Asymmetric mukaiyama-aldol reaction in aqueous media promoted by zinc-based chiral lewis acids

Article abstract of DOI:10.1002/adsc.200404314

Asymmetric aldol reactions in aqueous media have been realized by using zinc-based chiral Lewis acids. The aldol products have been obtained with high yield, diastereocontrol and a good level of enantioselectivity. The reactivity of both acetophenone and

Full text of DOI:10.1002/adsc.200404314

COMMUNICATIONS
Asymmetric Mukaiyama-Aldol Reaction in Aqueous Media
Promoted by Zinc-Based Chiral Lewis Acids
Jacek Mlynarski,* Joanna Jankowska#
Institute of Organic Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, 01-224 Warsaw, Poland
Fax: (þ48)-22-632-6681; e-mail: mlynar@icho.edu.pl
Received: October 12, 2004; Accepted: January 3, 2005
Abstract: Asymmetric aldol reactions in aqueous
media have been realized by using zinc-based chiral
Lewis acids. The aldol products have been obtained
with high yield, diastereocontrol and a good level
of enantioselectivity. The reactivity of both aceto-
phenone and propiophenone enol ether surrogates
was tested with a range of aldehydes.
of silicon enolates using an aqueous solution of formal-
dehyde.[8]
In this paper we demonstrate the utility of a zinc-
based chiral Lewis acid for the enantioselective Mu-
kaiyama-aldol reaction in aqueous media.
In spite of some promising examples of direct aldol
condensations promoted by chiral Zn-complexes[9] in
anhydrous solutions, little is know about the Zn-sup-
ported Mukaiyama-aldol condensation both in anhy-
drous and aqueous media. The initial attempt by Ko-
bayashiꢀs group employing a combination of Zn(OTf)2
and chiral crown ether can hardly by regarded as prom-
ising.[5] On the other hand, the good results obtained by
the same group with asymmetric silyl enol ether-based
Mukaiyama–Mannich type reactions in aqueous me-
Keywords: asymmetric catalysis; homogeneous catal-
ysis; Lewis acids; Mukaiyama-aldol reaction; N li-
gands; water
Among the Lewis acid-catalyzed carbon-carbon bond dia[10] strongly encourage intensive exploration of the
forming reactions, the aldol-type reaction of silyl enol subject.
ethers with carbonyl compounds (the Mukaiyama reac-
The successful application of a zinc salt to asymmetric
tion) has been recognized as one of the most valuable. reactions in water-containing solutions must rely on its
It is therefore not surprising that a large number of fit into the chiral ligand. Keeping in mind the well known
methods have been developed in order to achieve this kinship of the Zn2þ cation with N-donor ligands we de-
goal.[1]
cided to test various N-binding compounds (1–7) as po-
In recent years, stereoselective organic reactions in tential sources of chirality.
aqueousmediahave attracted a great deal ofattention.[2]
As for first-step substrates, we focused on benzalde-
To replace classical Lewis acids acting only in rigorously hyde and the Z-silyl enol ether 8. Preliminary experi-
dried solvents, their water-tolerant substitutes have ments revealed that the combination of zinc triflate
been screened with more or less satisfactory results in al- with pybox-type ligands was the couple of choice for
dol and aldol-type reactions.[3] The very first catalytic this process.
asymmetric aldol reactions in aqueous media were per-
When (S,S)-1 (22 mol %) and Zn(OTf)2 (20 mol %)
formed with Cu(OTf)2 as catalyst and bis(oxazolines) as were used, the reaction of benzaldehyde with silyl enol
chiral counterpart.[4] Other metals, e.g., lead[5] and par- ether 8 in THF/H2O (9/1) at 08C for 10 h gave the desired
ticularly gallium[6] have been used with some success in aldol adduct 10 in good yield and with good diastereo-
enantioselective aqueous Mukaiyama-aldol reactions. and enantioselectivity (Table 1, entry 1). Importantly,
Recently, Kobayashi and co-workers developed the similar yield and stereoselectivity were observed when
combination of RE(OTf)3 and chiral bis-pyridino-18- THF/H2O (1/1) was applied as solvent. The same reaction
crown-6 for the asymmetric Mukaiyama-aldol reaction in ethanol-water was visibly faster but less enantioselec-
in aqueous ethanol [ee up to 85% for aromatic enol tive (entry 1 vs. 11). From the practical point of view, it
ether when Pr(OTf)3 was used as Lewis acid].[7] In gen- was exciting to find that the reaction does not require a
eral, the state-of-the-art in this area provides aldol prod- large excess of silyl enol ether, and typically 1.2 equivs.
ucts in good ees (70–80%). The scope and limitation of are sufficient to obtain reasonable yields.
this reaction is still not well recognized, and for ee ex-
The key results from the reaction optimization are
ceeding 80% special ligands have to be chosen.[6,7] In summarized in Table 1. Among the many members of
general, the elaborated methods fail when aliphatic al- the pybox-type ligand family,[11] OH-pybox 2 revealed
dehydes are substrates, although Kobayashi recently better conversion with poor enantioselectivity and its
presented an excellent example of hydroxymethylation Ph analogue 3 failed miserably (Table 1, entries 5 and 6).
Adv. Synth. Catal. 2005, 347, 521–525
DOI: 10.1002/adsc.200404314
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
521

Jacek Mlynarski, Joanna Jankowska
COMMUNICATIONS
Table 1. Metal sources and some selected chiral ligands used in the initial studies.
Entry
Catalyst
Solvent
Temp. [ 8C]
Time [h]
Yield[a] (syn/anti)
Ee[c] syn
53 (S,S )[12]
10
–
a) ligand and anion effects
1
2
3
4
5
6
7
8
A1
B1
C1
D1
A2
A3
A4
A5
A6
A7
i
i
i
i
i
i
i
i
i
0
0–rt
0–rt
0
0
0
0
0
10
20
20
20
10
96
20
20
40
20
73 (9/1)
14
trace
53
82
56
67
59
25
58
50
ꢀ24
ꢀ13
ꢀ3
0
0
ꢀ27
9
10
0–rt
0
ii
b) solvent and temperature effects
11
12
13
14
15
16
17
18
19
20
A1
A1
A1
A1
A1
A1
A1
A1
A1
A1[d]
ii
ii
0
10
48
48
48
72
24
72
72
48
48
80 (9/1)
97[b]
39
62
73
65
69
69
75
77
75
72
ꢀ10
ꢀ10
ꢀ10
ꢀ20
ꢀ20
ꢀ20
ꢀ20
ꢀ25
ꢀ25
iþii (2: 1)
66[b]
iii
30
ii
ii
88 (95/5)
93 (95/5)[b]
43 (95/5)
34 (96/6)
86 (96/4)[b]
88 (93/7)[b]
iþii (1: 1)
iþii (2: 1)
iþii (1: 1)
iþii (1: 1)
[a]
[b]
[c]
[d]
Isolated yield.
2.0 equivs. of 8 were used.
The ees were determined on a Daicel Chiralpak AD-H column.
10 mol % of the catalyst was used.
The combination of box 7 with zinc was not as efficient
Next, we examined the effect of solvents in the reac-
as in the case of the Cu salt[4] (Table 1, entry 10). All oth- tion. Application of ethanol instead of THF gave a bet-
er amino ligands failed in the elaborated reaction.
ter conversion but affected the enantioselectivity (Ta-
Zinc fluorideas acatalystled toa somewhat poorer re- ble 1, entry 11). In this case, however, decreasing of re-
sults while the application of zinc acetate and iodide was action temperature was more promising, and at
simply disappointing.
ꢀ108C the ee improved visibly. At the same tempera-
522
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 521–525

Asymmetric Mukaiyama-Aldol Reaction
COMMUNICATIONS
Table 2. Mukaiyama-aldol condensation with the A1 catalytic system.
Entry
1
Aldehyde
Product
Time [h]
72
Yield[a] (syn/anti)
ee[b] (syn)
11a
77 (92/8)
85 (98/2)
62 (94/6)
86 (95/5)
70
2
3
4
11b
11c
11d
24
48
48
72
74
72
5
6
11e
11f
11g
24
79 (91/9)
72
45
74
72[c]
72
16 (syn only)
65 (95/5)
7
[a]
Isolated yield.
[b]
[c]
The ees were determined on a Daicel Chiralpak AD-H column.
Reaction temperature 08C.
ture the catalytic system A1 was insoluble in THF/H2O.
Surprisingly, the hindered o-anisaldehyde was more
Further decreasing of the reaction temperature resulted reactive as compared to its para counterpart, probably
in increases of both diastereo- and enantioselectivities, because of interaction of the methoxy function with
leading to 69% ee at ꢀ208C (entry 15).
the metal core of catalyst. Not only aromatic aldehydes
In this case, however, the main condensation product (entries 1–4) but also a,b-unsaturated aldehydes, e.g.,
10 was accompanied by some amount of a-hydroxypro- cinamaldehyde (entry 5) gave good yields as well as
piophenone (less than 10%). To find the optimal balance high stereoselectivity. Aliphatic aldehydes were good
between yield and stereoselectivity we have tested substrates as well in the presented catalytic system. Piv-
mixed solvents. In the case of a wet ethanol-THF mix- alaldehyde was obviously less reactive leading, however,
ture (1:1) we obtained 86% yield when 2 equivs. of 8 to the desired product only at 08C which resulted in the
were used. Substrate loading does not affect the ob- loss of reaction selectivity. Heptanal, in spite of a slightly
served ee (entries 17 and 19). We were delighted to lower reactivity, showed the same level of stereoselec-
find that the reaction proceeded in high yield with tivity as aromatic aldehydes. Retention of the ee level
good enantioselectivity even using 10 mol % of the cat- in the reaction with aliphatic aldehydes is a valuable ob-
alyst (entry 20). Five mol % also worked well, although servation assome previouslydemonstrated catalytic sys-
with slightly lower ee (70%).
tems lost stereosectivity with these important sub-
In water-based aldol reactions we observed a predom- strates.[4,6,7]
inance of syn-aldol products which is in full agreement
The reasonable outcome in the condensation of enol
with previously published results.[4–7] This, as a rule, is ether 8 encouraged us to test the reactivity of its shorter
in contrast with the analogous reactions run under anhy- analogue 12. Such a variant of the Mukaiyama-type re-
drous conditions where the anti-isomer is usually the action in aqueous media to the best of our knowledge
major product.[2a]
has never been presented in the literature. Substrate
Encouraged by the enhanced selectivity, we started to 12 was previously demonstrated to be moderately reac-
explore the reaction scope. In the catalytic asymmetric tive in Mukaiyama–Mannich reactions in aqueous solu-
aldol reaction using Zn(OTf)2 and 1 (A1 system), other tion with some success (ee up to 90%).[10]
aldehydes were tested, and the results are summarized
in Table 2.
Again, we tested combination of all three pybox li-
gands 1–3 with zinc triflate. In this case, because of
Adv. Synth. Catal. 2005, 347, 521–525
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
523

Jacek Mlynarski, Joanna Jankowska
COMMUNICATIONS
Table 3. Mukaiyama-aldol condensation of acetophenone si-
lyl enol ether.
scope and limitations of this process, especially with ali-
phatic substrates, are in progress in our laboratory.
Experimental Section
General Remarks
Entry Aldehyde
1
Ligand Product Yield[a] ee[b]
Reactions were controlled using TLC on silica alu-plates
[Merck (0.2 mm)]. All reagents and solvents were purified
and dried according to common methods. All organic solutions
were dried over Na2SO4. Reaction products were purified by
flash chromatography using Merck silica gel 60 (240–400
1
2
3
2
13a
13a
13a
13b
22
38
33
48
27 (S)[13]
64 (R)
17 (R)
43
2
3
4
1
mesh). The diastereomer ratios were determined using a H
NMR technique. Enantiomer ratios were measured on chiral
HPLC columns (Chiralpak AS and AD-H) with hexane/2-
propanol (9:1).
Typical Procedure for the Asymmetric Aldol Reaction
5
6
7
2
2
2
13c
13d
13e
85
15
60
56
56
77
To a solution of pybox ligand 1 (18 mg, 0.06 mmol) and zinc tri-
fluoromethanesulfonate (18 mg, 0.05 mmol) in THF-EtOH-
H2O (1/1/0.2; 1.5 mL) at ꢀ258C were added silyl enol ether 8
(250 mL, 1 mmol) and the appropriate aldehyde (0.5 mmol).
The whole solution was stirred for 24 h at the same tempera-
ture (maintaining the entire homogeneous reaction mixture
in the fridge at an appropriate temperature led to desired prod-
ucts with only slightly poorer results). The reaction mixture
was diluted with MTBE and washed with water and brine.
The organic solution was dried, evaporated to dryness and
the residue was purified by silica gel chromatography
(AcOEt/hexane, 1/4).
[a]
Isolated yield.
[b]
The ees were determined on a Daicel Chiralpak AS col-
umn.
the much lower reactivity of the substrate, the best re-
sults were obtained with as much as 3 equivs. of 12 at
08C when THF-H2O was used as medium. In ethanol-
based solvents we observed a faster hydrolysis of enol
ether substrate. Surprisingly, for the condensation of
12 the most promising source of chirality turned out to
be OH-pybox 2 (Table 3, entry2).
In the case of the condensation of 12 with aromatic al-
dehydes the same rules were observed favoring o-substi-
tuted aldehydes, e.g., o-anisaldehyde was a far more re-
active and selective substrate when compared to p-ani-
saldehyde.
Acknowledgements
Financial support by the Polish State Committee for Scientific
Research (KBN Grant 3 T09A 126 27) is gratefully acknowl-
edged. Author (JM) thanks the Alexander von Humboldt Foun-
dation for Return Fellowship and Knauer-HPLC system gener-
ous donation.
References and Notes
#
The author (JJ) is a student of Warsaw University of
In summary, we present the successful application of
zinc-based chiral Lewis acids to the enantioselective
Mukaiyama-aldol condensation in aqueous media. The
catalytic asymmetric aldol reaction using Zn(OTf)2
and chiral ligand proceeded with good to high yields
and stereoselectivities, when several aldehydes and
two silyl enol ethers were tested. The same stereoselec-
tivity was observed in wet to aqueous solutions (contain-
ing 10–50% water), and aliphatic aldehydes were found
to be acceptable substrates. To the best of our knowl-
edge this is the first example of an enantioselective var-
iant of this reaction executed with the zinc salt-pybox li-
gand combination. Further experiments to define the
Technology, and this work is a part of her M.Sc. Thesis.
[1] a) T. D. Machajewski, Ch.-H. Wong, Angew. Chem. Int.
Ed. 2000, 39, 1352–1374; b) C. Palomo, M. Oiarbide,
J. M. Garcia, Chem. Soc. Rev. 2004, 33, 65–75.
[2] a) U. M. Lindstrçm, Chem. Rev. 2002, 102, 2751–2772;
b) D. Sinou, Adv. Synth. Catal. 2002, 344, 221–237;
c) K. Manabe, S. Kobayashi, Chem. Eur. J. 2002, 8,
4095–4101; d) S. Kobayashi, K. Manabe, Acc. Chem.
Res. 2002, 35, 209–217.
[3] a) F. Fringuelli, O. Piermatti, F. Pizzo, in: Organic Syn-
thesis in Water, (Ed.: P. A. Grieco), Blackie Academic
& Professional, London, 1998, pp. 250–261; b) S. Ko-
bayashi, in: Organic Synthesis in Water, (Ed.: P. A.
524
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 521–525

Asymmetric Mukaiyama-Aldol Reaction
COMMUNICATIONS
Grieco), Blackie Academic & Professional, London,
1998, pp. 262–302.
[4] S. Kobayashi, S. Nagayama, T. Busujima, Tetrahedron,
1999, 55, 8739–8746.
[5] S. Nagayama, S. Kobayashi, J. Am. Chem. Soc. 2000, 122,
11531–11532.
[6] H.-J. Li, H.-Y. Tian, Y.-J. Chen, D. Wang, Ch.-J. Li,
Chem. Commun., 2002, 2994–2995.
Trost. H. Ito, E. Silcoff, J. Am. Chem. Soc. 2001, 123,
3367–3368; d) B. M. Trost. E. Silcoff, H. Ito, Org. Lett.
2001, 3, 2497–2500.
[10] a) S. Kobayashi, T. Hamada, K. Manabe, J. Am. Chem.
Soc. 2002, 124, 5640–5641; b) T. Hamada, K. Manabe,
S. Kobayashi, Angew. Chem. In. Ed. 2003, 42, 3927–
3930; c) T. Hamada, K. Manabe, S. Kobayashi, J. Am.
Chem. Soc. 2004, 126, 7768–7769.
[7] T. Hamada, K. Manabe, S. Ishikawa, S. Nagayama, M.
Shiro, S. Kobayashi, J. Am. Chem. Soc. 2003, 125,
2989–2996.
[8] S. Ishikawa, T. Hamada, K. Manabe, S. Kobayashi, J.
Am. Chem. Soc. 2004, 126, 12236–12237.
[11] G. Desimoni, G. Faita, P. Quadrelli, Chem. Rev., 2003,
103, 3119–3154.
[12] The absolute configuration of the alcohol 10 was deter-
mined by comparison of the HPLC analysis with litera-
ture data in ref.[7]
[9] a) N. Kumagai, S. Matsunaga, T. Kinoshita, S. Harada, S.
Okada, S. Sakamoto, K. Yamaguchi, M. Shibasaki, J.
Am. Chem. Soc. 2003, 125, 2169–2178; b) B. M. Trost. H.
Ito, J. Am. Chem. Soc. 2000, 122, 12003–12004; c) B. M.
[13] The absolute configuration of the alcohol 13a was deter-
mined by comparison of optical rotation with literature
data: S. H. Mashraqui, R. M. Kellogg, J. Org. Chem.
1984, 49, 2513–2516.
Adv. Synth. Catal. 2005, 347, 521–525
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
525