Mol 2 Net 2-Nitromethylacrylates as Useful Dinucleophiles for the Enantioselective Organocatalytic Michael / Henry Cascade Reaction

2-Nitromethylacrylates have proved to be suitable 1,3-dinucleophiles reacting with α,βunsaturated aldehydes in the presence of a secondary-amine catalyst to furnish Michael/Henry cascade products in moderate yields and with high enantioselectivities although with moderate diastereoselectivities. The reaction proceeds by iminium ion activation of the enal, which reacts regioselectively with the γ-carbon of the nitronate anion formed in situ, furnishing the desired cyclohexenes with three new stereocenters. Furthermore, and trying to avoid the diastereoselectivity issue, an efficient sequential Michael/Henry/dehydration reaction has been developed leading to enantiopure cyclohexadienes in moderate yields and excellent enantioselectivities.


Introduction
The Henry reaction has been associated to the Michael addition in domino type transformations, since it is an useful tool to form C-C bonds and also allows the preparation of a wide range of structurally different compounds due to the ability of the nitro group to be transformed into other nitrogen and oxygen containing functionalities. 1 The application of this cascade reaction to the synthesis of optically active complex products is unquestionable and SciForum http://sciforum.net/conference/mol2net-1 the examples described in the literature show its remarkable applicability making it a highly efficient tool for obtaining functionalized products with total diastereo-and enantiocontrol employing different types of catalysts.In this sense, it should be highlighted the area of organocatalysis, where enamine and iminium activation together with H-bonding catalysis strategy enable to carry out Michael/Henry cascade reactions with high efficiency and an exceptional level of control of the stereocenters formed in most cases. 2n this context, and taking into account the experience acquired in our research group in the field of organocatalytic Michael/Henry reactions, 3 we decided to focus our studies on developing a Michael/Henry cascade process employing 2-nitromethylacrylates as suitable functionalized Michael donors with α,β-unsaturated aldehydes under iminium activation (Scheme 1).
2-Nitromethylacrylates are expected to be active as carbon pronucleophiles in an initial Michael reaction because of the presence of two acidic protons in α position to the nitro group. 4oreover, the deprotonation of this compound 1 would lead to the generation of a resonancestabilized allylic anion, with potential to react either at C-α or at C-γ position (Scheme 1).Assuming that the selected pronucleophile can undergo a conjugate addition to the iminium ion resulting from the condensation of the Michael acceptor -an α,β-unsaturated aldehyde 2-and the aminocatalyst, the release of the catalyst after the initial Michael reaction would form a nitroaldehyde intermediate which is expected to undergo intramolecular Henry reaction in order to yield the desired product (compound 3 vs 4).Nevertheless, we have to be aware of the fact that if the catalyst is not released from the resulting Michael addition product, another intramolecular conjugate addition can take place between the enamine and the remaining α,βunsaturated ester moiety, leading to the formation of cyclopentanes 5 as shown in Scheme 1.With this goal in mind, we surveyed the behavior of compound 1a with trans-pentenal 2a as model system in the presence of a chiral secondary amine catalyst.In a first attempt, we carried out the reaction using diphenylprolinoltrimethyl silyl ether 4a as catalyst, in chloroform and at room temperature.When the compound 1a was completely consumed, the 1 H-NMR of the crude reaction mixture revealed that the cyclohexene 3a had been formed without observing other possible byproducts previously predicted (4 or 5).We concluded that the 2-nitromethylacrylate 1a was behaving as a 1,3-dinucleophile, leading to the polysubstituted cyclohexene after performing the projected cascade reaction through initial selective γ-addition to the intermediate iminium ion derived from pentenal, followed by catalyst releasing and intramolecular Henry reaction.The compound 3a was isolated by flash column chromatography in 36% yield as a mixture of three diastereoisomers (2.5:1:1), and with a 75% e.e. for the major diastereoisomer.

Scheme 2. Viability of the Michael/Henry reaction
These results made us explore the possibilities that this reaction could offer.The first parameter to study was the organocatalyst employed and therefore, we tested some different chiral secondary amines in the model reaction.Taking into account the high enantioselectivity achieved with prolinol derivative 4a, we decided to use other diphenylprolinol derivatives with a bulkier O-protecting group and also diarylprolinol derivatives containing bulkier aryl substituents, but the reaction did not progress in any case and the starting materials were recovered in all cases.As the other catalysts tested did not show any activity, we decided to study the influence of the additive (Table 1).Some examples in the literature, 1a made us think that Brønsted bases could be beneficial for the reaction, assuming that a base could help in the nucleophile formation by promoting the deprotonation of the substrate 1a, and later on it could also assist the intramolecular Henry reaction.Fortunately, when bases such as DBU, 1,1,3,3tetramethylguanidine, triethylamine and DMAP were used cyclohexene 3a was selectively obtained (entries 2-5).The highest value regarding the yield was achieved with triethylamine but it showed poor diastereoselectivity (entry 4) and the best enantiocontrol was obtained employing DMAP although the yield was not good enough (entry 5).http://sciforum.net/conference/mol2net-1Bifunctional Brønsted bases containing acidic H-donor sites were also tried as additives (entries 6-7) in order to evaluate their potential positive contribution to the reaction, but without giving better results than the ones obtained with triethylamine (entry 6).Finally, we decided to evaluate the use of phosphines as cocatalyst (entries 8-10) and in this sense, the use of a bulky phosphine like t Bu3P gave the compound 3a with good yield and an acceptable value of diastereomeric excess and the best enantioselectivity (entry 10).Moreover, it has to be pointed out that, in all cases the formation of other regioisomers was not detected by NMR analysis of the crude reaction mixture.
Selecting t Bu3P as the best additive, the optimization of the model reaction went on with the election of the most suitable solvent for the Michael/Henry transformation (Table 2).A battery of tests were run to evaluate the influence of different solvents, using polar as well as nonpolar ones, and in a general view, the diastereoand enantioselection of the cascade reaction was not influenced by the nature of the solvent (entries 1-7).The tests run revealed that chloroform (entry 1) remained being the most suitable solvent for the reaction in terms of overall yield and enantioselectivity.Finally, we tried to improve the diastereo-and enantiomeric ratio by lowering the temperature to -30º C but the results previously obtained at rt could not be improved (entry 8).http://sciforum.net/conference/mol2net-1The poor diastereoselectivity achieved in all reactions might lay in the second step of the cascade reaction.Taking into account that the C-3 stereocenter that comes from the Michael addition of the nucleophile to the iminium ion is expected to be efficiently controlled by the catalyst, it seems sensible to think that the lack of stereogenic control derives from the intramolecular Henry reaction in which the stereocenters C-5 and C-6 are created, after the release of the organocatalyst.Aware of this diastereoselection problem, we thought about possible transformations that could lead to a convenient process in which all diastereoisomers would converge into a single product.The dehydration of the cyclohexene 3a, obtaining cyclohexadiene 5a seemed a promising way to eliminate two sterocenters and to simplify the structure of the compound, maintaining the stereocenter created at the initial Michael reaction step.Therefore, a dehydration reaction was carried out, employing sodium methoxyde in methanol observing that the reaction took place in a quantitative way, and fortunately, being possible to carry out in a sequential fashion in an efficient manner (Scheme 3).http://sciforum.net/conference/mol2net-1

Scheme 3. Chemical manipulation of adduct 3a
With those results in our hands we decided to evaluate the importance of the substitution of the enals used as Michael acceptors and the substitution of the acrylate 1, in order to see the influence of this substitution, in the sequential Michael/Henry/dehydration reaction.As it can be seen in Table 3, overall yields obtained were around 30-40% in all cases in which linear aliphatic substituents were present at the βposition of the enal regardless the length of the chain.On the other hand, the enantiomeric excesses increased when longer chains were introduced (entries 1-3).In contrast, when a bulkier element, such as an i Pr group, was placed at the β position of the α,β-unsaturated aldehyde (entry 4) the yield decreased drastically although the enantioselection was very high.Similarly when the bulkier acrylate 1b was employed a poor yield of compound 5e was achieved although in very good enantiomeric excess.In a plausible mechanism proposed, a Michael addition of the deprotonated 2nitromethylacrylate 1 to the iminium ion resulting from the fusion between the aminocatalyst 4a and the α,β-unsaturated aldehyde 2, would take place.After the catalyst http://sciforum.net/conference/mol2net-1 is released, an intramolecular Henry cyclization occurs, achieving the cyclohexenes 3 which would undergo a sequential dehydration step to provide the corresponding cyclohexadienes 5.
General procedure for the Sequential Michael/Henry/dehydration reaction: The α,βunsaturated aldehyde 2a-d (0.50 mmol) was added to a solution of (S)-α,α-diphenyl-2-pyrrolidinemethanol trimethylsilyl ether 4a (0.05 mmol), tri-tertbutylphosphine (0.05 mmol) and the acrylate 1a-b (0.25 mmol) in chloroform (2 mL).The reaction was stirred at room temperature until full conversion.The solvent was evaporated under reduced pressure and after dissolving the reaction crude in dry MeOH (2 mL) it was added via canula to a solution of metallic sodium (2 mmol) in MeOH (4 mL) at 0 º C. The reaction mixture was allowed to reach room temperature and stirred it for 1 hour.Then a solution of H2SO4/MeOH (1:5) was added until pH≈5 was achieved and the solvent was evaporated under vacuum.H2O (20 mL) and CH2Cl2 (20 mL) were added and the aqueous layer was extracted with CH2Cl2 (3 × 20 mL).The organic extracts were dried over anhydrous Na2SO4 and the solvent was evaporated under vacuum.The crude was charged onto silica gel and subjected to flash column chromatography in order to achieve compound 5a-e.

Conclusions
To sum up, we have developed a methodology that involves an organocatalytic Michael/Henry cascade reaction under iminium activation between enals and 2-nitromethylacrylates 1a-c.The selected Michael donor, 2-nitromethylacrylate, has proved to be a suitable starting material for the reaction, acting as an appropriate 1,3-dinucleophile and has reacted with α,β-unsaturated aldehydes stereocontrolled by α,α-diphenylprolinol O-silylated catalyst, but showing poor diastereoselectivity.
In order to avoid this problem, we have designed a methodology involving Michael/Henry cascade reaction followed by a sequential dehydration leading to the enantiopure cyclohexadienes 5a-e in a moderate yield but good enantioselectivity.

Table 1 .
Influence of the additive in the cascade reaction.a a One equivalent of 1a and two equivalents of aldehyde 2a were used.b Determined by 1 H-NMR analysis for the mixture of diastereoisomers after flash column chromatography purification.c Referred to the mixture of diastereoisomers after flash column chromatography purification.d Calculated by HPLC for the major diastereoisomer.

Table 2 .
Influence of the solvent and temperature in the cascade reaction.a a One equivalent of 1a and two equivalents of aldehyde 2a were used.b Determined by 1 H-NMR analysis for the mixture of diastereoisomers after flash column chromatography purification.c Referred to the mixture of diastereoisomers after flash column chromatography purification.d Calculated by HPLC for the major diastereoisomer.