Enantioselective allylation of alkyl glyoxylates catalyzed by (salen)chromium(III) complexes

Article abstract of DOI:10.1016/j.tetlet.2004.05.088

The enantioselective addition of allylstannanes and allylsilanes to alkyl glyoxylates of type 1, catalyzed by chiral (salen)Cr(III) complexes 3, has been studied. We have found that the reaction proceeded smoothly for low loading (1-2mol%) of (1R,2R)-(sal

Full text of DOI:10.1016/j.tetlet.2004.05.088

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 5343–5346
Enantioselective allylation of alkyl glyoxylates catalyzed by
(salen)chromium(III) complexes
Piotr Kwiatkowski,a Wojciech Chaładajb and Janusz Jurczaka,b,
*
aInstitute of Organic Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, 01-224 Warsaw, Poland
bDepartment of Chemistry, Warsaw University, Pasteura 1, 02-093 Warsaw, Poland
Received 28 April2004; revised 14 May 2004; accepted 18 May 2004
Abstract—The enantioselective addition of allylstannanes and allylsilanes to alkyl glyoxylates of type 1, catalyzed by chiral
(salen)Cr(III) complexes 3, has been studied. We have found that the reaction proceeded smoothly for low loading (1–2 mol %) of
(1R,2R)-(salen)Cr(III)BF4 3a or (1R,2R)-(salen)Cr(III)ClO4 3c, and allyltributyltin under simple, undemanding conditions,
affording (R)-2-hydroxypent-4-enoic acid esters 2 in good yield (61–90%) and enantioselectivity (58–76% ee).
ꢀ 2004 Elsevier Ltd. All rights reserved.
12
The addition of allylic reagents to aldehydes and
ketones leads to homoallylic alcohols, which are very
important from the synthetic point of view. Numerous
allylic organometallics as well as various Lewis acids
are effective in these reactions.1 Investigation of the
enantioselective variant of these reactions was origi-
nated in the early 1990s by Yamamoto and co-workers,2
who used a chiralacyol xy borane catayl st derived from
tartaric acid and allyltrimethylsilane. Until now, many
efficient methods for enantioselective allylation have
been developed, although these methods apply almost
exclusively to simple aromatic and aliphatic aldehydes.3
In general, chiral Lewis acids are used as catalysts.
Among the most extensively studied catalysts are
BINOL complexes with Ti(IV),4 as well as BINOL/
Zr(IV),5 BINAP/Ag(I)6 and chiral bisoxazoline ligands
with Zn(II)7 and Rh(III).8 Allyltin and allylsilicon
derivatives are most frequently used as allylating agents
for the reactions catalyzed by chiral Lewis acids. Allyl-
trichlorosilanes are also employed for enantioselective
allylation, and their reactions are catalyzed by chiral
Lewis bases, for example, phosphoramide,9 formamide10
and N-oxides.11 Another efficient enantioselective allyl-
ation is the catalytic Nozaki–Hiyama–Kishi reaction
with allylic halides promoted by complexes of chro-
mium(II) with chiralsaeln
and bisoxazoline13 ligands.
Most of these procedures require anhydrous reaction
conditions, sometimes even oxygen-free reaction condi-
tions (as in the case of the Nozaki–Hiyama–Kishi
reaction).
Our studies concern the allylations of a particular group
of active aldehydes, glyoxylates 1 (Scheme 1), which
result in 2-hydroxypent-4-enoic acid esters 2, com-
pounds of significant synthetic interest.14 Currently, in
order to synthesize these compounds and their deriva-
tives in optically pure form, diastereoselective methods
15;16
using chiralauxiilaries
are mainly applied. Methods
using sultam derivatives of glyoxylic acid have been
developed in our research group16 however no efficient
method for the enantioselective allylation of glyoxylates
is currently known. This subject was investigated by
Mikami and co-workers17 using a BINOL-titanium
complex (10 mol %), but the results obtained for the
simple allylation of glyoxylates were poor, both in terms
of enantiomeric excess and in terms of yield. In the case
O
O
H
RO
RO
OH
O
1a R=Bun
1b R=Pri
1c R=But
1d R= Bn
Keywords: Asymmetric catalysis; Allylstannane; Carbonyl-addition
reaction; Chromium; Glyoxylate; Homoallylic alcohols; Salen.
* Corresponding author. Tel.: +48-22-6320578; fax: +48-22-6326681;
e-mail: jurczak@icho.edu.pl
2a-d
Scheme 1. Allylation of glyoxylates.
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.05.088

5344
P. Kwiatkowski et al. / Tetrahedron Letters 45 (2004) 5343–5346
Table 1. Results of the reactions of n-butyl glyoxylate (1a) with vari-
ous allylating reagents in the presence of (salen)CrBF4 complex (3a)a
of reactions with allyltrimethylsilane and allyltributyl-
tin, the ee values were 30% and 10%, respectively, and
the yields did not exceed 40%. Of interest was the fact
that the same catalytic system used for the reaction of
simple aliphatic and aromatic aldehydes with allyltri-
butyltin gave much better results (the ee was often above
90%).4
O
SnR3
O
Bun O
H
BunO
(salen)CrBF
CH2Cl2
4 (3a)
OH
O
2a
1a
Entry Allyl reagent
Time (h)
Yield (%)
Eeb (%)
Bun3SnAllyl
Me3SnAllyl
Ph3SnAllyl
Sn(Allyl)4
3
82
79
70
82
61
62
46
13
We thus decided to search for catalytic systems useful
for allylation of glyoxylates 1. We have chosen the salen
complexes of transition metals as candidate chiral Lewis
acid catalysts, since they are easy to handle and stable in
the presence of moisture and oxygen. The modelreac-
tion was the allylation of n-butylgyloxyalte ( 1a) with
allyltributyltin.18 Following preliminary screening of
chiral salen complexes of type 3 (Fig. 1) of the following
metals: Ti(IV), Cr(III), Mn(III), Fe(III), Co(II), Co(III),
Ni(II), Ni(III), Cu(II) and Al(III), it turned out that the
only enantiomerically efficient catalyst was the (salen)-
chromium(III) complex 3a. Although the remaining
complexes did catalyze the allylation, the ee was 10% at
best. Activation using the salen-metal complexes was
too low to enable efficient reaction of allyltrimethyl-
silane.
1
2
3
4
3
4
0.5
a The reactions were carried out using 1 mmolof n-butylgyloxy-
late (1a), 2 mol% of 3a and 1.2 equiv of R3SnAllyl (0.3 mmol of
Sn(Allyl)4) in 1 mL of CH2Cl2 at 20ꢁC.
b The enantiomeric excess was determined by GC on a capillary chiral
column b-dex 120 (permethyl-b-cyclodextrin).
improves the yield slightly (<40%), while the enantio-
selectivity remains practically the same.
The use of tin allylating reactants gave much better
results (Table 1). Their reactions proceeded at room
temperature in good yields. The best results were
observed for allyltrialkyltin derivatives, and the enan-
tiomeric excesses exceeded 60% (Table 1, entries 1 and
2).
(Salen)chromium(III) complexes of type 3 (Fig. 1) were
introduced to enantioselective catalysis by Jacobsen and
co-workers,19 and are known to be efficient catalysts for
hetero-Diels–Alder reactions,20 as well as for ring
opening of epoxides by trimethylsilyl azide19 amongst
other reactions.21 In our research group, we have used
these catalysts for [4+2] cycloaddition of various 1,3-
dienes to glyoxylates.22
When allyltriphenyltin was used, the enantioselectivity
decreased to 46% ee (entry 3). The reaction was most
rapid for tetraallyltin, but the enantiomeric excesses
obtained were much lower, not exceeding 13% ee (entry
4). Further studies were carried out using only com-
mercially available allyltributyltin.
As already mentioned, (salen)chromium complexes have
been successfully applied to enantioselective allylation
reactions using allyl halides, via the Nozaki–Hiyama–
Kishi reaction.12 In this reaction, the active catalytic
species is (salen)Cr(II), and the nature of the catalytic
cycles is a redox process. In our method, the cationic
complex (salen)Cr(III)þ is catalytically active, playing
the role of a typical Lewis acid.
The next stage of the study was an attempt to improve
the enantioselectivity of the reaction of n-butylgyloxy-
late with allyltributyltin. We investigated several factors
such as concentration, temperature, solvent, additives
and the counterion of the catalyst (Table 2).
The chromium complex 3a turned out to be too weak
as a Lewis acid to catalyze efficiently the reaction with
allyltrimethylsilane. The standard reaction conditions
resulted in low yields, below 15%, with enantiomeric
excesses of 30–40%. A significant amount of byproduct
is observed in this reaction. High pressure (10 kbar)
It turned out that when the reaction was conducted in
ꢀ
CH2Cl2, the presence of 4 A molecular sieves (entries 1
and 2 in Table 2) and the glyoxylate concentration
(entries 1 and 3) had practically no influence on the
enantiomeric excess of the product. Moreover, the rate
of addition of the allylating agent for the reaction based
on 1 mmolscael and the amount of cataylst (2 mo%l
and more) seemed to have no influence on enantiomeric
excess (entries 1 and 4).
To our surprise, lowering the temperature resulted in a
drop in enantioselectivity from 61% to 36% ee (entries
4–6). However, the reaction performed in boiling
CH2Cl2 proceeded much faster, while the drop in
enantioselectivity was insignificant (58% ee) (entry 7).
When the reactions were conducted in the presence of
amines or PPh3, we observed a slight improvement in
enantioselectivity. For example, a small amount of
lutidine caused increase in enantiomeric excess from
61% to 68% (see entries 1 and 8, respectively).
H
H
N
O
N
O
Cr
X
3a X=BF4, 3b X=Cl, 3c X=ClO4
Figure 1. (Salen)Cr(III)complexes.

P. Kwiatkowski et al. / Tetrahedron Letters 45 (2004) 5343–5346
5345
Table 2. Results of the enantioselective reaction of n-butyl glyoxylate (1a) with allyltributyltin catalyzed by chromium(III) complexes 3a–c and 4
under various reaction conditionsa
O
O
n
3a-c or 4
Bun O
H
SnBu3
Bun O
+
OH
2a
O
1a
Entry
Catalyst
Mol % of the
catalyst
Solvent
Concn of 1a
(mol/L)
Temp.
(ꢁC)
Time
(h)
Yield
(%)
Eeb
(%)
1
2
3
4
5
6
7
8
3a
3a+4 A molecular sieves
2
2
2
5
5
5
5
2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
1
20
20
3
1
82
82
81
80
78
76
77
80
61
62
61
61
51
36
58
68
ꢀ
1
3a
3a
0.1
1
20
20
5
1
3a
3a
1
5
24
24
0.25
4
1
)78 fi )20
40
20
3a
3a+2,6-lutidine
1
1
(2.5 mol%)
3a
9
10
2
2
MeNO2
MeNO2
1
1
20
20
3
3
79
80
70
70
3a+2,6-lutidine
(2.5 mol%)
11
12
13
14
15
3a
3a
3b
3c
4
0.2
2
MeNO2
No solvent
CH2Cl2
CH2Cl2
CH2Cl2
2
20
24
5
61
90
73
90
49
62
65
54
65
7
5 fi 20
20
20
2
1
1
1
3
2
3
2
20
3
a The reactions were carried out using 1 mmolof n-butyl glyoxylate (1a) and 1.2 mmol of allyltributyltin.
b The enantiomeric excess was determined by GC on a capillary chiral b-dex 120 column.
CH3
O
Of the solvents investigated, nitromethane appeared to
be more efficient than CH2Cl2 and raised the enantio-
meric excess to 70% (entry 9). The addition of lutidine to
the reaction conducted in MeNO2 caused no change
(entry 10). It is interesting that the reaction in MeNO2
proceeds in a biphasic system, because allyltributyltin is
poorly soluble thereby providing a relatively low and
constant concentration of the allylating agent in the
reaction medium. The reaction in MeNO2 can be effi-
N
Cr
O
Cl
4
Figure 2. Tridentate chromium(III) complex.
ciently catalyzed even by minor amounts of the catalyst
3a (0.2 mol%) accompanied by a decrease in enantio-
selectivity to 62% ee (entry 11).
reactions,24 we have tested its performance in the allyl-
ation reaction. This complex was unsatisfactory, and the
enantiomeric excess of product 2a was not greater than
7% (entry 15).
A great advantage of this reaction is that it can be
successfully performed with no solvent; a slight increase
in ee (to 65%) was observed compared to the reaction
conducted in CH2Cl2 (entries 1 and 12).
We have also carried out allylation using other glyoxy-
lates (R ¼ i-Pr, t-Bu, Bn) (Table 3). With respect to
enantioselectivity, the results were similar to those
We also tested the applicability of other chromium
complexes to this reaction. The commercially available
(salen)CrCl complex 3b, which served for the prepara-
tion of 3a and 3c, catalyzed the reactions less efficiently
and decreased the enantioselectivity to 54% (entry 13).
The best one, both in respect of yield and enantiomeric
excess, turned out to be the perchlorate complex 3c
(entry 14). We have also investigated chromium com-
plexes with diamines other than 1,2-diaminecyclohexane
such as 1,2-diphenylethylenediamine and 1,2-di-tert-
butylethylenediamine, but the enantioselectivities
obtained were lower.
Table 3. Influence of the alkyl substituent in glyoxylate 1 on the
reaction coursea
Entry
Glyoxylate Solvent
Yield (%)
Ee (%)
1
2
3
4
5
6
7
1a Bun
1a Bun
1b Pri
1b Pri
1c But
1c But
1d Bn
CH2Cl2
MeNO2
CH2Cl2
MeNO2
CH2Cl2
MeNO2
CH2Cl2
82
79
78
74
76
65
84
61
70
66
73
76
73
61
a The reactions were carried out using 1 mmol of alkyl glyoxylate (1a–
d), 2 mol% of 3a and 1.2 mmol of allyltributyltin, in 1 mL of solvent,
at 20 ꢁC for 3 h.
In the context of the recent work of Jacobsen concerning
the tridentate chromium(III) complex 4 (Fig. 2) and its
high efficiency in the hetero-Diels–Alder23 and the ene

5346
P. Kwiatkowski et al. / Tetrahedron Letters 45 (2004) 5343–5346
11. Nakajima, M.; Saito, M.; Shiro, M.; Hashimoto, S.-i.
J. Am. Chem. Soc. 1998, 120, 6419–6420.
12. (a) Bandini, M.; Cozzi, P. G.; Melchiorre, P.; Umani-
Ronchi, A. Angew. Chem., Int. Ed. 1999, 38, 3357–3359;
(b) Bandini, M.; Cozzi, P. G.; Umani-Ronchi, A. Tetra-
hedron 2001, 57, 835–843; (c) Berkessel, A.; Menche, D.;
obtained for n-butyl glyoxylate. In the case of the
reactions conducted in CH2Cl2, the enantiomeric excess
rises slightly (from 61% to 76% ee) along with the
increasing bulkiness of the R substituent in the glyoxy-
late. This effect is not as apparent for the reactions
conducted in MeNO2.
€
Sklorz, C. A.; Schroder, M.; Paterson, I. Angew. Chem.,
Int. Ed. 2003, 42, 1032–1035.
In all cases, when the catalysts 3a–c having (1R,2R)
configuration were used, the allylation product had the
(R) configuration. The absolute configuration of prod-
ucts 2a–d was determined by correlation with 1,2-pen-
tanediol.16a
13. Inoue, M.; Suzuki, T.; Nakada, M. J. Am. Chem. Soc.
2003, 125, 1140–1141.
14. (a) Dauben, W. G.; Hendricks, R. T.; Pandy, B.; Wu, S. C.
Tetrahedron Lett. 1995, 36, 2385–2388; (b) Rutjes, F. P. J.;
Kooistra, T. M.; Hiemstra, H.; Schoemaker, H. S. Synlett
1998, 192–194; (c) Zemribo, R.; Mead, K. T. Tetrahedron
Lett. 1998, 39, 3895–3898; (d) Macritchie, J. A.; Peakman,
T. M.; Silcock, A.; Willis, C. L. Tetrahedron Lett. 1998, 39,
In conclusion, we have developed an efficient,
undemanding method for the enantioselective allylation
of glyoxylates, which affords products with enantiomeric
excesses in the range of 58–76%. The reaction is highly
reproducible and not sensitive to external factors such as
oxygen or moisture. Moreover, it can be performed on
large scale.
ꢁ
ꢁ
7415–7418; (e) Dıaz, Y.; Bravo, F.; Castillon, S. J. Org.
Chem. 1999, 64, 6508–6511; (f) Ghosh, A. K.; Lei, H.
J. Org. Chem. 2000, 65, 4779–4781.
15. (a) Whitesell, J. K.; Bhattacharya, A.; Buchanan, C. M.;
Chen, H. H.; Deyo, D.; James, D.; Liu, C.-L.; Minton, M.
A. Tetrahedron 1986, 42, 2993–3001; (b) Macritchie, J. A.;
Silcock, A.; Willis, C. L. Tetrahedron: Asymmetry 1997, 8,
3895–3902; (c) Ramachandran, P. V.; Krzeminski, M. P.;
Reddy, M. V. R.; Brown, H. C. Tetrahedron: Asymmetry
1999, 10, 11–15; (d) Wang, D.; Wang, Z. G.; Wang, M.
W.; Chen, Y. J.; Liu, L.; Zhu, Y. Tetrahedron: Asymmetry
1999, 10, 327–338; (e) Loh, T.-P.; Xu, J. Tetrahedron Lett.
1999, 40, 2431–2434; (f) Jung, J. E.; Ho, H.; Kim, H.-D.
Tetrahedron Lett. 2000, 41, 1793–1796; (g) Yu, H.;
Ballard, C. E.; Wang, B. Tetrahedron Lett. 2001, 42,
1835–1838.
According to our knowledge, this reaction is the first
example of enantioselective allylation where the salen
complex is directly used as a Lewis acid (except for the
use of salen in the Nozaki–Hiyama–Kishi reaction). The
results seem to be a good starting point to further
optimization involving modification of the chiral ligand,
especially its 1,2-diamine moiety. Further studies to
improve the enantioselectivity of this reaction are
underway.
16. (a) Kiegiel, K.; Prokopowicz, P.; Jurczak, J. Synth.
Commun. 1999, 29, 3999–4005; (b) Czapla, A.; Chajewski,
A.; Kiegiel, K.; Bauer, T.; Wielogorski, Z.; Urbanczyk-
Lipkowska, Z.; Jurczak, J. Tetrahedron: Asymmetry 1999,
10, 2101–2111.
References and notes
17. Aoki, S.; Mikami, K.; Terada, M.; Nakai, T. Tetrahedron
1993, 49, 1783–1792.
18. Halligan, N. G.; Blaszczak, L. C. Org. Synth. 1990, 68,
104–108.
1. Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207–2293.
2. Furuta, K.; Mouri, M.; Yamamoto, H. Synlett 1991, 561–
562.
ꢁ
3. For a recent review, see: Denmark, S. E.; Fu, J. Chem.
Rev. 2003, 103, 2763–2793.
4. (a) Costa, A. L.; Piazza, M. G.; Tagliavini, E.; Trombini,
C.; Umani-Ronchi, A. J. Am. Chem. Soc. 1993, 115, 7001–
7002; (b) Keck, G. E.; Tarbet, K. H.; Geraci, L. S. J. Am.
Chem. Soc. 1993, 115, 8467–8468.
19. (a) Martınez, L. E.; Leighton, J. L.; Carsten, D. H.;
Jacobsen, E. N. J. Am. Chem. Soc. 1995, 117, 5897–5898;
(b) Leighton, J. L.; Jacobsen, E. N. J. Org. Chem. 1996,
61, 389–390.
ꢀ
20. Schaus, S. E.; Branalt, J.; Jacobsen, E. N. J. Org. Chem.
1998, 63, 403–405.
5. Bedeschi, P.; Casolari, S.; Costa, A. L.; Tagliavini, E.;
Umani-Ronchi, A. Tetrahedron Lett. 1995, 36, 7897–
7900.
6. Yanagisawa, A.; Nakashima, H.; Ishiba, A.; Yamamoto,
H. J. Am. Chem. Soc. 1996, 118, 4723–4724.
7. Cozzi, P. G.; Orioli, P.; Tagliavini, E.; Umani-Ronchi, A.
Tetrahedron Lett. 1997, 38, 145–148.
8. Motoyama, Y.; Narusawa, H.; Nishiyama, H. Chem.
Commun. 1999, 131–132.
9. Denmark, S. E.; Coe, D. M.; Pratt, N. E.; Griedel, B. D.
J. Org. Chem. 1994, 59, 6161–6163.
21. For a review of applications of (salen)Cr complexes in
asymmetric catalysis, see: Bandini, M.; Cozzi, P. G.;
Umani-Ronchi, A. Chem. Commun. 2002, 919–927.
22. (a) Kwiatkowski, P.; Asztemborska, M.; Caille, J.-C.;
Jurczak, J. Adv. Synth. Catal. 2003, 506–509; (b) Kwiat-
kowski, P.; Asztemborska, M.; Jurczak, J. Synlett, in press.
23. (a) Dossetter, A. G.; Jamison, T. F.; Jacobsen, E. N.
Angew. Chem., Int. Ed. 1999, 38, 2398–2400; (b) Gade-
mann, K.; Chavez, D. E.; Jacobsen, E. N. Angew. Chem.,
Int. Ed. 2002, 41, 3059–3061.
24. (a) Ruck, R. T.; Jacobsen, E. N. J. Am. Chem. Soc. 2002,
124, 2882–2883; (b) Ruck, R. T.; Jacobsen, E. N. Angew.
Chem., Int. Ed. 2003, 42, 4771–4773.
10. Iseki, K.; Mizuno, S.; Kuroki, Y.; Kobayashi, Y. Tetra-
hedron Lett. 1998, 39, 2767–2770.