A General Catalytic Enantioselective Transfer Hydrogenation Reaction

0 downloads 0 Views 4MB Size Report
Jul 30, 2016 - implemented with new disclosures and mechanistic insights results in a very general ... configurationally stable stereogenic center at the β-position of the nitro group. ... enzymes, guaranteeing access to both enantiomers of the reduced product, are known [8]. ..... Under these newly found conditions,.
molecules Article

A General Catalytic Enantioselective Transfer Hydrogenation Reaction of β,β-Disubstituted Nitroalkenes Promoted by a Simple Organocatalyst Luca Bernardi and Mariafrancesca Fochi Department of Industrial Chemistry “Toso Montanari”and INSTM RU Bologna, Alma Mater Studiorum, University of Bologna, V. Risorgimento 4, 40136 Bologna, Italy; [email protected] (L.B.); [email protected] (M.F.); Tel.: +39-051-209-3653 (L.B.); +39-051-209-3626 (M.F.) Academic Editors: Alessandro Palmieri and Derek J. McPhee Received: 28 June 2016; Accepted: 28 July 2016; Published: 30 July 2016

Abstract: Given its synthetic relevance, the catalytic enantioselective reduction of β,β-disubstituted nitroalkenes has received a great deal of attention. Several bio-, metal-, and organo-catalytic methods have been developed, which however are usually applicable to single classes of nitroalkene substrates. In this paper, we present an account of our previous work on this transformation, which implemented with new disclosures and mechanistic insights results in a very general protocol for nitroalkene reductions. The proposed methodology is characterized by (i) a remarkably broad scope encompassing various nitroalkene classes; (ii) Hantzsch esters as convenient (on a preparative scale) hydrogen surrogates; (iii) a simple and commercially available thiourea as catalyst; (iv) user-friendly procedures. Overall, the proposed protocol gives a practical dimension to the catalytic enantioselective reduction of β,β-disubstituted nitroalkenes, offering a useful and general platform for the preparation of nitroalkanes bearing a stereogenic center at the β-position in a highly enantioenriched form. A transition state model derived from control kinetic experiments combined with literature data is proposed and discussed. This model accounts and justifies the observed experimental results. Keywords: asymmetric catalysis; Hantzsch ester; H-bond; nitroalkanes; nitroalkenes; organocatalysis; reduction; thiourea; transfer hydrogenation

1. Introduction The enantioselective reduction of pro-chiral β,β-disubstituted nitroalkenes is a powerful synthetic transformation. It provides a straightforward access to optically active nitroalkanes carrying a configurationally stable stereogenic center at the β-position of the nitro group. These compounds can be easily converted to broadly useful chiral building blocks (e.g., enantioenriched β-chiral amines) exploiting the renowned synthetic versatility of the nitro moiety [1–3]. Accordingly, this reaction has received considerable attention from the synthetic chemistry community, leading to several efficient protocols based on different catalytic approaches and encompassing various nitroalkene substrates 1–4, providing access to enantioenriched nitroalkanes 6–9 (Scheme 1). “Ene”-reductases, mainly of the OYE (Old Yellow Enzyme) family, have been applied to the biocatalytic enantioselective reduction of β-alkyl-β-aryl nitroalkenes 1 [4–8]. In fact, α-methyl-β-nitrostyrene is a benchmark substrate for “ene”-reductases and stereocomplementary enzymes, guaranteeing access to both enantiomers of the reduced product, are known [8]. However, variations in the aryl and alkyl moieties of substrates 1 have only been partially addressed. Concurrently, enantioselective metal-catalyzed reductions of β,β-disubstituted nitroalkenes have been developed, employing different chiral complexes based on Cu, Rh, and Ir in hydrosilylation,

Molecules 2016, 21, 1000; doi:10.3390/molecules21081000

www.mdpi.com/journal/molecules

Molecules 2016, 21, 1000

2 of 14

transfer hydrogenation, and hydrogenation reactions [9–20]. Protocols featuring very broad substrate Molecules 2016, 21, 1000 2 of 14 scopes encompassing not only a variety of β-alkyl-β-aryl and β,β-dialkyl nitroalkenes 1 [9–16], but also β-amido [17–19] and and β-carbonyl [20] [20] counterparts 2–32–3 areareavailable. catalyzed their β-amido [17–19] β-carbonyl counterparts available.AALewis Lewis base base catalyzed reaction with with substrates substrates 22 has has also also been been reported reported [21]. [21]. hydrosilylation reaction

substrates 1–4 1–4 employed in catalytic enantioselective reductions, and Scheme 1. Nitroalkene Nitroalkene substrates employed in catalytic enantioselective reductions, corresponding nitroalkane products 6–9. 6–9. and corresponding nitroalkane products

An alternative alternative strategy strategy to to enantioselective enantioselective reduction attention, due to its its An reduction that that has has gained gained great great attention, due to elegance, simplicity, and practical practical features features (at (at least least on on aa preparative preparative scale), scale), is is the the organocatalytic organocatalytic elegance, simplicity, and transfer hydrogenation reaction with Hantzsch esters 10 as hydrogen surrogates [22–28]. this transfer hydrogenation reaction with Hantzsch esters 10 as hydrogen surrogates [22–28]. In In this context, exploring efficient coordination coordination of the nitro nitro group group by by double double hydrogen hydrogen context, exploring the the notion notion of of the the efficient of the bond donors [29–31], thioureas, and related catalysts 5a–f (Scheme 2) have been found to be efficient bond donors [29–31], thioureas, and related catalysts 5a–f (Scheme 2) have been found to be in the promotion of the enantioselective reduction of β,β-disubstituted nitroalkenes 1, 3, 4 with efficient in the promotion of the enantioselective reduction of β,β-disubstituted nitroalkenes 1, 3, Hantzsch esters 10 [32–40]. Catalyst 5a, a thiourea resulting from the combination of an N,N-diethyl 4 with Hantzsch esters 10 [32–40]. Catalyst 5a, a thiourea resulting from the combination of an tert-butylglycinamide residue on residue one side trans-1,2-cyclohexanediamine on the N,N-diethyl tert-butylglycinamide onand one an sideelaborated and an elaborated trans-1,2-cyclohexanediamine other, has been successfully employed in the reaction with nitroalkenes 1 and 3 [32,33]. The related on the other, has been successfully employed in the reaction with nitroalkenes 1 and 3 [32,33]. thiourea 5b, thiourea keeping 5b, the keeping same structural features of 5afeatures but bearing different The related the same structural of 5a abut bearingamino-acidic a different component (N,N-dimethylvalinamide), proved instead to be efficient in the of βamino-acidic component (N,N-dimethylvalinamide), proved instead to be efficient reduction in the reduction trifluoromethylnitroalkenes 4 [34]. The simpler structure 5c, wherein the cyclohexanediamine part of β-trifluoromethylnitroalkenes 4 [34]. The simpler structure 5c, wherein the cyclohexanediamine was replaced by the 3,5-bis(trifluoromethyl)phenyl group, and carrying an alcohol in place of the part was replaced by the 3,5-bis(trifluoromethyl)phenyl group, and carrying an alcohol in place amide moiety, was applied to substrates 1 and 3 giving results worse than 5a in terms of of the amide moiety, was applied to substrates 1 and 3 giving results worse than 5a in terms of enantioselectivities [35]. tested in in the the enantioselectivities [35]. The The unrelated unrelated paracyclophane paracyclophane catalysts catalysts 5d 5d and and 5e 5e were were also also tested reaction with a nitroalkene 1, but gave only low enantioselectivity [36]. Conversely, a chiral-at-metal reaction with a nitroalkene 1, but gave only low enantioselectivity [36]. Conversely, a chiral-at-metal pyrazole-amide double was applied applied very very successfully successfully to to the the transfer transfer pyrazole-amide double hydrogen hydrogen bond bond donor donor 5f 5f was hydrogenation reaction reaction with with nitroalkenes nitroalkenes 33 [37]. [37]. hydrogenation Herein, we report a general enantioselective transfer hydrogenation hydrogenation reaction reaction of of β,β-disubstituted β,β-disubstituted Herein, we report a general enantioselective transfer nitroalkenes with tert-butyl Hantzsch ester 10d, catalyzed by the simple and commercially available nitroalkenes with tert-butyl Hantzsch ester 10d, catalyzed by the simple and commercially available Jacobsen type type thiourea Jacobsen thiourea catalyst catalyst 5g 5g [41–43] [41–43] and and applicable applicable to to all all nitroalkene nitroalkene classes classes 1–4 1–4 (Scheme (Scheme 3). 3). Catalyst 5g, originally developed in the frame of the Strecker reaction, is considerably simpler to to Catalyst 5g, originally developed in the frame of the Strecker reaction, is considerably simpler synthesize than than 5a,b 5a,b or or 5f, 5f, the the previous previous most most successful successful structures, structures, that that moreover moreover were were only only applied applied synthesize to some nitroalkene classes (Scheme 2). In this paper, we first detail our previous development of the to some nitroalkene classes (Scheme 2). In this paper, we first detail our previous development of reaction with β-trifluoromethyl nitroalkenes 4 [39]. Then, providing experimental evidence on the the reaction with β-trifluoromethyl nitroalkenes 4 [39]. Then, providing experimental evidence on mode of action of of catalyst 5g5g ininthis to the mode of action catalyst thistransformation, transformation,we wedemonstrate demonstrateits its remarkable remarkable tolerance tolerance to variations in the substituents of the nitroalkene substrate (i.e., R1 and R2 in Scheme 3). By slightly varying reaction conditions, we show its application to nitroalkenes 2 [40], 1 and 3, ultimately providing a very general and broadly useful protocol for enantioselective transfer hydrogenation reactions of β,β-disubstituted nitroalkenes based on a readily available catalyst.

Molecules 2016, 21, 1000

3 of 14

variations in the substituents of the nitroalkene substrate (i.e., R1 and R2 in Scheme 3). By slightly varying reaction conditions, we show its application to nitroalkenes 2 [40], 1 and 3, ultimately providing a very general and broadly useful protocol for enantioselective transfer hydrogenation reactions of β,β-disubstituted nitroalkenes based on a readily available catalyst. Molecules 2016, 21,21, 1000 of 14 Molecules 2016, 1000 3 of3 14

Scheme Organocatalysts5a–f 5a–femployed employed in in transfer transfer hydrogenation hydrogenation reactions Scheme 2. 2. Organocatalysts reactionsofofβ,β-disubstituted β,β-disubstituted Scheme 2. Organocatalysts 5a–f employed transferscope. hydrogenation reactions of β,β-disubstituted nitroalkenes with Hantzschesters, esters, andtheir theirin substrate scope. nitroalkenes with Hantzsch and substrate nitroalkenes with Hantzsch esters, and their substrate scope.

Scheme 3. A general enantioselective transfer hydrogenation reaction of β,β-disubstituted nitroalkenes 1–4 with Hantzsch ester 10d catalyzed by the simple thiourea 5g. Scheme 3. transfer hydrogenation hydrogenation reaction reaction of of β,β-disubstituted β,β-disubstituted nitroalkenes Scheme 3. A A general general enantioselective enantioselective transfer nitroalkenes 1–4 with Hantzsch ester 10d catalyzed by the simple thiourea 5g. 1–4 with Hantzsch ester 10d catalyzed by the simple thiourea 5g. 2. Results and Discussion

2. Results and Discussion 2.1. Optimization and Development of the Transfer Hydrogenation Reaction with 2. Results and Discussion β-Trifluoromethylnitroalkenes 4 2.1. Optimization and Development of the Transfer Hydrogenation Reaction with 2.1. Optimization ofby thestudying Transfer the Hydrogenation Reaction withhydrogenation reaction of We startedand ourDevelopment investigation H-bond driven transfer β-Trifluoromethylnitroalkenes 4 β-Trifluoromethylnitroalkenes 4 β-trifluoromethyl nitroalkene 4a with Hantzsch ester 10a (Scheme 4) testing various typical H-bond We started our investigation by studying the H-bond driven transfer hydrogenation reaction of donor catalystsour (thioureas, ureas,by squaramides, diols, phosphoric etc.). We started investigation studying the H-bond drivenacids, transfer hydrogenation reaction of β-trifluoromethyl nitroalkene 4a with Hantzsch ester 10a (Scheme 4) testing various typical H-bond β-trifluoromethyl nitroalkene 4a with Hantzsch ester 10a (Scheme 4) testing various typical H-bond donor catalysts (thioureas, ureas, squaramides, diols, phosphoric acids, etc.). donor catalysts (thioureas, ureas, squaramides, diols, phosphoric acids, etc.). Even if most catalysts 5 were able to afford the desired β-trifluoromethyl nitroalkane 9a with good conversions (Table 1, entries 1–7) only the 1,2-diaminocyclohexane derived amido-thiourea 5o gave a promising enantioselectivity in the reaction (entry 8). Accordingly, variations in the two thiourea N-substituents were undertaken the reaction modularity this catalyst structure. Scheme 4. Test transferexploiting hydrogenation used of to identify an efficient catalyst 5. EvenScheme if most4.catalysts 5 were able to afford the desired Test transfer hydrogenation reaction used to β-trifluoromethyl identify an efficientnitroalkane catalyst 5. 9a with good conversions (Table 1, entries 1–7) only the 1,2-diaminocyclohexane derived amido-thiourea 5o gave a promising enantioselectivity into the reaction Accordingly, variations in the9atwo Even if most catalysts 5 were able afford the (entry desired8).β-trifluoromethyl nitroalkane with thiourea N-substituents were undertaken exploiting the modularity of this catalyst structure. good conversions (Table 1, entries 1–7) only the 1,2-diaminocyclohexane derived amido-thiourea 5o Catalyst 5p (entry 9) bearing the same 1,2-diaminocyclohexane moiety as 5o afforded the gave a promising enantioselectivity in the reaction (entry 8). Accordingly, variations in the two product with lower conversion and very poor and opposite selectivity, showing that enantioselectivity

thiourea N-substituents were undertaken exploiting the modularity of this catalyst structure. Catalyst 5p (entry 9) bearing the same 1,2-diaminocyclohexane moiety as 5o afforded the

2. Results and Discussion 2.1. Optimization and Development of the Transfer Hydrogenation Reaction with 4

β-Trifluoromethylnitroalkenes Molecules 2016, 21, 1000

4 of 14

We started our investigation by studying the H-bond driven transfer hydrogenation reaction of 1. β-trifluoromethyl nitroalkene 4a with Hantzsch esterscreening 10a (Scheme 4) testing various typical H-bond Table 1. Catalysts donor catalysts (thioureas, ureas, squaramides, diols, phosphoric acids, etc.). Molecules 2016, 21, 1000

4 of 14

is driven by the amide portion of the catalyst. Indeed, the simpler amido-thiourea catalyst 5g (entry 10) afforded results even better than 5o. 1. Table 1. Catalysts screening Scheme 4. 4. Test Test transfer transfer hydrogenation reaction used an Scheme hydrogenation reaction used to to identify identify an efficient efficient catalyst catalyst 5. 5.

Even if most catalysts 5 were able to afford the desired β-trifluoromethyl nitroalkane 9a with good conversions (Table 1, entries 1–7) only the 1,2-diaminocyclohexane derived amido-thiourea 5o gave a promising enantioselectivity in the reaction (entry 8). Accordingly, variations in the two thiourea N-substituents were undertaken exploiting the modularity of this catalyst structure. Catalyst 5p (entry 9) bearing the same 1,2-diaminocyclohexane moiety as 5o afforded the product with lower conversion and very poor and opposite selectivity, showing that enantioselectivity

Entry Entry

2 2 3 Catalyst Catalyst5 5 Conversion Conversion(%) (%) ee (%) ee 3 (%) 5h5h 97 97 +12 +12 1 1 5i5i 80 80 −9 ´9 2 2 3 3 5j5j >95>95 +20 +20 4 4 5k5k >95>95 +5 +5 5 5l 60 ´14 5l 5 60 −14 6 5m >95 ´17 5m5n >95 87 −17 ´24 7 6 5n5o 87 >95 −24 +48 8 7 9 8 5o5p >95 85 +48 ´15 10 5g >95 5p 9 85 −15 +77 1 Conditions: 4a (0.05 mmol), cat. 5 (10 mol %), toluene (0.5 mL, 0.1 M), 10a (0.06 mmol, 1.2 equiv), RT, 18–24 h; 5g 10 +77 2 19 3 >95

Determined on the crude mixture by

F-NMR analysis; Determined by chiral stationary phase HPLC.

Conditions: 4a (0.05 mmol), cat. 5 (10 mol %), toluene (0.5 mL, 0.1 M), 10a (0.06 mmol, 1.2 equiv), RT, 18–24 h; 2 Determined on the crude mixture by 19F-NMR analysis; 3 Determined by chiral stationary Catalyst 5p (entry 9) bearing the same 1,2-diaminocyclohexane moiety as 5o afforded the product phase HPLC. 1

with lower conversion and very poor and opposite selectivity, showing that enantioselectivity is driven Whereas the utility ofcatalyst. a 1,2-cyclohexanediamine catalystscatalyst (i.e., 5a, for the by the amide portion of the Indeed, the simpler derived amido-thiourea 5g Scheme (entry 10)2)afforded asymmetric of 5o. nitroalkenes with Hantzsch esters had been previously reported [32,33], the results even reactions better than notable outcome the simpler structure 5g was an derived unanticipated yet (i.e., gratifying result.2)Having Whereas the of utility of a 1,2-cyclohexanediamine catalysts 5a, Scheme for the identified catalyst 5g asofthe optimal, a with screening of reaction Hantzsch esters 10 was asymmetric reactions nitroalkenes Hantzsch estersconditions had been and previously reported [32,33], undertaken, more5g concentrated conditions and 40 °C, to ensure the notable performing outcome ofthe thereactions simplerunder structure was an unanticipated yetatgratifying result. full conversion (Table 2). The methyl, iso-butyl, and tert-butyl Hantzsch esters 10b–d were tested and Having identified catalyst 5g as the optimal, a screening of reaction conditions and Hantzsch esters 10 ˝ compared with 10a (entries 1–4). The tert-butyl 10d outclassed the other dihydropyridines was undertaken, performing the reactions underderivative more concentrated conditions and at 40 C, to ensure 10a–c in term of enantioinduction, affording 9a in 89% ee and for this reason it was selected for further exploration. The superior behavior the tert-butyl Hantzsch ester 10d compared to other esters, a common feature to many organocatalytic reactions, had been previously rationalized invoking not only increased nonbonding interactions between the Hantzsch ester ring and the catalyst-substrate complex, but also electronic factors related to the boat-like conformation of the tert-butyl Hantzsch

Molecules 2016, 21, 1000 Molecules 2016, 21, 1000

5 of 14 5 of 14

solubility in apolar solvents, to 10a–c.and Next, toluene Hantzsch was confirmed the most full conversion (Table 2). Thecompared methyl, iso-butyl, tert-butyl estersas10b–d wereappropriate tested and solvent after a short screening 4–7), and a satisfactory 94% ee value was dihydropyridines accomplished by compared with 10a (entries 1–4).(entries The tert-butyl derivative 10d outclassed the other cooling reaction mixture to −20 °C (entry 8).9a in 89% ee and for this reason it was selected for 10a–c inthe term of enantioinduction, affording A variation of the amide N-substituents catalystHantzsch 5 was then explored, by using the closely further exploration. The superior behavior the in tert-butyl ester 10d compared to other esters, related thioureas and 5r in the transfer hydrogenation at rationalized −20 °C in toluene. a common feature 5q to many organocatalytic reactions, had beenreaction, previously invokingThe not N,N-diethyl derivative 5q performed rather poorly, while the N-methyl-N-benzhydryl amide 5r did only increased nonbonding interactions between the Hantzsch ester ring and the catalyst-substrate not give any compared to 5g (entries 9–10). Theconformation use of α,α,α-trifluorotoluene solvent complex, butimprovement also electronic factors related to the boat-like of the tert-butyl as Hantzsch provided slightly better results than toluene,towhen catalyst 5gtert-butyl was employed (entry 11 vs.improved entry 8). ester 10d [23]. However, it is also important underline that ester 10d features Unluckily, a lower catalyst loading gave a small decrease in the enantiomeric excess of the product solubility in apolar solvents, compared to 10a–c. Next, toluene was confirmed as the most appropriate 9a (entryafter 12).a Reverting to 10 (entries mol % loading, better results by lowering solvent short screening 4–7), andslightly a satisfactory 94% ee were valueachieved was accomplished by ˝ reaction concentration (entry 13). cooling the reaction mixture to ´20 C (entry 8). Table 2. Reaction Reaction conditions conditions screening screening 11.. Table 2.

Entry 10 10 55 (mol %) Solvent (M) TT(°C) Conversion 22 (%) ee 33 (%) (mol %) Solvent (M) (˝ C) Conversion (%) ee (%) 10a 5g (10) 1 toluene (0.625) 40 >95 77 1 10a 5g (10) toluene (0.625) 40 >95 77 5g (10) 2 10b10b toluene (0.625) 40 >95 69 2 5g (10) toluene (0.625) 40 >95 69 5g (10) toluene 40 >95 70 3 3 10c10c 5g (10) toluene(0.625) (0.625) 40 >95 70 4 4 10d10d 5g (10) toluene(0.625) (0.625) 40 5g (10) toluene 40 >95 89 5 5 10d10d 5g (10) CH2Cl (0.13) 40 61 66 2 Cl22(0.13) 5g (10) CH 40 61 66 6 10d 5g (10) MTBE (0.13) 40 60 39 10d 5g (10) 6 MTBE (0.13) 40 60 39 7 10d 5g (10) THF (0.13) 40 60 4 5g (10) THF (0.13) 40 60 4 8 7 10d10d 5g (10) toluene (0.625) ´20 >95 94 5g (10) (10) toluene −20 >95 94 9 8 10d10d 5q toluene(0.625) (0.625) ´20 77 10 9 10d10d 5r (10) toluene(0.625) (0.625) ´20 >95 92 5q (10) toluene −20 >95 77 11 10 10d10d 5g (10) PhCF (0.625) ´20 >95 95 3 5r (10) toluene (0.625) −20 >95 92 12 10d 5g (5) PhCF3 (0.625) ´20 >95 92 5g (10) PhCF 3 (0.625) −20 >95 95 13 11 10d10d 5g (10) PhCF (0.3) ´20 >95 97 3 10d 5g (5) 12 PhCF3 (0.625) −20 >95 92 1 Conditions: 2 4a (0.05 mmol), cat. 5, solvent, 10 (0.06 mmol, 1.2 equiv), 18–24 h; Determined on the crude 3 Determined 10d analysis; 5g (10) 13 by 19 F-NMR PhCF (0.3)stationary −20 97 mixture by 3chiral phase HPLC. >95 Entry

Conditions: 4a (0.05 mmol), cat. 5, solvent, 10 (0.06 mmol, 1.2 equiv), 18–24 h; 2 Determined on the 19 analysis; 3 Determined by chiral phase HPLC. crude mixture A variation ofby the F-NMR amide N-substituents in catalyst 5 wasstationary then explored, by using the closely related

1

˝ in toluene. The N,N-diethyl thioureas and 5r in namely the transfer hydrogenation reaction, at ´20 These5q conditions, catalyst 5g (10 mol%) Hantzsch esterC10d, α,α,α-trifluorotoluene as derivative 5q performed rather poorly, while the N-methyl-N-benzhydryl amide 5r did not give any solvent, and −20 °C as reaction temperature, were applied on a preparative scale (Scheme 5) obtaining improvement compared to 5g (entries 9–10). The use of α,α,α-trifluorotoluene as solvent provided product 9a in 94% yield, with a small erosion in the enantioselectivity compared to the optimization slightly better(95% results than toluene, scale reaction instead of 97%). when catalyst 5g was employed (entry 11 vs. entry 8). Unluckily, a lower catalyst loading gave a small decrease in the enantiomeric excess of the product 9a (entry 12). Reverting to 10 mol % loading, slightly better results were achieved by lowering reaction concentration (entry 13). These conditions, namely catalyst 5g (10 mol%) Hantzsch ester 10d, α,α,α-trifluorotoluene as solvent, and ´20 ˝ C as reaction temperature, were applied on a preparative scale (Scheme 5) obtaining product 9a in 94% yield, with a small erosion in the enantioselectivity compared to the optimization scale reaction (95% instead of 97%). A study of the scope of the reaction was undertaken (Scheme 6). Similarly to derivative 4a, Scheme 5. Optimized reaction on preparative scale. the reactions with different substrates 4b–h bearing aromatic rings substituted at different positions with either electron donating or electron withdrawing groups furnished a series of β-aryl-β-trifluoro A study of the scope of the reaction was undertaken (Scheme 6). Similarly to derivative 4a, the reactions with different substrates 4b–h bearing aromatic rings substituted at different positions with

10d 5g (10) 6 MTBE (0.13) 40 60 39 10d 5g (10) 7 THF (0.13) 40 60 4 10d 5g (10) 8 toluene (0.625) −20 >95 94 10d 5q (10) 9 toluene (0.625) −20 >95 77 Molecules 2016, 21, 1000 6 of 14 10d 5r (10) 10 toluene (0.625) −20 >95 92 10d 5g (10) 11 PhCF3 (0.625) −20 >95 95 10d 5g (5) 12 PhCF3 (0.625) −20 >95 92 nitroalkanes 9b–h with excellent results (>70% yield and >93% ee). Very good yields and 10d 5g (10) 13 PhCF3 (0.3) −20 >95 97 enantioselectivities (93%–97% ee) can also be achieved when a 2-naphthyl and two heteroaromatic 1 Conditions: 4a (0.05 mmol), cat. 5, solvent, 10 (0.06 mmol, 1.2 equiv), 18–24 h; 2 Determined on the substituents are collocated in substrates 4i–k. Notably, the optimized reaction conditions could crude mixture by 19F-NMR analysis; 3 Determined by chiral stationary phase HPLC. also be applied successfully to substrates 4l and 4m bearing aliphatic chains, which furnished the expected 9l and 9m with very5g good results.Hantzsch The benefit using α,α,α-trifluorotoluene Theseadducts conditions, namely catalyst (10 mol%) esterof10d, α,α,α-trifluorotoluene as as reaction respect to toluenewere was applied demonstrated by performing few reactions using solvent, andmedium −20 °C aswith reaction temperature, on a preparative scale (Scheme 5) obtaining toluene solvent (Scheme results in brackets): in all the examined cases, thetoresults in terms of productas 9athe in 94% yield, with a6,small erosion in the enantioselectivity compared the optimization Molecules 2016, 21, 1000 6 of 14 enantioselectivity slightly lower. scale reaction (95%were instead of 97%). either electron donating or electron withdrawing groups furnished a series of β-aryl-β-trifluoro nitroalkanes 9b–h with excellent results (>70% yield and >93% ee). Very good yields and enantioselectivities (93%–97% ee) can also be achieved when a 2-naphthyl and two heteroaromatic substituents are collocated in substrates 4i–k. Notably, the optimized reaction conditions could also be applied successfully to substrates 4l and 4m bearing aliphatic chains, which furnished the expected adducts 9l and 9m with very good results. The benefit of using α,α,α-trifluorotoluene as reaction medium with respect to toluene was demonstrated by performing few reactions using toluene as the solvent (Scheme 6, results in brackets): in reaction all theon Scheme 5. Optimized onexamined preparativecases, scale. the results in terms of Scheme 5. Optimized reaction preparative scale. enantioselectivity were slightly lower. A study of the scope of the reaction was undertaken (Scheme 6). Similarly to derivative 4a, the reactions with different substrates 4b–h bearing aromatic rings substituted at different positions with

Scheme 6. 6. Scope of the the transfer transfer hydrogenation hydrogenation reaction reaction of of β-trifluoromethyl β-trifluoromethyl nitroalkenes nitroalkenes 4. 4. Scheme Scope of

2.2. Extension of the Transfer Hydrogenation Reaction Catalyzed by 5g to other Nitroalkenes 1–3 2.2. Extension of the Transfer Hydrogenation Reaction Catalyzed by 5g to Other Nitroalkenes 1–3 The relevant results obtained in the transfer hydrogenation reaction of β-trifluoromethyl The relevant results obtained in the transfer hydrogenation reaction of β-trifluoromethyl nitroalkenes 4 catalyzed by thiourea 5g prompted us to verify the applicability of the methodology nitroalkenes 4 catalyzed by thiourea 5g prompted us to verify the applicability of the methodology to other readily accessible β,β-disubstitued nitroolefins such us (E)-1-nitro-2-phenyl-1-propene 1a, βamino nitroolefines 2a and 2b, and β-nitroacrylate 3a (Figure 1).

Scheme 6. Scope of the transfer hydrogenation reaction of β-trifluoromethyl nitroalkenes 4.

2.2. Extension of the Transfer Hydrogenation Reaction Catalyzed by 5g to other Nitroalkenes 1–3 Molecules 2016, 21, 1000

7 of 14

The relevant results obtained in the transfer hydrogenation reaction of β-trifluoromethyl nitroalkenes 4 catalyzed by thiourea 5g prompted us to verify the applicability of the methodology other readily accessible β,β-disubstitued nitroolefinssuch suchusus(E)-1-nitro-2-phenyl-1-propene (E)-1-nitro-2-phenyl-1-propene 1a, 1a, βto to other readily accessible β,β-disubstitued nitroolefins β-amino nitroolefines 2a and 2b, and β-nitroacrylate 3a (Figure 1). amino nitroolefines 2a and 2b, and β-nitroacrylate 3a (Figure 1).

Molecules 2016, 21, 1000 Molecules 2016, 21, 1000

7 of 14

Figure 1–3studied. studied. Figure1. 1.Additional Additional nitroalkenes nitroalkenes 1–3

7 of 14

Substrates 1a and 3a had been previously studied and organocatalytic protocols based on Substrates 1a and 3a had been previously studied and organocatalytic protocols based on thioureas are present in the Scheme 2); on the contrary, prior to our work [40], β-amino Substrates 1a and 3aliterature had been(see previously studied organocatalytic on thioureas are present in the literature (see Scheme 2); on the and contrary, prior to our protocols work [40],based β-amino nitroolefines 2 were utilized only in asymmetric hydrosilylation [21] or in metal-catalyzed asymmetric thioureas are present in the literature (see Scheme 2); on the contrary, prior to our work [40], β-amino nitroolefines 2 were utilized only in asymmetric hydrosilylation [21] or in metal-catalyzed asymmetric hydrogenations [17–19]. nitroolefines 2 were utilized only in asymmetric hydrosilylation [21] or in metal-catalyzed asymmetric hydrogenations [17–19]. To our delight, after the reaction reactionconditions, conditions,thiourea thiourea5g5g was able promote hydrogenations [17–19]. To our delight, aftercarefully carefully adjusting adjusting the was able to to promote thethe transfer reaction of 2b in the thepresence presence Hantzsch ester 10d as hydrogen To ourhydrogenation delight, after carefully adjusting the conditions, thiourea 5g was10d able promote transfer hydrogenation reaction of 2a 2a and and 2breaction in ofofHantzsch ester asto hydrogen donor, affording the corresponding β-amino nitroalkanes 7a and 7b in 97% and 90% yield the transfer hydrogenation reactionβ-amino of 2a and 2b in the presence 10d as hydrogen donor, affording the corresponding nitroalkanes 7a and 7bofinHantzsch 97% and ester 90% yield respectively, respectively, with complete enantioselectivity (eenitroalkanes >99%) regardless of group the group donor, affording the corresponding β-amino 7a and 7b protecting in installed 97% and yield with complete enantioselectivity (ee >99%) regardless of the protecting on90% the installed amine respectively, with complete enantioselectivity (ee >99%) regardless of the protecting group installed onfunction the amine function (Scheme 7). Although a higher reaction temperature was required with (Scheme 7). Although a higher reaction temperature was required with these substratesthese 2 on the amine function 7). Althoughderivatives a higher reaction required these substrates 2 compared to(Scheme the trifluoromethyl 4, presumably duewas to stereoelectronic effects compared to the trifluoromethyl derivatives 4, presumably duetemperature to stereoelectronic effects with rendering substrates 2 compared the trifluoromethyl derivatives 4, presumably due toobserved stereoelectronic effects rendering amido substrates 2 less reactive 4, full enantioselectivity was even a lower amido substrates 2 lessto reactive than 4, fullthan enantioselectivity was observed even at a lower (5at mol %) rendering amido loading. substrates 2 less reactive than 4, full enantioselectivity was observed even at a lower (5 catalyst mol %) loading. catalyst (5 mol %) catalyst loading.

Scheme 7. Catalytic hydrogenationreactions reactions withsubstrates substrates 2a and 2b. Scheme Catalyticenantioselective enantioselective transfer transfer Scheme 7. 7. Catalytic enantioselective transfer hydrogenation hydrogenation reactions with with substrates 2a 2a and and 2b. 2b.

Eager further verify the capacity capacity of catalyst5g5g in inducing enantioselectivity inreaction the reaction Eagertoto enantioselectivity in in the of to further furtherverify verifythe the capacityofofcatalyst catalyst 5ginininducing inducing enantioselectivity the reaction of β,β-disubstitued β,β-disubstitued nitroolefin substrates with Hantzsch ester 10d, we moved to investigate nitroolefin substrates with Hantzsch ester 10d, we moved to investigate substrates 1a of β,β-disubstitued nitroolefin substrates with Hantzsch ester 10d, we moved to investigate substrates and 3a. of conditions (solvent, dilution,temperature, temperature, amount and 3a. A1ashort screening ofscreening reaction conditions (solvent, dilution, temperature, amount of Hantzsch substrates 1a and 3a.AAshort short screening of reaction reaction conditions (solvent, dilution, amount of ester Hantzsch ester indicated as asolvent a more morefor suitable solventfor for these substrates, 10d), indicated dichloromethane as a more suitable these substrates, and 0 and ´20 ˝ C of Hantzsch ester10d), 10d), indicated dichloromethane dichloromethane as suitable solvent these substrates, and 0optimal and −20 °C°Casasoptimal temperatures for 1a 1a and and 3a,respectively. respectively. Under these newly as reaction temperatures for 1a and 3a, respectively. Under these newlyUnder found conditions, and 0 and −20 optimalreaction reaction temperatures for 3a, these newly found conditions, catalyst 5g was indeed able to the corresponding nitroalkanes 6a and 8a catalyst 5g was indeed able provide the corresponding nitroalkanes 6a and 8a in high6ayields and found conditions, catalyst 5gto was indeed able to provide provide the corresponding nitroalkanes and 8a in in high yields and enantioselectivities, as depicted in Scheme 8. enantioselectivities, as depicted in Scheme 8. high yields and enantioselectivities, as depicted Scheme 8.

Scheme 8. 8. Catalytic enantioselective transfer hydrogenation hydrogenation reactions with with substrates 1a 1a and and 3a. 3a. Scheme Catalyticenantioselective enantioselective transfer transfer Scheme 8. Catalytic hydrogenationreactions reactions withsubstrates substrates 1a and 3a.

2.3. Proposed Reaction Model for the Transfer Hydrogenation Reactions Catalyzed by 5g with Nitroalkenes 2.3. Proposed Reaction Model for the Transfer Hydrogenation Reactions Catalyzed by 5g with Nitroalkenes Keeping the activation of nitro compounds by thioureas as the preliminary reasonable assumption, activation ofinsights nitro compounds by thioureas preliminary reasonable weKeeping aimed atthe gaining some on the mode of action as ofthe catalyst 5g in the reactions.assumption, We first weperformed aimed at some gaining some insights on the mode of action of catalyst 5g in the reactions. Wethe first simple control experiments, comparing in the reaction between 4a and 10d

Molecules 2016, 21, 1000

8 of 14

2.3. Proposed Reaction Model for the Transfer Hydrogenation Reactions Catalyzed by 5g with Nitroalkenes Keeping the activation of nitro compounds by thioureas as the preliminary reasonable assumption, we aimed at gaining some insights on the mode of action of catalyst 5g in the reactions. We first performed some simple control experiments, comparing in the reaction between 4a and 10d the activity 2016, of catalyst Molecules 21, 1000 5g vs. simpler achiral thiourea derivatives 5s and 5t (Scheme 9). 8 of 14 t-BuO 2C

Ph NO2

F 3C

+

CO2t-Bu

10d CF 3

S N

H Molecules 2016, 21, 1000 O

NO2 9a

CF3

N

Ph F 3C

N H

4a

Ph

5 (10 mol%) tol (0.125 M) 0 °C

CF3

CF 3

S N H

CF3

N H

5g

S N H

CF 3

F 3C

N H

N H

CF 3

8 of 14

5s

5t 5 (10 mol%) Ph tol (0.125 M) Scheme Control thiourea catalysts. Scheme 9. 9. Control experiments experiments using using0 different different thiourea catalysts. °C NO2 + NO2 F 3C F 3C N H 19 F-NMR evolution of 4athe reactions was 10d conveniently monitored 9aby Ph

t-BuO 2C

CO2t-Bu

The spectroscopy. Whereas thiourea 5s should 5g (pKa ca. 12.57 in DMSO [44]), 5t bearing CF3feature a similar acidity toCF CF3 CF 3 3 two aryl Ph substituents is considerably more acidic (pKa ca. 8.5 in DMSO [44]). Considering only the S S S coordination of the nitro group by the thiourea moiety acting as a general acid catalyst, and neglecting N N N CF conformational effects, an increase in acidity Nshould toNan increase inCFcatalyst 3 CF 3correspond F 3C N N roughly 3 H H H H H H activity [45].O However, as shown in Figure 2, catalyst 5g is considerably more efficient in reaction 5g 5s 5t promotion than 5s and 5t. Furthermore, the less acidic 5s was found to be slightly more effective Scheme 9. Control experiments using different thiourea catalysts. than 5t.

Figure 2. Evolution over time (19F-NMR) of the reactions performed with catalysts 5g, 5s, and 5t.

These reactions were found to obey pseudo-second order kinetics according to the following equation [46]: ln



t

(1)

where: [4a]0 = 0.125 M; [10d]0 = 0.1875 M; [9a] = (0.125

conversion) M

(2)

Figure 2. Evolution Evolution over time time of performed coefficients. with catalysts catalystsTheir 5g, 5s, 5s,slopes and 5t. 5t.are the 19F-NMR) Figure 2. over F-NMR) of the the reactions reactions performed with 5g, and The straight lines reported in((Figure 3 display good correlation rate constant kobs (in M−1·h−1) for the reactions, from which a more quantitative comparison of the These reactions were found to obey pseudo-second order kinetics according to the following activity displayed by were the three catalysts 5g,pseudo-second 5s, and 5t can be deduced. These reactions found to obey order kinetics according to the following equation [46]: equation [46]: r4as0 ˆ pr10ds0 ´ r9asq ln (1) lnp q “ pr10ds0 ´ r4as0 q ˆ kobs t t (1) r10ds0 ˆ pr4as0 ´ r9asq 19

where: where: r4as0 “ 0.125 M; r10ds0 “ 0.1875 M; r9as “ p0.125 ˆ conversionq M [4a]0 = 0.125 M; [10d]0 = 0.1875 M; [9a] = (0.125 conversion) M

(2) (2)

The straight lines reported in Figure 3 display good correlation coefficients. Their slopes are the rate constant kobs (in M−1·h−1) for the reactions, from which a more quantitative comparison of the activity displayed by the three catalysts 5g, 5s, and 5t can be deduced.

ln



t

(1)

where: Molecules 2016, 21, 1000

[4a]0 = 0.125 M; [10d]0 = 0.1875 M; [9a] = (0.125

conversion) M

9 of 14

(2)

The straight lines reported in Figure 3 display good correlation coefficients. Their slopes are the The straight inthe Figure 3 display good correlation slopes are −1·h−1) for rate constant kobs lines (in Mreported reactions, from which a more coefficients. quantitativeTheir comparison of the the ´1 ¨h´1 ) for the reactions, from which a more quantitative comparison of the rate constant k (in M obs by the three catalysts 5g, 5s, and 5t can be deduced. activity displayed activity displayed by the three catalysts 5g, 5s, and 5t can be deduced.

Figure 3. Pseudo-second order rate constants for the three catalysts 5g, 5s, and 5t. Figure 3. Pseudo-second order rate constants for the three catalysts 5g, 5s, and 5t.

Molecules 2016, 21, 1000

9 of 14

Together with with the thepreviously previously determined mode of action of catalyst 5g Strecker in the Strecker determined mode of action of catalyst 5g in the reaction reaction [41,42], these experiments themoiety amide moiety of this catalyst effectively participates [41,42], these experiments suggest suggest that thethat amide of this catalyst effectively participates in the in the coordination/stabilization of a reaction transition leading themajor majorenantiomer enantiomer of coordination/stabilization of a reaction transition statestate (TS)(TS) leading to tothe product 9. The pseudo-second order kinetics followed by the reactions (Figure 3) indicates as a first approximation areare notnot relevant (i.e., Curtin-Hammett control). Thus, it can approximation that thatinteractions interactionsprior priortotoTSTS relevant (i.e., Curtin-Hammett control). Thus, it be that, along with the coordination of the Lewis acidic thiourea to the negatively cansurmised be surmised that, along with the coordination of the Lewis acidic moiety thiourea moiety to the charged nitro moiety, the moiety, amide oxygen acts oxygen as a Lewis this TS, the N-H proton negatively charged nitro the amide actsbase as a in Lewis basecoordinating in this TS, coordinating the of theproton Hantzsch having a positive charge the previously determined most N-H of ester the Hantzsch ester having a density. positive Applying charge density. Applying the previously stable conformation of catalyst 5g [41], these considerations the model depicted 10. determined most stable conformation of catalyst 5g [41], result these in considerations result in Scheme the model Overall, dipolar TS is complementary functionalities depicted ain Scheme 10.effectively Overall, stabilized a dipolar by TSthe is electrostatically effectively stabilized by the electrostatically present in the catalyst structure,present in a way reminding mode ofinaction some macromolecular complementary functionalities in the catalystthe structure, a wayofreminding the mode of enzymes [47]. Amacromolecular coordination ofenzymes both dipole (negativeof atboth the nitro moiety and(negative positive at the action of some [47].charges A coordination dipole charges Hantzsch ester) might be more efficient than an exclusive of the negative of the nitro moiety and positive at the Hantzsch ester) mightcoordination be more efficient than anpart exclusive dipole (nitro moiety), justifying lower activity offered by “mono-functional” 5s and by 5t. coordination of the negative partthe of the dipole (nitro moiety), justifying the lower catalysts activity offered After TS, an irreversible proton-transfer leading to the products 9 and the pyridine “mono-functional” catalysts 5s and 5t. follows, After TS, an irreversible proton-transfer follows,co-product. leading to We previously using an α-substituted nitroalkene giving a pro-chiral nitronate the have products 9 anddemonstrated the pyridine[40], co-product. We have previously demonstrated [40], using an intermediate, that catalystgiving 5g is not able to exert significant stereocontrol in this5g proton-transfer α-substituted nitroalkene a pro-chiral nitronate intermediate, that catalyst is not able to process [48,49]. stereocontrol in this proton-transfer process [48,49]. exert significant

Scheme state model model for for the the reduction reduction of of nitroalkenes nitroalkenes 44 catalyzed catalyzed by by 5g. 5g. Scheme 10. 10. Transition Transition state

The soundness of this model is confirmed by computational work on the transfer hydrogenation reaction with the related catalyst 5b [34]. In line with the established mode of action of thiourea catalysts in this type of reactions [50], enantioinduction is not due to repulsive interactions between catalysts and substrates, but rather to the good 3D geometrical fit between the polar functionalities of the catalyst and a TS giving the (R)-enantiomer of 9. Apparently, alternative TSs leading to (S)-9 are not matching well the 3D structure of the catalyst, and cannot thus be stabilized/promoted. The

Molecules 2016, 21, 1000

10 of 14

The soundness of this model is confirmed by computational work on the transfer hydrogenation reaction with the related catalyst 5b [34]. In line with the established mode of action of thiourea catalysts in this type of reactions [50], enantioinduction is not due to repulsive interactions between catalysts and substrates, but rather to the good 3D geometrical fit between the polar functionalities of the catalyst and a TS giving the (R)-enantiomer of 9. Apparently, alternative TSs leading to (S)-9 are not matching well the 3D structure of the catalyst, and cannot thus be stabilized/promoted. The tert-butyl moiety serves to control the 3D conformation of the catalyst. Furthermore, catalyst/substrates interactions in this TS are limited to the nitro and N-H groups, the nitroalkene β-substituents do not play an obvious role. Indeed, the reaction with nitroalkenes 4 proved to be applicable with very good results to both aromatic and aliphatic substrates (R in Scheme 10). Besides, these considerations justify the remarkable performances of this catalyst with the different nitroalkenes 1–4, as well as the attack of the hydride to the same pro-chiral face [51] of the nitroolefins, irrespective of the nature of the β-substituents (Scheme 11).21, 1000 Molecules 2016, 10 of 14

Scheme face of of 1–4 1–4 according according to to the the TS TS model. model. Scheme 11. 11. Attack Attack of of the the hydride hydride to to the the same same pro-chiral pro-chiral face

3. Materials 3. Materialsand andMethods Methods Analytical grade commercially available reagents were were used as received, unless Analytical gradesolvents solventsand and commercially available reagents used as received, otherwise stated. CH 2Cl2 for the catalytic reactions was filtered on a plug of basic alumina before use. unless otherwise stated. CH2 Cl2 for the catalytic reactions was filtered on a plug of basic alumina before Hantzsch esters 10a–c [52,53] andand 10d10d [40] were Chiral use. Hantzsch esters 10a–c [52,53] [40] wereprepared preparedfollowing followingliterature literature procedures. procedures. Chiral thiourea 5g was prepared as reported in the literature, or purchased from commercial sources [41– thiourea 5g was prepared as reported in the literature, or purchased from commercial sources [41–43]. 43]. (E)-1-Nitro-2-phenyl-1-propene 1a [4], β-acylamino nitroolefins and β-t-butyloxycarbonylamino (E)-1-Nitro-2-phenyl-1-propene 1a [4], β-acylamino nitroolefins and β-t-butyloxycarbonylamino nitroolefins 22 [40], nitroolefins [40], ethyl ethyl Z-3-nitro-2-phenylacrylate Z-3-nitro-2-phenylacrylate3a 3a[33], [33],and andtrifluoromethylated trifluoromethylatednitroalkenes nitroalkenes4 [54], were prepared following literature procedures. 4 [54], were prepared following literature procedures. (S)-(1-Nitropropan-2-yl)benzene 6a: in a screw cap 1.5 mL vial, to a solution of (S)-(1-Nitropropan-2-yl)benzene 6a: in a screw cap 1.5 mL vial, to a solution of (E)-1-nitro-2-phenyl-1(E)-1-nitro-2-phenyl-1-propene 1a (24.5 mg, 0.15 mmol) in dichloromethane (0.3 mL, 0.5 M), catalyst propene 1a (24.5 mg, 0.15 mmol) in dichloromethane (0.3 mL, 0.5 M), catalyst 5g (7.6 mg, 0.015 mmol, 5g (7.6 mg, 0.015 mmol, 0.1 equiv) and Hantzsch ester 10d (69.5 mg, 0.225 mmol, 1.5 equiv) were 0.1 equiv) and Hantzsch˝ester 10d (69.5 mg, 0.225 mmol, 1.5 equiv) were sequentially added at 0 °C. sequentially added at 0 C. The vial was saturated with nitrogen and closed with the cap, then the The vial was saturated with nitrogen and closed with the cap, then the reaction mixture was stirred reaction mixture was stirred for 70 h at 0 ˝ C. The resulting mixture was directly purified by column for 70 h at 0 °C. The resulting mixture was directly purified by column chromatography eluting with chromatography eluting with diethyl ether/n-hexane 1:15 (v/v), to afford the title compound in diethyl ether/n-hexane 1:15 (v/v), to afford the title compound in quantitative yield (24.6 mg). The quantitative yield (24.6 mg). The enantiomeric excess of 6a was determined by CSP HPLC analysis enantiomeric excess of 6a was determined by CSP HPLC analysis (Chiralcel OJ-H, flow = 0.75 mL/min, (Chiralcel OJ-H, flow = 0.75 mL/min, eluent: n-hexane/i-PrOH 90:10, λ = 234 nm, tmaj = 20.2 min, eluent: n-hexane/i-PrOH 90:10, λ = 234 nm, tmaj = 20.2 min, tmin = 22.2 min). The obtained spectroscopic tmin = 22.2 min). The obtained spectroscopic data were in accord with those previously published [32]. data were in accord with those previously published [32].

Asymmetric transfer hydrogenation β-amino nitroalkanes 2: in a screw cap round bottom vial, to a stirred solution of 2 (0.15 mmol), toluene (510 μL, 0.3 M), catalyst 5g (3.9 mg, 0.0075 mmol, 0.05 equiv), and Hantzsch ester 10d (56 mg, 0.18 mmol, 1.2 equiv) were added. The vial was saturated with nitrogen and closed with the cap. The reaction mixture was stirred for 14 h at 40 °C. The resulting mixture was purified by column chromatography to afford product 7. The spectroscopic data have been

Molecules 2016, 21, 1000

11 of 14

Asymmetric transfer hydrogenation β-amino nitroalkanes 2: in a screw cap round bottom vial, to a stirred solution of 2 (0.15 mmol), toluene (510 µL, 0.3 M), catalyst 5g (3.9 mg, 0.0075 mmol, 0.05 equiv), and Hantzsch ester 10d (56 mg, 0.18 mmol, 1.2 equiv) were added. The vial was saturated with nitrogen and closed with the cap. The reaction mixture was stirred for 14 h at 40 ˝ C. The resulting mixture was purified by column chromatography to afford product 7. The spectroscopic data have been previously published [40]. (S)-N-(2-Nitro-1-phenylethyl)acetamide 7a: the title compound was prepared according to the above procedure and purified by silica gel chromatography (eluent ethyl acetate/petroleum ether 4:1). White solid; Yield 97%. The enantiomeric excess was determined by CSP HPLC analysis using a Daicel Chiralcel OJH column; n-hexane/2-propanol 9:1; 0.75 mL/min; λ = 254 nm; tR (major) = 35.3 min; tR (minor) = 42.8 min; ee >99%. tert-Butyl (S)-(2-nitro-1-phenylethyl)carbamate 7b: the title compound was prepared according to the above procedure and purified by silica gel chromatography (eluent CH2 Cl2 /petroleum ether 6:1). White solid; Yield 90%. The enantiomeric excess was determined by CSP HPLC analysis using a Daicel Chiralcel OJH column; n-hexane/2-propanol 9:1; 0.75 mL/min, λ = 254 nm; tR (major) = 22.6 min; tR (minor) = 25.7 min; ee >99%. Asymmetric transfer hydrogenation of β-trifluoromethyl nitroalkenes 4: in a screw cap round bottom vial, to a solution of nitroalkene 4 (0.15 mmol) in trifluorotoluene (0.5 mL, 0.3 M), catalyst 5g (7.6 mg, 0.015 mmol, 0.1 equiv), and Hantzsch ester 10d (56 mg, 0.18 mmol, 1.2 equiv) were added at ´20 ˝ C. The reaction mixture was stirred for 24 h at ´20 ˝ C. The resulting mixture was purified by column chromatography eluting with diethyl ether/n-hexane 2:50 (v/v), to afford the trifluoromethyl nitroalkane 9. (R)-1,1,1-Trifluoro-2-phenyl-3-nitropropane 9a: in a screw cap round bottom vial, to a solution of nitroalkene 4a (870 mg, 4.0 mmol) in trifluorotoluene (12 mL, 0.3 M), catalyst 5g (202 mg, 0.4 mmol, 0.1 equiv) and Hantzsch ester 10d (1.51 g, 4.8 mmol, 1.2 equiv) were added at ´20 ˝ C. The reaction mixture was stirred for 24 h at ´20 ˝ C. The resulting mixture was purified by column chromatography eluting with diethyl ether/n-hexane 2:50 (v/v), to afford the trifluoromethyl nitroalkane 9a. The obtained spectroscopic data were in accordance with those previously published [39]. Yield: 94% (823 mg); pale yellow oil; The enantiomeric excess was determined by CSP HPLC analysis using a Daicel Chiralcel OJH column; n-hexane/2-propanol 9:1; 0.75 mL/min; λ = 254 nm. tR (major) = 24.4 min; tR (minor) = 31.0 min; ee 95%. Kinetic studies on the asymmetric transfer hydrogenation of β-trifluoromethyl nitroalkene 4a: in a screw cap round bottom vial, to a solution of nitroalkene 4a (21.8 mg, 0.1 mmol), toluene (800 µL, 0.125 M), catalyst 5 (0.01 mmol, 0.1 equiv), and Hantzsch ester 10d (46.4 g, 0.15 mmol, 1.5 equiv) were added at 0 ˝ C. The mixture was stirred at the same temperature. The reaction evolution was followed by sampling the mixture with a Pasteur pipette, followed by immediate dilution in CDCl3 , followed by 19 F-NMR analysis: product 9a: ´69 ppm (d, J = 8.5 Hz, 3F); nitroalkene 4a: ´67 ppm (s, 3F). 4. Conclusions To summarize, we have presented a very general methodology for organocatalytic enantioselective transfer hydrogenation reactions of β,β-disubstituted nitroalkenes, making use of a simple thiourea catalyst, and Hantzsch esters as convenient hydrogen donors. It is the first time that a single catalyst can be used with all nitroalkene substrate classes; previous reports [32–37] were all focused on specific catalysts tailored for specific substrates. Some kinetic studies confirmed the generally accepted mode of action of this type of catalysts, allowing us to put forward a reasonable transition state model accounting for the remarkable substrate generality. We envision this synthetic platform might be useful for further synthetic applications and other studies related to thiourea based on catalytic asymmetric transformations [55]. Acknowledgments: We acknowledge financial support from the University of Bologna (RFO program). We are grateful to our co-workers involved in the previous development of the reaction with nitroalkenes 2 and 4, whose

Molecules 2016, 21, 1000

12 of 14

names can be found in the reference section, and to Giacomo Foli for preliminary results with nitroalkenes 1a and 3a. Takashi Ooi is gratefully acknowledged for suggesting to use of a thiourea like 5s in the kinetic studies. Author Contributions: M.F. conceived the project. L.B. and M.F. designed and performed the experiments, analyzed the data, and wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3. 4. 5.

6. 7.

8.

9. 10. 11. 12. 13. 14. 15.

16. 17. 18. 19.

20.

Ono, N. The Nitro Group in Organic Synthesis; John Wiley & Sons: New York, NY, USA, 2001. Ballini, R.; Gabrielli, S.; Palmieri, A.; Petrini, M. Nitroalkanes as key compounds for the synthesis of amino derivatives. Curr. Org. Chem. 2011, 15, 1482–1506. [CrossRef] Ballini, R.; Petrini, M. The nitro to carbonyl conversion (Nef reaction): New perspectives for a classical transformation. Adv. Synth. Catal. 2015, 357, 2371–2402. [CrossRef] Ohta, H.; Kobayashi, N.; Ozaki, K. Asymmetric reduction of nitro olefins by fermenting bakers’ yeast. J. Org. Chem. 1989, 54, 1802–1804. [CrossRef] Hall, M.; Stueckler, C.; Kroutil, W.; Macheroux, P.; Faber, K. Asymmetric bioreduction of activated alkenes using cloned 12-oxophytodienoate reductase isoenzymes OPR-1 and OPR-3 from lycopersicon esculentum (tomato): A striking change of stereoselectivity. Angew. Chem. Int. Ed. 2007, 46, 3934–3937. [CrossRef] [PubMed] Toogood, H.S.; Scrutton, N.S. New developments in ‘ene’-reductase catalyzed biological hydrogenations. Curr. Opin. Chem. Biol. 2014, 19, 107–115. [CrossRef] [PubMed] Fryszkowska, A.; Fisher, K.; Gardiner, J.M.; Stephens, G.M. Highly enantioselective reduction of β,β-disubstituted aromatic nitroalkenes catalyzed by clostridium sporogenes. J. Org. Chem. 2008, 73, 4295–4298. [CrossRef] [PubMed] Bertolotti, M.; Brenna, E.; Crotti, M.; Gatti, F.G.; Monti, D.; Parmeggiani, F.; Santangelo, S. Substrate scope evaluation of the enantioselective reduction of β-alkyl-β-arylnitroalkenes by old yellow enzymes 1–3 for organic synthesis applications. ChemCatChem 2016, 8, 577–583. [CrossRef] Czekelius, C.; Carreira, E.M. Catalytic enantioselective conjugate reduction of β,β-disubstituted nitroalkenes. Angew. Chem. Int. Ed. 2003, 42, 4793–4795. [CrossRef] [PubMed] Czekelius, C.; Carreira, E.M. Enantioselective reduction of nitroalkene mixtures by in situ equilibration. Org. Process Res. Dev. 2007, 11, 633–636. [CrossRef] Soltani, O.; Ariger, M.A.; Carreira, E.M. Transfer hydrogenation in water: Enantioselective, catalytic reduction of (E)-β,β-disubstituted nitroalkenes. Org. Lett. 2009, 11, 4196–4198. [CrossRef] [PubMed] Li, S.; Huang, K.; Zhang, J.; Wu, W.; Zhang, X. Rh-catalyzed highly enantioselective hydrogenation of nitroalkenes under basic conditions. Chem. Eur. J. 2013, 19, 10840–10844. [CrossRef] [PubMed] Zhao, Q.; Li, S.; Huang, K.; Wang, R.; Zhang, X. A novel chiral bisphosphine-thiourea ligand for asymmetric hydrogenation of β,β-disubstituted nitroalkenes. Org. Lett. 2013, 15, 4014–4017. [CrossRef] [PubMed] Yu, Y.-B.; Cheng, L.; Li, Y.-P.; Fu, Y.; Zhu, S.-F.; Zhou, Q.-L. Enantioselective iridium-catalyzed hydrogenation of β,β-disubstituted nitroalkenes. Chem. Commun. 2016, 52, 4812–4815. [CrossRef] [PubMed] Tang, Y.; Xiang, J.; Cun, L.; Wang, Y.; Zhu, J.; Liao, J.; Deng, J. Chemoselective and enantioselective transfer hydrogenation of β,β-disubstituted nitroalkenes catalyzed by a water-insoluble chiral diamine-rhodium complex in water. Tetrahedron Asymmetry 2010, 21, 1900–1905. [CrossRef] Li, S.; Huang, K.; Cao, B.; Zhang, J.; Wu, W.; Zhang, X. Highly enantioselective hydrogenation of β,β-disubstituted nitroalkenes. Angew. Chem. Int. Ed. 2012, 51, 8573–8576. [CrossRef] [PubMed] Li, P.; Zhou, M.; Zhao, Q.; Wu, W.; Hu, X.; Dong, X.-Q.; Zhang, X. Synthesis of chiral β-amino nitroalkanes via rhodium-catalyzed asymmetric hydrogenation. Org. Lett. 2016, 18, 40–43. [CrossRef] [PubMed] Yan, Q.; Liu, M.; Kong, D.; Zi, G.; Hou, G. Highly efficient iridium-catalyzed asymmetric hydrogenation of β-acylamino nitroolefins. Chem. Commun. 2014, 50, 12870–12872. [CrossRef] [PubMed] Zhou, M.; Dong, D.; Zhu, B.; Geng, H.; Wang, Y.; Zhang, X. Rhodium-catalyzed enantioselective hydrogenation of β-acylamino nitroolefins: A new approach to chiral β-amino nitroalkanes. Org. Lett. 2013, 15, 5524–5527. [CrossRef] [PubMed] Li, S.; Xiao, T.; Li, D.; Zhang, X. First iridium-catalyzed highly enantioselective hydrogenation of β-nitroacrylates. Org. Lett. 2015, 17, 3782–3785. [CrossRef] [PubMed]

Molecules 2016, 21, 1000

21. 22. 23. 24. 25. 26. 27.

28. 29. 30. 31. 32. 33. 34.

35.

36. 37.

38. 39.

40. 41.

42. 43. 44.

13 of 14

Liu, X.-W.; Yan, Y.; Wang, Y.-Q.; Wang, C.; Sun, J. Highly enantioselective reduction of β-amino nitroolefins with a simple N-sulfinyl urea as bifunctional catalyst. Chem. Eur. J. 2012, 18, 9204–9207. [CrossRef] [PubMed] Wang, D.; Astruc, D. The golden age of transfer hydrogenation. Chem. Rev. 2015, 115, 6621–6686. [CrossRef] [PubMed] Ouellet, S.G.; Walji, A.M.; MacMillan, D.W.C. Enantioselective organocatalytic transfer hydrogenation reactions using Hantzsch esters. Acc. Chem. Res. 2007, 40, 1327–1339. [CrossRef] [PubMed] Rossi, S.; Benaglia, M.; Massolo, E.; Raimondi, L. Organocatalytic strategies for enantioselective metal-free reductions. Catal. Sci. Technol. 2014, 4, 2708–2723. [CrossRef] Zheng, C.; You, S.-L. Transfer hydrogenation with Hantzsch esters and related organic hydride donors. Chem. Soc. Rev. 2012, 41, 2498–2518. [CrossRef] [PubMed] Herrera, R.P. Organocatalytic transfer hydrogenation and hydrosilylation reactions. Top. Curr. Chem. 2016, 374, 29. [CrossRef] Rueping, M.; Dufour, J.; Schoepke, S.J. Advances in catalytic metal-free reductions: From bio-inspired concepts to applications in the organocatalytic synthesis of pharmaceuticals and natural products. Green Chem. 2011, 13, 1084–1105. [CrossRef] Bernardi, L.; Fochi, M.; Comes Franchini, M.; Ricci, A. Bioinspired organocatalytic asymmetric reactions. Org. Biomol. Chem. 2012, 10, 2911–2922. [CrossRef] [PubMed] Taylor, M.S.; Jacobsen, E.N. Asymmetric catalysis by chiral hydrogen-bond donors. Angew. Chem. Int. Ed. 2006, 45, 1520–1543. [CrossRef] [PubMed] Miyabe, H.; Takemoto, Y. Discovery and application of asymmetric reaction by multi-functional thioureas. Bull. Chem. Soc. Jpn. 2008, 81, 785–795. [CrossRef] Zhang, Z.; Schreiner, P.R. (Thio)urea organocatalysis—What can be learnt from anion recognition? Chem. Soc. Rev. 2009, 38, 1187–1198. [CrossRef] [PubMed] Martin, N.J.A.; Ozores, L.; List, B. Organocatalytic asymmetric transfer hydrogenation of nitroolefins. J. Am. Chem. Soc. 2007, 129, 8976–8977. [CrossRef] [PubMed] Martin, N.J.A.; Cheng, X.; List, B. Organocatalytic asymmetric transfer hydrogenation of β-nitroacrylates: Accessing β2 -amino acids. J. Am. Chem. Soc. 2008, 130, 13862–13863. [CrossRef] [PubMed] Massolo, E.; Benaglia, M.; Orlandi, M.; Rossi, S.; Celentano, G. Enantioselective organocatalytic reduction of β-trifluoromethyl nitroalkenes: An efficient strategy for the synthesis of chiral β-trifluoromethyl amines. Chem. Eur. J. 2015, 21, 3589–3595. [CrossRef] [PubMed] Schneider, J.F.; Lauber, M.B.; Muhr, V.; Kratzer, D.; Paradies, J. Readily available hydrogen bond catalysts for the asymmetric transfer hydrogenation of nitroolefins. Org. Biomol. Chem. 2011, 9, 4323–4327. [CrossRef] [PubMed] Schneider, J.F.; Falk, F.C.; Fröhlich, R.; Paradies, J. Planar-chiral thioureas as hydrogen-bond catalysts. Eur. J. Org. Chem. 2010, 2265–2269. [CrossRef] Chen, L.-A.; Xu, W.; Huang, B.; Ma, J.; Wang, L.; Xi, J.; Harms, K.; Gong, L.; Meggers, E. Asymmetric catalysis with an inert chiral-at-metal iridium complex. J. Am. Chem. Soc. 2013, 135, 10598–10601. [CrossRef] [PubMed] Anderson, J.C.; Koovits, P.J. An enantioselective tandem reduction/nitro-Mannich reaction of nitroalkenes using a simple thiourea organocatalyst. Chem. Sci. 2013, 4, 2897–2901. [CrossRef] Martinelli, E.; Vicini, A.C.; Mancinelli, M.; Mazzanti, A.; Zani, P.; Bernardi, L.; Fochi, M. Catalytic highly enantioselective transfer hydrogenation of β-trifluoromethyl nitroalkenes. An easy and general entry to optically active β-trifluoromethyl amines. Chem. Commun. 2015, 51, 658–660. [CrossRef] [PubMed] Ferraro, A.; Bernardi, L.; Fochi, M. Organocatalytic enantioselective transfer hydrogenation of β-amino nitroolefins. Adv. Synth. Catal. 2016, 358, 1561–1565. [CrossRef] Zuend, S.J.; Jacobsen, E.N. Mechanism of amido-thiourea catalyzed enantioselective imine hydrocyanation: Transition state stabilization via multiple non-covalent interactions. J. Am. Chem. Soc. 2009, 131, 15358–15374. [CrossRef] [PubMed] Zuend, S.J.; Coughlin, M.P.; Lalonde, M.P.; Jacobsen, E.N. Scaleable catalytic asymmetric Strecker syntheses of unnatural α-amino acids. Nature 2009, 461, 968–971. [CrossRef] [PubMed] CAS 959979-30-7, Sigma-Aldrich code 693316. Jakab, G.; Tancon, C.; Zhang, Z.; Lippert, K.M.; Schreiner, P.R. (Thio)urea organocatalyst equilibrium acidities in DMSO. Org. Lett. 2012, 14, 1724–1727. [CrossRef] [PubMed]

Molecules 2016, 21, 1000

45. 46. 47. 48.

49.

50.

51. 52.

53. 54.

55.

14 of 14

Jensen, K.H.; Sigman, M.S. Systematically probing the effect of catalyst acidity in a hydrogen-bond-catalyzed enantioselective reaction. Angew. Chem. Int. Ed. 2007, 47, 4748–4750. [CrossRef] [PubMed] Cornish-Bowden, A. Fundamentals of Enzyme Kinetics; Portland Press: London, UK, 1995. Warshel, A.; Sharma, P.K.; Kato, M.; Xiang, Y.; Liu, H.; Olsson, M.H.M. Electrostatic basis for enzyme catalysis. Chem. Rev. 2006, 106, 3210–3235. [CrossRef] [PubMed] Romanini, S.; Galletti, E.; Caruana, L.; Mazzanti, A.; Himo, F.; Santoro, S.; Fochi, M.; Bernardi, L. Catalytic asymmetric reactions of 4-substituted indoles with nitroethene: A direct entry to Ergot alkaloid structures. Chem. Eur. J. 2015, 21, 17578–17582. [CrossRef] [PubMed] Liu, C.; Han, P.; Wu, X.; Tang, M. The mechanism investigation of chiral phosphoric acid-catalyzed Friedel-Crafts reactions—How the chiral phosphoric acid regains the proton. Comp. Theor. Chem. 2014, 1050, 39–45. [CrossRef] Knowles, R.R.; Jacobsen, E.N. Attractive noncovalent interactions in asymmetric catalysis: Links between enzymes and small molecule catalysts. Proc. Natl. Acad. Sci. USA 2010, 107, 20678–20685. [CrossRef] [PubMed] It should be noted that the Re-Si descriptors for the pro-chiral face of nitroalkenes 1–4 are not relevant to this discussion, since they are related to the CIP priorities of the substituents. Leutbecher, H.; Greiner, G.; Amann, R.; Stolz, A.; Beifuss, U.; Conrad, J. Laccase-catalyzed phenol oxidation. Rapid assignment of ring-proton deficient polycyclic benzofuran regioisomers by experimental 1 H-13 C long-range coupling contants and DFT-predicted product formation. Org. Biomol. Chem. 2011, 9, 2667–2673. [CrossRef] [PubMed] Roomi, M.W. Porphyria-inducing activity of a series of pyridine and dihydropyridine compounds. Investigation in a cell culture system. J. Med. Chem. 1975, 18, 457–460. [CrossRef] [PubMed] Gao, J.-R.; Wu, H.; Xiang, B.; Yu, W.-B.; Han, L.; Jia, Y.-X. Highly enantioselective construction of trifluoromethylated all-carbon quaternary stereocenters via nickel-catalyzed Friedel-Crafts alkylation reaction. J. Am. Chem. Soc. 2013, 135, 2983–2986. [CrossRef] [PubMed] Foli, G.; Sasso D’Elia, C.; Fochi, M.; Bernardi, L. Reversible modulation of the activity of thiourea catalysts with anions: A simple approach to switchable asymmetric catalysis. RSC Adv. 2016, 6, 66490–66494. [CrossRef]

Sample Availability: Samples of the compounds 7b and 9a are available from the authors. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).