Nickel-Catalyzed Asymmetric Reductive Cross-Coupling between Heteroaryl Iodides and α-Chloronitriles

Abstract

A Ni-catalyzed asymmetric reductive cross-coupling of heteroaryl iodides and α-chloronitriles has been developed. This method furnishes enantioenriched α,α-disubstituted nitriles from simple organohalide building blocks. The reaction tolerates a variety of heterocyclic coupling partners, including pyridines, pyrimidines, quinolines, thiophenes, and piperidines. The reaction proceeds under mild conditions at room temperature, and precludes the need to pre-generate organometallic nucleophiles.

Graphical abstract

In recent years, Ni-catalyzed reductive cross-coupling reactions have experienced a surge of development.¹ These transformations forge C–C bonds between two organic electrophiles, employing a stoichiometric reductant (usually Zn⁰ or Mn⁰) to turn over the Ni catalyst. An appealing aspect of this chemistry is that sec-alkyl electrophiles are competent reaction partners,² and in some cases, these reactions can be rendered enantioselective by use of an appropriate chiral ligand.³ However, the scope of C(sp³) electrophiles used in these asymmetric transformations has so far been limited to α-substituted benzyl chlorides. Moreover, the asymmetric cross-coupling of heteroaryl electrophiles – a substrate class that would be of high value for medicinal chemists⁴ – has proven challenging.⁵ In this communication, we report the Ni-catalyzed asymmetric reductive cross-coupling between α-chloronitriles and heteroaryl iodides, a reaction that provides access to a variety of enantioenriched heterocyclic products.

In considering the development of new C(sp³) electrophiles for Ni-catalyzed reductive cross-coupling reactions, we became interested in the use of α-chloronitriles. Nitriles are valuable synthetic intermediates that serve as precursors to amines, carboxylic acids, carboxamides, aldehydes, ketones, and alcohols.⁶ The cyano group is also found in a number of natural products and medicinal compounds.⁷ However, there are few transition metal-catalyzed cross-coupling methods to directly prepare enantioenriched α,α-disubstituted nitriles. In 2010, Falck and coworkers published a Pd-catalyzed stereospecific Suzuki cross-coupling of α-cyanohydrin triflates.⁸ Two years later, Fu and Choi reported a highly enantioselective Ni-catalyzed Negishi coupling between racemic α-bromonitriles and arylzinc reagents.^9,10 However, neither report included heteroaryl nucleophiles as part of their substrate studies.

We envisioned that a Ni-catalyzed asymmetric reductive cross-coupling between α-chloronitriles and heteroaryl halides could provide access to a complementary scope of synthetically useful products. Furthermore, α-chloronitriles have not been developed as C(sp³) electrophiles for Ni-catalyzed reductive cross-coupling reactions (even in the racemic sense).¹¹ Thus, the successful development of this transformation would expand the scope of reductive cross-coupling reactions to new and synthetically versatile classes of electrophiles.

We began our investigations with the coupling between α-chloronitrile 1a and 3-iodoquinoline (2). When the reaction was conducted in DMA with chiral BOX ligand L1 and TMSCl to activate Mn⁰, no product was formed (Table 1, entry 2). Rather, the α-chloronitrile was rapidly consumed, generating the hydrodehalogenation product 4-phenylbutyronitrile (4). A solvent screen revealed that 3a forms in trace yield and 38% ee when dioxane is used as solvent (entry 3); chiral BiOX ligand L2 provided slightly improved yield (entry 4). We hypothesized that more electron-rich ligands might accelerate the rate of oxidative addition of 2 to a LNi(0) complex, relative to the rate of hydrodehalogenation and decomposition reactions of 1, thereby improving the yield of 3a.¹² Consistent with this hypothesis, phosphino-oxazoline (PhOX) ligands were found to provide improved reactivity, with BnPhOX L5 furnishing 3a in 52% yield and 77% ee (entry 7). Further ligand optimization identified DMMB-PhOX L6 as providing the best combination of yield and selectivity (entry 1). A study of additional reaction parameters revealed that: 1) use of Zn⁰ instead of Mn⁰ provides the product in lower yield and ee (entry 12), and 2) 2-bromo-4-phenylbutanenitrile (5) suffers from facile hydrodehalogenation and elimination under the reaction conditions, providing 3a in only 9% yield and 84% ee (entry 16). Additives that have been shown to improve the yields of reductive cross-electrophile coupling were also investigated (entries 13–15).^3a,5a The addition of NaBF₄ provided 3 in comparable yield and improved selectivity (entry 14); however, further studies revealed that for many substrates, NaBF₄ provides no added benefits. Erring toward the use of fewer reagents, NaBF₄ was only added for the cross-coupling of certain more challenging substrates, as indicated in the following Tables. The exact role of NaBF₄ in these transformations is unknown at this time.^5a

Table 1.

Optimization of reaction conditions.^a

Entrya	Ligand	Deviation from Standard Conditions	Yield 3 (%)b	Yield 4 (%)b	ee 3 (%)c
1	L6	None	78	20	84
2	L1	DMA instead of dioxane	0	62	–
3	L1	–	<5	32	38
4	L2	–	17	22	9
5	L3	–	25	40	83
6	L4	–	0	32	–
7	L5	–	52	35	77
8	L6	No Ni	0	0	–
9	L6	No Mn0	0	0	–
10	L6	No TMSCl	<5	<5	82
11	–	No Ligand	4	23	0
12	L6	Zn0 instead of Mn0	25	32	10
13	L6	TFA (0.4 equiv)	48	37	78
14	L6	NaBF4 (1.0 equiv)	76	24	90
15	L6	NaI (0.25 equiv)	71	29	84
16	L6	RBrCN (5) instead of 1	9	36	84

^aReactions conducted under inert atmosphere on 0.1 mmol scale for 14 h.

^bDetermined by ¹H NMR versus an internal standard.

^cDetermined by SFC using chiral stationary phase.

Having optimized the reaction parameters for the coupling between 1a and 2, we sought to probe the scope of the heteroaryl partner (Table 2). We were pleased to find that a variety of heteroaryl iodides undergo cross-coupling to furnish the α,α-disubstituted nitriles in good yields and with high enantioinduction. The reaction demonstrates good chemoselectivity, with no coupling observed at the 2-position of 2-bromo- or 2-chloro-5-iodopyridine (see products 7a and 7b).¹³ Whereas substitution meta to the iodide was tolerated (7f), a decrease in yield was observed when the substituent was ortho to the iodide (7g). Iodo-pyridines or -pyrimidines lacking substitution at C2 were poor substrates, presumably due to the increased Lewis basicity of the nitrogen. A variety of C2-substituted pyrimidines, as well as 2-iodothiophene and a 6-imidazopyridine also undergo cross-coupling, delivering products in good yield and with high enantioinduction (7h–7o). Importantly, many of the products were easily recrystallized to afford highly enantioenriched (>95% ee) material with excellent recovery. For less-reactive heteroaryl iodides, competitive hydrodehalogenation of 1a resulted in decreased yields; this could be mitigated in most cases by using two equivalents of the iodide partner.¹⁴

Table 2.

Scope of heteroaryl iodide.^a

^aReaction conducted on 0.2 mmol scale. Isolated yields are provided; ee is determined by SFC using chiral stationary phase. Values in parentheses are yield and ee following a single recrystallization of the product.

^b2.0 equiv heteroaryl iodide used.

^c1.0 equiv NaBF₄ added.

We also investigated the scope of the α-chloronitrile (1). In general, less sterically-encumbered substrates provide products in good yield and modest enantioselectivity, while more bulky substrates provide the product with good enantioselectivity and slightly more modest yields. Nonetheless, the reaction exhibits notable functional group tolerance, including carbamates (3h and 3i), esters (3f) and a primary alkyl chloride (3g). Recrystallization of nitrile 3e provided crystals suitable for X-ray diffraction analysis, which allowed us to assign the stereochemistry as the (S)-configuration.¹⁵

The enantioenriched α,α-disubstituted nitriles produced in this reaction serve as versatile synthetic intermediates. For example, hydrogenation of 2-piperidyl-pyrimidine 7k under standard conditions provides phenethylamine 8 in excellent yield and with no erosion of ee (Scheme 2). The two step sequence involving cross-coupling and hydrogenation represents an straightforward approach to the synthesis of this bioactive class of molecule. The same substrate can be subjected to Pt-catalyzed hydrolysis¹⁶ to afford carboxamide 9 in high yield and with complete stereoretention. Alternatively, reduction of thiophene-containing nitrile 7n with DIBAL-H furnished the enantioenriched aldehyde 10 in excellent yield and with minor erosion of ee.

Scheme 2.

Derivatization of α,α-disubstituted nitriles.

Several experiments were conducted to interrogate the potential mechanism of this transformation. To probe whether the oxidative addition of the α-chloronitrile proceeds by a radical pathway, cyclopropyl-containing substrate 11 was prepared and subjected to the reaction conditions (Scheme 3). Ring-opened coupling product 12 was obtained in 21% yield as a 1:1 mixture of cis and trans isomers, consistent with a radical intermediate.¹⁷ None of the corresponding cyclopropane-containing product was observed. Despite this evidence for a radical intermediate, the reaction proceeds with comparable efficiency in the presence of 50 mol % of common radical inhibitors, such as 2,6-bis(1,1-dimethylethyl)-4-methylphenol (BHT) or dihydroanthracene (DHA).¹⁸ The latter finding is inconsistent with cage-escaped radicals expected in a radical chain mechanism, although more studies are needed to fully elucidate the reaction pathway.¹⁹

Scheme 3.

Mechanistic experiments.

In conclusion, a Ni-catalyzed asymmetric reductive cross-coupling between α-chloronitriles and heteroaryl iodides has been developed. A new chiral PHOX ligand was identified that provides α,α-disubstituted nitriles in good yields and with high enantioinduction. This is the first example of a Ni-catalyzed asymmetric reductive cross-coupling reaction that tolerates N- and S-heterocyclic coupling partners, and demonstrates the feasibility of developing related transformations of electrophiles containing Lewis basic functional groups. The development of new asymmetric reductive cross-coupling reactions as well as mechanistic investigations are the subject of ongoing research in our laboratory.

Scheme 1.

Transition metal-catalyzed cross-coupling reactions of α-cyano electrophiles.

Table 3.

Scope of α-chloronitriles.^a

^aReaction conducted on 0.2 mmol scale. Isolated yields are provided; ee is determined by SFC using chiral stationary phase. Values in parentheses are yield and ee following a single recrystallization of the product.

^b2.0 equiv heteroaryl iodide used.

^c1 equiv NaBF₄ added.