Abstract
A series of single enantiomer, 2,4-cis-disubstituted amino azetidines were synthesised and used as ligands for copper-catalysed Henry reactions of aldehydes with nitromethane. Optimisation of ligand substituents and the reaction conditions was conducted. The enantiomeric excess of the formed products was highest when alkyl aldehydes were employed in the reaction (>99% e.e.). The absolute stereochemistry of one representative azetidine derivative salt was determined by analysis of the Flack parameter of an XRD single crystal structure. The origin of selectivity in catalysis was investigated computationally, revealing the importance of the amino-substituent in determining the stereochemical outcome. A racemic platinum complex of a cis-disubstituted azetidine is examined by XRD single crystal structure analysis with reference to its steric parameters, and analogies to the computationally determined copper complex catalyst are drawn. A preliminary example of the use of a cis-disubstituted azetidine scaffold in thiourea H-bonding catalyst is noted in the supporting information.
Similar content being viewed by others
Introduction
Enzymes are often excellent catalysts that are able to achieve very high levels of stereoselectivity. One property of enzymes that has been implicated in their capacity to deliver highly stereoselective reaction outcomes is that their active sites are concave, well-defined, cavities1. When flexible arrays of ligands are arranged around a metal centre they can be constrained into similar concave shapes but few ligands are inherently concave in a rigid fashion. Upon surveying the crystal structures of some cis-azetidine derivatives it was reasoned that the amino-azetidine scaffold might be one such ligand. The cis-ring geometry of ligand 1 makes it inherently concave and in the chelation complex 2, the R1-substiuent should point over the metal (Fig. 1) and offer a rigid platform to potentially strongly influence the stereoselectivity of an asymmetric reaction catalysed by such a complex. As a result of an on-going interest in the synthesis of nitrogen-containing heterocycles2,3,4,5,6,7, co-authors of this report have identified a protocol for delivery of 2,4-cis-disubstituted azetidine derivatives as single diastereoisomers8,9.
Arbitrary depiction of the complexation of 1 with a square-planar metal to give complex 2.
Chelating-N,N′ C2-symmetric ligands are well-developed as stereoselectivity-engendering motifs in asymmetric catalysis, following the introduction of anionic semicorrin and related N,N′-ligands by Pfaltz10 and others11. Corey et al. employed a neutral analogue (3) as a ligand for an iron(III)-catalysed Diels-Alder reaction12. Evans and co-workers deployed 3 and related ligands for copper-catalysed transformations13. This work deeply impacted the field of asymmetric catalysis, providing a platform for both asymmetric reaction development and understanding of the underlying principles of asymmetric Lewis acid catalysis14,15,16. C2-symmetric N,N′-chelating bis-amine (4) and bis-imine ligands (Fig. 2) have been popularised by Kobayashi and co-workers (Including research co-authored by one of the authors of this report)17,18,19,20,21, for catalysis of a range of asymmetric transformations. Whilst ligands and complexes displaying C2-symmetry offer some advantages, including ease of synthetic access, the virtues of C1-symmetric asymmetric catalysts remain strong, with numerous reports detailing and contrasting them22,23,24,25,26.
C2-Symmetric bisoxazoline (3) and diamine (4) ligands.
Reasoning that the shape of the chiral pocket described by a cis-disubstituted azetidine N,N′-ligand has the potential to offer a unique C1-symmetric catalytic platform, and that access to single enantiomer azetidines has already been established, we posed the question: Can 2,4-cis-disubstituted azetidine N,N′-ligands be developed to deliver high enantiomeric excess in transition metal-catalysed carbon-carbon bond forming reactions?
The stereochemical utility of cis-disubstituted azetidines 1 may be probed by examining their use as ligands in asymmetric catalysis, as such the copper-catalysed Henry reaction was selected for this investigation (Fig. 3)27. C2-Symmetric28,29,30,31,32,33,34,35, and C1-symmetric N,N′-ligands36,37 have previously been use to engender asymmetry in the Henry reaction38,39,40,41,42. Therefore, the Henry reaction is ideal for demonstrating tractability and stereochemical scope of 1-type ligands in asymmetric catalysis.
General scheme for the additon of nitromethane to an aldehyde (5) under control of a chiral diamine ligated copper catalyst, to furnish a nitro-aldol adduct (6).
Results and Discussion
In order to evaluate and optimise the stereochemical induction potential of cis-disubstituted azetidines 1 in the Henry reaction of Fig. 3, a total of fifteen single enantiomer ligands, 1a-o (Fig. 4), were prepared8 (For full details see supporting information).
Single enantiomer azetidines 1a-o deployed as ligands for asymmetric copper-catalysed Henry reactions.
The relative (cis) and absolute stereochemistry of the HI salt of ligand 1d were determined by single crystal X-ray diffraction structure analysis (Fig. 5), and reference to the Flack parameter determined as −0.017(2), thus confirming (2S, 4R) stereochemistry43.
Chemical structure (left) and crystal structure (right) of (2S, 4R) 1d, Flack parameter -0.017(2) thus confirming stereochemical assignment.
Since simple alkyl amine derivatives of 1 (positions R3 and R4, Fig. 1) were chronologically first available to this project as single enantiomers, azetidine 1a was selected for initial investigation (Table 1, entry 1). In the reaction of para-nitrobenzaldehyde (5a) with nitromethane 5 mol% loading of a 1:1 mixture of copper(II)acetate.(H2O) and ligand 1a mixture, a promising 95% conversion and 35% e.e. of 6a was obtained. Increasing the ring substituent size from phenyl to 1-naphthyl (1b) resulted in lower conversion and racemic product (Table 1, entry 2, 86% conv., 1% e.e.). Similarly adding a tertiary butyl group to the azetidine’s phenyl substituent (1c) also lowered the conversion and gave very low enantiomeric excess in contrast to the first ligand tested (Table 1, entry 3, 90% conv., 9% e.e.; versus entry 1). Electron withdrawing substituents 4-nitro- (1d, Table 1, entry 4) and 4-chloro- (1e, Table 1, entry 5) on the ring substituent phenyl group showed good conversion (95 and 96% respectively). Only 25% e.e. in 6a was obtained with 1d as ligand, whereas an improved e.e. of 47% was observed for use of 1e (Table 1, entries 4 and 5). Thus, the 4-chlorophenyl derivative is the most effective 2-position substituent among those initially tested (1e). Switching the 4-chloro- and phenyl substituents (R2 and R3, Table 1, entry 6 1f, versus entry 5 1e), gave slightly lower conversion (82%) and e.e. (32%). Changing the aldehyde to benzaldehyde (5b) in the reaction facilitated by ligand 1a (Table 1, entry 7) gave a lower conversion than using 4-nitrobenzaldehyde (5a), which is expected on electronic grounds (5a being more prone to nucleophilic attack than 5b); pleasingly, the enantiomeric excess increased to 77%. Both ligands 1e and 1f also resulted in forming catalysts capable of delivering product 6a with similar outcomes, Table 1, entries 8 (88% conv., 71% e.e.) and 9 (77% conv., 77% e.e.) respectively. Changing position R3 to tertiary butyl (1g), 3,5-bis-trifluoromethyl-phenyl (1h), 4-tertiary butyl-phenyl (1i) and 4-methoxy-phenyl (1j) (Table 1, entries 10 to 13) did not improve the conversion or e.e. beyond that observed in Table 1, entry 7. Based on these results ligand 1e was retained for further study.
Choosing the addition of nitromethane to benzaldehyde (5b) and employing ligand 1e various copper sources were next investigated (Table 2, entries 1 to 10). To confirm a ligand accelerated reaction is occurring and no background reaction is compromising enantiomeric excess, the following reactions were performed: (i) in the absence of ligand 1e and copper source; (ii) in the presence of 5 mol% ligand 1e and the absence of a copper source; and (iii) in the absence of ligand 1e and in the presence of a copper source (Cu(OAc)2·H2O, 5 mol%); Table 2 entries 1, 2 and 3 respectively. After 16 hours, at room temperature product (6b) was not observed in all three cases, confirming no background or ligand-free metal catalysed reactions are leading to measurable amounts of product under the standard conditions used. Upon comparing catalyst loading (1 mol% and 10 mol%; Table 2 entries 4 and 6 respectively, versus 5 mol% entry 5), a lower loading reduced the rate of product formation (21% conversion after 48 h), whilst increasing the loading offered no observed advantage, as such 5 mol% loading was employed for all subsequent copper-catalysed Henry reactions in this report. Copper(I)acetate as a metal source (Table 2 entry 7) offered no advantage in terms of conversion to product 6b and lowered the enantiomeric excess (compared to Table 2 entry 5). The use of CuSO4.5H2O, CuCl2.2H2O, and Cu(OTf)2 as copper sources (Table 2 entries 8, 9 and 10 respectively) failed to deliver any product 6b. Next the effect of solvent was investigated (Table 2 entries 11 to 17, versus entry 5). Methanol and iso-propanol gave 96 and 78% conversion (Table 2 entry 11 and 12 respectively), both delivered product 6b in 32% e.e. Tetrahydrofuran and diethyl ether both gave essentially racemic product in 43 and 73% conversion (Table 2 entries 13 and 14 respectively). Toluene offered no advantage (Table 2 entry 15; 46% conversion and 39% e.e.). Dichloromethane and acetonitrile gave acceptable conversions, Table 2 entries 16 (88%) and 17 (86%) respectively, whilst enantiomeric excess was low (21% and 7% respectively). The initial, literature-informed selections of Cu(OAc)2·H2O as the copper source, and ethanol as the solvent are confirmed as superior. As expected, the enantioselectivity follows a temperature dependant trend, across a 0 to 60 °C (Table 2 entries, 18, 5, 19 and 20; 0 °C, 18 °C, 40 °C and 60 °C respectively). The highest enantiomeric excess, of 79%, was observed at the lowest temperature (0 °C, entry 18) although 48 h was required to reach 93% conversion. As expected the highest temperature gave an improved conversion but compromised the enantioselectivity somewhat (98% conversion and 51% e.e., entry 20).
The results displayed in Tables 1 and 2 lead to the interim conclusion that optimal conditions determined thus far, when the exocyclic amine = -NHiPr, (and with reference to substituent numbering of Fig. 1) are: R1 = 4-chloro-phenyl; R2 = benzyl; ethanol as solvent; and Cu(OAc)2·H2O as metal source. Thus, for further screening and optimisation, the same solvent and metal source were retained at 5 mol% catalyst loading. Reactions were run for 16 hours at room temperature (18 °C, for convenience, whilst recognising that e.e. may be improved by lowering the reaction temperature and extending the time). Comparing the use of iso-propylamine functionalised ligand 1e (Table 1, entries 5 and 8) in the reactions of aldehydes 5a and 5b to those mediated by the n-propylamine congener (Table 3, entries 1 and 2) there are no obvious advantages or differences between branched (1e) and linear (1k) secondary alkyl amine functionalities. Whereas the N-benzyl secondary amine analogue (1l) offered an increased enantiomeric excess for the same reactions (Table 3, entries 3 and 4), where products 6a and 6b were obtained in 81% and 87% e.e. respectively. Tertiary amine derivative 1m (pyrrolidine substituent 1m, Table 3, entries 5 and 6) gave circa. 80% conversion but the enantiomeric excesses of products 6a and 6b were only 5 and 28% respectively. Having now identified the N-benzyl derivative 1l as the best ligand to this point, electron -poor (1n 4-chlorobenzyl) and -rich ligand analogues (1o 4-methoxybenzyl) were compared in the catalysed formation of 6b (Table 3, entry 4, ligand 1l versus entries 7 and 8, ligands 1n and 1o respectively). Among these three ligands compared, the electron-rich analogue 1o was inferior (64% conversion and 66% e.e.). Ligand 1n was also slightly inferior to ligand 1l in the same 6b-forming reaction. Ligand 1l was therefore identified as the best of the ligands prepared in this report and used to demonstrate substrate scope (Table 4).
Since the Henry reaction between aldehydes and nitromethane is a widely used protocol for the delivery of 2-nitro-ethanol derivatives, which are in turn often reductively transformed into corresponding methylene amino alcohols, for use in biology-facing applications, the addition of nitromethane to ten aldehydes under optimal conditions was investigated (Substrate scope in the nucleophile part was not investigated in this study. Future work will include investigation of diastereoselectivity of nitromethylene derivatives. Since amino-methylation is an important reaction in the synthesis of drug-like molecules it is a suitable reaction to assess the capacity of cis-azetidines to deliver asymmetric products in this first study). Recognising that ligand 1l gave slightly higher conversion than ligand 1e at room temperature, for laboratory operational ease, 50 h reactions at 0 °C were conducted in anticipation of achieving good to high yield across the set.
Under the optimised conditions, aldehydes 5a and 5b (Table 4, entries 1 and 2 respectively) provided the corresponding products 6a and 6b in 98% and 91% conversion and 91% and 95% enantiomeric excess respectively. When 4-methoxybenzaldehyde 5c was subjected to the optimised reaction conditions (Table 4, entry 3) only 66% conversion to product 6c was observed, whilst enantiomeric excess was a reasonable 92%. The use of electron-poor 4-chlorobenzaldehyde (5d) resulted in good conversion and enantiomeric excess in product 6d (95% conv., 95% e.e., Table 4, entry 4). Benzaldehyde derivatives bearing 4-alkyl substituents (5e 4-methyl and 5f 4-tert-butyl, Table 4, entry 5 and entry 6 respectively) gave acceptable conversions of 73% and 85% respectively, enantiomeric excesses were on a par with the preceding four table entries (6e and 6f both 93% e.e.). Aldehyde 5g, 1-naphthylaldehyde (Table 4, entry 7), gave 83% conversion and 93% enantiomeric excess. Since this is similar to the level of conversion and enantiomeric excess when 5f was employed (4-tert-butyl-phenyl aldehyde, Table 4, entry 6), an extended aromatic surface, an additional ring, offers no divergence from an additional bulky alkyl group suggesting steric parameters rather than π-interactions may lie at the origin of the observed selectivity. To test this hypothesis alkyl aldehydes cyclohexyl aldehyde 5i and pivaldehyde 5j were tested (Table 4, entries 9 and 10 respectively). To our delight the reactions to form 6i and 6j proceeded with excellent conversion and in excellent enantiomeric excess, Table 4, entry 9 and 10 (6i, 98% conv., >99% e.e. and 6j, 99% conv., >99% e.e. respectively), thus confirming π-interactions are not required in order to achieve high conversion and selectivity.
According to analysis using the LLAMA web tool44, from the University of Leeds (UK), this small set of products, 6a-j all lie within Lipinski-space, molecular weight less than 500 and logP less than 5, furthermore 5 of the 10 products accessed fall within lead-like space. Whilst far from exhaustive in substrate scope this does confirm the reactions giving good to excellent enantiomeric excess for C-C bond forming reactions are giving rise to lead-like compounds that may be of utility in drug discovery. Future work in this area will include direct access to libraries of enantiopure small molecules by extending substrate scope in both nucleophile and electrophile and developing novel asymmetric transformations facilitated by the cis-substituted azetidine scaffold. In order to achieve these ambitions a deeper understanding of the stereochemical influence of the ligand-metal manifold is required, as such a stereochemical rationale was next sought through transition state modelling.
Stereochemical rationale and transition state modelling
To explore the properties of the ligands and their metal complexes, quantum mechanical calculations were undertaken. Previous studies have shown that the M06-2×/6-31G* (with the LANL basis set and pseudo-potential on copper) level of theory is suitable for studying the copper catalysed Henry reaction and have described the structures of minima and transition states for the process45. These protocols have been adapted for use in this study of 2,4-cis-disubstituted azetidine ligands 1 in the asymmetric copper-catalysed Henry reaction of this study (All calculations were performed in Gaussian09 (Gaussian 09, Revision A.02). Minima and transition states were characterised by calculation of vibrational frequencies)46,47,48,49,50. A model azetidine-containing ligand (1pC, Fig. 6), with reduced complexity, is deployed in the initial calculations with copper(II)acetate as the metal source, using the aforementioned protocol, and structures of this complex were obtained by editing structures optimised by Das et al.51, resulting in the computed structure given in Fig. 7.
Structure of ligands 1pC and 1qC investigated computationally in this study.
Representations of the minimised structure of a copper(II)acetate derived complex of 1pC.
After ligand complexation, the next step along the reaction coordinate is for one or other of the acetate ligands of the formed complex to be exchanged with an anion of nitromethane, several ligand arrangements were considered. These calculations revealed that the lowest energy conformation (Fig. 8, i) is where the right hand acetate group from Fig. 7 (as drawn) is replaced with nitromethane anion, and the left hand (as drawn) acetate retains a H-bonding interaction with the ligand’s NH proton. All other conformations (Fig. 8, ii, iii and iv) involve loss of the hydrogen bond between acetate and the NH of the azetidine ligand and are more than 4 kcal/mol higher in energy. This hydrogen bonding interaction with an acetate is a key influence that ensures a significant preference for replacing only one specific acetate ligand. When the ligand conformation is altered such that the NHMe presents a methyl group towards the acetate rather than a proton, the energy increases by 4.0 kcal/mol (Fig. 8, ii) and a six-coordinate pseudo-octahedral geometry on the copper centre is imposed by a now bidentate acetate-copper interaction. When the nitromethane anion replaces the hydrogen bonding acetate the pseudo-octahedral geometry and hydrogen bonded geometries (Fig. 7, acetate on left side as drawn resulting in structures depicted as Fig. 8 iii and iv) have the same energy suggesting that the nitromethane anion forms a weaker hydrogen bond (than acetate), which is approximately equivalent to its interaction with metal. Alternative conformations including those with variations in the azetidine ring were all prohibitively high in energy and result in the azetidine dissociating from the copper. The acetate group has a strong preference for forming a copper-oxygen bond and a hydrogen bond and this dictates a preference for displacing the acetate lacking a hydrogen bond with nitromethane anion. The cyclic binding mode for a ligand containing an O = X-O− (X = C-R or S(O)-R) functionality is in congruence with analogous complexes (featuring sulfonyl groups) previously described by calculations and observed in XRD single crystal structures51,52,53,54,55,56.
Four minima and free energies in kcal/mol for the complexes calculated to be formed upon exchange of an acetate from 1pC with a nitromethane anion.
Having established a strong preference for replacing one acetate group and for a single backbone conformation of the azetidine ligand, with some variation permitted in the positioning of the benzyl group on the exocyclic nitrogen atom, transition states for the reaction between the complex of nitromethane anion and benzaldehyde were investigated. For these initial calculations azetidine ligand 1pC was used, resulting in the overall profile depicted in Fig. 9 51.
Calculated reaction coordinates and free energies in kcal/mol for the 1pC-catalyed reaction of benzaldehyde (5b) with the anion of nitromethane. Two diastereomeric transition states leading to opposite enantiomers of product 6b depicted, free energies in parenthesis include solvation in THF.
Importantly, the model system described in Fig. 9 incorrectly predicts the stereochemical outcome observed experimentally (1a-o). The higher energy of the two transition states depicted as TS1-1pC-I (Fig. 9) adopts a six-membered chair-like orientation, with the phenyl group of the benzaldehyde derived part in an axial position, this higher transition state would lead to the experimentally observed (S) product (Fig. 10, left). Whereas the computationally predicted outcome, in this case, resulting from the lower energy transition state TS1-1pC-II (Fig. 9), which adopts a boat-like orientation, would be the incorrect (R) enantiomer (Fig. 10, right). Placing a phenyl group in the equatorial position in TS1-1pC-I is not possible because this space is filled by the phenyl of ligand 1pC. Clearly a feature not considered up to this point is critical in determining the stereochemical outcome of 1-ligated copper-catalysed Henry reactions.
Drawings highlighting the axially substituted six-memebered chair-like transition state TS-1pC-I (left upper) and the eclipsed boat-like transition state (TS-1qC-II), and the products to which the lead (lower images respectively). The disconnect between the experimentally observed and the computationally predicted products, using this model, are highlighted.
Three possibilities were considered: (i) Alternative coordination geometries around the copper; (ii) the inclusion of the benzyl substituent (simplified to a methyl in the previous calculation) on the exocyclic nitrogen (1pC); and (iii) solvation effects. Addressing each in turn: (i) Several alternative coordination geometries were studied but none were found to be lower in energy. (ii) Swapping the HN-methyl group of 1pC to an HN-benzyl substituent (1qC) increases the computational complexity but results in an interesting observation. First the conformations of 1qC (Fig. 6) upon complexation with copper(II) acetate were explored, using the protocol described above. The lowest energy structure (Fig. 11, left) involves the phenyl of the HN-benzyl substituent projected towards the copper atom; whereas an alternative with the phenyl directed away from the metal centre is 1.1 kcal/mol higher in energy (Fig. 11, right). In both structures copper adopts a pseudo-square-based pyramidal coordination geometry, and in the lowest energy structure the phenyl of the benzyl occupies the region of space that a sixth ligand might otherwise occupy in an octahedral complex. The cis-stereochemistry of the ligand creates a structure with a concave cavity which envelops the copper. The acetate hydrogen bond and preferential displacement of the non-hydrogen bonding acetate, described previously, were assumed to operate in this larger system.
Two calculated minima and free energies in kcal/mol for the calculated copper(II)acetate-derived complex of 1qC.
Calculated reaction coordinates and free energies in kcal/mol for the 1qC-catalyed reaction of benzaldehyde (5b) with the anion of nitromethane. Two diastereomeric transition states (TS-1qC-I and TS-1qC-II) leading to opposite enantiomers of product 6b depicted, free energies in parenthesis include solvation in THF.
Employing ligand 1qC in the computed reaction of nitromethane and benzaldehyde led to the computationally predicted reaction outcome preference being in agreement with the experimentally observed (S)-stereoisomer of 6a (Fig. 12). Whilst this benzyl group is oriented away from copper in the lowest energy transition states, it is exerting an important influence on the reaction outcome. The lowest energy transition states for attack on the Si and Re faces of the aldehyde involve a square pyramidal arrangement of ligands around the copper ion with the azetidine nitrogen occupying the apical position (Fig. 13). In both TS1qC-I and TS1qC-II the nitromethane anion occupies the rear right coordination site as drawn (Fig. 13 upper left and right respectively), likely because this places the anionic nitro-oxygen trans to the electronegative oxygen of the acetate. Were the nitromethane anion and aldehyde swapped such that, in both cases, the aldehyde lies at the rear of the complex (not drawn, see ESI tables of coordinates), the nitromethane anions are then cis to the acetate and result in transition states with activation barriers of 20.3 (leading to (R)-product) and 29.3 kcal/mol (leading to (S) product) respectively. Furthermore, the face of the aldehyde presented to the nitromethane anion is governed by a minimisation of a steric clash between the phenyl substituents of the aldehyde and the ligand. In the favoured TS (Fig. 13 left upper and lower) the phenyl ring of the aldehyde is orientated in an endo fashion (towards the metal centre), minimising steric interaction with the concave ligand architecture. Whereas, in the higher energy scenario the phenyl group of the aldehyde lies exo (in a direction away from the metal) but experiences more of a clash with the ligand’s phenyl substituent (right side of images as drawn in Fig. 13 right upper and lower). (iii) Increasing the polarity of the solvent (using solvation single points with the settings for THF c.f. Table 1, entry 13) is computed to decrease the difference in free energy between transition states in each system and therefore suggests that more polar solvents are likely to be detrimental for selectivity, which is broadly in line with experimental evidence.
Representations of the two diastereomeric transition states TS-1qC-I (left upper and lower) and TS-1qC-II (right upper and lower), from Fig. 12, leading to the formation of enantiomers of 6b.
The origin of the change in selectivity upon replacing N-Me (ligand 1pC) with N-Bn (ligand 1qC) in the model is a subtle change for many of the interactions involved in each of the transition states, but the tightness of binding of the amine to the metal and the impact of this on the ligand trans to it is likely key. In each of the key transition states, the ligand trans to the amine is the forming oxy anion (the aldehyde oxygen). In the transition states leading to the observed product, the N-Cu distance is 2.05 Å for both ligands and the Cu-O distance is also unchanged at 1.99 Å. By contrast, in the transition state leading to the disfavoured (R) product, the N-Cu distance is 2.04 Å for N-Me (ligand 1-I) and 2.03 Å for N-Bn (ligand 1qC); this small change causes a larger corresponding change in the Cu-O distance which increases from 1.92 Å for N-Me to 1.96 Å for N-Bn. The N-Me system is better able to stabilise the transition state leading to the (R) product than is N-Bn and hence the computed stereoselectivity inverts to that agreeing with experimental observations.
For both ligands studied computationally (1pC and 1qC), the geometry of the transition state leading to the S isomer is broadly the same. This is a six-membered chair-like structure in which the phenyl of the benzaldehyde is placed in an axial position. The space that an equivalent equatorial substituent would fill is occupied by the phenyl of the ligand. The geometry of the transition state leading to the R isomer is also largely unchanged when the ligand is changed. This is a boat transition state which permits the phenyl of the aldehyde to be positioned in an equatorial-like position. This is the trade-off that determines selectivity: A chair-like transition state with an axial aldehyde substituent or a boat-like transition state with an equatorial, eclipsed, aldehyde substituent. The axial position, in this case is not as strongly disfavoured as might be expected, because the transannular axial positions around the transition state are occupied by the other oxygen of the nitro group and the empty space that would otherwise be occupied by a sixth ligand on copper (making it octahedral). This axial positioning is therefore not strongly disfavoured (Fig. 14).
Drawings highlighting the axially substituted six-memebered chair-like transition state TS-1qC-I (left upper) and the eclipsed boat-like transition state (TS-1qC-II), and the products to which the lead (lower images respectively). The agreement between the experimentally observed and the computationally predicted products, using this model, are highlighted.
Model metal complex X-ray diffraction crystal structure
During the course of developing azetidine derivatives as chiral ligands for copper-catalysed Henry reactions various attempts to form, isolate and study azetidine metal complexes were made. Whilst crystals suitable for X-ray diffraction structure determination were not isolated from attempts using various copper salts, solvents and techniques, it was possible to form a small number of crystals of a square planar platinum(II) chloride complex from racemic ligand and K2PtCl4. Whilst platinum complexes are not the main topic of study in this report the crystal structure of 7 (Fig. 15) might be instructive for understanding steric interactions and corroborating selectivity hypotheses. Details of this and a small number of additional complexes of platinum and palladium will be reported in detail in a later report. Herein, the crystal structure of complex (rac)-7 is used to inform discussion.
Representation of one molecule from the single crystal XRD structure of platinum complex 7. Elipsoids plot at 50% probability, and rendered using Ortep-III for Windows and PovRay v3.7.
From the crystal structure of 7 it is clear the 2,4-cis stereochemistry of the azetidine ligand imparts a rigid geometry presenting the 2-substiuent out over the metal atom, in one quadrant of the front view of the molecule (as depicted in Fig. 15). The N-benzyl groups of the ring and the secondary amine part point down, as depicted, occupying two other quadrants of the molecule. This leaves the fourth, top right, quadrant open and presenting and N-H group as a potential H-bond donor. The N,N′-1,2-chelation and structural features observed in this model complex help to picture features that may be common across the ligand family and between various transition metal centres.
To explore the degree to which the rigid structure of the ligand is retained, even in a complex with a different metal and other ligands, the structure of complex 7 is overlaid with that found in the calculations for the lowest energy transition state (Fig. 13, TS-1qC-I) and diacetate complex (Cu[1qC-(OAc) 2 ]), and shown in Fig. 16.
Larger central lower image, overlay of the structures (indecated and colour coded above) of: Ligth blue, XRD-derived platinum(II) complex 7; light green, the calcualted structural minima for the coppe(II) acetate complex of 1qC; magenta, the calculted transtion state, leading to the experimentally observed stereochemical outcome, TS-1qC-I.
The common substructures components of these complexes are depicted, and colour coded: The platinum complex 7 is shown in light blue, copper diacetate complex is shown in green (Cu[1qC-(OAc) 2 ]) and the copper transition state is shown in magenta (TS-1qC-I). Apart from a slight twist of the phenyl ring attached to the azetidine ring, much of the shared ligand architecture is rigid and invariant among the calculated copper complex structures and the experimentally observed platinum complex, the shared, rigid concave form of the ligand is retained across the three systems. Variation is restricted to the sidechain CH2N-benzyl group which adjusts to accommodate the preference of the platinum centre. That part can adapt to best complement the reactant complex or the transition state, as discussed earlier. The stable and isolable platinum complex provides a reasonable approximation for evaluating steric parameters and effects in 1-mediated transition metal catalysis.
Inspired by publications from Nolan57, Cavallo58 and their respective co-workers a steric description of the crystal structure of platinum complex 7 was sought, and examination using the free web tool SambVca 2.0 was attempted59. The corresponding PDB file was uploaded to SambVca for analysis (one molecule of 7 from the unit cell of the crystal structure, with chlorides removed). An overall buried volume was determined to be 62.1%Vbur, and as surmised from visual inspection of the crystal structure, the quadrant map generated (Fig. 17) confirms the cis-geometry of the ligand to imparts steric constraint or pressure in three of the four quadrants leaving the NH-containing quadrant essentially vacant.
Left: Image defining axis, obtained from X-ray crystal structure of 7, chlorides removed. Right: SambVca 2.0 generated buried volume plot depicting 62.1%VBur SW 69.6%; NW 67.2%; NE 43.3%; SE 68.4%, derived from a molecule of 11 in its X-ray crystal structure.
Notably the two largest (or most buried) quadrants correspond to the azetidine ring 2-phenyl group and the NH-benzyl group, occupying the North West and South-East quadrants as drawn (67.2%Vbur and 68.4%Vbur, respectively). The NH-benzyl group exerts the largest steric pressure around the coordination centre that corroborates the importance of including it in the preceding calculations of this report and helps to further rationalise the structure-selectivity relationships witnessed earlier.
Conclusions
In conclusion, the 2,4-cis-disubstitution pattern about azetidines 1 has been demonstrated as a useful chiral scaffold about which to build ligands for asymmetric copper-catalysed Henry reactions. Through optimisation of ligand structure and conditions a system capable of delivering enantiopure (to the limits of the used analysis method) products from the addition of nitromethane to alkyl aldehydes. Furthermore, whilst a preliminary result, the same cis-substituted azetidine scaffold was capable of engendering asymmetry under an organocatalytic manifold (see supporting information). Computational analysis was employed to help rationalise the stereochemical outcome observed in the copper catalysed reactions herein. Initially, simplified ligand systems failed to correctly predict the observed stereochemical outcome, but inclusion of a critical group permitted the corroboration of computation and experimentation, feeding into new models of C-C bond formation. We plan to design new ligands drawing on the knowledge acquired here and deploy them across a wider substrate scope and in more metal-mediated reactions. The initial organocatalytic result will likely be followed up in a future project within the team.
Data Availability
Procedures for ligand synthesis protocols, tables of data, spectrums and CIF files may be found among the additional supporting material (Citations to synthetic procedures, materials and methods contained within the supporting information that are not already cited in the main text are given here to acknowledge there importance to the information reported in the main text)60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94. A version of this manuscript was lodged with a pre-print server prior to peer evaluation of this manuscript95.
References
Reck, L. M., Haberhauer, G. & Lüning, U. Enantiopure Chiral Concave 1,10-Phenanthrolines. Eur. J. Org. Chem. 2016, 1119–1131, https://doi.org/10.1002/ejoc.201501289 (2016).
Feula, A. & Fossey, J. S. Base and solvent dependency of an oxidative retro-alkylation of secondary and tertiary benzylamines. RSC Adv. 3, 5370–5373, https://doi.org/10.1039/c3ra40228b (2013).
Hama Salih, M. A., Male, L., Spencer, N. & Fossey, J. S. [gamma]-Lactams and furan bispyrrolidines via iodine mediated cyclisation of homoallylamines. Org. Chem. Front. 2, 1445–1449, https://doi.org/10.1039/C5QO00183H (2015).
Brittain, W. D. G., Buckley, B. R. & Fossey, J. S. Kinetic resolution of alkyne-substituted quaternary oxindoles via copper catalysed azide-alkyne cycloadditions. Chem. Commun. 51, 17217–17220, https://doi.org/10.1039/C5CC04886A (2015).
Brittain, W. D. G. et al. The Bull-James assembly as a chiral auxiliary and shift reagent in kinetic resolution of alkyne amines by the CuAAC reaction. Org. Biomol. Chem. 14, 10778–10782, https://doi.org/10.1039/C6OB01623E (2016).
Zhai, W. et al. “Click-fluors”: triazole-linked saccharide sensors. Org. Chem. Front. 3, 918–928, https://doi.org/10.1039/C6QO00171H (2016).
Zhai, W., Male, L. & Fossey, J. S. Glucose selective bis-boronic acid click-fluor. Chem. Commun. 53, 2218–2221, https://doi.org/10.1039/C6CC08534B (2017).
Feula, A. et al. Synthesis of azetidines and pyrrolidines via iodocyclisation of homoallyl amines and exploration of activity in a zebrafish embryo assay. Org. Biomol. Chem. 11, 5083–5093, https://doi.org/10.1039/c3ob41007b (2013).
Feula, A., Male, L. & Fossey, J. S. Diastereoselective preparation of azetidines and pyrrolidines. Org. Lett. 12, 5044–5047 (2010).
Pfaltz, A. From Corrin Chemistry to Asymmetric Catalysis - A Personal Account. Synlett 1999, 835–842, https://doi.org/10.1055/s-1999-3122 (1999).
Lowenthal, R. E., Abiko, A. & Masamune, S. Asymmetric catalytic cyclopropanation of olefins: bis-oxazoline copper complexes. Tetrahedron Lett. 31, 6005–6008, https://doi.org/10.1016/S0040-4039(00)98014-6 (1990).
Corey, E. J., Imai, N. & Zhang, H. Y. Designed catalyst for enantioselective Diels-Alder addition from a C2-symmetric chiral bis(oxazoline)-iron(III) complex. J. Am. Chem. Soc. 113, 728–729, https://doi.org/10.1021/ja00002a081 (1991).
Evans, D. A., Woerpel, K. A., Hinman, M. M. & Faul, M. M. Bis(oxazolines) as chiral ligands in metal-catalyzed asymmetric reactions. Catalytic, asymmetric cyclopropanation of olefins. J. Am. Chem. Soc. 113, 726–728, https://doi.org/10.1021/ja00002a080 (1991).
Evans, D. A., Murry, J. A., Vonmatt, P., Norcross, R. D. & Miller, S. J. C-2-Symmetric Cationic Copper(Ii) Complexes as Chiral Lewis-Acids - Counterion Effects in the Enantioselective Diels-Alder Reaction. Angew. Chem. Int. Edt. Engl. 34, 798–800, https://doi.org/10.1002/anie.199507981 (1995).
Rasappan, R., Laventine, D. & Reiser, O. Metal-bis(oxazoline) complexes: From coordination chemistry to asymmetric catalysis. Coord. Chem. Rev. 252, 702–714, https://doi.org/10.1016/j.ccr.2007.11.007 (2008).
Johnson, J. S. & Evans, D. A. Chiral Bis(oxazoline) Copper(II) Complexes: Versatile Catalysts for Enantioselective Cycloaddition, Aldol, Michael, and Carbonyl Ene Reactions. Acc. Chem. Res. 33, 325–335, https://doi.org/10.1021/ar960062n (2000).
Kobayashi, S., Matsubara, R., Nakamura, Y., Kitagawa, H. & Sugiura, M. Catalytic, asymmetric Mannich-type reactions of N-acylimino esters: Reactivity, diastereo- and enantioselectivity, and application to synthesis of N-acylated amino acid derivatives. J. Am. Chem. Soc. 125, 2507–2515 (2003).
Fossey, J. S., Matsubara, R., Kiyohara, H. & Kobayashi, S. Heterochiral triangulo nickel complex as evidence of a large positive nonlinear effect in catalysis. Inorg. Chem. 47, 781–783, https://doi.org/10.1021/Ic7017727 (2008).
Fossey, J. S., Matsubara, R., Vital, P. & Kobayashi, S. A C-2-symmetric nickel diamine complex as an asymmetric catalyst for enecarbamate additions to butane-2,3-dione. Org. Biomol. Chem. 3, 2910–2913 (2005).
Berthiol, F., Matsubara, R., Kawai, N. & Kobayashi, S. Catalytic asymmetric Michael reactions with enamides as nucleophiles. Angew. Chem., Int. Ed. 46, 7803–7805 (2007).
Matsubara, R. & Kobayashi, S. Enamides and Enecarbamates as Nucleophiles in Stereoselective C–C and C–N Bond-Forming Reactions. Acc. Chem. Res. 41, 292–301, https://doi.org/10.1021/ar700098d (2008).
Pfaltz, A. & Drury, W. J. Design of chiral ligands for asymmetric catalysis: From C-2-symmetric P,P- and N,N-ligands to sterically and electronically nonsymmetrical. P,N-ligands. Proc. Natl. Acad. Sci. USA 101, 5723–5726, https://doi.org/10.1073/pnas.0307152101 (2004).
Nishiyama, H., Soeda, N., Naito, T. & Motoyama, Y. Single-chiral bis(oxazolinyl)pyridine (pybox). Efficiency in asymmetric catalytic cyclopropanation. Tetrahedron Asymmetr. 9, 2865–2869, https://doi.org/10.1016/S0957-4166(98)00303-6 (1998).
Suenobu, K., Itagaki, M. & Nakamura, E. Reaction Pathway and Stereoselectivity of Asymmetric Synthesis of Chrysanthemate with the Aratani C1-Symmetric Salicylaldimine−Copper Catalyst. J. Am. Chem. Soc. 126, 7271–7280, https://doi.org/10.1021/ja031524c (2004).
Fraile, J. M. et al. Theoretical Insights into the Role of a Counterion in Copper-Catalyzed Enantioselective Cyclopropanation Reactions. Chem. Eur. Jour. 10, 758–765, https://doi.org/10.1002/chem.200305161 (2004).
Cornejo, A. et al. Asymmetric versus C2-Symmetric Ligands: Origin of the Enantioselectivity in Ruthenium–Pybox-Catalyzed Cyclopropanation Reactions. Angew. Chem., Int. Ed. 44, 458–461, https://doi.org/10.1002/anie.200461418 (2005).
Ono, N. In The Nitro Group in Organic Synthesis 30–69 (John Wiley & Sons, Inc., 2002).
Chunhong, Z., Liu, F. & Gou, S. Application of chiral N,N′-dialkyl-1,2-cyclohexanediamine derivatives in asymmetric copper(II)-catalyzed Henry reactions. Tetrahedron Asymmetr. 25, 278–283, https://doi.org/10.1016/j.tetasy.2013.12.017 (2014).
Kowalczyk, R., Sidorowicz, Ł. & Skarżewski, J. Asymmetric Henry reaction catalyzed by chiral secondary diamine-copper(II) complexes. Tetrahedron Asymmetr. 19, 2310–2315, https://doi.org/10.1016/j.tetasy.2008.09.032 (2008).
El-Asaad, B., Métay, E., karamé, I. & Lemaire, M. Chiral N-arylated diamine – Copper complexes catalyzed asymmetric Henry reaction. Molecular Catalysis 435, 76–81, https://doi.org/10.1016/j.mcat.2017.03.017 (2017).
Xu, H. & Wolf, C. Synthesis of chiral tertiary trifluoromethyl alcohols by asymmetric nitroaldol reaction with a Cu(ii)-bisoxazolidine catalyst. Chem. Commun. 46, 8026–8028, https://doi.org/10.1039/C0CC02378G (2010).
Selvakumar, S., Sivasankaran, D. & Singh, V. K. Enantioselective Henry reaction catalyzed by C2-symmetric chiral diamine-copper(II) complex. Org. Biomol. Chem. 7, 3156–3162, https://doi.org/10.1039/B904254G (2009).
Cruz, H., Aguirre, G., Madrigal, D., Chávez, D. & Somanathan, R. Enantioselective nitromethane addition to brominated and fluorinated benzaldehydes (Henry reaction) catalyzed by chiral bisoxazoline–copper(II) complexes. Tetrahedron Asymmetr. 27, 1217–1221, https://doi.org/10.1016/j.tetasy.2016.10.007 (2016).
Evans, D. A. et al. A New Copper Acetate-Bis(oxazoline)-Catalyzed, Enantioselective Henry Reaction. J. Am. Chem. Soc. 125, 12692–12693, https://doi.org/10.1021/ja0373871 (2003).
Bandini, M., Piccinelli, F., Tommasi, S., Umani-Ronchi, A. & Ventrici, C. Highly enantioselective nitroaldol reaction catalyzed by new chiral copper complexes. Chem. Commun., 616–618, https://doi.org/10.1039/B613019D (2007).
Blay, G., Hernandez-Olmos, V. & Pedro, J. R. A catalytic highly enantioselective direct synthesis of 2-bromo-2-nitroalkan-1-ols through a Henry reaction. Chem. Commun., 4840–4842, https://doi.org/10.1039/B809739A (2008).
Atoholi Sema, H., Bez, G. & Karmakar, S. Asymmetric Henry reaction catalysed by L-proline derivatives in the presence of Cu(OAc)2: isolation and characterization of an in situ formed Cu(II) complex. Appl. Organomet. Chem. 28, 290–297, https://doi.org/10.1002/aoc.3123 (2014).
Boruwa, J., Gogoi, N., Saikia, P. P. & Barua, N. C. Catalytic asymmetric Henry reaction. Tetrahedron Asymmetr. 17, 3315–3326, https://doi.org/10.1016/j.tetasy.2006.12.005 (2006).
Palomo, C., Oiarbide, M. & Laso, A. Recent Advances in the Catalytic Asymmetric Nitroaldol (Henry) Reaction. Eur. J. Org. Chem. 2007, 2561–2574, https://doi.org/10.1002/ejoc.200700021 (2007).
White, J. D. & Shaw, S. A New Catalyst for the Asymmetric Henry Reaction: Synthesis of β-Nitroethanols in High Enantiomeric Excess. Org. Lett. 14, 6270–6273, https://doi.org/10.1021/ol3030023 (2012).
Tanaka, K. et al. Solvent-dependent strong asymmetric amplification in the catalytic enantioselective Henry reaction using the trans-N,N[prime or minute]-bis-biphenyl-4-ylmethyl-cyclohexane-1,2-diamine-CuCl2 complex. Chem. Commun. 51, 7907–7910, https://doi.org/10.1039/C5CC02302E (2015).
Chen, Z.-N., Wang, K., Cui, D. & Wu, A. The new chemical insight for understanding the mechanism of Henry reaction over Cu(II) catalyst. Chemical Physics Letters 673, 7–10, https://doi.org/10.1016/j.cplett.2017.01.066 (2017).
Flack, H. On enantiomorph-polarity estimation. Acta Crystallogr. Sect. A 39, 876–881, https://doi.org/10.1107/S0108767383001762 (1983).
Colomer, I. et al. A divergent synthetic approach to diverse molecular scaffolds: assessment of lead-likeness using LLAMA, an open-access computational tool. Chem. Commun. 52, 7209–7212, https://doi.org/10.1039/C6CC03244C (2016).
Blay, G., Domingo, L. R., Hernández-Olmos, V. & Pedro, J. R. New Highly Asymmetric Henry Reaction Catalyzed by CuII and a C1-Symmetric Aminopyridine Ligand, and Its Application to the Synthesis of Miconazole. Chem. Eur. Jour. 14, 4725–4730, https://doi.org/10.1002/chem.200800069 (2008).
Frisch, M. J. et al. Gaussian 09 Revision A.02, Wallingford, CT, Gausian, Inc. (2016).
Zhao, Y. & Truhlar, D. G. The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 120, 215–241, https://doi.org/10.1007/s00214-007-0310-x (2008).
Zhao, Y. & Truhlar, D. G. Density Functionals with Broad Applicability in Chemistry. Acc. Chem. Res. 41, 157–167, https://doi.org/10.1021/ar700111a (2008).
Tomasi, J., Mennucci, B. & Cammi, R. Quantum Mechanical Continuum Solvation Models. Chem. Rev. 105, 2999–3094, https://doi.org/10.1021/cr9904009 (2005).
Hariharan, P. C. & Pople, J. A. The influence of polarization functions on molecular orbital hydrogenation energies. Theor. Chim. Acta 28, 213–222, https://doi.org/10.1007/bf00533485 (1973).
Das, A. et al. Asymmetric Henry reaction of trifluoromethyl ketone and aldehyde using Cu(II)-complex: computational study offers the origin of enantioselectivity with varied size of catalysts. Tetrahedron 71, 5229–5237, https://doi.org/10.1016/j.tet.2015.06.033 (2015).
Panov, I., Drabina, P., Padělková, Z., Šimůnek, P. & Sedlák, M. Highly Enantioselective Nitroaldol Reactions Catalyzed by Copper(II) Complexes Derived from Substituted 2-(Pyridin-2-yl)imidazolidin-4-one Ligands. J. Org. Chem. 76, 4787–4793, https://doi.org/10.1021/jo200703j (2011).
Spingler, B. & Da Pieve, C. Induction of B- to Z-DNA transition by copper and zinc complexes with C(15) substituted macrocyclic pentaaza ligands. Dalton Trans., 1637–1643, https://doi.org/10.1039/B500317B (2005).
Wu, S.-S. et al. Synthesis, crystal structure and interaction with DNA and HSA of (N,N′-dibenzylethane-1,2-diamine) transition metal complexes. J. Inorg. Biochem. 102, 2026–2034, https://doi.org/10.1016/j.jinorgbio.2008.08.005 (2008).
Medina-Molner, A. & Spingler, B. When two metal centres are needed instead of one: exclusive induction of Z-DNA by dinuclear metal complexes. Chem. Commun. 48, 1961–1963, https://doi.org/10.1039/C2CC16483C (2012).
Zhang, G., Yashima, E. & Woggon, W.-D. Versatile Supramolecular Copper(II) Complexes for Henry and Aza-Henry Reactions. Adv. Synth. Catal. 351, 1255–1262, https://doi.org/10.1002/adsc.200900069 (2009).
Clavier, H. & Nolan, S. P. Percent buried volume for phosphine and N-heterocyclic carbene ligands: steric properties in organometallic chemistry. Chem. Commun. 46, 841–861, https://doi.org/10.1039/B922984A (2010).
Poater, A. et al. SambVca: A Web Application for the Calculation of the Buried Volume of N-Heterocyclic Carbene Ligands. Eur J Inorg Chem 2009, 1759–1766, https://doi.org/10.1002/ejic.200801160 (2009).
Falivene, L. et al. SambVca 2. A Web Tool for Analyzing Catalytic Pockets with Topographic Steric Maps. Organometallics 35, 2286–2293, https://doi.org/10.1021/acs.organomet.6b00371 (2016).
Fossey, J. S. & Brittain, W. D. G. The CASE 2014 symposium: Catalysis and sensing for our environment, Xiamen 7th-9th November 2014. Org. Chem. Front. 2, 101–105, https://doi.org/10.1039/C4QO90045F (2015).
Payne, D. T., Fossey, J. S. & Elmes, R. B. P. Catalysis and Sensing for our Environment (CASE2015) and the Supramolecular Chemistry Ireland Meeting (SCI 2015): Dublin and Maynooth, Ireland. 8th–11th July. Supramol. Chem. 28, 921–931, https://doi.org/10.1080/10610278.2016.1150595 (2016).
Coles, S. J. & Gale, P. A. Changing and challenging times for service crystallography. Chem. Sci. 3, 683–689 (2012).
Ruano, J. L. G., Fernandez, I., Catalina, M. D. & Cruz, A. A. Asymmetric aziridination by reaction of chiral N-sulfinylimines with sulfur ylides: Stereoselectivity improvement by use of tert-butylsulfinyl group as chiral auxiliary. Tetrahedron-Asymmetr. 7, 3407–3414 (1996).
Ellman, J. A., Owens, T. D. & Tang, T. P. N-tert-Butanesulfinyl imines: Versatile intermediates for the asymmetric synthesis of amines. Acc. Chem. Res. 35, 984–995, https://doi.org/10.1021/Ar020066u (2002).
Plobeck, N. & Powell, D. Asymmetric synthesis of diarylmethylamines by diastereoselective addition of organometallic reagents to chiral N-tert-butanesulfinimines: switchover of diastereofacial selectivity. Tetrahedron Asymmetr. 13, 303–310, https://doi.org/10.1016/S0957-4166(02)00099-X (2002).
Sugiura, M., Hirano, K. & Kobayashi, S. α-Aminoallylation of Aldehydes with Ammonia: Stereoselective Synthesis of Homoallylic Primary Amines. J. Am. Chem. Soc. 126, 7182–7183, https://doi.org/10.1021/ja049689o (2004).
Ramachandran, P. V. & Burghardt, T. E. Highly diastereoselective and enantioselective preparation of homoallylic amines: Application for the synthesis of beta-amino acids and gamma-lactams. Chem.-Eur. J. 11, 4387–4395, https://doi.org/10.1002/chem.200401295 (2005).
Sugiura, M., Mori, C. & Kobayashi, S. Enantioselective Transfer Aminoallylation: Synthesis of Optically Active Homoallylic Primary Amines. J. Am. Chem. Soc. 128, 11038–11039, https://doi.org/10.1021/ja064106r (2006).
Sun, X.-W., Xu, M.-H. & Lin, G.-Q. Room-Temperature Highly Diastereoselective Zn-Mediated Allylation of Chiral N-tert-Butanesulfinyl Imines: Remarkable Reaction Condition Controlled Stereoselectivity Reversal. Org. Lett. 8, 4979–4982, https://doi.org/10.1021/ol062216x (2006).
Sheldrick, G. M. A short history of SHELX. Acta Crystallogr., Sect. A A64, 112–122, https://doi.org/10.1107/S0108767307043930 (2008).
Sun, X.-W., Liu, M., Xu, M.-H. & Lin, G.-Q. Remarkable Salt Effect on In-Mediated Allylation of N-tert-Butanesulfinyl Imines in Aqueous Media: Highly Practical Asymmetric Synthesis of Chiral Homoallylic Amines and Isoindolinones. Org. Lett. 10, 1259–1262, https://doi.org/10.1021/ol8001514 (2008).
Dolomanov, O. V., Bourhis, L. J., Gildea, R. J., Howard, J. A. K. & Puschmann, H. OLEX2: a complete structure solution, refinement and analysis program. J. Appl. Crystallogr. 42, 339–341, https://doi.org/10.1107/S0021889808042726 (2009).
Almansa, R., Collados, J. F., Guijarro, D. & Yus, M. Asymmetric synthesis of α- and β-amino acids by diastereoselective addition of triorganozincates to N-(tert-butanesulfinyl)imines. Tetrahedron Asymmetr. 21, 1421–1431, https://doi.org/10.1016/j.tetasy.2010.03.046 (2010).
Robak, M. T., Herbage, M. A. & Ellman, J. A. Synthesis and Applications of tert-Butanesulfinamide. Chem. Rev. 110, 3600–3740, https://doi.org/10.1021/cr900382t (2010).
Soltani, O., Ariger, M. A., Vázquez-Villa, H. & Carreira, E. M. Transfer Hydrogenation in Water: Enantioselective, Catalytic Reduction of α-Cyano and α-Nitro Substituted Acetophenones. Org. Lett. 12, 2893–2895, https://doi.org/10.1021/ol1008894 (2010).
Lai, G. et al. Highly Enantioselective Henry Reactions in Water Catalyzed by a Copper Tertiary Amine Complex and Applied in the Synthesis of (S)-N-trans-Feruloyl Octopamine. Chem. Eur. Jour. 17, 1114–1117, https://doi.org/10.1002/chem.201002915 (2011).
MacDonald, F. K., Carneiro, K. M. M. & Pottie, I. R. The first example of trimethylsilyl methylenenitronate reacting with aldehydes under an apparent Mukaiyama nitro-aldol reaction. Tetrahedron Lett. 52, 891–893, https://doi.org/10.1016/j.tetlet.2010.12.056 (2011).
Subba Reddy, B. V. & George, J. Enantioselective Henry reaction catalyzed by a copper(II) glucoBOX complex. Tetrahedron Asymmetr. 22, 1169–1175, https://doi.org/10.1016/j.tetasy.2011.06.012 (2011).
Vázquez-Villa, H., Reber, S., Ariger, M. A. & Carreira, E. M. Iridium Diamine Catalyst for the Asymmetric Transfer Hydrogenation of Ketones. Angew. Chem., Int. Ed. 50, 8979–8981, https://doi.org/10.1002/anie.201102732 (2011).
Boobalan, R., Lee, G.-H. & Chen, C. Copper Complex of Aminoisoborneol Schiff Base Cu2(SBAIB-d)2: An Efficient Catalyst for Direct Catalytic Asymmetric Nitroaldol (Henry) Reaction. Adv. Synth. Catal. 354, 2511–2520, https://doi.org/10.1002/adsc.201200337 (2012).
Fernandes, R. A. & Chaudhari, D. A. Development of the First Menthane-Based Chiral Bis(π-allylpalladium) Catalysis: Asymmetric Allylation of Imines. Eur. J. Org. Chem. 2012, 1945–1952, https://doi.org/10.1002/ejoc.201101588 (2012).
Fernandes, R. A. & Nallasivam, J. L. Enantioselective allylation of imines catalyzed by newly developed (-)-beta-pinene-based pi-allylpalladium catalyst: an efficient synthesis of (R)-alpha-propylpiperonylamine and (R)-pipecolic acid. Org. Biomol. Chem. 10, 7789–7800, https://doi.org/10.1039/c2ob26188j (2012).
Li, J.-L., Liu, L., Pei, Y.-N. & Zhu, H.-J. Copper(II)-containing C 2 -symmetric bistetracarboline amides in enantioselective Henry reactions. Tetrahedron 70, 9077–9083, https://doi.org/10.1016/j.tet.2014.10.005 (2014).
Moragas, T., Churcher, I., Lewis, W. & Stockman, R. A. Asymmetric Synthesis of Trisubstituted Aziridines via Aza-Darzens Reaction of Chiral Sulfinimines. Org. Lett. 16, 6290–6293, https://doi.org/10.1021/ol502967x (2014).
Petrone, D. A., Yoon, H., Weinstabl, H. & Lautens, M. Additive Effects in the Palladium-Catalyzed Carboiodination of Chiral N-Allyl Carboxamides. Angew. Chem., Int. Ed. 53, 7908–7912, https://doi.org/10.1002/anie.201404007 (2014).
Sheldrick, G. M. SHELXT - Integrated space-group and crystal-structure determination. Acta Crystallogr A 71, 3–8, https://doi.org/10.1107/S2053273314026370 (2015).
Taylor, M. S. & Jacobsen, E. N. Asymmetric catalysis by chiral hydrogen-bond donors. Angew. Chem., Int. Ed. 45, 1520–1543, https://doi.org/10.1002/anie.200503132 (2006).
Yu, X. H. & Wang, W. Hydrogen-bond-mediated asymmetric catalysis. Chem.-Asian J. 3, 516–532, https://doi.org/10.1002/asia.200700415 (2008).
Menguy, L. & Couty, F. Azetidine-derived bifunctional organocatalysts for Michael reactions. Tetrahedron-Asymmetr. 21, 2385–2389, https://doi.org/10.1016/j.tetasy.2010.08.003 (2010).
Zhang, L. et al. Novel Cinchona-Aminobenzimidazole Bifunctional Organocatalysts. Adv. Synth. Catal. 351, 3063–3066, https://doi.org/10.1002/adsc.200900787 (2009).
CrysAlisPro. Version 1.171.36.28, Agilent Technologies (2013).
Hooft, R. W. W. COLLECT Data Collection Software. COLLECT Data Collection Software (1998, ed. Nonius, B.V., Delft, The Netherlands).
Sheldrick, G. M. SADABS. (Bruker AXS Inc., 2007).
Burla, M. C. et al. SIR2004: an improved tool for crystal structure determination and refinement. J. Appl. Crystallogr. 38, 381–388, https://doi.org/10.1107/S002188980403225X (2005).
Yoshizawa, A., Feula, A., Male, L., Leach, A. G. & Fossey, J. S. Rigid and concave, 2,4-cis-substituted azetidine derivatives: A platform for asymmetric catalysis. ChemRxiv, https://doi.org/10.26434/chemrxiv.5887243.v5887241, https://doi.org/10.26434/chemrxiv.5887243.v1 (2018).
Acknowledgements
The University of Birmingham are thanked for support, including travel funds permitting AY and AF to undertake overseas training placements within the CASE network, and funds for AY to attend a crystallography summer school at the University of Durham. AGL thanks Liverpool John Moores University for support. JSF would like to thank the Royal Society for an Industrial Fellowship and the EPSRC for funding (EP/J003220/1). Dr Paul W. Davies is thanked for helpful suggestions during the preparation of this manuscript. Dr Etienne Baranoff is thanked for helpful suggestions during the course of conducting this research. Yiming Zhao is thanked for help in finalising the manuscript. Dr Chi Tsang and Dr. Peter Ashton are thanked for assistance with mass spectrometry. Dr Cécile S. Le Duff and Dr. Neil Spencer gave advice on aspects of NMR spectroscopy underpinning the preliminary findings. Dr Philip Craven is thanked for providing valuable insight and training relating to informatics.
Author information
Authors and Affiliations
Contributions
All authors contributed critically to conducting the research presented through making suggestions, analysing data and providing intellectual input into the direction of the investigation. Individual contributions in alphabetical order: A.F.: Initially evidenced the Ellman auxiliary approach to asymmetric azetidine synthesis, reported elsewhere, conducted the experiments of the organocatalysis section, see ESI, prepared complex 7 and recorded and analysed the data corresponding to these aspects; J.S.F.: Conceived and led the project, made critical decisions, interpreted data, undertook the analysis by SambVca and wrote the majority of the text in the manuscript; A.G.L.: Performed all of the computational analyses and data interpretation pertaining to copper catalysis and wrote aspects of the manuscript; The EPSRC UK National Crystallography Service at the University of Southampton are thanked for the collection of the crystallographic data of compound 7; L.M.: Analysed the single crystal X-ray diffraction study of compound 7; A.Y.: Synthesised single enantiomer ligands 1a to 1o, conducted all copper-catalysed reactions, analysed the data thereof and wrote aspect of the manuscript.
Corresponding authors
Ethics declarations
Competing Interests
The authors declare no competing interests.
Additional information
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Yoshizawa, A., Feula, A., Male, L. et al. Rigid and concave, 2,4-cis-substituted azetidine derivatives: A platform for asymmetric catalysis. Sci Rep 8, 6541 (2018). https://doi.org/10.1038/s41598-018-24784-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-018-24784-3
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
