Tetrahedron Letters 48 (2007) 2405–2408
Enantioselective glyoxylate-ene reactions catalysed by (salen)chromium(III) complexes Wojciech Chaładaj,a Piotr Kwiatkowski,a Jakub Majera and Janusz Jurczaka,b,* a
Institute of Organic Chemistry, Polish Academy of Sciences, 01-224 Warsaw, Poland b Department of Chemistry, Warsaw University, 02-093 Warsaw, Poland Received 14 December 2006; revised 9 January 2007; accepted 18 January 2007 Available online 24 January 2007
Abstract—An enantioselective carbonyl-ene reaction of alkyl glyoxylates with various 1,1-disubstituted oleﬁns, catalysed by chiral (salen)Cr(III)BF4 complexes, has been studied. We found that a chromium complex bearing adamantyl substituents at the 3,3 0 -positions of the salicylidene moiety catalysed the reaction with much greater selectively than the classic Jacobsen-type catalyst. The reaction proceeded eﬀectively under undemanding conditions in the presence of 2 mol % of the catalyst in an acceptable yield and with 59–92% ee. Ó 2007 Elsevier Ltd. All rights reserved.
The carbonyl-ene reaction of glyoxylates with oleﬁns is a method of choice for the construction of a-hydroxy acids containing a double bond at the c,d-position, which are important precursors for synthetic chemistry. The enantioselective version of this simple and atomeconomic reaction has been intensively investigated since 1989, when Mikami et al. applied BINOL–titanium(IV) complexes as catalysts.1,2 The literature also describes other chiral catalytic systems utilised for this reaction,2,3 mainly based on bis-oxazoline (Cu,4,5 Sc6) and bisphosphine (Ni,7 Pd,7,8 Pt9) ligands. BINOL– Ti(IV)1 as well as bis-oxazoline–Cu(II) complexes5 seem to be the most eﬃcient and general catalytic systems for the enantioselective glyoxylate-ene reaction. Taking into account our previous investigations on asymmetric catalytic hetero-Diels–Alder reactions10 and allylation of aldehydes,11 we focused our attention on salen–chromium(III) complexes 1a.12 So far, there is no example of the application of C2-symmetrical Jacobsen complexes to the ene reaction, although Yamada et al.13 demonstrated that the related chiral b-ketoiminato cobalt(III) eﬃciently catalysed the ene reaction of phenylglyoxal with 1,1-disubstituted alkenes. Furthermore, chiral Cr(III) tridentate Schiﬀ-base complexes were employed by Jacobsen in the ene-type reactions
of 2-methoxy- or 2-TMSO-propene with aromatic and aliphatic aldehydes.14 Recently, we have shown that the simple introduction of a bulky group at the 3- and 3 0 -positions of the salicylidene moiety of salen chromium(III) complexes increased the enantioselectivity of the allylation of aldehydes15 and hetero-Diels–Alder (HDA) reactions.16 Introduction of the adamantyl group to give complex 1b led to excellent selectivity (up to 98% ee), especially in the reaction of alkyl glyoxylates with cyclohexa-1,3-diene.16 Exploring the range of dienes, we found that, in the case of 2,3-dimethylbuta-1,3-diene, the use of catalyst 1b did not improve the selectivity of the HDA reaction; instead the chemoselectivity was shifted towards the ene product (Scheme 1). Moreover, the enantiomeric purity of the homoallylic alcohols obtained increased markedly.
N Cr + O O BF 4
1a * Corresponding author. Tel.: +48 22 343 23 30; fax: +48 22 632 66 81; e-mail: [email protected]
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tetlet.2007.01.104
N Cr + O O BF 4
This result prompted us to examine the sterically modiﬁed Jacobsen’s chromium complexes in the ene reaction.
W. Chaładaj et al. / Tetrahedron Letters 48 (2007) 2405–2408
1a or 1b
6 : 4
3 : 7
Scheme 1. Chemoselectivity of the (salen)Cr(III)-catalysed reaction of 2,3-dimethylbuta-1,3-diene with glyoxylate.
The (salen)chromium(III)-catalysed reaction of ethyl glyoxylate with isobutylene (2a) in toluene was investigated ﬁrst (Scheme 2). Indeed, increasing the steric hindrance at the 3,3 0 -positions of the salicylidene moiety of the catalyst exerted a beneﬁcial eﬀect on the enantioselectivity (Table 1). In the series of complexes 1c–e with dialkylphenylmethyl groups, the selectivity increased with the size of the alkyl substituent (entries 2–4). The ortho substituents on the aromatic groups (1i) further improved the selectivity, in contrast to those in the para position (1f–h) (compare entries 5–7 and 8). The best selectivity was obtained using catalyst 1b bearing large adamantyl groups (entry 9).
THF, acetonitrile and nitromethane), toluene seemed to be the best medium in terms of the enantioselectivity. Further optimisation of the reaction conditions (summarised in Table 2) showed that none of the reaction variables such as temperature, concentration or catalyst loading had a signiﬁcant inﬂuence on the enantioselectivity (Table 2, entries 1–6). Nonetheless, with higher temperature and catalyst loading, better yields were obtained. Finally, the alcohol residue in the glyoxylate was changed, which resulted in a drop in enantioselectivity for more branched groups (Table 2, see entries 1, 7 and 8). It is interesting that extension of the reaction time over 24 h did not improve the yield signiﬁcantly.
It is worth mentioning that catalyst 1b not only improved the selectivity of the model reaction from 15% to 79% ee (compared to the classic Jacobsen catalyst), but it is also readily accessible on a multigram scale from inexpensive commercial chemicals via a short, eﬃcient synthesis.16
The catalytic system proved to be generally applicable for the reaction of 1,1-disubstituted oleﬁns 2a–h under the optimised conditions, leading to the corresponding products of type 3 in moderate to good yields and selectivities, when 2 mol % of 1b was applied as a catalyst (Table 3).17 Methyl alkyl disubstituted oleﬁnes 2b–c reacted predominantly at the less hindered methyl group, with regioselectivities higher for the more branched substituents (entries 2 and 3). Moreover, the enantioselectivity was higher for more branched substrates (entries 1, 3 and 4), with the best result of 92% ee obtained for 2,3,3-trimethyl-1-butene (2d). 2-Arylpropenes 2e–f were slightly less eﬀective substrates in the reaction, leading to products with enantioselectivities around 60% ee (entries 5 and 6). Methylenecyclohexane (2g) reacted with a selectivity comparable to a-methylstyrene, whereas the more reactive methylenecyclopentane (2h) gave nearly a racemate (entries 7 and 8). Fortunately, the selectivity improved with a decrease in temperature (entry 9).
Having the best catalyst in hand, we decided to optimise the reaction conditions. Out of the solvents tested (methylene chloride, toluene, methyl tert-butyl ether,
N N Cr + O O BF4-
Scheme 2. The trial reaction. a
Table 1. Results of the trial reaction as a function of catalyst structure Entry
1 2 3 4 5 6 7 8 9 a
1a 1c 1d 1e 1f 1g 1h 1i 1b
t-Bu C(Me)2Ph C(Et)2Ph C(n-Pr)2Ph C(Me)2C6H4p-Cl C(Me)2C6H4p-OMe C(Me)2C6H4p-Me C(Me)2C6H4o-Me Adamantyl
53 48 50 36 51 52 50 42 49
15 35 53 71 31 36 36 48 79
Reaction conditions: 1 mol % of catalyst 1a–i, 1 mmol of ethyl glyoxylate, 3 mmol of oleﬁn in 1 ml of toluene, rt, 24 h. b Determined by GC on a chiral capillary b-dex 120 column.
The absolute conﬁguration of selected products (3a, 3g) was determined by reduction to the corresponding diols and subsequent comparison with a sample of known conﬁguration (Scheme 3). The reference diols were obtained by reduction of the corresponding (2R)-bornane-10,2-sultam derivatives, the conﬁguration of which was unambiguously proved via X-ray crystallography.18 The conﬁguration of the product turned out to be (S) whenever the (S,S)-catalyst was used. The result is in agreement with our earlier observations of the salen-catalysed reactions of glyoxylates10,11,15,16 and matches the previously presented model of chirality transfer.10c,11b To conclude, we have shown that the sterically modiﬁed complex 1b catalysed the ene reaction of alkyl glyoxylates with good enantioselectivities (up to 92% ee), while
W. Chaładaj et al. / Tetrahedron Letters 48 (2007) 2405–2408
Table 2. Results of the trial reaction under various reaction conditionsa
Mol % of 1b
Concentration of the glyoxylate
Alkyl in the glyoxylate
1 2 3 4 5 6 7 8
1 2 5 1 1 1 1 1
1 1 1 1 1 0.5 1 1
20 20 20 4 25 20 20 20
Et Et Et Et Et Et n-Bu i-Pr
49 64 79 32 7 47 50 45
79 79 79 81 80 79 76 42
Reaction conditions: 1b as catalyst, 1 mmol alkyl glyoxylate, 3 mmol of oleﬁn in toluene, 24 h. Determined by GC on a chiral capillary b-dex 120 column.
Table 3. Reaction of various oleﬁns under optimised conditionsa
R= Me 2a n-Pr 2b i-Pr 2c
t-Bu 2d 2e Ph p-ClC6H4 2f
N Cr + O
or 2 mol% of 1b
( )n n=2 n=1
1 2 3 4 5 6 7 8 9e
2a (Me) 2b (n-Pr) 2c (i-Pr) 2d (t-Bu) 2e (Ph) 2f (C6H4p-Cl) 2g 2h 2h
64 64 65 63 56 68 69 70 58
— 72:28 89:11 — — — — — —
79 89 (55)c 84 (62)c 92 61d 59d 63 4 30
Reaction conditions: 2 mol % of catalyst 1 b, 1 mmol ethyl glyoxylate, 1.5 mmol of oleﬁn (3 mmol in the case of 2a) in 1 ml of toluene, rt, 24 h. Determined by GC on a chiral capillary b-dex 120 column. c ee of the minor regioisomer given in parentheses. d Determined by HPLC on a Chiracel AS-H column. e The reaction was carried out at 25 °C. b
N S O
This work was ﬁnancially supported by the Ministry of Education and Science Grant PBZ-KBN-118/T09/16. Financial backing for P.K. from the Foundation for Polish Science is gratefully acknowledged. References and notes
configuration confirmed by X...