close

Вход

Забыли?

вход по аккаунту

?

Catalytic Enantioselective 1 3-Dipolar Cycloaddition Reaction of Azomethine Ylides and Alkenes The Direct Strategy To Prepare Enantioenriched Highly Substituted Proline Derivatives.

код для вставкиСкачать
Highlights
DOI: 10.1002/anie.200501074
Asymmetric Catalysis
Catalytic Enantioselective 1,3-Dipolar Cycloaddition
Reaction of Azomethine Ylides and Alkenes: The Direct
Strategy To Prepare Enantioenriched Highly Substituted
Proline Derivatives**
Carmen Njera* and Jos M. Sansano*
Keywords:
asymmetric catalysis · azomethine ylides ·
chiral ligands · cycloaddition · Lewis acids
The interest in asymmetric synthesis is
continuously increasing owing to the
continuous demand for enantiomerically enriched molecules. The main
challenge in this area is to create the
maximum number of stereogenic centers in one reaction step by employing
the minimum number of reagents. From
both an economic and a synthetic point
of view, asymmetric catalysis[1] can constitute a powerful tool, especially for
pericyclic reactions, where the relative
and absolute configuration of several
carbon atoms can be established almost
simultaneously.
One example is the 1,3-dipolar cycloaddition of azomethine ylides (from
imines A) and alkenes B,[2] which allows
the stereoselective synthesis of pyrrolidines or proline derivatives C
(Scheme 1). Among such molecules of
type C, which can be obtained through
longer and more sophisticated routes,
are very important pharmaceuticals and
natural alkaloids, organocatalysts, and
building blocks in organic synthesis.[3] In
this cycloaddition, the particular struc[*] Prof. Dr. C. N6jera, Dr. J. M. Sansano
Departamento de Qu;mica Org6nica
Universidad de Alicante
Apartado 99, 03080 Alicante (Spain)
Fax: (+ 34) 965-903-549
E-mail: cnajera@ua.es
jmsansano@ua.es
[**] We thank the Spanish Ministerio de Ciencia y Tecnolog;a (BQU2001-0724-C02 and
CTQ 2004-00808/BQU), the Generalitat
Valenciana (CTIOIB/2002/320 and GRUPOS 03/134), and the University of Alicante for financial support.
6272
Scheme 1. EWG = electron-withdrawing group.
dipole A; b) by attaching a chiral auxiliary to the EWG2 of the alkene B; c) by
employing a chiral Lewis acid capable of
chelating both components A and B.
According to the last strategy, and considering all of the previous advantages
of these metalloazomethine ylides,
Grigg and co-workers employed a stoichiometric amount of the catalytic system ephedrine derivative 1/cobalt(ii)
chloride in the reaction of the dipole
precursor 3 with the dipolarophile 4
(Y = CO2Me, used as solvent) giving an
excellent ee in the case of the enantiomer 5 a with total endo-selectivity
(Scheme 2, Table 1).[4, 5] However, the
stoichiometric catalyst formed by bisphosphine 2/AgOTf (Tf = trifluoro-
ture of the azomethine ylide promotes a
characteristic frontier molecular orbital
(FMO) controlled reactivity with alkenes, which can lead to the formation
of four stereocenters with high stereoselectivity. There are several procedures
for generating azomethine ylides, of
which the metalation of imino esters A
is the most widely used in organic
chemistry.[2a,c] The benefits of the in situ
preparation of metalloazomethine
dipoles
from a-imino esters A
in basic media, such as
working at room temperature with a highly
coordinated transition
state and starting from
easily available compounds, make this reaction much more attractive (Scheme 1).
The asymmetric version of this 1,3-dipolar
cycloaddition reaction
can be accomplished following different strategies: a) by attaching a
chiral auxiliary to the
imino (R1) or to the
electron-withdrawing
Scheme 2. R1 = 2-naphthyl, R2 : see Table 1; 4 as solvent when
1
group (EWG ) of the Y = CO2Me, Bn = benzyl.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6272 – 6276
Angewandte
Chemie
Table 1: Asymmetric catalytic 1,3-dipolar cycloaddition using stoichiometric catalyst systems
(see Scheme 2).
R2
Y
Lig.[a] Metal
salt[a]
H CO2Me 1
Me COMe 2
Me SO2Ph 2
CoCl2
AgOTf
AgOTf
Yield
[%]
5, ee
[%]
84
83
84
5 a, 96
5 a’, 70
5 a’, 70
[a] 100 mol %.
methanesulfonyl) afforded the other
enantiomer, endo-5 a’, albeit with lower
enantioselection.[5] The high enantiomeric excesses obtained in these examples can be justified, presumably, by the
presence of the compact transition
states 6 and 7, respectively.
Despite these attractive precedents,
the first asymmetric catalytic 1,3-dipolar
cycloaddition reaction, in which substoichiometric amounts of a chiral metallic
complex were used, was not described
until 2002.[6] Zhang and co-workers
screened several chiral bisphosphine
ligands such as, binap ((R-) or (S)-2,2’bis(diphenylphosphanyl)-1,1’-binaphthyl), Me-DuPhos ([( )-1,2-bis-(2R,5R)2,5-dimethylphospholano]benzene),
PennPhos (P,P’-1,2-phenylenebis(endo2,5-dialkyl-7-phosphobicyclo[2.2.1]heptane) and bicp ((R,R)- or (S,S)-2,2’-bis(diphenylphosphanyl)dicyclopentane),
with silver acetate, and obtained very
poor enantioselectivities and even diastereoselectivities, in the cyclization reaction of imino esters 3 and dimethyl
maleate.[6] However, Trost>s ligand 8[7]
provided a promising higher enantioselectivity; the weak interaction of the
nitrogen atom of the bisamide seems to
be very important in the assembly of the
chiral ligand with the silver cation
(Scheme 3).[6]
A similar ligand 9, which incorporates an additional element of planar
chirality (two ferrocene units), was designed to impart different steric and
stereoelectronic properties on the product (Scheme 3). The best results for the
endo-10 adducts were obtained with
these ligands when the aryl group was
3,5-dimethylphenyl and dimethyl maleate was used (Table 2).[6] Surprisingly,
the adduct 10 was also formed in good
yield and noticeable enantiomeric purity when the substituent R1 of the
dipolarophile tert-butyl acrylate (Table 2).[6]
However, it has been demonstrated
that chiral bisphosphine ligands form a
more suitable complex with copper triflate than with silver acetate. Komatsu
and co-workers observed reverse exoselectivity using chiral bisphosphine/
copper(ii) triflate at 40 8C.[9] In most
of the cases, the exo/endo ratio exceeded
95:5 using N-phenylmaleimide (NPM),
the higher enantioselection being achieved with the catalytic systems formed
by (R)-binap 11 or (R)-SegPhos 12/
copper(ii) triflate (Scheme 4). With other dipolarophiles, such as dimethyl fumarate or fumaroniTable 2: Asymmetric catalytic 1,3-dipolar cycloaddition ustrile, the endo-adduct (e.g. 14)
ing substoichiometric amounts of a chiral metal complex
was formed in higher propor(see Scheme 3).
tion (Table 3).
1
R
Dipolarophile
Yield [%]
ee [%]
The authors proposed a
Ph
methyl maleate
87
87
plausible mechanism for the
4-(NC)C6H4
methyl maleate
90
86
cycloaddition reaction and for
methyl maleate
98
97
2-Naph[a]
the explanation of the obiPr
methyl maleate
82
70
tained
diastereoselection
[b]
methyl maleate
82
81
Cy
(Scheme 2) by which the azoPh
methyl acrylate
90
60
methine ylide 15 would be
Ph
tert-butyl acrylate
85
93
87
79
Ph
NMM[c]
generated under basic media,
[a] Naph = naphthyl, [b] Cy = cyclohexyl, [c] NMM = N- and in turn this would react
with NPM to give the exomethyl maleimide.
precursor 3 was a cyclohexyl or
isopropyl group. In previous
work, it has been pointed out
that the Michael-type addition
reaction of the dipole to the
electrophilic alkene is the expected pathway and, normally,
no further cyclization occurred
when R1 is an alkyl substituent.[8] Very good enantioselection was also exhibited by the
Scheme 4. R1: see Table 3, R2 = H; SegPhos = (4,4’-bi-1,3-benzodioxole)-5,5’-diylbis(diphenylphosphane).
Scheme 3. R1: see Table 2, R2 = H, R3, R4 : see dipolarophiles listed in Table 2, Ar = 3,5-dimethylphenyl.
Angew. Chem. Int. Ed. 2005, 44, 6272 – 6276
transition state 16, because here the
steric interactions between NPM and
the bisphosphine ligand are less than in
the endo-transition state 16 (Scheme 5).
This hypothesis was also supported by
ZINDO calculations.[9]
Focusing on the silver(i)-catalyzed
enantioselective cycloaddition developed by Zhang and co-workers, Schreiber and co-workers studied the catalytic
1,3-dipolar cycloaddition reaction of
azomethine ylides with a series of chiral
monophosphines bearing a donor
group; the P,N-ligand quinap 17 is an
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6273
Highlights
Table 3: Asymmetric catalytic 1,3-dipolar cycloaddition with catalyst systems comprising the chiral
bisphosphines 11 and 12 and Cu(OTf)2 (see Scheme 4).
R1
Ligand
Dipolarophile
Yield [%]
4-(MeO)C6H4
Ph
11
12
NPM
NPM
Ph
11
dimethyl fumarate
exo/endo
eeexo [%]
83
78
> 95:5
89:11
87
72
80
36:64
77
Cycloadduct
Scheme 5.
excellent chelating molecule for the
silver cation (see chelate 7 in
Scheme 2).[10] The catalyst loading
(3 mol %) was very similar to that reported in previous examples (see
above). When substrates 3 were allowed
to react with tert-butyl acrylate, tertbutyl crotonate, and tert-butyl cinnamate (Scheme 6, Table 4), the enantioselectivities achieved were very high.
Nevertheless, the reaction between the
imino ester 3 and dimethyl maleate
furnished compound 19 in 60 % ee. In
all these cycloadditions the endo-selectivity was overwhelmingly predominant,
except when tert-butyl cinnamate was
employed as dipolarophile. In addition,
a-substituted esters 3 (incorporating
methyl, isobutyl, benzyl, and 3-indolylmethyl groups) were also evaluated in
this transformation using a catalyst
loading of 10 mol %; proline derivatives
19 with a quaternary carbon atom in the
a-position were obtained in high yields
and very good ee values.
The same enantioselection and
yields were achieved by Carreira and
co-workers with the ligand O-(S)-pinap
18 under the reaction conditions detailed in Table 4 (R1 = 4-(NC)C6H4),[11]
namely 3 mol % of the catalytic mixture
18/AgOAc was employed at 40 8C.
This result also confirmed the efficient
coordination of the silver cation to the
chiral moiety of (S)-quinap 17 and O(S)-pinap 18, as was shown in the model
7 (Scheme 2).
Jørgensen and co-workers demonstrated that chiral bisoxazolines 20–22
(Scheme 7) were suitable ligands for the
standard 1,3-dipolar cycloaddition reac-
Scheme 6. R1, R2, R3 : see Table 4; (S)-pinap = (S)-1-[2-(diphenylphosphanyl)-1-naphthyl]phthalazine, (S)-quinap = (S)-1-[2-(diphenylphosphanyl)-1-naphthyl]isoquinoline.
Table 4: Asymmetric catalytic 1,3-dipolar cycloaddition with catalyst systems comprising the P,Nligands 17 and 18 and AgI (see Scheme 6).
R1
R2
R3
Ligand
Ph
Ph
Ph
4-(NC)C6H4
4-(NC)C6H4
H
Me
iBu
H
H
Me
H
H
H
H
17
17
17
17
18
T [8C]
20
20
20
45
40
Dipolarophile
Yield [%]
ee [%]
tert-butyl crotonate
tert-butyl acrylate
tert-butyl acrylate
tert-butyl acrylate
tert-butyl acrylate
97
98[a]
97[a]
92
94
84
80
84
96
95
[a] By using 10 mol % of the catalyst system 17/AgI.
6274
www.angewandte.org
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Scheme 7. Ph-box = 2,2’-isopropylidenebis[(4S)-4-phenyl-2-oxazoline], Phdbfox = (R,R)-4,6-dibenzofurandiyl-2,2’-bis(4phenyloxazoline), tBu-box = 2,2’-isopropylidenebis[(4S)-4-tert-butyl-2-oxazoline].
Angew. Chem. Int. Ed. 2005, 44, 6272 – 6276
Angewandte
Chemie
tions of azomethine ylides and electrophilic alkenes when zinc(ii) triflate,
rather than copper(ii) triflate, was used
as Lewis acid.[12] The reaction of the
imino ester 3 in basic media with several
dipolarophiles occurred at room temperature with excellent diastereoselectivity (endo-products 23 were exclusively formed) in very good yields and very
high ee values, although an improvement of the enantioselectivity was observed when the reaction was carried
out at 20 8C (Scheme 8, Table 5). The
Scheme 8. R1, R3 : see Table 5, R2 = H.
catalyst loading employed in this reaction is higher than that employed in the
previously described cases. These reaction conditions were independent of the
quantity of base used and were very
sensitive to the bulkiness of the substituents anchored to the dipolarophile
(the ee value decreased when tert-butyl
acrylate was used as the dipolarophile).
Based on the absolute configuration of
the N-tosylated adducts 23 (determined
by X-ray structure analysis), the authors
proposed an intermediate 24 in which
the azomethine ylide coordinates with
the ZnII-tBu-box catalyst to form an 18electron complex with a tetrahedral
arrangement of the ligand around the
metal center (Scheme 8).[12]
In summary, if we evaluate all these
aspects we determine that the metalcatalyzed 1,3-dipolar cycloaddition re-
action of azomethine ylides is a very
difficult reaction because too many
parameters have to be controlled. For
example, the coordination between the
metal center and the chiral ligand has to
be stronger than the coordination of the
metal center and the 1,3-dipole; however, the interaction with the final
pyrrolidine should be weaker to allow
the regeneration of the chiral catalyst.
The chiral domain must differentiate the
two enantiotopic faces of the 1,3-dipole
on attack of the dipolarophile. The
substituents of the imine and
the ester groups, as well as the
structure of the electron-deficient olefin are crucial for the
course of the enantioselection.
Finally, the solvent and the temperature play important roles in
fine-tuning this asymmetric reaction and have to be considered carefully.
The five contributions described previously[6, 9–12] constitute very good pieces of chemistry, each of which has some
intrinsic drawbacks that need
to be corrected. Which ligand/
metal cation couple can ensure a very
good enantioselection? From the work
done so far, it is seen that N,P-ligands/
AgI, P,P-ligands/CuII , and N,N-ligands/
ZnII are, a priori, suitable combinations
for this enantioselective cycloaddition.
The perfect asymmetric reaction, for
which yields and enantiomeric enrichment are excellent, independent of the
structure of the 1,3-dipole and dipolarophile, has not been reported yet. A fast
reaction, at room temperature, in the
presence of a recoverable chiral ligand
without any significant loss of efficiency
or activity of the catalyst system, is very
desirable. Thus, this area remains fascinating and much work still has to be
done in the pursuit of the perfect
catalytic system.[13]
Published online: September 19, 2005
Table 5: Asymmetric catalytic 1,3-dipolar cycloaddition with a catalyst system comprising chiral
bisoxazoline 21 and Zn(OTf)2 (see Scheme 8).
R1
R3
T [8C]
Dipolarophile
Yield [%]
ee [%]
Ph
4-BrC6H4
2-Naph
Ph
2-Naph
H
H
H
CO2Me
CO2Me
0
methyl acrylate
methyl acrylate
methyl acrylate
dimethyl fumarate
dimethyl fumarate
95
89
84
78
84
78
94
91
76
90
20
20
20
0
Angew. Chem. Int. Ed. 2005, 44, 6272 – 6276
[1] Comprehensive Asymmetric Catalysis,
Vol. 1–3, Suppl. 1–2 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer,
Heidelberg, 2004.
[2] For recent reviews of 1,3-dipolar cycloaddition reactions of azomethine ylides,
see: a) C. NLjera, J. M. Sansano, Curr.
Org. Chem. 2003, 7, 1105 – 1150; b) Synthetic Applications of 1,3-Dipolar Cycloaddition Chemistry Towards Heterocycles and Natural Products (Eds.: A.
Padwa, W. H. Pearson), Wiley, New
York, 2003; c) S. Kanemasa, Synlett
2002, 1371 – 1387; d) K. V. Gothelf in
Cycloaddition Reactions in Organic Synthesis (Eds.: S. Kobayashi, K. A. Jørgensen). Wiley-VCH, Weinheim, 2002,
pp. 211 – 245.
[3] For reviews about synthesis an applications of pyrrolidine derivatives, see:
a) Y. Cheng, Z.-T. Huang, M.-X. Wang,
Curr. Org. Chem. 2004, 8, 325 – 351;
b) W. Notz, F. Tanaka, C. F. Barbas III,
Acc. Chem. Res. 2004, 37, 580 – 591;
c) F.-X. Felpin, J. Lebreton, Eur. J. Org.
Chem. 2003, 3693 – 3712; d) W. H. Pearson, P. Stoy, Synlett 2003, 903 – 921;
e) W. H. Pearson, Pure Appl. Chem.
2002, 74, 1339 – 1347; f) Pharmaceuticals, Vol. 1–4 (Ed.: J. L. McGuire), Wiley-VCH, Weinheim, 2000.
[4] P. Allway, R. Grigg, Tetrahedron Lett.
1991, 32, 5817 – 5820.
[5] R. Grigg, Tetrahedron: Asymmetry 1995,
6, 2475 – 2486.
[6] J. M. Longmire, B. Wang, X. Zhang, J.
Am. Chem. Soc. 2002, 124, 13 400 –
13 401.
[7] B. M. Trost, D. L. Van Vranken, C. Bingel, J. Am. Chem. Soc. 1992, 114, 9327 –
9329.
[8] J. Casas, R. Grigg, C. NLjera, J. M.
Sansano, Eur. J. Org. Chem. 2001,
1971 – 1982.
[9] a) Y. Oderaotoshi, W. Cheng, S. Fujitomi, Y. Kasano, S. Minakata, M. Komatsu, Org. Lett. 2003, 5, 5043 – 5046; b)
ZINDO: Modified INDO (intermediate
neglect of the differential overlap) program. It is a semiempirical molecular
orbital program for studying the spectroscopic properties of a wide range of
compounds, such as organic and inorganic molecules, polymers, and organometallic complexes.
[10] C. Chen, X. Li, S. L. Schreiber, J. Am.
Chem. Soc. 2003, 125, 10 174 – 10 175.
[11] T. F. KnMpfel, P. Aschwanden, T. Ichikawa, T. Watanabe, E. M. Carreira, Angew. Chem. 2004, 116, 6097 – 6099; Angew. Chem. Int. Ed. 2004, 43, 5971 –
5973.
[12] A. S. Gothelf, K. V. Gothelf, R. G. Hazell, K. A. Jørgensen, Angew. Chem.
2002, 114, 4410 – 4412; Angew. Chem.
Int. Ed. 2002, 41, 4236 – 4238.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6275
Highlights
[13] Note added in proof: After this Highlight had been written, Pfalz and coworkers reported the catalytic enantioand diastereoselective [3+2] cycloaddi-
6276
www.angewandte.org
tion of azomethine ylides 3 using a chiral
phosphanylazoline–AgI complex as the
catalyst. Inter- and intramoleculare cyclizations were achieved with up to
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
99 % ee; R. Stohler, F. Wahl, A. Pfaltz,
Synthesis 2005, 1431 – 1436.
Angew. Chem. Int. Ed. 2005, 44, 6272 – 6276
Документ
Категория
Без категории
Просмотров
2
Размер файла
823 Кб
Теги
prepare, reaction, cycloadditions, proline, direct, strategy, enantioselectivity, alkenes, dipolar, derivatives, enantioenriched, catalytic, substituted, ylide, highly, azomethine
1/--страниц
Пожаловаться на содержимое документа