DOI: 10.1002/chem.201501044
Communication
& N-Heterocyclic Carbenes
Enantioselective Synthesis of Tertiary Propargylic Alcohols under N-Heterocyclic Carbene Catalysis Eduardo Snchez-Dez, Maitane Fernndez, Uxue Uria, Efraim Reyes,* Luisa Carrillo, and Jose L. Vicario*[a] Abstract: A straightforward procedure to carry out the enantioselective benzoin reaction between aldehydes and ynones by employing a chiral N-heterocyclic carbene (NHC) as catalyst was developed. Under the optimized reaction conditions, these ynones undergo a clean and selective 1,2-addition with the catalytically generated Breslow intermediate, not observing any byproduct arising from competitive Stetter-type reactivity. This procedure allows the preparation of tertiary alkynyl carbinols as highly enantioenriched materials, which have the remarkable potential to be used as chiral building blocks in organic synthesis.
Enantioenriched tertiary propargylic alcohols incorporating a carbonyl moiety at the quaternary stereocenter are a particularly valuable class of chiral building blocks in organic synthesis, which are widely employed in the preparation of several bioactive molecules and natural products.[1] The classical strategy used for building up the propargyl alcohol scaffold in a stereocontrolled fashion relies on the nucleophilic addition of acetylides to aldehydes or ketones directed by a chiral metal catalyst.[2] In the case of adducts containing a tertiary alcohol next to the carbonyl substituent, the procedure would require the use of an a-keto carbonyl compound (typically an a-keto ester) as the electrophilic counterpart (Scheme 1, path a). As an alternative method to access this particular scaffold, we herein propose the use of a catalytic enantioselective benzoin reaction between an aldehyde and an alkynone through the known ability of N-heterocyclic carbenes (NHC) to activate aldehydes, enabling their synthetic use as umpoled acyl anion equivalents (Scheme 1, path b).[3] The benzoin reaction,[4] a classic transformation carried out under NHC catalysis, has experienced impressive advances since the initial works on catalytic enantioselective versions focusing on the self-condensation of aldehydes.[5] Most recent progress in the field even shows the
Scheme 1. Access to a-hydroxycarbonyl compounds containing a tertiary propargyl carbinol moiety.
possibility of carrying out the enantioselective cross-benzoin reaction between two different aldehydes.[6] However, the possibility of using ketones as the electrophilic counterpart in enantioselective cross-benzoin reactions is still very limited. The majority of the work in this area is focused on intramolecular variants, with several reports showing highly efficient protocols that proceed in good yield and enantioselectivities.[7] Nevertheless, the only examples described to date of intermolecular cross-benzoin reactions between an aldehyde and a ketone comprise the use of highly electrophilic ketones, such as a-ketoesters or trifluoromethyl ketones.[8, 9] However, and despite these advances, enantioselectivities typically range around 70–80 % ee, with only a few cases showing enantiomeric excesses higher than 90 %. The proposed transformation presents important challenges that have to be faced: on the one hand, the NHC-catalyzed enantioselective cross-benzoin reaction shows an intrinsic chemoselectivity issue, arising from the tendency of aldehydes to undergo self-condensation. On the other hand, the bidentate electrophilic nature of alkynones entails that the activated Breslow intermediate could react with the ynone either in a 1,2fashion, yielding the desired benzoin product,[10] or through a competitive 1,4-addition pathway, which would eventually lead to the formation of Stetter-type products (Scheme 1).[11] In this regard, optimized conditions have to be found that allow to control these chemo- and regioselectivity challenges, in ad-
[a] E. Snchez-Dez, Dr. M. Fernndez, Dr. U. Uria, Prof. E. Reyes, Prof. L. Carrillo, Prof. J. L. Vicario Department of Organic Chemistry II University of the Basque Country (UPV/EHU) P.O. Box 644, 48080 Bilbao (Spain) Fax: (+ 34) 94-601-2748 E-mail: [email protected] [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201501044. Chem. Eur. J. 2015, 21, 1 – 6
These are not the final page numbers! ÞÞ
1
2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Communication highest level of enantiocontrol.[13] Proline-derived precatalyst 3 d was only poorly active in this case (entry 4), while the related derivative 3 e performed slightly better to afford the product in still a low yield, but with a moderate level of enantiocontrol (entry 5). A similar behavior was noted when triazolium salt 3 f derived from cis-1-amino indanol was employed (entry 6). However, when varying the N-aryl substituent of this catalyst from pentafluorophenyl to a mesityl group (precatalyst 3 g), a-hydroxyketone 4 a was isolated in excellent 94 % ee, albeit still with a rather low isolated yield and together with significant amounts of the self-benzoin byproduct (entry 7). We also tested the performance of precatalyst 3 h in which the methyl groups of the N-aryl substituent are replaced with chlorine substituents (entry 8); however, this transformation furnished the desired product with a slightly lower enantioselectivity. Once 3 g had been identified as the most efficient precatalyst in terms of enantiocontrol, we proceeded to optimize other parameters of the reaction to improve the yield of this initial result, mainly trying to avoid the occurrence of the selfbenzoin side reaction. First, we evaluated different solvents (entries 9–11), observing that lower yields were obtained when the reaction was run in other more polar solvents (entries 9 and 10). Benzene, on the other hand, offered a very similar result to toluene in terms of yield, without affecting the enantioselectivity; importantly, NMR analysis of the crude reaction mixture showed that the formation of the self-benzoin byproduct was minimized in this case (entry 11). This observation prompted us to test the reaction at lower temperatures in benzene, with the aim to obtain a highly chemoselective transformation. We found that lowering the temperature to 15 8C resulted in a completely chemo- and highly enantioselective reaction (entry 12) for which an excellent yield could also be achieved by slightly increasing the catalyst loading (entry 13). Having established a robust experimental protocol for the reaction, we next proceeded to explore the scope of the reaction and its performance for the preparation of differently substituted tertiary carbinols 4. As shown in Scheme 2, the reaction showed a good tolerance to modifications at the acetylenic position of the ynone reagent, both aryl and alkyl substituents can be introduced (compounds 4 a–e). Next, we explored the performance towards modifications of the aldehyde donor in the reaction with ynone 2 a; in most of the cases, the corresponding a-hydroxyketones (4 f–j) were obtained in good yields and up to 99 % ee. The reaction also proceeded well when different ynone acceptors were tested (compounds 4 k– l). In all cases, the formation of self-benzoin or Stetter-type byproducts was never observed. Moreover, the reaction proved to be even wider in scope, showing that less electrophilic alkynones were also suitable substrates in this transformation, which is in deep contrast to previous reports in which the benzoin reaction under NHC-catalysis involving ketones as the acceptor always required highly electrophilic substrates. This is exemplified by the synthesis of compounds 4 m–q¸ which were obtained cleanly in good yields and high enantioselectivities. Using more challenging substrates like propanal or butanal as the donor (adducts 4 s and 4 t) or an ynone with a bulkier ethyl substituent at the carbonyl group (adduct 4 r) selectively
dition to the standard stereochemical issues that have to be faced if an enantioselective process is also desired. With all these aspects under consideration, we started our work by surveying the viability of the proposed transformation, taking the reaction between aldehyde 1 a and trifluoromethyl alkynone 2 a as a useful working model (Table 1). First, we
Table 1. Screening for the best experimental conditions.[a]
Entry
Catalyst
Base
Solvent
1 2 3 4 5 6 7 8 9 10 11 12 13[f]
3a 3b 3c 3d 3e 3f 3g 3h 3g 3g 3g 3g 3g
DBU DBU DBU K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3
toluene toluene toluene toluene toluene toluene toluene toluene CH2Cl2 THF benzene benzene[e] benzene[e]
T [8C] RT RT RT RT RT RT RT RT RT RT RT 15 15
Yield [%][b]
ee [%][c]
16
Communication
& N-Heterocyclic Carbenes
Enantioselective Synthesis of Tertiary Propargylic Alcohols under N-Heterocyclic Carbene Catalysis Eduardo Snchez-Dez, Maitane Fernndez, Uxue Uria, Efraim Reyes,* Luisa Carrillo, and Jose L. Vicario*[a] Abstract: A straightforward procedure to carry out the enantioselective benzoin reaction between aldehydes and ynones by employing a chiral N-heterocyclic carbene (NHC) as catalyst was developed. Under the optimized reaction conditions, these ynones undergo a clean and selective 1,2-addition with the catalytically generated Breslow intermediate, not observing any byproduct arising from competitive Stetter-type reactivity. This procedure allows the preparation of tertiary alkynyl carbinols as highly enantioenriched materials, which have the remarkable potential to be used as chiral building blocks in organic synthesis.
Enantioenriched tertiary propargylic alcohols incorporating a carbonyl moiety at the quaternary stereocenter are a particularly valuable class of chiral building blocks in organic synthesis, which are widely employed in the preparation of several bioactive molecules and natural products.[1] The classical strategy used for building up the propargyl alcohol scaffold in a stereocontrolled fashion relies on the nucleophilic addition of acetylides to aldehydes or ketones directed by a chiral metal catalyst.[2] In the case of adducts containing a tertiary alcohol next to the carbonyl substituent, the procedure would require the use of an a-keto carbonyl compound (typically an a-keto ester) as the electrophilic counterpart (Scheme 1, path a). As an alternative method to access this particular scaffold, we herein propose the use of a catalytic enantioselective benzoin reaction between an aldehyde and an alkynone through the known ability of N-heterocyclic carbenes (NHC) to activate aldehydes, enabling their synthetic use as umpoled acyl anion equivalents (Scheme 1, path b).[3] The benzoin reaction,[4] a classic transformation carried out under NHC catalysis, has experienced impressive advances since the initial works on catalytic enantioselective versions focusing on the self-condensation of aldehydes.[5] Most recent progress in the field even shows the
Scheme 1. Access to a-hydroxycarbonyl compounds containing a tertiary propargyl carbinol moiety.
possibility of carrying out the enantioselective cross-benzoin reaction between two different aldehydes.[6] However, the possibility of using ketones as the electrophilic counterpart in enantioselective cross-benzoin reactions is still very limited. The majority of the work in this area is focused on intramolecular variants, with several reports showing highly efficient protocols that proceed in good yield and enantioselectivities.[7] Nevertheless, the only examples described to date of intermolecular cross-benzoin reactions between an aldehyde and a ketone comprise the use of highly electrophilic ketones, such as a-ketoesters or trifluoromethyl ketones.[8, 9] However, and despite these advances, enantioselectivities typically range around 70–80 % ee, with only a few cases showing enantiomeric excesses higher than 90 %. The proposed transformation presents important challenges that have to be faced: on the one hand, the NHC-catalyzed enantioselective cross-benzoin reaction shows an intrinsic chemoselectivity issue, arising from the tendency of aldehydes to undergo self-condensation. On the other hand, the bidentate electrophilic nature of alkynones entails that the activated Breslow intermediate could react with the ynone either in a 1,2fashion, yielding the desired benzoin product,[10] or through a competitive 1,4-addition pathway, which would eventually lead to the formation of Stetter-type products (Scheme 1).[11] In this regard, optimized conditions have to be found that allow to control these chemo- and regioselectivity challenges, in ad-
[a] E. Snchez-Dez, Dr. M. Fernndez, Dr. U. Uria, Prof. E. Reyes, Prof. L. Carrillo, Prof. J. L. Vicario Department of Organic Chemistry II University of the Basque Country (UPV/EHU) P.O. Box 644, 48080 Bilbao (Spain) Fax: (+ 34) 94-601-2748 E-mail: [email protected] [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201501044. Chem. Eur. J. 2015, 21, 1 – 6
These are not the final page numbers! ÞÞ
1
2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Communication highest level of enantiocontrol.[13] Proline-derived precatalyst 3 d was only poorly active in this case (entry 4), while the related derivative 3 e performed slightly better to afford the product in still a low yield, but with a moderate level of enantiocontrol (entry 5). A similar behavior was noted when triazolium salt 3 f derived from cis-1-amino indanol was employed (entry 6). However, when varying the N-aryl substituent of this catalyst from pentafluorophenyl to a mesityl group (precatalyst 3 g), a-hydroxyketone 4 a was isolated in excellent 94 % ee, albeit still with a rather low isolated yield and together with significant amounts of the self-benzoin byproduct (entry 7). We also tested the performance of precatalyst 3 h in which the methyl groups of the N-aryl substituent are replaced with chlorine substituents (entry 8); however, this transformation furnished the desired product with a slightly lower enantioselectivity. Once 3 g had been identified as the most efficient precatalyst in terms of enantiocontrol, we proceeded to optimize other parameters of the reaction to improve the yield of this initial result, mainly trying to avoid the occurrence of the selfbenzoin side reaction. First, we evaluated different solvents (entries 9–11), observing that lower yields were obtained when the reaction was run in other more polar solvents (entries 9 and 10). Benzene, on the other hand, offered a very similar result to toluene in terms of yield, without affecting the enantioselectivity; importantly, NMR analysis of the crude reaction mixture showed that the formation of the self-benzoin byproduct was minimized in this case (entry 11). This observation prompted us to test the reaction at lower temperatures in benzene, with the aim to obtain a highly chemoselective transformation. We found that lowering the temperature to 15 8C resulted in a completely chemo- and highly enantioselective reaction (entry 12) for which an excellent yield could also be achieved by slightly increasing the catalyst loading (entry 13). Having established a robust experimental protocol for the reaction, we next proceeded to explore the scope of the reaction and its performance for the preparation of differently substituted tertiary carbinols 4. As shown in Scheme 2, the reaction showed a good tolerance to modifications at the acetylenic position of the ynone reagent, both aryl and alkyl substituents can be introduced (compounds 4 a–e). Next, we explored the performance towards modifications of the aldehyde donor in the reaction with ynone 2 a; in most of the cases, the corresponding a-hydroxyketones (4 f–j) were obtained in good yields and up to 99 % ee. The reaction also proceeded well when different ynone acceptors were tested (compounds 4 k– l). In all cases, the formation of self-benzoin or Stetter-type byproducts was never observed. Moreover, the reaction proved to be even wider in scope, showing that less electrophilic alkynones were also suitable substrates in this transformation, which is in deep contrast to previous reports in which the benzoin reaction under NHC-catalysis involving ketones as the acceptor always required highly electrophilic substrates. This is exemplified by the synthesis of compounds 4 m–q¸ which were obtained cleanly in good yields and high enantioselectivities. Using more challenging substrates like propanal or butanal as the donor (adducts 4 s and 4 t) or an ynone with a bulkier ethyl substituent at the carbonyl group (adduct 4 r) selectively
dition to the standard stereochemical issues that have to be faced if an enantioselective process is also desired. With all these aspects under consideration, we started our work by surveying the viability of the proposed transformation, taking the reaction between aldehyde 1 a and trifluoromethyl alkynone 2 a as a useful working model (Table 1). First, we
Table 1. Screening for the best experimental conditions.[a]
Entry
Catalyst
Base
Solvent
1 2 3 4 5 6 7 8 9 10 11 12 13[f]
3a 3b 3c 3d 3e 3f 3g 3h 3g 3g 3g 3g 3g
DBU DBU DBU K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3
toluene toluene toluene toluene toluene toluene toluene toluene CH2Cl2 THF benzene benzene[e] benzene[e]
T [8C] RT RT RT RT RT RT RT RT RT RT RT 15 15
Yield [%][b]
ee [%][c]
16
