Nickel-Catalyzed Asymmetric Hydrogenation of Cyclic Sulfamidate Imines: Efficient Synthesis of Chiral Cyclic Sulfamidates.

Chiral cyclic sulfamidates are useful building blocks to construct compounds, such as chiral amines, with important applications. Often these compounds can only be generated through expensive precious metal catalysts. Here, Ni(OAc)2/(S, S)-Ph-BPE-catalyzed highly efficient asymmetric hydrogenation of cyclic sulfamidate imines was successfully developed, affording various chiral cyclic sulfamidates with high yields and excellent enantioselectivities (up to 99% yield, >99% enantiomeric excess [ee]). This Ni-catalyzed asymmetric hydrogenation on a gram scale has been achieved with only 0.1 mol% catalyst loading in 99% yield with 93% ee. Other types of N-sulfonyl ketimines were also hydrogenated well to obtain the corresponding products with >99% conversion, 96%-97% yields, and 97%->99% ee. In addition, this asymmetric methodology could produce other enantioenriched organic molecules, such as chiral β-fluoroamine, amino ether, and phenylglycinol. Moreover, a reasonable catalytic cycle was provided according to the deuterium-labeling studies, which could reveal a possible mechanism for this Ni-catalyzed asymmetric hydrogenation.

Introduction

Efficienpan>t synthesis of chiral cyclic sulfamidates has attracted great attention in the past decades, owing to their versatilities working as valuable intermediates for the construction of some important organic compounds and bioactive molecules (Aguilera and Fernandez-Mayoralas, 1996, Williams et al., 2003, Bower et al., 2004, Bower et al., 2007a, Bower et al., 2007b, Bower et al., 2007c, Bower et al., 2010, Jamieson et al., 2009, Lorion et al., 2010, Megia-Fernandez et al., 2011, Boulton et al., 1999, Wei and Lubell, 2000, Espino et al., 2001, Cohen and Halcomb, 2001, Cohen and Halcomb, 2002, Atfani et al., 2001, Nicolaou et al., 2002, Meléndez and Lubell, 2003, Ni et al., 2007, Rönnholm et al., 2007, Baig et al., 2010, Baig et al., 2011, Venkateswarlu et al., 2014, Albu et al., 2016, Su et al., 2016, Chen et al., 2014, Kong et al., 2015, Liu et al., 2017, Wu et al., 2018). For example, ring-opening reactions of chiral cyclic sulfamidates can offer convenient and efficient access to chiral amines, amino alcohols, amino acids, and their derivatives (Boulton et al., 1999, Wei and Lubell, 2000, Espino et al., 2001, Cohen and Halcomb, 2001, Cohen and Halcomb, 2002, Atfani et al., 2001, Nicolaou et al., 2002, Meléndez and Lubell, 2003, Ni et al., 2007, Rönnholm et al., 2007, Baig et al., 2010, Baig et al., 2011, Venkateswarlu et al., 2014, Albu et al., 2016, Su et al., 2016, Chen et al., 2014, Kong et al., 2015, Liu et al., 2017, Wu et al., 2018). So far the asymmetric catalytic synthetic methods of chiral cyclic sulfamidates were mainly focused on transition metal-catalyzed asymmetric intramolecular amidation of sulfamate esters (Liang et al., 2002, Liang et al., 2004, Fruit and Mueller, 2004, Zhang et al., 2005, Zalatan and Du Bois, 2008, Lin et al., 2008, Ichinose et al., 2011), additions of organoboron reagents to cyclic imines (Chen et al., 2014, Chen et al., 2018, Kong et al., 2015, Liu et al., 2017, Wu et al., 2018, Nishimura et al., 2012, Nishimura et al., 2013, Luo et al., 2012a, Luo et al., 2012b, Wang et al., 2013, Hepburn et al., 2013, Wang and Xu, 2013, Zhang et al., 2016a), and asymmetric reduction of cyclic ketimines (Wang et al., 2008, Yu et al., 2009, Kang et al., 2010, Lee et al., 2011, Lee et al., 2012, Han et al., 2011, Liu et al., 2019, Itsuno et al., 2014, Seo et al., 2015, Kim et al., 2018).

Asymmetric catalytic hydrogenation of prochiral unsaturated compounds has emerged as a powerful and effective approach for the construction of chiral compounds, which has made tremendous progress (Knowles, 1983, Noyori and Takaya, 1990, Noyori and Ohkuma, 2001, Tang and Zhang, 2003, Blaser et al., 2003, Cui and Burgess, 2005, Minnaard et al., 2007, Zhang et al., 2007, Zhang et al., 2016b, Johnson et al., 2007, Zhou, 2007, Roseblade and Pfaltz, 2007, Fleury-Bregeot et al., 2010, Xie et al., 2011, Xie et al., 2012, Wang et al., 2012, Chen et al., 2013, Verendel et al., 2014, Zhang et al., 2016b). Most of these powerful catalytic systems typically depended on scarce and precious transition metals, such as Ru, Rh, Ir, and Pd, which faced difficulties like limited resource, high cost, and environmental contamination. Therefore, it is important and necessary to devote much effort to developing cheap, earth-abundant, first-row transition metal catalytic systems. Recently, the Fe-, Co-, and Ni-catalyzed asymmetric hydrogenation of prochiral unsaturated compounds has received great attention, which shows the great potential of first-row transition metals in catalytic asymmetric (transfer) hydrogenation (Morris, 2009, Morris, 2015, Chirik, 2015, Li et al., 2014, Li et al., 2015, Li et al., 2017, Bauer and Knölker, 2015, Sui-Seng et al., 2008, Zhou et al., 2011, Monfette et al., 2012, Friedfeld et al., 2013, Friedfeld et al., 2016, Lagaditis et al., 2014, Sonnenberg et al., 2014, Lu et al., 2015, Chen et al., 2016, Hamada et al., 2008, Hibino et al., 2009, Dong et al., 2012, Yang et al., 2014, Yang et al., 2016, Guo et al., 2015, Xu et al., 2015, Shevlin et al., 2016, Gao et al., 2017, Zhao et al., 2019). Among these catalytic methodologies, Ni-catalyzed asymmetric hydrogenation is still in early stage, and there are a few related studies at present (Li et al., 2015, Li et al., 2017, Hamada et al., 2008, Hibino et al., 2009, Dong et al., 2012, Yang et al., 2014, Yang et al., 2016, Guo et al., 2015, Xu et al., 2015, Shevlin et al., 2016, Gao et al., 2017, Zhao et al., 2019). In 2008 and 2009, Hamada and co-workers reported Ni-catalyzed asymmetric hydrogenation of α-amino-β-ketoester hydrochlorides and substituted aromatic α-aminoketone hydrochlorides through dynamic kinetic resolution (Hamada et al., 2008, Hibino et al., 2009). In 2016, Chirik and co-workers discovered Ni-catalyzed asymmetric hydrogenation of α,β-unsaturated esters (Shevlin et al., 2016). Recently, our group reported Ni-catalyzed asymmetric hydrogenation of functionalized enamides with excellent results (Gao et al., 2017, Li et al., 2017). Despite some progress having been made, it is still quite urgent to explore the wide range of substrate generality, high reactivity, excellent stereoselectivity, and high turnover numbers (TON) for the Ni-catalyzed asymmetric hydrogenation.

Asymmetric hydrogenation of cyclic sulfamidate imines is a direct and effective access to chiral sulfamidates. Zhou and co-workers established highly efficient Pd-catalyzed enantioselective hydrogenation of cyclic sulfamidate imines with excellent results (Wang et al., 2008). To the best of our knowledge, there is no example about cheap transition metal Ni-catalyzed asymmetric hydrogenation of cyclic sulfamidate imines. Herein, we successfully realized the highly efficient Ni-catalyzed asymmetric hydrogenation of cyclic sulfamidate imines to afford chiral cyclic sulfamidates with high yields and excellent enantioselectivities (Scheme 1, up to 99% yield, >99% enantiomeric excess [ee]), and the gram-scale hydrogenation can be easily achieved with only 0.1 mol% catalyst loading (TON = 1,000).

Scheme 1

Asymmetric Hydrogenation of Cyclic Sulfamidate Imines

Results

Optimization Reaction Conditions

We started initial investigation of the Ni(OAc)2-catalyzed asymmetric hydrogenation of model substrate 4-phenyl-5H-1,2,3-oxathiazole 2,2-dioxide 1a to evaluate a variety of important chiral diphosphine ligands (Figure 1) under 60 atm H2 at 80°C in MeOH for 24 h. As shown in Table 1, full conversion and good to excellent enantioselectivities were obtained with (S)-Binapine and (S, S)-Ph-BPE as ligand (>99% conversion, 86%–92% ee, Table 1, entries 1 and 4). Although high catalytic activity was achieved, very poor enantioselective control was afforded in the presence of (Rc, Sp)-DuanPhos and (S, S)-Me-DuPhos (Table 1, entries 2 and 3). In addition, ligands (R, S)-WalPhos, (S)-SegPhos, and (S)-BINAP did not work in this reaction; no reaction was observed (Table 1, entries 5–7). Therefore, (S, S)-Ph-BPE was revealed to be superior with the best enantioselectivity (>99% conversion, 92% ee, Table 1, entry 4). To our delight, the same result can be achieved when the catalyst loading of Ni(OAc)2/(S, S)-Ph-BPE was decreased from 5.0 mol% to 1.0 mol% (Table 1, entry 8).

Figure 1

The Structure of Chiral Diphosphine Ligands

Table 1

Screening Ligands for Ni-Catalyzed Asymmetric Hydrogenation of 4-Phenyl-5H-1,2,3-oxathiazole 2,2-dioxide (1a)

Entry	Ligand	Conversion (%)a	ee (%)b
1	(S)-Binapine	>99	86
2	(Rc, Sp)-DuanPhos	>99	−2
3	(S, S)-Me-DuPhos	>99	−13
4	(S, S)-Ph-BPE	>99	92
5	(R)-WalPhos	NR	NA
6	(S)-SegPhos	NR	NA
7	(S)-BINAP	NR	NA
8c	(S, S)-Ph-BPE	>99	92

NR, no reaction; NA, not available.

Unless otherwise noted, all reactions were carried out with a Ni(OAc)2/ligand/substrate 1a (0.1 mmol) ratio of 1:1.1:20 in 1.0 mL MeOH under 60 atm H2 at 80°C for 24 h. The configuration of 2a was determined by comparing the optical rotation data with those reported in the literature (Wang et al., 2008, Kang et al., 2010, Lee et al., 2011).

Conversion was determined by 1H NMR analysis.

ee was determined by chiral high-performance liquid chromatography analysis.

1.0 mol% catalyst loading.

1.0 mol% n class="Chemical">catalyst loading.

Inspired by the promising results, the Ni(OAc)2/(S, S)-Ph-BPE-catalyzed asymmetric hydrogenation of model substrate 4-phenyl-5H-1,2,3-oxathiazole 2,2-dioxide 1a was carried out in different solvents. We found that moderate to high conversions and excellent enantioselectivities were obtained in several kinds of alcoholic solvents, such as MeOH, EtOH, PrOH, CF3CH2OH, and (CF3)2CHOH (62%–>99% conversions, 91%–94% ee, Table 2, entries 1–5). Poor conversions and moderate to good enantioselectivities were provided in nonprotic solvents, such as tetrahydrofuran (THF), toluene, ethyl acetate, and 1,4-dioxane (7%–22% conversions, 58%–87% ee, Table 2, entries 7–10), and this reaction did not work in dichloromethane (Table 2, entry 6). Therefore, CF3CH2OH was selected as the best solvent to provide full conversion and the highest enantioselectivity (>99% conversion, 94% ee, Table 2, entry 4).

Table 2

Screening Solvents for Ni-Catalyzed Asymmetric Hydrogenation of 4-Phenyl-5H-1,2,3-oxathiazole 2,2-dioxide (1a)

Entry	Solvent	Conversion (%)a	ee (%)b
1	MeOH	>99	92
2	EtOH	83	91
3	iPrOH	62	92
4	CF3CH2OH	>99	94
5	(CF3)2CHOH	>99	93
6	CH2Cl2	NR	NA
7	THF	17	87
8	Toluene	22	82
9	Ethyl acetate	12	86
10	1,4-dioxane	7	58

NR, no reaction; NA, not available.

Unless otherwise noted, all reactions were carried out with a Ni(OAc)2/(S, S)-Ph-BPE/substrate 1a (0.1 mmol) ratio of 1:1.1:100 in 1.0 mL solvent under 60 atm H2 at 80°C for 24 h; the catalyst was pre-complexed in MeOH (0.1 mL for each reaction vial).

Conversion was determined by 1H NMR analysis.

ee was determined by chiral high-performance liquid chromatography analysis.

Substrate Scope Study

After establishing the optimized reaction conditions, we sought to examine the substrate scope generality of this Ni-catalyzed asymmetric hydrogenation of cyclic sulfamidate imines. As listed in Table 3, the Ni-catalyzed asymmetric hydrogenation of a variety of aryl-substituted cyclic sulfamidate imines could proceed smoothly, affording the desired hydrogenation products (2a-2l) with full conversions, high yields, and excellent enantioselectivities (>99% conversion, 94%–99% yields, 91%–>99% ee). Diverse aryl-substituted cyclic sulfamidate imines bearing electron-donating (1b-1f) or electron-withdrawing (1g-1l) substituents worked well in this asymmetric hydrogenation. It is worth noting that the hydrogenation product 2i is an important intermediate for the synthesis of the enantiomer of piperazinone acid, which was one of the two main molecular motifs in clinical candidate MK-3207 (McLaughlin et al., 2013). In addition, the position of substituted group on the phenyl ring was also investigated; whether the substituted groups are on the ortho-, meta-, or para-position of the phenyl ring, these asymmetric reductions proceeded efficiently with excellent results. Interestingly, cyclic sulfamidate imines with substituents in ortho-position on the phenyl ring (1b, 1e, 1g) can provide chiral cyclic sulfamidates (2b, 2e, 2g) with higher enantioselectivities. When the phenyl ring was replaced with 2-naphthyl group, the substrate (1m) performed well with 97% yield and 92% ee. Moreover, the heteroaromatic substrate (1n) was hydrogenated with moderate reactivity and excellent enantioselectivity (65% conversion, 55% yield, 95% ee). It is noteworthy that the alkyl substrates (1o-1p) worked smoothly in this asymmetric hydrogenation, providing the desired products (2o-2p) with good to excellent results (>99% conversion, 96%–98% yields, and 83%–92% ee).

Table 3

Substrate Scope Study for Ni-Catalyzed Asymmetric Hydrogenation of Cyclic Sulfamidate Imines

Unless otherwise noted, all reactions were carried out with a Ni(OAc)2/(S, S)-Ph-BPE/substrate 1 (0.1 mmol) ratio of 1:1.1:100 in 1.0 mL CF3CH2OH under 60 atm H2 at 80°C for 24 h. Conversion was determined by 1H NMR analysis. Yield is isolated yield. The ee value was determined by high-performance liquid chromatography on a chiral column.

Superscript letter ‘a’ indicates S/C = 20, 36 h.

Encouraged by these promising reaction results, other types of ketimines were employed in this catalytic system. As shown in Scheme 2, the acetophenone and 2,3-dihydro-1H-inden-1-one-derived N-sulfonyl ketimines 1q and 1r worked efficiently under optimized reaction conditions; the corresponding hydrogenation products 2q and 2r were obtained with full conversion, high yields, and excellent enantioselectivities (>99% conversion, 96%–97% yields, 97%–>99% ee).

Scheme 2

The Ni-Catalyzed Asymmetric Hydrogenation of Other N-Sulfonyl Ketimines

Synthetic Application

The synthetic application potentiality of this Ni-catalyzed asymmetric hydrogenation was demonstrated by the gram-scale transformation. The asymmetric reduction of model substrate 1a on the 6-mmol scale proceeded well in the presence of just 0.1 mol% catalyst loading (S/C = 1,000), affording product 2a in 99% yield with 93% ee, which showed that our catalytic system had excellent catalytic activity (Scheme 3). In addition, >99% ee can be easily achieved in CH2Cl2/hexane through simple crystallization.

Scheme 3

Gram-Scale Asymmetric Hydrogenation of 1a with High TON

To reveal the great utility of this methodology, some derivatization reactions of hydrogenation product 2a were conducted (Scheme 4). The tert-butoxycarbonyl (Boc) group was easily introduced on the nitrogen atom of hydrogenation product 2a to prepare compound 3 without loss of enantiomeric purity (Kang et al., 2010). Also, it was treated with tetrabutylammonium fluoride to give enantioenriched β-fluoroamine 4 in 77% yield (Wu et al., 2018, Nishimura et al., 2013). In addition, compound 3 went through nucleophilic attack of 4-methoxyphenol, which led to chiral amino ether 5 in 76% yield (Wu et al., 2018, Nishimura et al., 2013). The hydrogenation product 2a could also be efficiently reduced with LiAlH4 to generate (S)-phenylglycinol 6 in 87% yield and without loss of ee value (>99% ee) (Chen et al., 2014, Liu et al., 2017), which is the key intermediate to construct chiral cyclic carbamate Evans' auxiliary (Jnoff et al., 2014) and bisoxazoline ligand (S,S)-Ph-Box (Corey et al., 1991, Cornejo et al., 2005, Ouhamou, 2010).

Scheme 4

Synthetic Transformations of Product 2a

Discussion

Mechanism Study

To explore the possible reaction mechanism for this Ni-catalyzed asymmetric hydrogenation, a series of isotopic labeling studies were conducted (Scheme 5). The cyclic sulfamidate imine 1a was hydrogenated with 25 atm D2 in CF3CH2OH; the deuterium atom was solely added at the benzylic position and partly at the nitrogen atom of the product. In addition, this reduction was repeated in the presence of H2 and CF3CH2OD, and we found that the deuterium atom was just partly located at the nitrogen atom. Our hydrogenation product 2a was dissolved and stirred in CF3CH2OD, and the deuterium atom was detected to be partly incorporated at the N-H position, which showed that proton exchange should occur in this process. These results suggested that the H atom at the benzylic position of the hydrogenation product was solely from H2.

Scheme 5

Deuterium-Labeling Experiments

(A) The hydrogenation with D2 in CF3CH2OH.

(B) The hydrogenation with H2 in CF3CH2OD.

(C) The product 2a stirring in CF3CH2OD.

n class="Chemical">Deuterium-Labeling Expen class="Chemical">riments

Based on these observations and previous studies (Shevlin et al., 2016, Gao et al., 2017), the possible catalytic mechanism of this transformation was presented in Scheme 6. The hydrogen was involved in heterolytic cleavage to form [Ni]-H intermediate (II) (Korstanje et al., 2015, Ashby and Halpern, 1991), and it then went through ligand exchange with cyclic sulfamidate imine 1a, followed by enantioselective conjugated addition of [Ni]-H to C=N bond of imine to provide intermediate (TSIII). Subsequent protonation by AcOH released the product 2a. The N-H group of product 2a has the possibility of undergoing H-exchange with CF3CH2OH (or AcOH) to generate compound 2a’. To our delight, the deuterium-labeling experimental observations above are consistent with this possible catalytic cyclic pathway.

Scheme 6

Proposed Catalytic Cycle for the Ni-Catalyzed Asymmetric Hydrogenation of 1a

Conclusion

In conclusion, the Ni-catalyzed asymmetric hydrogenation of cyclic sulfamidate imines was successfully realized, affording a variety of chiral cyclic sulfamidates with high yields and excellent enantioselectivities (up to 99% yield, >99% ee, and 1,000 TON). Other types of N-sulfonyl ketimines worked well to give the corresponding hydrogenation products with full conversion, 96%–97% yields, and 97%–>99% ee. In addition, this asymmetric methodology owned great synthetic utility through various product derivations to construct some important enantioenriched organic molecules, such as chiral β-fluoroamine, amino ether, and phenylglycinol. Moreover, a reasonable catalytic cycle was provided to reveal a possible mechanism for this Ni-catalyzed asymmetric hydrogenation based on the deuterium-labeling studies. Further investigations on the detailed mechanisms of Ni-catalyzed asymmetric hydrogenation strategy are in progress in our laboratory.

Limitations of the Study

The six-memben class="Chemical">red n class="Chemical">cyclic sulfamidate imine was not suitable in this methodology.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.