Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Jul 31;146(32):22629–22641. doi: 10.1021/jacs.4c07117

Enantioselective Nitrene Transfer to Hydrocinnamyl Alcohols and Allylic Alcohols Enabled by Systematic Exploration of the Structure of Ion-Paired Rhodium Catalysts

Nicholas J Hodson 1, Shotaro Takano 1, Alexander Fanourakis 1, Robert J Phipps 1,*
PMCID: PMC11328136  PMID: 39083568

Abstract

graphic file with name ja4c07117_0011.jpg

This work describes highly enantioselective nitrene transfer to hydrocinnamyl alcohols (benzylic C–H amination) and allylic alcohols (aziridination) using ion-paired Rh (II,II) complexes based on anionic variants of Du Bois’ esp ligand that are associated with cinchona alkaloid-derived chiral cations. Directed by a substrate hydroxyl group, our previous work with these complexes had not been able to achieve high enantioselectivity on these most useful short-chain compounds, and we overcame this challenge through a combination of catalyst design and modified conditions. A hypothesis that modulation of the linker between the anionic sulfonate group and the central arene spacer might provide a better fit for shorter chain length substrates led to the development of a new biaryl-containing scaffold, which has allowed a broad scope for both substrate classes to be realized for the first time. Furthermore, we describe a systematic structural “knockout” study on the cinchona alkaloid-derived chiral cation to elucidate which features are crucial for high enantioinduction. De novo synthesis of modified scaffolds led to the surprising finding that for high ee the quinoline nitrogen of the alkaloid is crucial, although its location within the heterocycle could be varied, even leading to a superior catalyst. The free hydroxyl is also crucial and should possess the naturally occurring diastereomeric configuration of the alkaloid. These findings underline the privileged nature of the cinchona alkaloid scaffold and provide insight into how these cations might be used in other catalysis contexts.

1. Introduction

Nitrogen-containing molecules are ubiquitous, and it is important to continue to develop and refine amination methods. Nitrenes are, in principle, attractive reactive nitrogen-based intermediates, which can engage in a range of reactions.1 Free nitrenes themselves, however, are typically too reactive to allow controlled, selective intermolecular processes. In contrast, catalytic approaches that proceed via metal nitrenoids have proven incredibly versatile in permitting a broad range of C–H amination and alkene aziridination processes through insertion into C–H bonds and addition to alkenes, respectively.2 A number of metals are proficient at forming metal nitrenoids, and among these, rhodium (II,II) paddlewheel complexes have proven very effective, and Rh-catalyzed amination and aziridination have become mainstream synthetic methods, aided by the popularization of Du Bois’ Rh2(esp)2 as an active and functional group tolerant catalyst.3 A continuing challenge in the field is that of developing generally applicable ligand strategies for asymmetric nitrene transfer.2 An established approach for rhodium paddlewheels involves the replacement of the achiral carboxylate ligands with chiral variants,4 as demonstrated in early work on amination of benzylic C–H bonds from Hashimoto et al.,5 Müller et al.,6 Davies and Reddy,7 and Du Bois and Zalatan8 and followed by recent important intermolecular contributions from Dauban et al. (Figure 1a, left).9 Very recently, Miller et al. have developed novel peptide-derived carboxylate ligands, which, through a well-defined hydrogen bonding network, form a chiral pocket in which benzylic C–H amination occurs (Figure 1a, center).10−12 Applied to aziridination, chiral Rh carboxylate complexes have been recently showcased by Dauban and co-workers as being very effective on trisubstituted styrenes and terminal aliphatic alkenes.13,14 Using a distinct approach, Rh(III) Cp* complexes have also been used productively in enantioselective allylic amination15 and aziridination16 reactions. Stereoinduction using chiral carboxylate complexes generally relies on repulsive interactions alone. In comparison, substrate-directed catalysis involves a functional group on the substrate interacting with the catalyst to assist the organization at the transition state and can offer substantial advantages so long as the functional group is common and synthetically useful.17 This has been explored by Bach and co-workers in the context of Rh-catalyzed enantioselective nitrene transfer: a dual hydrogen bonding interaction was designed between a lactam substrate and a modified version of Rh2(esp)2 appended with its own chiral lactam (Figure 1a, right).18 This permitted C–H amination with a maximum 74% ee(18a) and was subsequently also applied to aziridination for which the ee values were higher, but the substrate variability was limited to specific motifs.18b We recently disclosed a conceptually distinct strategy for substrate-directed asymmetric nitrene transfer whereby a derivative of Rh2(esp)2, rendered anionic by the addition of sulfonate groups, is ion-paired with a chiral cinchona alkaloid-derived cation.19 It is the latter that provides the chiral environment in which nitrene transfer occurs, and we envisioned that the quaternized cinchona scaffold would provide rich opportunities for engaging in attractive noncovalent interactions with the substrate (Figure 1b).20,21 We anticipated that a hydroxyl group of the substrate may hydrogen bond with the sulfonate group of the ligand to create a high level of organization in this ternary system. In our initial proof-of-concept study, we demonstrated that such ion-paired Rh complexes were effective for enantioselective C–H amination at the benzylic position of 4-phenylbutanol and analogues (as depicted in Figure 1b).19a However, shortening or lengthening the alkyl chain separating the alcohol and the arene was detrimental. In particular, amination of hydrocinnamyl alcohol occurred in only 78% ee, a serious limitation given that amination of this broader substrate class leads to γ-amino alcohols, important chiral building blocks present in various pharmaceutical agents, as well as direct precursors to β-amino acids upon oxidation (Figure 1c, upper). We subsequently applied our ion-paired catalysts to the aziridination of alkenyl alcohols and obtained excellent results on a range of substrates including homoallylic alcohols and those with longer chains between the alkene and the alcohol.19b As before, a significant limitation was that the shortest chain length substrates, allylic alcohols, gave only a moderate ee (Figure 1c, lower). Yet, these would arguably be the most useful to produce small molecule chiral building blocks for synthesis (in analogy with the Sharpless asymmetric epoxidation for oxygen transfer, as opposed to nitrogen transfer). We hypothesized that these limitations might be addressed through judicious catalyst modification. The modularity of our ion-paired catalysts bestows several opportunities for variation both on the chiral cation and the achiral anionic dicarboxylate ligand. Until this point, we had made relatively limited variations, which focused on (1) the quaternizing benzyl group on the cation and (2) the geminal dialkyl groups adjacent to the carboxylate on the anionic ligand scaffold (Figure 1b, blue). Two key areas of the catalyst remain unexplored (Figure 1b, green). The first was the linker between the added sulfonate group and the backbone benzene ring in the “esp” ligand, which had always been a simple methylene group. We hypothesized that modifying this to reduce the distance between the sulfonate group and a rhodium center may enable shorter chain substrates to be better accommodated. The second is related to the array of structural features provided by nature in the cinchona alkaloid scaffold. Until this point, these had not been altered substantially and it was unclear which of these were crucial for selectivity and whether some could even be detrimental. These included the free hydroxyl group, the methoxy on the quinoline, the ethyl group on the quinuclidine, and, perhaps most intriguingly, the basic nitrogen of the quinoline ring. We herein describe a detailed study where these features are systematically knocked out to gain insight into those essential for high ee. Ultimately, we have been successful in identifying a modified catalyst scaffold, which allows valuable shorter-chain substrates to deliver high enantioselectivities in both nitrene-transfer processes: amination of hydrocinnamyl alcohols and aziridination of allylic alcohols (Figure 1d). During this process, we gained valuable insights into the features of the cation important for stereoinduction and identified an isomerized cation, which was combined with the new ligand scaffold to obtain an additive superior ee outcome. We envisage these insights will be of use to others who apply this privileged catalyst motif to enantioselective transition metal catalysis.

Figure 1.

Figure 1

Open in a new tab

Background and approach in this study.

2. Results and Discussion

2.1. Anionic Ligand Design and Effect of Additives

Our prior studies on aziridination uncovered several improvements to the original reaction conditions used in our C–H amination of aryl butanols.19a This included the use of a more soluble oxidant (C6F5IO vs PhIO), which allowed a lower reaction temperature (− 35 °C vs −25 °C). We also identified the beneficial impact on ee of a substoichiometric additive, C6F5I(OTFA)2, which served to release trifluoroacetic acid as the reaction progressed. Finally, the use of a Rh complex ligated by two pyridine molecules also gave an important increase in ee.19b Collectively these changes comprise what we now refer to as the “modified reaction conditions” (Figure 2, top scheme). Although the exact role of the latter two alterations was unclear, we speculated that they may modulate a binding equilibrium between the cation-based quinoline nitrogen and the axial coordination sites on the Rh-dimer. Throughout this article, we use a catalyst naming convention that highlights the four variables for a given Rh complex (Figure 2): (1) letters A-D represent the substitution α to the ligand carboxylate groups, (2) numerals I–V represent the different linkers between sulfonate and arene, (3) Cat1–5 represents which chiral cations are ion-paired to the Rh complex, and (4) “pyr” indicates that the complex is axially ligated by two pyridine molecules. Before evaluating novel ligand scaffolds, we first examined the modified reaction conditions on the amination of hydrocinnamyl alcohol using our previously reported catalysts (Figure 2). Using Rh2(A-I)2·(Cat1)2·(pyr)2, the aminated product 2a was obtained in 88% ee (Figure 2, chart, I (modified conditions)), considerably higher than the 73% ee obtained with the analogous complex under the original C–H amination conditions (Figure 2, chart, I (original conditions)). We evaluated the modified conditions with the corresponding complexes in which the α-carboxylate alkyl groups are cyclized in differing ring sizes (B–D, right-hand box of the chart) and saw an ee increase of between 10 and 20% in all cases compared with the original conditions (blue line versus orange line), peaking with scaffold C at 89% ee. For this complex Rh2(C–I)2·(Cat1)2·(pyr)2 a detailed analysis of the effect of each change moving from the original to modified conditions was conducted (chart, pink insert). The use of C6F5IO at lower temperatures gave only a slight increase in ee (73 to 75%). The inclusion of C6F5I(OTFA)2 as an additive afforded a similar ee value (75 to 74%); however, using a pyridine-ligated Rh complex in addition to C6F5IO afforded a large increase (75 to 87% ee). Finally, the inclusion of the C6F5I(OTFA)2 additive to afford the full modified conditions gave a small further increase to 89% ee.

Figure 2.

Figure 2

Open in a new tab

Schematic detail of structural variations in anionic Rh dimer complexes, chiral cations used in this study, and effects of previously used versus modified reaction conditions. aAntipode of 2a obtained with DHQ-derived Cat2.

Although we were pleased that the modified reaction conditions significantly improved ee in the C–H amination of hydrocinnamyl alcohol, we sought further improvement through catalyst design to allow the highest levels of selectivity to be obtained. In our prior work, the methylene linker between the central arene and sulfonate group was consistent across all catalysts (Figure 2, catalyst scaffolds, I). We hypothesized that modulation of the distance between the pendant sulfonate group and the Rh–Rh axis may enable shorter chain substrates to be accommodated with high selectivity. To probe this, we designed a series of scaffolds in which the linker between the arene and the sulfonate was systematically varied. The linker was removed entirely in II, directly attaching the sulfonate group to the arene. In III and IV an ortho-substituted arene is used as the linker, with IV featuring an additional methylene spacer before the sulfonate. A naphthyl group is used to extend the aromatic linker in V while retaining rigidity. Once the dicarboxylic acids of each were accessed (using prototypical scaffold A featuring gem-dimethyl groups), the dimer assembly and ion-exchange steps were carried out to obtain the final ion-paired catalysts. For completeness, we synthesized dihydroquinidine (DHQD)- and dihydroquinine (DHQ)-derived cations (Cat1 and Cat2) for all scaffolds using the same quaternizing group to probe any enantioselectivity gap between the two pseudoenantiomeric cations for each. The outcomes of these catalysts in the amination of hydrocinnamyl alcohol are shown in the left-hand box of the chart. The poorest outcome was obtained with naphthyl-linked V which afforded 2a in only 62% ee using Cat1 (vs 88% with the original methylene-linked scaffold I). Scaffold II, bearing no linker, also performed poorly giving only 71% ee. We were pleased to observe that biaryl scaffold III gave an increase in ee to 92%, constituting the best catalyst so far. Extending this linker by an additional methylene in IV resulted in a drop in ee to 79%. Interestingly, the relative performance of DHQD- and DHQ-derived catalysts varied significantly between scaffolds: for V, II, and IV the DHQ-derived complexes gave higher ee than the DHQD-derived, in contrast to original scaffold I where DHQD was superior (the relative ordering of complexes in terms of ee remained unchanged). Gratifyingly, the new biaryl scaffold III was unique in exhibiting no ee difference between the two pseudoenantiomeric complexes: 92% ee with cation Cat1 and (−)92% ee with cation Cat2. This represents a considerable advantage, as each product enantiomer is now accessible with equally high ee (a tedious process of devinylation of the alkaloid was required to access the same magnitude for substrates examined in our previous work). With a new optimal scaffold identified in III, the various cyclic surrogates for the gem-dimethyl group (B–D) were synthesized and tested (right-hand box, purple line). Only the cyclopentyl analogue C was competitive in ee, but a lower yield of 2a meant that the simpler gem-dimethyl complex (Rh2(A-III)2·(Cat1)2·(pyr)2) was taken forward. A brief survey of quaternizing groups on the cation (inset Table) showed that removal of the peripherial tBu groups (Cat3, entry 2) and enlargement to SiEt3 groups (Cat4, entry 3) were both slighly detrimental to ee. Interestingly, the use of NH2Tces as the amine source in place of the perfluorinated sulfamate ester used thus far gave similarly high levels of enantioselectivity (entry 4), which could be slightly improved to 93% ee by switching to cation Cat4 (entry 5). The use of NH2Tces as an alternative aminating agent is advantageous due to its simple and mild deprotection conditions.22 The optimal ion-paired catalyst Rh2(A-III)2·(Cat4)2·(pyr)2 shall now be referred to as Rh1 for brevity.

2.2. Scope Evaluation for C–H Amination

With NH2Tces as an aminating agent, we evaluated the scope of substituted hydrocinnamyl alcohols using Rh1 (Scheme 1). We were pleased to observe that high reactivity and enantioselectivity could be obtained for a range of ortho-substituted substrates, which have proven very challenging in other benzylic C–H amination protocols, including in our previous work. This encompassed ortho methyl, methoxy, chloro, and bromo substituents (2c–2f). Meta-substitution was well tolerated, including alkyl groups (2g, 2h) and inductively (2i) as well as conjugatively (2j) withdrawing groups. A naphthyl substrate was compatible (2k) as were a range of para-substituted hydrocinnamyl alcohols (2l-2o). Lower yields for 2i and 2o can be attributed to the strongly withdrawing CF3 group deactivating the benzylic C–H bonds toward nitrene insertion. Substitution on the alkyl chain could be incorporated on the central carbon, with 2p and 2q obtained in very high ee (98 and 97%) despite the amination occurring on a very hindered methylene. In contrast, trace amounts of 2p and 2q were afforded when Rh2(esp)2 was used as the catalyst. Tertiary alcohol directing groups could also be used with only a minimal impact on ee (2r and 2s). The antipodal product 2b-ent was obtained using the DHQ-derived version of Rh1, Rh2(A-III)2·(Cat5)2·(pyr)2 (from herein abbreviated as Rh2), in high ee. Finally, the Tces protecting group was removed to give 3 from which the absolute configuration could be determined by comparison with the literature and all others assigned by analogy. A benefit of the Tces protecting group was that amino alcohols 2 were often solids, and a simple recrystallization procedure could access 2b in >99% ee.

Scheme 1. Enantioselective C–H Amination of Hydrocinnamyl Alcohols,

Scheme 1

Open in a new tab

ee determined by chiral SFC analysis of the purified amino alcohol product.

NHTces = HNSO3CH2CCl3.

Using 2.0 mol % catalyst at −25 °C.

We next investigated heterocyclic substrates, focusing first on sulfur heterocycles (Scheme 2a). A 2-substituted thiophene gave excellent results (2t) as did both 5- and 2-substituted benzothiophenes (2u, 2v). In moving to more electron-rich heteroarenes, we observed an intriguing switch in chemoselectivity using Rh1 compared with Rh2(esp)2 (Scheme 2b). For pyrrole-containing 1w, aziridination on the heteroarene followed by ring-opening afforded 4a as the major product with Rh2(esp)2. In contrast, Rh1 gave benzylic amination as the major product in high ee (2w). The low mass balance of the reaction is attributed to the apparent decomposition of side products. The same trend was observed for benzofuran-containing 1x and furan-containing 1y. In both cases, the benzylic amination products 2x and 2y were obtained with Rh1, in contrast to the dearomatized products 4b and 4c obtained with Rh2(esp)2. These divergent chemoselectivity outcomes highlight the powerful control our ion-paired catalysts have on the reaction outcome is not limited to enantioselectivity, a benefit of the substrate-directed strategy employed which presumably directs the reaction to a site that would not be typically functionalized using a nondirected approach. We next sought to challenge the control of diastereoselectivity as well as enantioselectivity in achiral secondary alcohol 1z (Scheme 2c). Control of both metrics was very high, with essentially a single diastereomer obtained in excellent yield and ee (2z). Next, we examined how the catalyst might deal with substrates featuring an existing, defined stereocenter. Commercially available (R)-1za, which gave 2zaa in moderate d.r. of 4:1 using Rh2(esp)2, gave 2zaa in >20:1 d.r. using Rh1, indicating a strong match between intrinsic selectivity and the catalyst. In contrast, the use of pseudoenantiomeric Rh2 (featuring the DHQ-derived cation Cat5 rather than DHQD-derived Cat4) gave 2zab in 7:1 d.r. in favor of the previously minor diastereomer, demonstrating the powerful ability of our ion-paired catalyst to override an existing stereocenter. We also evaluated (R)-1zb, featuring an ester rather than a methyl at the alcohol stereocenter and with the opposite absolute stereochemistry. In this case, the matched catalyst was Rh2, which gave 2zbb in 15:1 d.r. in the same sense as Rh2(esp)2 (3:1 d.r.). With Rh1, the overriding effect of the catalyst was not as strong, but the previously minor diastereomer 2zba could be obtained in 2:1 d.r. The 1,3-amino alcohol relationship in our products provides the opportunity to access valuable azetidine scaffolds through the facile intramolecular displacement of the alcohol. To showcase this, we utilized a mesylation/cyclization procedure to access several azetidines 5a–5d bearing different substitution patterns (Scheme 2d). Tertiary alcohol 2s was cyclized under Mitsunobu conditions to afford the spirocycle 5e, with a lower yield in this case due to competing elimination pathways.

Scheme 2. Heterocyclic and Secondary Alcohol Substrates.

Scheme 2

Open in a new tab

As the new catalyst scaffold III combined with the modified reaction conditions had been successful on hydrocinnamyl alcohols, we questioned whether other chain lengths from our initial study may also be improved.19a We found that the updated conditions together with the new catalyst Rh1 improved both yield and ee on 4-arylbutanol substrates (Scheme 3, 7a–7c). For the longer chain 5-phenylpentanol, the “original” conditions and catalyst gave 7d with only 76% ee.19a Use of the new biaryl catalyst Rh1 and modified conditions increased this to 84% ee. However, we found that ee could be improved to 92% by using the previously identified benzyl sulfonated catalyst Rh2(B-I)2·(Cat1)2·(pyr)2. Several substrates were examined successfully (7d–7f). This catalyst matching reinforces that the biaryl linker is most suitable for shorter chain lengths, while the original benzyl linker is still optimal for longer chains, in line with our original hypothesis.

Scheme 3. Application of Modified Catalytic System to Aryl Butanols and Aryl Pentanols.

Scheme 3

Open in a new tab

Previously reported with H2NSO3CH2C3F7 as the aminating agent instead of H2NTces, under the “original” reaction conditions: catalyst (1.0 mol %), PhIO (2.0 equiv), H2NSO3CH2C3F7 (1.2 equiv), 1,3-difluorobenzene (0.2 M), −25 °C.

2.3. Aziridination of Allylic Alcohols

In our previous aziridination work, Rh2(B-I)2·(Cat4)2·(pyr)2 was the optimal catalyst for nitrene transfer to homoallylic, bishomoallylic, and trishomoallylic alkenyl alcohol substrates.19b Disappointingly, we observed a significant drop in ee for allylic alcohols, arguably the most useful class of the set (Figure 1c). Given the prevalence and utility of the allylic alcohol motif, we decided to evaluate the most promising of our new catalyst scaffolds for their enantioselective aziridination and selected the trisubstituted alkene prenol for optimization (8a) (Scheme 4). Across α-carboxylate groups A–D, our previous benzyl sulfonate scaffold I delivered aziridine 9a in the range 55–66% ee (Scheme 4, chart). We thus explored the biaryl scaffold III and immediately observed an improvement. Across groups A–D, III consistently gave 10–20% higher ee than the counterparts with I, peaking with A-III, which gave 76% ee. This greater increase compared to the shorter-chain C–H amination may reflect the lower flexibility of the allylic alcohol, meaning that the biaryl linker is more important to effectively position the chiral pocket for allylic alcohol aziridination. The use of Tces as aminating agent gave inferior enantioselectivity in this case (see SI for details). As in the C–H amination, extending III to give scaffold IV resulted in poorer performance, giving 9a in only 47% ee. Returning to III, the slightly larger chiral cation Cat4 gave a small improvement in ee (Table, entry 2 vs 1). Employing 2.0 mol % of catalyst at −78 °C using Rh1 enabled an ee of 88% (entry 3), and we found that high reactivity could be maintained by using a larger excess of aminating agent and a longer reaction time (entry 4).

Scheme 4. Catalyst Comparison on Allylic Alcohol Aziridination.

Scheme 4

Open in a new tab

Having identified suitable conditions for the aziridination of prenol, we next explored the scope of the reaction on related trisubstituted allylic alcohols using the optimal catalyst Rh1 (Scheme 5). Longer alkyl chains were well tolerated (9b) as well as a cyclohexyl ring (9c) and a larger cyclododecyl (9d) ring. Additional functional groups could be included at the 4-position of the cyclohexyl ring; a gemdifluoro (9e), 4H-pyran (9f), and protected ketone (9g) also gave highly enantioenriched aziridines. Rh1 could accommodate a trans-alkyl substrate with good levels of ee (9h); however, competing oxidation of the allylic alcohol to the corresponding enal, presumably occurring due to slower aziridination, precluded a high yield under the present conditions. The corresponding cis isomer also gave poor yield but lower ee (see SI for details). Although yields were high, styrenyl allylic alcohols did not give good enantioselectivity outcomes under similar conditions (see SI for details). High enantioselectivity was maintained for cyclopentyl and cycloheptyl examples; however, degradation of the aziridine upon isolation was observed, and these were transformed to the corresponding thiophenol adducts 10a and 10b to enable characterization. Next, we examined the ability of Rh1 to control site selectivity as well as enantioselectivity in the aziridination of nerol and geraniol. Although Rh2(esp)2 also has a slight preference for the proximal alkene b in both cases, Rh1 gave considerably higher ratios of 16:1 (geraniol) and 18:1 (nerol) while affording the desired aziridines (9i and 9j) in high ee.23

Scheme 5. Scope of Allylic Alcohol Aziridination.

Scheme 5

Open in a new tab

In all cases, ee was determined by chiral SFC analysis of the purified thiophenol adduct.

Degradation of these aziridines occurred upon attempted purification, see the Supporting Information for details.

1.05 equiv of sulfamate ester used.

2.4. Evaluation of Cation Structural Features

Having thoroughly evaluated changes to the achiral portion of the complex, we turned our attention to the systematic variation of the chiral cation. Throughout our studies so far using ion-paired catalysts, we have relied on chiral cations derived from naturally occurring quinine and quinidine. Due to the good performance in a range of substrates our ventures into scaffold variation had been mainly limited to the nature of the aromatic N-quaternizing group. However, there are several distinctive structural features possessed by these alkaloids, and we were intrigued to probe their importance in a systematic manner by “building up” the features from a simpler version to gauge the impact on selectivity. These features comprise the following: the methoxy group on the quinoline ring, the ethyl group on the quinuclidine, the free hydroxyl group adjacent to the quaternized quinuclidine, and, perhaps most challengingly, the basic nitrogen of the quinoline (Figure 3a). The latter in particular demands critical assessment, as the inclusion of a basic heteroarene into the cation architecture would certainly be a counterintuitive move in any de novo design due to the possibility of deleterious ligation of the metal center. In our first study involving these catalysts, we established that axial ligation of the Rh dimer (via quinoline N-ligation) was occurring in solution, as evidenced by distinctive bands in the UV–visible spectra.19a In subsequent work on asymmetric aziridination, we found that the use of a Rh complex ligated by two molecules of pyridine increased ee by ∼10%.19b We found a similar increase in enantioselectivity through the addition of a substoichiometric amount of a weak acid additive, specifically C6F5I(OTFA)2, which releases trifluoroacetic acid in situ. We supposed that these two factors together might influence a putative binding equilibrium between the Rh complex and cation. For this reason, we were particularly intrigued to remove the quinoline nitrogen from the cation to assess whether this would negate the requirement for the various additives and even improve selectivity; perhaps, these were needed to “correct” problems stemming from the presence of the quinoline nitrogen in the parent alkaloid?

Figure 3.

Figure 3

Open in a new tab

Synthesis of structurally varied Cinchona alkaloid analogues.

Addressing this question necessitated a de novo synthetic approach to nonheteroaromatic cinchona analogues. Lygo and co-workers previously asked similar questions in the context of enolate alkylation using phase transfer catalysis and devised a route, based on Sharpless Asymmetric Dihydroxylation (AD),24 to a simplified cinchona scaffold, which lacked the vinyl group and permitted the quinoline to be replaced with various arenes at an early stage.25,26 We have modified Lygo’s approach to use Suzuki-Miyaura coupling for the introduction of the aromatic fragment,27 with the final route shown in detail for the synthesis of cation Cat6 in which the 6-methoxyquinoline of the quinine/quinidine alkaloids is replaced by a naphthalene (Figure 3b). This route allowed access to the following further cations, depicted in Figure 3c: Cat7, which lacks the quinoline nitrogen but retains the methoxy on the naphthalene, and Cat8, which isomerizes the 4-substituted quinoline of the alkaloids to a 5-substituted analogue. We attempted to access the corresponding isoquinoline analogue of Cat8 but the Sharpless AD delivered only moderate ee in this case, precluding its evaluation. In addition to these de novo synthesized scaffolds, we accessed others of interest from the derivatization of the alkaloids (Figure 3d, see the SI for details). This included versions in which the ethyl and methoxy groups are removed (Cat9), just the ethyl is (Cat10), a variant in which an n-butyl group is introduced at the quinoline C2 position (Cat11), one in which the hydroxyl group is methylated (Cat12) and one in which the hydroxyl stereochemistry is inverted (Cat13).

For consistent comparison, all of the above were quaternized using the 3,5-bis-tertbutyl-substituted m-terphenyl benzylating group possessed by original cations Cat1 and Cat2 at the outset of this work. Similarly, all were paired with the anionic Rh dimer scaffold Rh2(A-I)2 which bears a gem-dimethyl as the geminal dialkyl group and a benzyl sulfonate linker. However, the method of chiral ion-pairing required variation depending on the cation used (Figure 3b, lower). Typically, we perform cation exchange by simply stirring a slight excess of the sodium salt of the bis-sulfonated Rh complex with chiral cation bromide in a biphasic solvent mixture. The correct ratio between the two is usually easily obtained in the organic layer and the exchange is associated with a striking solution color change from green to red, consistent with the axial ligation of the Rh complex, presumably by the quinoline nitrogen of the cation. This proceeded as normal for cations in which the heteroaromatic nitrogen was present and unencumbered. In contrast, cations Cat6, Cat7, and Cat11 did not give complete ion exchange under these conditions and necessitated the formation of the silver salt of the Rh complex to drive the exchange to completion by precipitation of silver bromide. Furthermore, the solutions for the latter remained green after the exchange. These observations together suggest that when the quinoline nitrogen is not present (Cat6, Cat7) or is encumbered (Cat11), axial ligation by the cation is precluded.

We evaluated the Rh complexes bearing these cations in C–H amination of hydrocinnamyl alcohol, evaluating both pyridine-ligated and nonligated Rh complexes for each to examine if there was any divergence (Figure 4). In the first instance, the trifluoroacetic acid-releasing additive C6F5I(OTFA)2 was included (Figure 4b). We began with Cat6, the most simple cation scaffold in our study, in which the three intrinsic alkaloid features are “knocked out”: the basic quinoline nitrogen, the methoxy group at the quinoline 6-position and the quinuclidine-located ethyl. This complex gave a very poor outcome, delivering only 7% ee for the pyridine-ligated catalyst and 30% for the nonligated. Interestingly for the pyridine-ligated catalyst, the yield was also unusually low, suggesting that the pyridine may remain strongly bound in the absence of competitive binding from the cation (see the SI for yield information). We next introduced the methoxy group to the naphthalene in Cat7 and saw a small improvement with the ligated complex but overall was still poor. We next introduced the basic nitrogen to form a quinoline, but now, a 5-substituted quinoline than the 4-isomer of the natural alkaloids (Cat8). This gave a very large jump (91% ee with ligated complex and 82% with nonligated) and was in fact higher than the natural quinoline isomer Cat9, which delivered 84 and 48% ee for ligated and nonligated, respectively.28 Returning the methoxy group to the 6-position of the quinoline (Cat10) and adding the ethyl group to the quinuclidine (Cat1) gave minor increases to reach 89% ee (ligated) and 76% (nonligated). Having established that the quinoline nitrogen was crucial for high ee, seemingly irrespective of position, we next evaluated the 2-n-butyl-substituted cation Cat11. We anticipated that binding to Rh would be severely impaired (as evidenced by the lack of color change of the solution upon cation ion exchange) but that the quinoline should retain its basicity, potentially allowing a distinction between a binding effect and a basicity effect. Very interestingly, the ee was only slightly reduced in direct comparison with Cat1, at 84% (pyridine-ligated), which could be attributed to the quinoline nitrogen acting beneficially as a base. Finally, we evaluated cation Cat12 in which the hydroxyl is methylated and were surprised to see that this favored formation of the opposite product enantiomer with moderate ee (−44% and −24%, ligated and nonligated). Such a dramatic shift hints at a very important role for the cation hydroxyl group in the presumed network of noncovalent interactions at the transition state.29Cat13 differs from Cat1 by being epimeric at the hydroxyl group and this also gave the opposite enantiomer in −77 and −61% ee for ligated and nonligated, respectively. Although relatively high, these ee values are still 10–15% lower than those with the natural diastereomeric configuration possessed by the alkaloids.

Figure 4.

Figure 4

Open in a new tab

Evaluation of Rh complexes containing (a) cations Cat1 & Cat6-Cat13. (b) C–H Amination of hydrocinnamyl alcohol with 10 mol % of C6F5I(OTFA)2 as additive. (c) C–H Amination of hydrocinnamyl alcohol without C6F5I(OTFA)2 additive. aAntipode of 2a obtained due to Cat10 having the opposite aminoalcohol configuration compared with the others.

We also performed an analogous analysis for aziridination, using a bishomoallylic alcohol that had been optimized extensively in our previous study.19b Very similar trends were observed across the cations, confirming for this reaction too the clear need for the quinoline nitrogen, irrespective of the specific isomer (Cat8 or Cat9) used (see the SI for details).

The outcome of this investigation demonstrated that the quinoline nitrogen possessed by the cation is essential for obtaining high ee in this chemistry. Furthermore, the precise location of this basic nitrogen seems unimportant as complexes containing Cat8 and Cat9 provided very similar enantioselectivities in direct comparisons, with the non-natural isomer Cat8 even providing slightly better results (vide infra). The free hydroxyl with the correct stereochemistry was similarly crucial. The remaining features of the natural cinchona scaffold (methoxy on the quinoline ring, ethyl on the quinuclidine) provided small degrees of tuning, which varies from one substrate to another, the effect being nowhere near as dramatic. The tolerance of the location of the quinoline nitrogen combined with the fact that ee is largely retained when an alkyl group is introduced adjacent to it led us to suppose that the quinoline nitrogen effect may be acting as a weak base, which is needed to increase ee. In line with this, we have established that the acid-releasing additive C6F5I(OTFA)2 in substoichiometric amounts boosts ee in this system and others.19c For this reason, we returned to the original C–H amination reaction and evaluated all of the complexes in the absence of C6F5I(OTFA)2 (Figure 4c). Informative trends here are more challenging to elucidate, but it is clear that inclusion of the quinoline nitrogen has a less pronounced effect on ee under these conditions. The quinoline effect is substantial in direct comparison between Cat6 and Cat9, albeit only for the pyridine-ligated complex. For the analogous comparison between Cat7 and Cat10, there was little impact on ee with nitrogen inclusion and this held true for both ligated and nonligated complexes. What we had previously deemed “ancillary” features of the cation in the presence of C6F5I(OTFA)2 now seem to have a larger effect in its absence. For instance, the inclusion of the quinoline methoxy reduces ee by around 20% (Cat9 to Cat10), but the inclusion of the quinuclidine ethyl group increased it by a remarkable 30% (Cat10 to Cat1). Although high ee values can be obtained in the absence of the C6F5I(OTFA)2 additive, tolerance of variations in the cation structure is greater in its presence, allowing high ee values to be obtained more generally. This leads to a more general, highly enantioselective reaction in its presence, and we believe that protonation of the quinoline nitrogen seems likely to play a role, potentially impacting the conformation of the cation.30 Our studies have shown that introducing the Rh-complex with axially ligating pyridines has a consistently beneficial effect (as seen during optimization, Figure 2 graph inset, and reinforced here) of typically 10–20% ee in most cases.31 A summary of the key findings in this cation exploration is given in Scheme 6a.

Scheme 6. Summary of Cation Exploration Findings and Amination Using a Catalyst Containing Cat8 and Biaryl Linker.

Scheme 6

Open in a new tab

Finally, we were intrigued by the observation that the complex containing Cat8, in which the nitrogen of the quinoline has been transposed to the non-natural position, provided an edge in ee terms compared with Cat9, its direct comparator containing the quinoline isomer produced by nature. In the amination of 2a it gave 91% ee when combined with scaffold I, which possessed the original benzyl sulfonate linker. Given the increase we had seen in the present work with the improved biaryl scaffold III, we wondered whether combining novel cation Cat8 with the new linker might raise enantioselectivity even further. Upon testing this, we were pleased to find that 2a could be obtained in 95% ee, the highest we have obtained for this compound so far throughout our studies, demonstrating that the improvements to scaffold and cation identified during this study are additive (Scheme 6b). We then selected two substrates from our main scope, 2k and 2s, that gave slightly lower than optimal ee using Rh1 and re-examined them with the complex incorporating Cat8. We were very happy to find that, in both cases, the ee values were increased, from 84 to 89% and 83 to 93% (Scheme 6c). While the synthesis of Cat8 is significantly longer compared with the alkaloid-derived Cat4 that is possessed by Rh1, we hope that this demonstration will be of value in certain circumstances where Rh1 may provide insufficient selectivity.

3. Conclusions

In this study, we have developed an effective catalyst system for enantioselective nitrene transfer to hydrocinnamyl alcohols (C–H amination) and allylic alcohols (aziridination). This expansion of highly enantioselective C–H amination and aziridination allows access to small, densely functionalized chiral building blocks that we anticipate will be of great utility in synthesis. These reactions are enabled by the directing effect of the substrate alcohol combined with highly effective ion-paired Rh complexes that we have developed that incorporate cinchona alkaloid-derived cations. In our previous work, these shorter chain length substrates gave inadequate enantioselectivity and we have developed a superior catalyst scaffold in which the sulfonate group is separated from the catalyst backbone by an ortho-biaryl linker. Both reactions lead to valuable chiral building blocks with excellent enantioselectivities which are amenable to further functionalization for a range of possible applications, for example, in the synthesis of highly substituted azetidines. A key component of the ion-paired catalysts, which had not been previously explored, beyond variation of the N-quaternizing group, was the cinchona alkaloid-derived chiral cation; we had until this point retained many natural features of the alkaloid without rigorous analysis. We carried out a “knockout” and systematic reinclusion study, which included investigating the quinoline nitrogen of the alkaloid through de novo asymmetric synthesis. While features such as the 6-methoxy on the quinoline and the ethyl of the quinuclidine had a minor impact, the free hydroxyl group, with the stereochemistry possessed by the natural alkaloid, was found to be profoundly important. Just as important was the presence of the quinoline nitrogen although its precise position was unimportant and steric obstruction of it was inconsequential. While these observations are difficult to reconcile into a unified stereochemical model, it is clear that the complex environment provided by the quaternized alkaloid provides rich opportunities for productive interactions with the substrate during enantiodetermining nitrene transfer and under the optimal conditions the basic nitrogen of the cation plays an important role. It is somewhat remarkable that the chemical structure provided by nature turns out to have several key features required for a purely synthetic reaction. On the other hand, the cinchona alkaloids have surprised practitioners of asymmetric catalysis time and time again with the diversity of situations in which they have been highly effective. We hope these insights will be of use in the further application of these complexes to asymmetric nitrene transfer and beyond, for example in carbene transfer methodology.

Acknowledgments

We are grateful to the EPSRC for a PhD studentship (N.J.H.), R&D Center of Mitsui Chemicals, Inc. (S.T.), Cambridge Trust and Wolfson College Cambridge for a Vice-Chancellor’s & Wolfson College Scholarship (A.F.), and to the European Research Council under the Horizon 2020 Program (starting grant no. 757381). We are grateful to Dr Kieran J. Paterson and Dr Benjamin D. Williams for performing several steps towards the synthesis of Rh2(A-II)2·(Cat1)2·(pyr)2.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.4c07117.

  • Additional optimization, full experimental details, and characterization data for compounds (PDF)

The authors declare no competing financial interest.

Supplementary Material

ja4c07117_si_001.pdf (35.1MB, pdf)

References

  1. a Dequirez G.; Pons V.; Dauban P. Nitrene Chemistry in Organic Synthesis: Still in Its Infancy?. Angew. Chem., Int. Ed. 2012, 51, 7384–7395. 10.1002/anie.201201945. [DOI] [PubMed] [Google Scholar]; b Wentrup C. Carbenes and Nitrenes: Recent Developments in Fundamental Chemistry. Angew. Chem., Int. Ed. 2018, 57, 11508–11521. 10.1002/anie.201804863. [DOI] [PubMed] [Google Scholar]
  2. a Müller P.; Fruit C. Enantioselective Catalytic Aziridinations and Asymmetric Nitrene Insertions into CH Bonds. Chem. Rev. 2003, 103, 2905–2920. 10.1021/cr020043t. [DOI] [PubMed] [Google Scholar]; b Collet F.; Lescot C.; Dauban P. Catalytic C–H amination: the stereoselectivity issue. Chem. Soc. Rev. 2011, 40, 1926–1936. 10.1039/c0cs00095g. [DOI] [PubMed] [Google Scholar]; c Uchida T.; Katsuki T. Asymmetric Nitrene Transfer Reactions: Sulfimidation, Aziridination and C–H Amination Using Azide Compounds as Nitrene Precursors. Chem. Rec. 2014, 14, 117–129. 10.1002/tcr.201300027. [DOI] [PubMed] [Google Scholar]; d Hayashi H.; Uchida T. Nitrene Transfer Reactions for Asymmetric C–H Amination: Recent Development. Eur. J. Org. Chem. 2020, 2020, 909–916. 10.1002/ejoc.201901562. [DOI] [Google Scholar]; e Ju M.; Schomaker J. M. Nitrene transfer catalysts for enantioselective C–N bond formation. Nat. Rev. Chem. 2021, 5, 580–594. 10.1038/s41570-021-00291-4. [DOI] [PubMed] [Google Scholar]; f Fanourakis A.; Phipps R. J. Catalytic, asymmetric carbon–nitrogen bond formation using metal nitrenoids: from metal–ligand complexes via metalloporphyrins to enzymes. Chem. Sci. 2023, 14, 12447–12476. 10.1039/D3SC04661C. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. a Espino C. G.; Fiori K. W.; Kim M.; Du Bois J. Expanding the Scope of C–H Amination through Catalyst Design. J. Am. Chem. Soc. 2004, 126, 15378–15379. 10.1021/ja0446294. [DOI] [PubMed] [Google Scholar]; b Fiori K. W.; Du Bois J. Catalytic Intermolecular Amination of C–H Bonds: Method Development and Mechanistic Insights. J. Am. Chem. Soc. 2007, 129, 562–568. 10.1021/ja0650450. [DOI] [PubMed] [Google Scholar]; c Roizen J. L.; Harvey M. E.; Du Bois J. Metal-Catalyzed Nitrogen-Atom Transfer Methods for the Oxidation of Aliphatic C–H Bonds. Acc. Chem. Res. 2012, 45, 911–922. 10.1021/ar200318q. [DOI] [PMC free article] [PubMed] [Google Scholar]; d Roizen J. L.; Zalatan D. N.; Du Bois J. Selective Intermolecular Amination of C-H Bonds at Tertiary Carbon Centers. Angew. Chem., Int. Ed. 2013, 52, 11343–11346. 10.1002/anie.201304238. [DOI] [PMC free article] [PubMed] [Google Scholar]; e Bess E. N.; DeLuca R. J.; Tindall D. J.; Oderinde M. S.; Roizen J. L.; Du Bois J.; Sigman M. S. Analyzing Site Selectivity in Rh2(esp)2-Catalyzed Intermolecular C–H Amination Reactions. J. Am. Chem. Soc. 2014, 136, 5783–5789. 10.1021/ja5015508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Hansen J.; Davies H. M. L. High symmetry dirhodium(II) paddlewheel complexes as chiral catalysts. Coord. Chem. Rev. 2008, 252, 545–555. 10.1016/j.ccr.2007.08.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Yamawaki M.; Tsutsui H.; Kitagaki S.; Anada M.; Hashimoto S. Dirhodium(II) tetrakis[N-tetrachlorophthaloyl-(S)-tert-leucinate]: a new chiral Rh(II) catalyst for enantioselective amidation of C-H bonds. Tetrahedron Lett. 2002, 43, 9561–9564. 10.1016/S0040-4039(02)02432-2. [DOI] [Google Scholar]
  6. a Nägeli I.; Baud C.; Bernardinelli G.; Jacquier Y.; Moraon M.; Müller P. Rhodium(II)-Catalyzed CH Insertions with 6aphenyl-λ3-iodane. Helv. Chim. Act. 1997, 80, 1087–1105. [Google Scholar]; b Fruit C.; Müller P. Intramolecular Asymmetric Amidations of Sulfonamides and Sulfamates Catalyzed by Chiral Dirhodium(II) Complexes. Helv. Chim. Act. 2004, 87, 1607–1615. 10.1002/hlca.200490148. [DOI] [Google Scholar]; c Fruit C.; Müller P. Asymmetric transfer of nitrenes catalyzed by chiral dirhodium(II) using aromatic sulfamate esters. Tetrahedron: Asym. 2004, 15, 1019–1026. 10.1016/j.tetasy.2004.01.015. [DOI] [Google Scholar]
  7. Reddy R. P.; Davies H. M. L. Dirhodium Tetracarboxylates Derived from Adamantylglycine as Chiral Catalysts for Enantioselective C–H Aminations. Org. Lett. 2006, 8, 5013–5016. 10.1021/ol061742l. [DOI] [PubMed] [Google Scholar]
  8. Zalatan D. N.; Du Bois J. A Chiral Rhodium Carboxamidate Catalyst for Enantioselective C–H Amination. J. Am. Chem. Soc. 2008, 130, 9220–9221. 10.1021/ja8031955. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. a Nasrallah A.; Boquet V.; Hecker A.; Retailleau P.; Darses B.; Dauban P. Catalytic Enantioselective Intermolecular Benzylic C(sp3)–H Amination. Angew. Chem., Int. Ed. 2019, 58, 8192–8196. 10.1002/anie.201902882. [DOI] [PubMed] [Google Scholar]; b Nasrallah A.; Lazib Y.; Boquet V.; Darses B.; Dauban P. Catalytic Intermolecular C(sp3)–H Amination with Sulfamates for the Asymmetric Synthesis of Amines. Org. Process Res. Dev. 2020, 24, 724–728. 10.1021/acs.oprd.9b00424. [DOI] [Google Scholar]; c Brunard E.; Boquet V.; Saget T.; Sosa Carrizo E. D.; Sircoglou M.; Dauban P. Catalyst-Controlled Intermolecular Homobenzylic C(sp3)–H Amination for the Synthesis of β-Arylethylamines. J. Am. Chem. Soc. 2024, 146, 5843–5854. 10.1021/jacs.3c10964. [DOI] [PubMed] [Google Scholar]
  10. van den Heuvel N.; Mason S. M.; Mercado B. Q.; Miller S. J. Aspartyl β-Turn-Based Dirhodium(II) Metallopeptides for Benzylic C(sp3)–H Amination: Enantioselectivity and X-ray Structural Analysis. J. Am. Chem. Soc. 2023, 145, 12377–12385. 10.1021/jacs.3c03587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. For examples of peptides as ligands for Rh(II/II) complexes, see:; a Sambasivan R.; Ball Z. T. Metallopeptides for Asymmetric Dirhodium Catalysis. J. Am. Chem. Soc. 2010, 132, 9289–9291. 10.1021/ja103747h. [DOI] [PubMed] [Google Scholar]; b Ball Z. T. Designing Enzyme-like Catalysts: A Rhodium(II) Metallopeptide Case Study. Acc. Chem. Res. 2013, 46, 560–570. 10.1021/ar300261h. [DOI] [PubMed] [Google Scholar]
  12. For selected examples of asymmetric intermolecular C-H amination through nitrene transfer catalyzed by other metals, see:; a Zhou X.-G.; Yu X.-Q.; Huang J.-S.; Che C.-M. Asymmetric amidation of saturated C–H bonds catalysed by chiral ruthenium and manganese porphyrins. Chem. Commun. 1999, 2377–2378. 10.1039/a907653k. [DOI] [Google Scholar]; b Kohmura Y.; Katsuki T. Mn(salen)-catalyzed enantioselective C-H amination. Tetrahedron Lett. 2001, 42, 3339–3342. 10.1016/S0040-4039(01)00427-0. [DOI] [Google Scholar]; c Nishioka Y.; Uchida T.; Katsuki T. Enantio- and Regioselective Intermolecular Benzylic and Allylic C-H Bond Amination. Angew. Chem., Int. Ed. 2013, 52, 1739–1742. 10.1002/anie.201208906. [DOI] [PubMed] [Google Scholar]; d Prier C. K.; Zhang R. K.; Buller A. R.; Brinkmann-Chen S.; Arnold F. H. Enantioselective, intermolecular benzylic C–H amination catalysed by an engineered iron-haem enzyme. Nat. Chem. 2017, 9, 629–634. 10.1038/nchem.2783. [DOI] [PMC free article] [PubMed] [Google Scholar]; e Jin L.-M.; Xu P.; Xie J.; Zhang X. P. Enantioselective Intermolecular Radical C–H Amination. J. Am. Chem. Soc. 2020, 142, 20828–20836. 10.1021/jacs.0c10415. [DOI] [PMC free article] [PubMed] [Google Scholar]; f Annapureddy R. R.; Jandl C.; Bach T. A Chiral Phenanthroline Ligand with a Hydrogen-Bonding Site: Application to the Enantioselective Amination of Methylene Groups. J. Am. Chem. Soc. 2020, 142, 7374–7378. 10.1021/jacs.0c02803. [DOI] [PubMed] [Google Scholar]; g Jia Z.-J.; Gao S.; Arnold F. H. Enzymatic Primary Amination of Benzylic and Allylic C(sp3)–H Bonds. J. Am. Chem. Soc. 2020, 142, 10279–10283. 10.1021/jacs.0c03428. [DOI] [PubMed] [Google Scholar]; h Athavale S. V.; Gao S.; Liu Z.; Mallojjala S. C.; Hirschi J. S.; Arnold F. H. Biocatalytic, Intermolecular C–H Bond Functionalization for the Synthesis of Enantioenriched Amides. Angew. Chem., Int. Ed. 2021, 60, 24864–24869. 10.1002/anie.202110873. [DOI] [PMC free article] [PubMed] [Google Scholar]; i Liu Z.; Qin Z.-Y.; Zhu L.; Athavale S. V.; Sengupta A.; Jia Z.-J.; Garcia-Borràs M.; Houk K. N.; Arnold F. H. An Enzymatic Platform for Primary Amination of 1-Aryl-2-alkyl Alkynes. J. Am. Chem. Soc. 2022, 144, 80–85. 10.1021/jacs.1c11340. [DOI] [PMC free article] [PubMed] [Google Scholar]; j Xu P.; Xie J.; Wang D.-S.; Zhang X. P. Metalloradical approach for concurrent control in intermolecular radical allylic C–H amination. Nat. Chem. 2023, 15, 498–507. 10.1038/s41557-022-01119-4. [DOI] [PMC free article] [PubMed] [Google Scholar]; k Ahmed H.; Ghosh B.; Breitenlechner S.; Feßner M.; Merten C.; Bach T. Intermolecular Enantioselective Amination Reactions Mediated by Visible Light and a Chiral Iron Porphyrin Complex. Angew. Chem., Int. Ed. 2024, e202407003 10.1002/anie.202407003. [DOI] [PubMed] [Google Scholar]
  13. Boquet V.; Nasrallah A.; Dana A. L.; Brunard E.; Di Chenna P. H.; Duran F. J.; Retailleau P.; Darses B.; Sircoglou M.; Dauban P. Rhodium(II)-Catalyzed Enantioselective Intermolecular Aziridination of Alkenes. J. Am. Chem. Soc. 2022, 144, 17156–17164. 10.1021/jacs.2c07337. [DOI] [PubMed] [Google Scholar]
  14. For selected examples of asymmetric intermolecular aziridination through nitrene transfer catalyzed by other metals, see: 14.; a 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. 1991, 113, 726–728. 10.1021/ja00002a080. [DOI] [Google Scholar]; b Li Z.; Conser K. R.; Jacobsen E. N. Asymmetric alkene aziridination with readily available chiral diimine-based catalysts. J. Am. Chem. Soc. 1993, 115, 5326–5327. 10.1021/ja00065a067. [DOI] [Google Scholar]; c Li Z.; Quan R. W.; Jacobsen E. N. Mechanism of the (Diimine)copper-Catalyzed Asymmetric Aziridination of Alkenes. Nitrene Transfer via Ligand-Accelerated Catalysis. J. Am. Chem. Soc. 1995, 117, 5889–5890. 10.1021/ja00126a044. [DOI] [Google Scholar]; d Lai T.-S.; Che C.-M.; Kwong H.-L.; Peng S.-M. Catalytic and asymmetric aziridination of alkenes catalysed by a chiral manganese porphyrin complex. Chem. Commun. 1997, 2373–2374. 10.1039/a706395d. [DOI] [Google Scholar]; e Omura K.; Uchida T.; Irie R.; Katsuki T. Design of a robust Ru(salen) complex: aziridination with improved turnover number using N-arylsulfonyl azides as precursors. Chem. Commun. 2004, 2060–2061. 10.1039/b407693a. [DOI] [PubMed] [Google Scholar]; f Subbarayan V.; Ruppel J. V.; Zhu S.; Perman J. A.; Zhang X. P. Highly asymmetric cobalt-catalyzed aziridination of alkenes with trichloroethoxysulfonyl azide (TcesN3). Chem. Commun. 2009, 4266–4268. 10.1039/b905727g. [DOI] [PubMed] [Google Scholar]; g Kim C.; Uchida T.; Katsuki T. Asymmetric olefin aziridination using a newly designed Ru(CO)(salen) complex as the catalyst. Chem. Commun. 2012, 48, 7188–7190. 10.1039/c2cc32997b. [DOI] [PubMed] [Google Scholar]; h Jin L.-M.; Xu X.; Lu H.; Cui X.; Wojtas L.; Zhang X. P. Effective Synthesis of Chiral N-Fluoroaryl Aziridines through Enantioselective Aziridination of Alkenes with Fluoroaryl Azides. Angew. Chem., Int. Ed. 2013, 52, 5309–5313. 10.1002/anie.201209599. [DOI] [PMC free article] [PubMed] [Google Scholar]; i Chan K.-H.; Guan X.; Lo V. K.-Y.; Che C.-M. Elevated Catalytic Activity of Ruthenium(II)–Porphyrin-Catalyzed Carbene/Nitrene Transfer and Insertion Reactions with N-Heterocyclic Carbene Ligands. Angew. Chem., Int. Ed. 2014, 53, 2982–2987. 10.1002/anie.201309888. [DOI] [PubMed] [Google Scholar]; j Farwell C. C.; Zhang R. K.; McIntosh J. A.; Hyster T. K.; Arnold F. H. Enantioselective Enzyme-Catalyzed Aziridination Enabled by Active-Site Evolution of a Cytochrome P450. ACS Cent. Sci. 2015, 1, 89–93. 10.1021/acscentsci.5b00056. [DOI] [PMC free article] [PubMed] [Google Scholar]; k Subbarayan V.; Jin L.-M.; Cui X.; Zhang X. P. Room temperature activation of aryloxysulfonyl azides by [Co(II)(TPP)] for selective radical aziridination of alkenes via metalloradical catalysis. Tetrahedron Lett. 2015, 56, 3431–3434. 10.1016/j.tetlet.2015.01.186. [DOI] [PMC free article] [PubMed] [Google Scholar]; l Riart-Ferrer X.; Sang P.; Tao J.; Xu H.; Jin L.-M.; Lu H.; Cui X.; Wojtas L.; Zhang X. P. Metalloradical activation of carbonyl azides for enantioselective radical aziridination. Chem. 2021, 7, 1120–1134. 10.1016/j.chempr.2021.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. a Fukagawa S.; Kojima M.; Yoshino T.; Matsunaga S. Catalytic Enantioselective Methylene C(sp3)–H Amidation of 8-Alkylquinolines Using a Cp*RhIII/Chiral Carboxylic Acid System. Angew. Chem., Int. Ed. 2019, 58, 18154–18158. 10.1002/anie.201911268. [DOI] [PubMed] [Google Scholar]; b Harris R. J.; Park J.; Nelson T. A. F.; Iqbal N.; Salgueiro D. C.; Bacsa J.; MacBeth C. E.; Baik M.-H.; Blakey S. B. The Mechanism of Rhodium-Catalyzed Allylic C–H Amination. J. Am. Chem. Soc. 2020, 142, 5842–5851. 10.1021/jacs.0c01069. [DOI] [PubMed] [Google Scholar]
  16. a Wang J.; Luo M.-P.; Gu Y.-J.; Liu Y.-Y.; Yin Q.; Wang S.-G. Chiral CpxRhodium(III)-Catalyzed Enantioselective Aziridination of Unactivated Terminal Alkenes. Angew. Chem., Int. Ed. 2024, 63, e202400502 10.1002/anie.202400502. [DOI] [PubMed] [Google Scholar]; b Gross P.; Im H.; Laws D. III; Park B.; Baik M.-H.; Blakey S. B. Enantioselective Aziridination of Unactivated Terminal Alkenes Using a Planar Chiral Rh(III) Indenyl Catalyst. J. Am. Chem. Soc. 2024, 146, 1447–1454. 10.1021/jacs.3c10637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. a Hoveyda A. H.; Evans D. A.; Fu G. C. Substrate-directable chemical reactions. Chem. Rev. 1993, 93, 1307–1370. 10.1021/cr00020a002. [DOI] [Google Scholar]; b Bhadra S.; Yamamoto H. Substrate Directed Asymmetric Reactions. Chem. Rev. 2018, 118, 3391–3446. 10.1021/acs.chemrev.7b00514. [DOI] [PubMed] [Google Scholar]
  18. a Höke T.; Herdtweck E.; Bach T. Hydrogen-bond mediated regio- and enantioselectivity in a C–H amination reaction catalysed by a supramolecular Rh(ii) complex. Chem. Commun. 2013, 49, 8009–8011. 10.1039/c3cc44197k. [DOI] [PubMed] [Google Scholar]; b Zhong F.; Bach T. Enantioselective Construction of 2,3-Dihydrofuro[2,3-b]quinolines through Supramolecular Hydrogen Bonding Interactions. Chem. - Eur. J. 2014, 20, 13522–13526. 10.1002/chem.201404617. [DOI] [PubMed] [Google Scholar]
  19. a Fanourakis A.; Williams B. D.; Paterson K. J.; Phipps R. J. Enantioselective Intermolecular C–H Amination Directed by a Chiral Cation. J. Am. Chem. Soc. 2021, 143, 10070–10076. 10.1021/jacs.1c05206. [DOI] [PMC free article] [PubMed] [Google Scholar]; b Fanourakis A.; Hodson N. J.; Lit A. R.; Phipps R. J. Substrate-Directed Enantioselective Aziridination of Alkenyl Alcohols Controlled by a Chiral Cation. J. Am. Chem. Soc. 2023, 145, 7516–7527. 10.1021/jacs.3c00693. [DOI] [PMC free article] [PubMed] [Google Scholar]; c Paterson K. J.; Dahiya A.; Williams B. D.; Phipps R. J. Tertiary Amides as Directing Groups for Enantioselective C–H Amination using Ion-Paired Rhodium Complexes. Angew. Chem., Int. Ed. 2024, e202317489 10.1002/ange.202317489. [DOI] [PubMed] [Google Scholar]
  20. For initial application of the strategy to iridium catalysis, see:; a Genov G. R.; Douthwaite J. L.; Lahdenperä A. S. K.; Gibson D. C.; Phipps R. J. Enantioselective remote C–H activation directed by a chiral cation. Science 2020, 367, 1246–1251. 10.1126/science.aba1120. [DOI] [PubMed] [Google Scholar]; b Ermanis K.; Gibson D. C.; Genov G. R.; Phipps R. J. Interrogating the Crucial Interactions at Play in the Chiral Cation-Directed Enantioselective Borylation of Arenes. ACS Catal. 2023, 13, 13043–13055. 10.1021/acscatal.3c03384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. For pioneering work on a related, charge-inverted approach for palladium catalysis, see:; a Ohmatsu K.; Ito M.; Kunieda T.; Ooi T. Ion-paired chiral ligands for asymmetric palladium catalysis. Nat. Chem. 2012, 4, 473–477. 10.1038/nchem.1311. [DOI] [PubMed] [Google Scholar]; b Ohmatsu K.; Ito M.; Kunieda T.; Ooi T. Exploiting the Modularity of Ion-Paired Chiral Ligands for Palladium-Catalyzed Enantioselective Allylation of Benzofuran-2(3H)-ones. J. Am. Chem. Soc. 2013, 135, 590–593. 10.1021/ja312125a. [DOI] [PubMed] [Google Scholar]
  22. Guthikonda K.; Du Bois J. A Unique and Highly Efficient Method for Catalytic Olefin Aziridination. J. Am. Chem. Soc. 2002, 124, 13672–13673. 10.1021/ja028253a. [DOI] [PubMed] [Google Scholar]
  23. Additionally, significant amounts of over-amination were found in the reactions with Rh2(esp)2 which complicated analysis of site-selectivity.
  24. Kolb H. C.; VanNieuwenhze M. S.; Sharpless K. B. Catalytic Asymmetric Dihydroxylation. Chem. Rev. 1994, 94, 2483–2547. 10.1021/cr00032a009. [DOI] [Google Scholar]
  25. a Lygo B.; Crosby J.; Lowdon T. R.; Wainwright P. G. Enantio- and diastereoselective synthesis of all four possible stereoisomers of 2-(phenylhydroxymethyl)quinuclidine. Tetrahedron Lett. 1997, 38, 2343–2346. 10.1016/S0040-4039(97)00310-9. [DOI] [Google Scholar]; b Lygo B.; Crosby J.; Lowdon T.; Wainwright P. G. Asymmetric approaches to 2-hydroxymethylquinuclidine derivatives. Tetrahedron 1999, 55, 2795–2810. 10.1016/S0040-4020(99)00050-2. [DOI] [Google Scholar]
  26. For a later, related approach, see:; Furukawa K.; Katsukawa M.; Nuruzzaman M.; Kobayashi Y. Synthesis of a Series of Structural Analogues of the Cinchona Alkaloids. Heterocycles 2007, 74, 159–166. 10.3987/COM-07-S(W)28. [DOI] [Google Scholar]
  27. a Raheem I. T.; Goodman S. N.; Jacobsen E. N. Catalytic Asymmetric Total Syntheses of Quinine and Quinidine. J. Am. Chem. Soc. 2004, 126, 706–707. 10.1021/ja039550y. [DOI] [PubMed] [Google Scholar]; b Dominguez-Molano P.; Bru G.; Salvado O.; Maza R. J.; Carbó J. J.; Fernández E. Transborylation of alkenylboranes with diboranes. Chem. Commun. 2021, 57, 13361–13364. 10.1039/D1CC05815K. [DOI] [PubMed] [Google Scholar]
  28. The ee value obtained for the nonligated complex seemed abnormally low but repitition of the reaction twice yielded identical outcomes therefore we conclude that this is a genuine effect.
  29. In our studies using similar chiral cations in enantioselective borylation we had observed decreased ee using the methylated cation, but the decrease was much smaller than that seen here (ref (20a)).
  30. We recently found that enantioselective C-H amination using amide directing groups could be perfomed with high ee, for which the presence of the C6F5I(OTFA)2 additive was crucial (ref (19c)). Control experiments suggested that the presence of the additive allowed hydrogen bond acceptor groups to be effective directors, including a methyl ether. We have evaluated amination of the corresponding methyl ether of 2a and found that very low reactivity (<5% yield) was obtained under the optimized conditions using the new biaryl catalyst Rh1 suggesting that this is not an effective directing group in this case (see the SI for details).
  31. Sharland J. C.; Wei B.; Hardee D. J.; Hodges T. R.; Gong W.; Voight E. A.; Davies H. M. L. Asymmetric synthesis of pharmaceutically relevant 1-aryl-2-heteroaryl- and 1,2-diheteroarylcyclopropane-1-carboxylates. Chem. Sci. 2021, 12, 11181–11190. 10.1039/D1SC02474D. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja4c07117_si_001.pdf (35.1MB, pdf)

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES