Readily Available Chiral Benzimidazoles-Derived Guanidines as Organocatalysts in the Asymmetric α-Amination of 1,3-Dicarbonyl Compounds

The synthesis and the evaluation as organocatalysts of new chiral guanidines derived from benzimidazoles in the enantioselective α-amination of 1,3-dicarbonyl compounds using di-t-butylazodicarboxylate as aminating agent is herein disclosed. The catalysts are readily synthesized through the reaction of 2-chlorobezimidazole and a chiral amine in moderate-to-good yields. Among all of them, those derived from (R)-1-phenylethan-1-amine (1) and (S)-1-(2-naphthyl)ethan-1-amine (3) turned out to be the most efficient for such asymmetric transformation, rendering good-to-high yields and moderate-to-good enantioselectivities for the amination products.

In the last decade, our research group has been involved in the synthesis and application of benzimidazole-derivatives as bifunctional organocatalysts able to activate different 1,3-dicarbonyl compounds through hydrogen-bond interactions in different asymmetric organocatalyzed transformations [30][31][32][33][34]. Thus, although such benzimidazoles have recently demonstrated to be efficient catalysts for the electrophilic amination of dicarbonyl compounds [35], we decided to further extend the investigations in this difficult transformation by synthesizing new and readily available guanidines derived from benzimidazoles and evaluating their catalytic performance (Figure 1).

Results and Discussion
Firstly, a series of new different guanidine organocatalysts were synthesized following a straightforward methodology developed in our research group [30][31][32][33][34][35].Thus, 2-chlorobenzimidazole was reacted with the corresponding chiral amine in the presence of Et3N (3 equiv.) at 200 °C (pressure tube) for 24 h (Figure 2).Under these conditions, the corresponding products (1-10) were obtained in moderate-to-good yields.As expected, lower yields were achieved when bulky aliphatic amines were employed (5)(6)(7).For the synthesis of benzimidazole 10, the corresponding 2chloromehtylbenzimidazole was used instead.Once the guanidine derivatives were synthesized, their performance as organocatalysts in the asymmetric electrophilic amination of 1,3-dicarbonyl compounds was evaluated.In the search for optimal conditions and catalyst, and based on our previous experience in this transformation [35], the reaction between ethyl 2-oxocyclopenthanecarboxylate and di-t-butylazodicarboxylate in toluene

Results and Discussion
Firstly, a series of new different guanidine organocatalysts were synthesized following a straightforward methodology developed in our research group [30][31][32][33][34][35].Thus, 2-chlorobenzimidazole was reacted with the corresponding chiral amine in the presence of Et3N (3 equiv.) at 200 °C (pressure tube) for 24 h (Figure 2).Under these conditions, the corresponding products (1-10) were obtained in moderate-to-good yields.As expected, lower yields were achieved when bulky aliphatic amines were employed (5)(6)(7).For the synthesis of benzimidazole 10, the corresponding 2chloromehtylbenzimidazole was used instead.Once the guanidine derivatives were synthesized, their performance as organocatalysts in the asymmetric electrophilic amination of 1,3-dicarbonyl compounds was evaluated.In the search for optimal conditions and catalyst, and based on our previous experience in this transformation [35], the reaction between ethyl 2-oxocyclopenthanecarboxylate and di-t-butylazodicarboxylate in toluene Once the guanidine derivatives were synthesized, their performance as organocatalysts in the asymmetric electrophilic amination of 1,3-dicarbonyl compounds was evaluated.In the search for optimal conditions and catalyst, and based on our previous experience in this transformation [35], the reaction between ethyl 2-oxocyclopenthanecarboxylate and di-t-butylazodicarboxylate in toluene at 25 • C was chosen as a model reaction using 10 mol % of guanidine-derivative loading (Table 1).Thus, catalysts bearing an aromatic ring in the amine moiety (1-4) produced amination product 13a in high conversions and good-to-high enantioselectivities (Table 1, entries 1-4), whereas those bearing an aliphatic moiety (5-7) gave rise to 13a in low or moderate conversions and ee's (Table 1, entries 5-7).The influence of the hydrogens in the guanidine moiety was evaluated with catalyst 8 lacking one of them and bearing a tertiary amine part.As can be observed, although the results were quite good, the presence of an additional hydrogen-which could play an important role in the substrate activation-seems to be crucial in order to achieve high conversions and enantioselectivities (Table 1, entry 8).By the contrary, the introduction of an extra activation point by using an aminoalcohol moiety as in catalyst 9 turned out to be detrimental for the reaction results (Table 1, entry 9).Catalyst 10 lacks conjugation between nitrogens in the guanidine moiety, and gave rise to both low conversion and enantioselection for compound 13a.at 25 °C was chosen as a model reaction using 10 mol % of guanidine-derivative loading (Table 1).Thus, catalysts bearing an aromatic ring in the amine moiety (1-4) produced amination product 13a in high conversions and good-to-high enantioselectivities (Table 1, entries 1-4), whereas those bearing an aliphatic moiety (5-7) gave rise to 13a in low or moderate conversions and ee's (Table 1, entries 5-7).The influence of the hydrogens in the guanidine moiety was evaluated with catalyst 8 lacking one of them and bearing a tertiary amine part.As can be observed, although the results were quite good, the presence of an additional hydrogen-which could play an important role in the substrate activation-seems to be crucial in order to achieve high conversions and enantioselectivities (Table 1, entry 8).By the contrary, the introduction of an extra activation point by using an aminoalcohol moiety as in catalyst 9 turned out to be detrimental for the reaction results (Table 1, entry 9).Catalyst 10 lacks conjugation between nitrogens in the guanidine moiety, and gave rise to both low conversion and enantioselection for compound 13a.According to the catalyst screening results, we decided to further optimize the reaction conditions with guanidine 1, since it performed the best in terms of both conversion and enantioselectivity for compound 13a.In this regard, the influence of the solvent, temperature, and catalyst loading was assessed (Table 2).As can be observed, although good conversions were reached, none of the solvents tested produced an amelioration in the enantioselectivity (Table 2, entries 1-4).Then, once toluene was selected as the optimal solvent, lowering the reaction temperature was the next parameter assayed.Thus, it was found that while ee barely changed at 0 °C, the conversion dropped down to 87% (Table 2, entry 5).Finally, a decrease in the catalyst loading (5 mol %) was evaluated not only with catalyst 1 but also with 3, which also gave satisfactory results in the amination reaction.However, a slight drop in the enantioselectivity of the process was observed ( According to the catalyst screening results, we decided to further optimize the reaction conditions with guanidine 1, since it performed the best in terms of both conversion and enantioselectivity for compound 13a.In this regard, the influence of the solvent, temperature, and catalyst loading was assessed (Table 2).As can be observed, although good conversions were reached, none of the solvents tested produced an amelioration in the enantioselectivity (Table 2, entries 1-4).Then, once toluene was selected as the optimal solvent, lowering the reaction temperature was the next parameter assayed.Thus, it was found that while ee barely changed at 0 • C, the conversion dropped down to 87% (Table 2, entry 5).Finally, a decrease in the catalyst loading (5 mol %) was evaluated not only with catalyst 1 but also with 3, which also gave satisfactory results in the amination reaction.However, a slight drop in the enantioselectivity of the process was observed (Table 2, entries 6 and 7).The influence of the substituent in the diazocompounds was also tested.In this regard, diethyland dibenzylazodicarboxylate were screened, but both of them gave rise to the corresponding amination products in lower yields and ee's (see Supporting Information for details).
After having screened several parameters, the optimal reaction conditions chosen were the use of benzimidazole 1 (10 mol %) as catalyst, toluene as solvent, and 25 °C as temperature.In addition, since catalyst 3 performed similarly to 1, it was also selected for testing the substrate scope of the reaction (Figure 3).Thus, cyclic five-membered 1,3-dicarbonyl compounds were firstly evaluated.As mentioned, good results were achieved for 13a.However, when substrates 11b and 11c were tested, the corresponding products 13b and 13c were only isolated in moderate-to-good yields and enantioselectivities [36].Unfortunately, six-membered dicarbonyl compounds turned out to be unreactive under these conditions [36].Next, indanone-derived β-keto esters were assayed.As can be observed in Figure 2, increasing the steric bulk at the alkyl substituent of the ester moiety resulted in lower enantioselectivities and yields, obtaining the best enantioselection for methyl ester derivative 13f (60% ee).As somehow expected, the more-reactive 2-acetyl-1-indanone gave rise to the amination product 13j in moderate enantioselectivity, although in unexpectedly slightly lower yield than its keto ester analogue.Next, tetralone derivatives were taken into account.Contrary to the behavior observed with their non-benzocondensed analogues, very high enantioselectivities were achieved along with good yields.Again, 13k, bearing a methyl ester moiety, was superior in terms of ee in front of 13l (85% and 80% ee, respectively with catalyst 1).Lower yields and enantioselectivities were obtained for product 13m when the more reactive diketone was employed, being the drop even more accused in this case.Finally, open-chained 1.3-keto esters-which normally fail with most of the published catalysts [35]-were assayed.To our surprise, the use of ethyl benzoyl acetate gave rise to the corresponding amination product in good ee's, although with poor yields.Despite this, encouraged by the good enantioselectivities reached for 13n we decided to test another acyclic ketoester: methyl 3-oxo-2-methylbutanoate. Disappointingly, the reaction failed completely with both catalysts.It is important to mention that with the aim of obtaining better results a decrease in reaction temperature was assayed in the majority of the cases where compounds 13 were achieved with moderate ee's and good yields.However, with the exception of compound 13j, where the ee increased significantly, no substantial enhancement of optical purity was observed and, in addition, a drop in the yield was observed in most cases.The influence of the substituent in the diazocompounds was also tested.In this regard, diethyland dibenzylazodicarboxylate were screened, but both of them gave rise to the corresponding amination products in lower yields and ee's (see Supporting Information for details).
After having screened several parameters, the optimal reaction conditions chosen were the use of benzimidazole 1 (10 mol %) as catalyst, toluene as solvent, and 25 • C as temperature.In addition, since catalyst 3 performed similarly to 1, it was also selected for testing the substrate scope of the reaction (Figure 3).Thus, cyclic five-membered 1,3-dicarbonyl compounds were firstly evaluated.As mentioned, good results were achieved for 13a.However, when substrates 11b and 11c were tested, the corresponding products 13b and 13c were only isolated in moderate-to-good yields and enantioselectivities [36].Unfortunately, six-membered dicarbonyl compounds turned out to be unreactive under these conditions [36].Next, indanone-derived β-keto esters were assayed.As can be observed in Figure 2, increasing the steric bulk at the alkyl substituent of the ester moiety resulted in lower enantioselectivities and yields, obtaining the best enantioselection for methyl ester derivative 13f (60% ee).As somehow expected, the more-reactive 2-acetyl-1-indanone gave rise to the amination product 13j in moderate enantioselectivity, although in unexpectedly slightly lower yield than its keto ester analogue.Next, tetralone derivatives were taken into account.Contrary to the behavior observed with their non-benzocondensed analogues, very high enantioselectivities were achieved along with good yields.Again, 13k, bearing a methyl ester moiety, was superior in terms of ee in front of 13l (85% and 80% ee, respectively with catalyst 1).Lower yields and enantioselectivities were obtained for product 13m when the more reactive diketone was employed, being the drop even more accused in this case.Finally, open-chained 1.3-keto esters-which normally fail with most of the published catalysts [35]-were assayed.To our surprise, the use of ethyl benzoyl acetate gave rise to the corresponding amination product in good ee's, although with poor yields.Despite this, encouraged by the good enantioselectivities reached for 13n we decided to test another acyclic ketoester: methyl 3-oxo-2-methylbutanoate. Disappointingly, the reaction failed completely with both catalysts.It is important to mention that with the aim of obtaining better results a decrease in reaction temperature was assayed in the majority of the cases where compounds 13 were achieved with moderate ee's and good yields.However, with the exception of compound 13j, where the ee increased significantly, no substantial enhancement of optical purity was observed and, in addition, a drop in the yield was observed in most cases.In order to gain a deeper knowledge about the reaction mechanism, nonlinear effect experiments were conducted to determine whether one or more catalyst molecules are involved in the catalytic cycle [37].According to the observed results, it was determined that such a nonlinear effect did not occur in the studied reaction, pointing towards the involvement of a single catalyst molecule operating in the catalytic cycle (see Supporting Information for details about the experiments).Thus, a hypothetical model for the catalytic cycle is depicted in Figure 4, which is based on previous computational and experimental studies carried out in our research group employing benzimidazolederived catalysts in asymmetric conjugate additions.Thus, guanidine benzimidazole-derived (R)-1 could act as a bifunctional organocatalyst, acting initially as a base.Then, the 1,3-dicarbonyl compound enolate formed could be coordinated with the organocatalyst through hydrogen bonding, as depicted in intermediate A. A possible π-π stacking interaction would explain the good results observed in the benzocondensed β-keto esters, and could not be ruled out.Afterwards, the protonated guanidine group can activate the azodicarboxylate and hence facilitate the enantioselective attack of the enolate (intermediate B), releasing the corresponding amination product and regenerating (R)-1.It is important to remark that (R)-configured amination product was In order to gain a deeper knowledge about the reaction mechanism, nonlinear effect experiments were conducted to determine whether one or more catalyst molecules are involved in the catalytic cycle [37].According to the observed results, it was determined that such a nonlinear effect did not occur in the studied reaction, pointing towards the involvement of a single catalyst molecule operating in the catalytic cycle (see Supporting Information for details about the experiments).Thus, a hypothetical model for the catalytic cycle is depicted in Figure 4, which is based on previous computational and experimental studies carried out in our research group employing benzimidazole-derived catalysts in asymmetric conjugate additions.Thus, guanidine benzimidazole-derived (R)-1 could act as a bifunctional organocatalyst, acting initially as a base.Then, the 1,3-dicarbonyl compound enolate formed could be coordinated with the organocatalyst through hydrogen bonding, as depicted in intermediate A. A possible π-π stacking interaction would explain the good results observed in the benzocondensed β-keto esters, and could not be ruled out.Afterwards, the protonated guanidine group can activate the azodicarboxylate and hence facilitate the enantioselective attack of the enolate (intermediate B), releasing the corresponding amination product and regenerating (R)-1.It is important to remark that (R)-configured amination product was obtained when (R)-1 is employed.This assumption was taken by comparison of experimental evidence (chiral HPLC charts and optical rotation) with those previously reported in the literature for the same products.As expected, when (S)-3 was the catalyst employed, the opposite absolute configuration was observed in the amination product.evidence (chiral HPLC charts and optical rotation) with those previously reported in the literature for the same products.As expected, when (S)-3 was the catalyst employed, the opposite absolute configuration was observed in the amination product.

Materials and Methods
All reagents were purchased from commercial sources and used without further purification.Substrates that were not commercially available were synthesized according to known literature procedures [19,22,35].IRs were recorded on a JASCO FT-IR 4100 LE (Pike Miracle ATR) (Jasco Analítica Spain, Madrid, Spain), and only the structurally most relevant peaks are listed.NMR spectra were performed on a Bruker AC-300 or Bruker Avance-400 400 (Bruker Corporation, Billerica, MA, USA) using CDCl3 as solvent and tetramethylsilane (TMS) as internal standard unless otherwise stated.Low-resolution electron impact (EI) mass spectra were obtained at 70 eV on Agilent GC/MS-5973N apparatus equipped with a HP-5MS column (Agilent technologies, 30 m × 0.25 mm).Optical rotations were measured on a JASCO P-1030 Polarimeter (Jasco Analítica Spain, Madrid, Spain) with a 5 cm cell (c given in g/100 mL).Enantioselectivities were determined by HPLC analysis (Agilent 1100 Series HPLC) equipped with a G1315B diode array detector and a Quat Pump G1311A (Agilent Technologies, Santa Clara, CA, USA) equipped with the corresponding Daicel chiral column, and the retention time of the major enantiomer is highlighted in bold.Analytical thin layer chromatography (TLC) was performed on Merck silica gel plates and the spots visualized with UV light at 254 nm (Merck Millipore, Billerica, MA, USA).Flash chromatography employed Merck silica gel 60 (0.040-0.063 mm) (Merck Millipore).

General Procedure for the Synthesis of Benzimidazole-Based Organocatalysts 1-10
In a sealed pressure tube, a mixture of 2-chloro-1H-benzo[d]imidazole (0.5 mmol), the corresponding chiral amine 1 (2.5 equiv.), and Et3N (3 equiv.) was heated at 195-200 °C during 24 h.The reaction mixture was then allowed to reach room temperature.After the addition of H2O (5 mL), the mixture was extracted with CH2Cl2 (3 × 5 mL).The combined organic phases were dried over MgSO4 and evaporated under reduced pressure to give the corresponding crude products, which were purified by flash chromatography (Hexane/EtOAc/MeOH).For the synthesis of benzimidazole

Materials and Methods
All reagents were purchased from commercial sources and used without further purification.Substrates that were not commercially available were synthesized according to known literature procedures [19,22,35].IRs were recorded on a JASCO FT-IR 4100 LE (Pike Miracle ATR) (Jasco Analítica Spain, Madrid, Spain), and only the structurally most relevant peaks are listed.NMR spectra were performed on a Bruker AC-300 or Bruker Avance-400 400 (Bruker Corporation, Billerica, MA, USA) using CDCl 3 as solvent and tetramethylsilane (TMS) as internal standard unless otherwise stated.Low-resolution electron impact (EI) mass spectra were obtained at 70 eV on Agilent GC/MS-5973N apparatus equipped with a HP-5MS column (Agilent technologies, 30 m × 0.25 mm).Optical rotations were measured on a JASCO P-1030 Polarimeter (Jasco Analítica Spain, Madrid, Spain) with a 5 cm cell (c given in g/100 mL).Enantioselectivities were determined by HPLC analysis (Agilent 1100 Series HPLC) equipped with a G1315B diode array detector and a Quat Pump G1311A (Agilent Technologies, Santa Clara, CA, USA) equipped with the corresponding Daicel chiral column, and the retention time of the major enantiomer is highlighted in bold.Analytical thin layer chromatography (TLC) was performed on Merck silica gel plates and the spots visualized with UV light at 254 nm (Merck Millipore, Billerica, MA, USA).Flash chromatography employed Merck silica gel 60 (0.040-0.063 mm) (Merck Millipore).

General Procedure for the Asymmetric Amination of 1,3-Dicarbonyl Compounds
In an open-air tube at room temperature (25 • C) the corresponding 1,3-dicarbonyl compound (0.1 mmol) was added to a solution of organocatalyst (0.01 mmol, 10 mol %) in toluene (1 mL).After 5 min, di-tert-butylazodicarboxilate (0.11 mmol, 1.1 equiv.) was added in one portion and the reaction was then allowed to react for 15 h.After this time, water (5 mL) and ethyl acetate were added, and then the aqueous layer was re-extracted twice (2 × 10 mL).The combined organic phases were dried (MgSO 4 ) and evaporated under reduced pressure.Finally, the reaction crude was purified by column chromatography on silica gel or preparative TLC using hexane/ethyl acetate mixtures as eluent.The analytical data shown below correspond to those enantioenriched products (≥20% ee) as representative compounds.All the compounds are described in the literature.Therefore, only 1 H-NMR, MS (EI) and enantiomeric excess determination conditions are listed.

Conclusions
In summary, in this work we have described the synthesis and application of new chiral guanidines derived from benzimidazoles as organocatalysts for the asymmetric electrophilic α-amination of 1,3-dicarbonyl compounds employing di-t-butylazodicarboxylate as aminating agent.In general, the new catalysts are synthesized in good yields in a straightforward one-step reaction.In addition, the amination products were obtained in good yields and enantioselectivities varying from moderate to high when 10 mol % of catalysts 1 and 3 were employed.Although a more comprehensive study about the reaction mechanism is necessary, a bifunctional role of the catalyst is postulated.Supplementary Materials: General remarks and NMR copies of representative catalysts as well as 1 H-NMR and HPLC charts of amination compounds are provided.

Figure 3 .
Figure 3. Scope of the reaction.

Figure 3 .
Figure 3. Scope of the reaction.

Table 2 .
Optimization of the reaction parameters.a

Table 2 .
Optimization of the reaction parameters.a a Unless otherwise stated, general conditions were 11a (0.15 mmol), 12 (1.1 equiv.), 1 or 3 (5 or 10 mol %) in the corresponding solvent (1 mL) at the corresponding temperature for 15 h.b Conversions determined by 1 H-NMR from the crude reaction mixture.c Determined by chiral HPLC analysis (Daicel Chiralpak IA, see Experimental Section for details).