Primary-tertiary diamine-catalyzed Michael addition of ketones to isatylidenemalononitrile derivatives

Introduction

The Michael reaction is one of the fundamental carbon–carbon bond forming reactions in organic synthesis, since a plethora of carbon nucleophiles and activated olefins could be expected to give versatile arrangements [1-8]. Among the various Michael acceptors, the oxindole-based Michael acceptors (Figure 1) are considered as valuable electrophiles for catalytic Michael reactions, as these provide a viable approach to procure 3,3'-disubstituted oxindole frameworks [9-16]. The oxindole framework bearing a tetra-substituted carbon stereocenter at C-3 is a privileged heterocyclic motif that is present in a large variety of bioactive natural products and a series of pharmaceutically active compounds [17]. In recent years, isatins derived Michael acceptors, isatylidenemalononitriles, have attracted considerable attention as novel substrates for the enantioselective synthesis of 3,3'-disubstituted oxindoles [18-22]. The organocatalytic Michael addition of ketones to isatylidenemalononitriles via enamine-catalysis has emerged as a useful tool for the synthesis of 3,3'-disubstituted oxindoles, which serve as important precursors to procure various structurally diverse spirooxindoles [23,24].

Over the years, many chiral organocatalysts have been developed and explored for Michael reactions [1-8]. Recently, aminocatalysts – in particular those bearing a primary amine moiety – have been found to catalyze a variety of carbon–carbon bond-forming reactions [25-30]. Small peptides derived from acyclic amino acids, primary-secondary diamines, Cinchona-based primary amines, and thioureas with a primary amine functionality etc., have found many successful applications in Michael addition reactions via an iminium–enamine catalysis [31-37]. A few applications of primary-tertiary diamine in aldol reactions have been published [39-43]. To the best of our knowledge, however, the catalytic potential of amino acids derived primary-tertiary diamine organocatalysts for Michael reaction via enamine activation has not been investigated so far [38]. With readily available and inexpensive natural amino acids as a chiral source, we developed very simple primary-tertiary diamine organocatalysts (Figure 2) for asymmetric aldol reactions [44,45]. We describe herein that a similar catalyst system could also efficiently catalyze the asymmetric Michael addition reaction between ketones 2 and isatylidenemalononitrile derivatives 3 to procure highly functionalized 3,3'-disubstituted oxindoles 4 which could easily be transformed into spirooxindoles.

Results and Discussion

Initially, the Michael addition of acetone (2a) to isatylidenemalononitrile (3a) catalyzed by chiral diamine 1a (10 mol %) with trichloroacetic acid (10 mol %) as an additive under mild conditions at room temperature was investigated (Scheme 1).

The Michael adduct 4a was isolated in 95% yield and 69% ee (Table 1, entry 1). Encouraged by the outcome of the preliminary reaction, we optimized the reaction conditions by studying the effect of different solvents and their amount as well as the effect of acid additives on this transformation (Tables 1–3). In our previous studies [10], the diamine 1a afforded the best result for aldol reactions with water as a solvent. Consequently, the above transformation was performed with water as a solvent. The Michael adduct 4a was obtained in good yield of 87% and a moderate enantioselectivity of 55% ee (Table 1, entry 2). It was planned to study the effect of different organic solvents on the stereochemical outcome of this reaction. Interestingly, the reaction of 2a with 3a performed in tetrahydrofuran (THF) gave Michael adduct 4a in good yield of 91% and a higher enantioselectivity of 84% ee (Table 1, entry 3). Other etheral solvents, such as 1,4-dioxane and methyl tert-butyl ether (MTBE), provided Michael adduct 4a in 89% and 90% yield and 89% ee and 87% ee, respectively (Table 1, entries 4 and 5). In toluene, 4a was obtained in 89% yield and 90% ee (Table 1, entry 6). The chlorinated solvents such as dichloromethane, chloroform and 1,2-dichloroethane (DCE) gave 4a in 90%, 89% and 91% yield and 90% ee, 88% ee and 91% ee, respectively (Table 1, entries 7–9). The utilization of methanol as a solvent led to the isolation of Michael adduct 4a in good yield but with a low enantioselectivity of 12% ee (Table 1, entry 11). Dimethylformamide was also found to be an inferior solvent for this reaction (Table 1, entry 12). Thus, 1,2-dichloroethane emerged as the solvent of choice for this transformation and was used for all further optimization studies (Table 1, entry 9).

In order to see the effect of the reaction concentration, the amount of 1,2-dichloroethane was varied (Table 2). The use of 0.5 mL of 1,2-dichloroethane afforded 4a in good yield of 93% and enantioselectivity of 91% ee (Table 2, entry 2). The reaction carried out with 0.75 mL, 1.0 mL and 1.5 mL of 1,2-dichloroethane provided 4a with an enhanced enantioselectivity of 92% ee, 93% ee and 95% ee, respectively (Table 2, entries 3–5). On performing the reaction in 2.0 mL of 1,2-dichloroethane, 4a was obtained in 88% yield and increased enantioselectivity of 96% ee after a long reaction time of 96 hours (Table 2, entry 6). There was a small difference in the enantioselectivity of the product 4a on performing the reaction in 1.5 mL and 2.0 mL of 1,2-dichloroethane, but the rate of the reaction was faster when a lower amount of solvent was used. So, we decided to employ 1.5 mL of 1,2-dichloroethane as a solvent for the further optimization experiments.

In order to obtain a highly enantioselective transformation, the effect of different acid additives on the enantioselectivity of 4a was studied (Table 3). The reaction was performed with 3,5-dinitrobenzoic acid and chloroacetic acid afforded 4a in good yield of 91% and 90%; and moderate enantioselectivity of 58% ee and 60% ee, respectively (Table 3, entries 1 and 2). The reaction carried out with strong acids such as trifluoromethanesulfonic acid (TsOH) and trifluoroacetic acid (TFA) gave 4a in good yields of 86% and 89% and high enantioselectivities of 97% ee and 96% ee (Table 3, entries 3 and 4). The application of L-camphorsulfonic acid as an additive resulted in the isolation of 4a in 92% yield and an enantioselectivity of 98% ee after 30 hours (Table 3, entry 5). The D-camphorsulfonic acid turned out to be the best acid additive providing 4a in high yield of 93% and excellent enantioselectivity of 99% ee after a reaction time of 24 hours (Table 3, entry 6) [46].

Even though an excellent level of enantioselectivity of the product was observed with catalyst 1a we screened different diamine catalysts in our quest for a superior catalyst. L-Isoleucine derived primary-tertiary diamine catalysts having piperidinyl 1b and pyrrolidinyl 1c groups gave Michael product 4a in high yield (>90%) and excellent enantioselectivity (98% ee each) (Table 4, entries 2 and 3). The primary-tertiary diamine catalysts characterized by an acyclic tertiary amine, such as a N,N-dioctyl (1d) group, gave 4a in 89% yield and 96% ee (Table 4, entry 4). The L-leucine, L-valine and L-phenylalanine derived primary-tertiary diamine catalysts (1e–1g) also provide the Michael adduct 4a in good yield (92–94%) and an excellent level of enantioselectivity (97–98% ee) (Table 4, entries 5–7). All primary-tertiary diamine catalysts 1a–1g gave 4a in high yield (89–94%) with an excellent level of enantioselectivity (96–99% ee). In contrast, the primary-secondary diamine 1h catalyst afforded Michael adduct 4a in 10% yield after a long reaction time (Table 4, entry 8). The screening study highlights the importance of the primary-tertiary diamine skeleton in the catalysis of the addition of acetone (2a) to isatylidenemalononitrile (3a). Thus, the best reaction conditions consist of 10 mol % of catalyst 1a, 10 mol % of D-camphorsulfonic acid as an additive and 1.5 mL of 1,2-dichloroethane at room temperature providing Michael adduct 4a in 93% yield and 99% ee.

Once the optimized reaction conditions have been found, the substrate scope was investigated by using different ketones 2a–c and isatylidenemalononitrile derivatives 3a–i (Scheme 2). The methodology was found to be suitable for both N-substituted isatylidenemalononitrile and N–H isatylidenemalononitrile derivatives. Acetone (2a) reacts well with various N–H isatylidenemalononitrile derivatives (3a–f) providing corresponding Michael adducts 4a–f in excellent enantioselectivity (96–99% ee) after a reaction time of 24–36 hours. The 5-fluoroisatylidenemalononitrile (3b), 5-chloroisatylidenemalononitrile (3c), 5-bromoisatylidenemalononitrile (3d) and 5-iodoisatylidenemalononitrile (3e) gave corresponding Michael adducts 4b–e in 93%, 95%, 94% and 91% yield and 98% ee, 98% ee, 98% ee and 99% ee, respectively (Table 5, entries 2–5). The reaction of 5,7-dibromoisatylidenemalononitrile (3f) with acetone (2a) proceeds with a high yield of 92% and a high enantioselectivity of 96% ee (Table 5, entry 6). The N-substituted isatylidenemalononitriles 3g–i react slowly with acetone (2a) to afford Michael adducts 4g–i in good yield (85–87%) and good enantioselectivity (88–92% ee) (Table 5, entries 7–9). Acetone (2a) reacts well with N-allyl isatylidenemalononitrile derivatives 3g and 3h to provide the respective Michael adducts 4g and 4h in 85% and 87% yield and 89% ee and 92% ee, respectively (Table 5, entries 7 and 8). Using N-benzyl isatylidenemalononitrile (3i), the Michael adduct 4i was isolated in 86% yield and 88% ee (Table 5, entry 9). A recently reported similar reaction catalyzed by Cinchona alkaloid-based primary amine catalyst requires high catalyst loading and is only suitable for N-unprotected isatylidenemalononitrile derivatives [5]. In contrast, our methodology is suitable for both N-unprotected and N-protected isatylidenemalononitrile derivatives and is highly enantioselective. Next, the substrate scope of the reaction was extended to different acyclic ketones 2b and 2c. Under the optimized conditions, the Michael addition of methyl isobutyl ketone (2b) and 2-octanone (2c) with 3a provided Michael adducts 4j and 4k in 24% and 41% yield and 97% ee and 96% ee, respectively (Table 5, entries 10 and 11). Due to the low reactivity of these ketones, these experiments were carried out after a slight modification of the optimized conditions, i.e., a higher catalyst loading of 20 mol % of 1a and 1.0 mL of 1,2-dichloroethane as a solvent. The 20 mol % of 1a catalyzes the Michael addition of 2b with 3a providing Michael adduct 4j in 80% yield and 96% ee after a reaction time of 7 days (Table 5, entry 12). The 2-octanone (2c) gave Michael adduct 4k in 85% yield and 97% ee after a reaction time of 6 days (Table 5, entry 13). The R absolute configuration of Michael adducts was assigned by comparing the HPLC chromatograms of Michael adducts with that reported in the literature [23,24].

Next, we studied the multicomponent version of this reaction. Acetone, isatin and malononitrile react in one pot providing Michael product 4a in a good yield of 80% and a high enantioselectivity of 98% ee (Scheme 3). The slightly lower yield of the one-pot process compared to the stepwise process was due to the competing reaction of isatin and acetone to form aldol adduct 5 (8% yield). The reaction involves the initial formation of isatylidenemalononitrile by Knoevenagel condensation of isatin with malononitrile followed by the addition of acetone to provide Michael adduct 4a. Thus, catalyst 1a also finds its successful application in the multicomponet version of this reaction without compromising enantioselectivity – albeit with a slight loss in yield.

We further demonstrated that Michael adducts could be transformed into spirooxindoles by following a simple strategy. The reduction of Michael adduct 4a with sodium borohydride in ethanol followed by spontaneous cyclization gave spirooxindole product 6 in 90% yield, 82:18 dr and 98% ee (Scheme 4).

Conclusion

In conclusion, we successfully demonstrated the use of the very simple primary-tertiary diamine catalyst 1a in combination with D-CSA as an additive for the enantioselective catalysis of a Michael addition of acetone to isatylidenemalononitriles. A three component process involving a domino Knoevenagel–Michael sequence was developed. 3,3'-Disubstituted oxindole could be transformed into spirooxindoles by reduction with NaBH₄.