; Klotz, P.;
Rh(CO)2acac (1 mol%)
biphephos (2 mol%)
MeI
LiHMDS
OTBS
Mann, A. Tetrahedron Lett. 2005, 46, 1897. (b) Billet, M.;
Schoenfelder, A.; Klotz, P.; Mann, A. Tetrahedron Lett.
2002, 43, 1453. (c) Billet, M.; Klotz, P.; Mann, A.
Tetrahedron Lett. 2001, 42, 631.
6b
NBoc
Me
H2/CO (1:1), 5 bar
THF, 65 °C, 18 h
THF–DMF
H
–10 °C, 88%
9
(6) Orientating experiments with different protecting groups on
the alcohol were performed with TBS (5e/6e = 85:15), Bn
(5e/6e = 75:25), Me (5e/6e = 70:30) using conditions from
Table 1, entry 1.
O
PPh3
11
CHO
OTBS
(7) (a) Keck, G. E.; Murry, J. A. J. Org. Chem. 1991, 56, 6606.
(b) Batey, R. A.; Thadani, A. N.; Smil, D. V.; Lough, A. J.
Synthesis 2000, 990.
NBoc
16 h, 65 °C, 77%
H
Me
10
(8) Typical Procedure for an Aza-Sakurai–Hosomi
Reaction: In a dry flask under argon were introduced 4
(0.756 mmol) and R2NH2 (0.756 mmol) in anhydrous
CH2Cl2 (4 mL, 0.2 M). The mixture was cooled to 0 °C by
means of an ice bath. Allylsilane (0.765 mmol) was then
added followed by a dropwise addition of a 1 M solution of
BF3·OEt2 in CH2Cl2 (0.756 mmol). The reaction was stirred
for 1 h at 0 °C before addition of sat. NaHCO3. The organic
layer was extracted with CH2Cl2 (3 ×) and dried over
Na2SO4. The solvent was removed under reduced pressure
and the residue was purified by silica gel chromatography.
Spectroscopic data for 6a: IR(film): 2952, 2928, 2856, 1696,
1508, 1250 cm–1. 1H NMR (200 MHz, CDCl3): d = 7.23–
7.39 (m, 10 H), 5.70–5.80 (m, 1 H), 5.05–5.21 (m, 5 H), 4.82
(dd, J = 2.0, 9.1 Hz, 1 H), 3.80–3.91 (m, 1 H), 2.26–2.40 (m,
2 H), 1.93 (ddd, J = 2.9, 10.0, 13.8 Hz, 1 H), 1.68 (br d, J =
O
OTBS
Ph
HCl, i-PrOH, 60 °C
72%
( )-2
NBoc
H
Me
12
Scheme 4 Preparation of ( )-2
vored diastereomer.17 Finally ( )-allo-lobeline was pre-
pared for the first time in seven steps (from benzaldehyde)
with an overall yield of 25%.
Thus we have reported that implementation of the aza-
Sakurai–Hosomi three-component reaction (aSH-R3C)
on 1,3-O-protected aldehydes delivered homoallylamines
in good yields and with moderate anti 1,3-diastereoselec-
tivities in an acyclic stereocontrol. The anti-allylamines
6a and 6b were used for the expeditious syntheses of two
piperidine alkaloids ( )-allo-sedamine (1a; 7 steps) and
( )-allo-lobeline (2; 7 steps), respectively, by using a cy-
clohydrocarbonylation or a one-pot hydroformylation–
Wittig reaction. We are currently focusing our efforts on
the rapid access to other biomolecules using hydroformy-
lation.
9.1 Hz, 1 H), 0.88 (s, 9 H), 0.05 (s, 3 H), –0.26 (s, 3 H). 13
C
NMR (75 MHz, CDCl3): d = 155.9 (C), 145.0 (C), 136.9 (C),
134.5 (CH), 128.5 (CH), 128.3 (CH), 128.1 (CH), 128.0
(CH), 127.3 (CH), 125.9 (CH), 117.8 (CH2), 72.7 (CH), 66.5
(CH2), 48.4 (CH), 44.4 (CH2), 39.7 (CH2), 25.8 (Me), 18.1
(C), –4.5 (Me), –5.1 (Me). HRMS (ESI, positive, HCOOLi):
m/z [M + Li] calcd for C23H39NO3Si: 446.2698; found:
446.2690.
(9) For a review on hydroformylation, see: Breit, B.; Seiche, W.
Synthesis 2001, 1.
(10) (a) Cuny, G. D.; Buchwald, S. L. J. Am. Chem. Soc. 1993,
115, 2066. (b) The structure of Biphephos is shown in
Figure 2.
Acknowledgment
MeO
OMe
This work was supported by the Ministère délégué à l’Enseigne-
ment Supérieur et à la Recherche (T.S., E.A., M.D.). The authors
thank Prof. Eric Marcioni et al. (ULP) for GC–MS analysis, and
Patrick Wehrung and Pascale Buisine (IFR 85) for HRMS spectro-
metry.
t-Bu
O
O
t-Bu
P
P
O
O
O
O
References and Notes
(1) (a) Bloch, R. Chem. Rev. 1998, 98, 1407. (b) Kobayashi, S.;
Hirano, K.; Sugiura, M. Chem. Commun. 2005, 14.
(c) Sugiura, M.; Hirano, K.; Kobayashi, S. J. Am. Chem. Soc.
2004, 126, 7182. (d) Li, S.-W.; Batey, R. A. Chem.
Commun. 2004, 1382. (e) Vilaivan, T.; Winotapan, C.;
Bauphavichit, V.; Shimada, T.; Ohfune, Y. J. Org. Chem.
2005, 70, 3464.
(2) For a review of the history, chemistry, and biology of
Lobelia alkaloids, see: (a) Felpin, F.-X.; Lebreton, J.
Tetrahedron 2004, 60, 10127. (b) Felpin, F.-X.; Lebreton, J.
J. Org. Chem. 2002, 67, 9192. (c) Cossy, J.; Willis, C.;
Bellosta, V.; BouzBouz, S. J. Org. Chem. 2002, 67, 1982.
(3) Mihovilovic, M. D.; Stanetty, P. Angew. Chem. Int. Ed.
2007, 46, 8612; and references therein.
Figure 2
(11) Typical Procedure for Hydroformylation: In a dry
Schlenk glassware under argon were introduced
Rh(CO)2acac (1 mol%) and anhydrous degassed THF (1
mL). Biphephos (2 mol%) was added and CO evolution was
observed. Subsequent addition of homoallylic amide (and
PPTS) was performed. The mixture was transferred via a
syringe in a dry stainless autoclave under argon. The
glassware was rinsed with anhydrous degassed THF (3 ×) to
reach a final concentration of 0.04 M. The autoclave was
purged (3 ×) with H2/CO (1:1) before setting the pressure at
5 bar. The autoclave was heated at 65 °C (internal
(4) Veenstra, S. J.; Schmid, P. Tetrahedron Lett. 1997, 38, 997.
temperature) by means of an oil bath. Once the reaction was
Synlett 2008, No. 18, 2859–2863 © Thieme Stuttgart · New York

LETTER
Expeditious Syntheses of ( )-allo-Sedamine and ( )-allo-Lobeline
2863
finished, the autoclave was depressurized and the solvent
CH), 42.7 (CH2), 26.0 (Me), 25.5, 24.7 (rotamers, CH2), 18.2
(C), 17.8, 17.4 (rotamers, CH2), –4.5, –4.8 (rotamers, Me).
HRMS (ESI, positive, HCOOLi): m/z [M + Li] calcd for
C26H37NO3Si: 458.2698; found: 458.2685.
was removed under reduced pressure. The residue was
purified by silica gel chromatography. Spectroscopic data
for 7: IR(film): 2951, 2927, 2854, 1703, 1651, 1324, 1089,
1060, 833 cm–1. 1H NMR (200 MHz, CDCl3): d = 7.15–7.36
(m, 10 H), 6.79 (br d, J = 8.2 Hz, 0.5 H, rotamers), 6.66 (br
d, J = 8.2 Hz, 0.5 H, rotamers), 5.16 (s, 2 H), 4.94 (m, 0.5 H,
rotamers), 4.81 (m, 1 H), 4.65 (m, 0.5 H, rotamers), 4.45 (m,
0.5 H, rotamers), 4.14 (m, 0.5 H, rotamers), 1.72–2.13 (m,
6 H), 0.87, 0.85 (s, 9 H, rotamers), 0.1 (s, 1.5 H, rotamers),
–0.03 (s, 1.5 H, rotamers), –0.24, –0.26 (s, 3 H, rotamers).
13C NMR (50 MHz, CDCl3): d = 153.2, 152.9 (rotamers, C),
145.3, 144.6 (rotamers, C), 136.5, 136.4 (rotamers, C), 128.6
(CH), 128.5 (CH), 128.2 (CH), 128.0 (CH), 127.3 (CH),
126.2, 126.1 (rotamers, CH), 124.3, 123.7 (rotamers, CH),
106.2 (CH), 73.8 (CH), 67.4 (CH2), 49.1, 48.8 (rotamers,
(12) Ojima, I.; Tzamarioudaki, M.; Eguchi, M. J. Org. Chem.
1995, 60, 7078.
(13) Borah, J. C.; Boruwa, J.; Barua, N. C. Lett. Org. Chem. 2006,
3, 140.
(14) (a) Evans, D. A.; Dart, J. M.; Duffy, J. L.; Yang, M. G. J. Am.
Chem. Soc. 1996, 118, 4322. (b) Reetz, M. Tetrahedron
Lett. 1984, 25, 729.
(15) Birman, V. B.; Jiang, H.; Li, X. Org. Lett. 2007, 9, 3237.
(16) Breit, B.; Zahn, S. K. Tetrahedron 2005, 61, 6171.
(17) Compère, D.; Marazano, C.; Das, B. H. J. Org. Chem. 1999,
64, 4528.
Synlett 2008, No. 18, 2859–2863 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.