Homogeneous Organic Reductant Based on 4,4’-^tBu₂-2,2’-Bipyridine for Cross-Electrophile Coupling

Abstract

The synthesis of a new homogeneous reductant based on 4,4’-^tBu₂-2,2’-bipyridine, ^tBu-OED₄, is reported. ^tBu-OED₄ was prepared on a multigram scale in two steps from inexpensive and commercially available starting materials, with no chromatography required for purification. ^tBu-OED₄ has a reduction potential of −1.33 V (vs Ferrocenium/Ferrocene) and is soluble in a range of common organic solvents. We demonstrate that ^tBu-OED₄ can facilitate Ni/Co dual-catalyzed C(sp²)–C(sp³) cross-electrophile coupling reactions and is highly functional group tolerant. ^tBu-OED₄ is expected to be a valuable addition to the set of homogeneous reductants available for organic transformations.

Graphical Abstract

• New homogeneous reductant synthesized in two steps from 4,4’-^tBu₂-2,2’-bipyridine

• Reduction potential of reductant is −1.33 V (vs Ferrocenium/Ferrocene)

• New reductant can be utilized in Ni-catalyzed cross-electrophile coupling

• Reaction is compatible with substrates relevant to medicinal chemistry

Introduction

Ni-catalyzed cross-electrophile coupling (XEC) reactions are a powerful method for forming C(sp²)–C(sp³) and C(sp³)–C(sp³) bonds from a range of abundant and stable functional groups including aryl and alkyl halides, (pseudo)halides, alcohols, carboxylic acids, and amines.¹ To generate a new C–C bond in XEC, an external source of electrons is required. Heterogeneous reductants such as Mn⁰ and Zn⁰ are the most common electron sources because they are inexpensive, and practical to handle and store. However, heterogeneous reductants are challenging to use on scale due to irreproducibility in reactions kinetics^2,3 and the need for toxic amide-based solvents.⁴ Further, they can be challenging to implement in high throughput experimentation (HTE), and it is often difficult to elucidate the mechanisms of reactions performed with heterogeneous electron sources. As an alternative to heterogeneous reductants for XEC, strategies based on electro⁵- and photochemistry⁶ have been developed. Both approaches have distinct strengths and weaknesses and have led to the discovery of new transformations, but they can be challenging to implement on scale and use in HTE.

In principle, transformations promoted by homogeneous reductants are straightforward to execute on both large and small scale because the reductants are soluble, but limitations associated with the preparation, stability, and cost of homogeneous reductants have restricted their use in XEC.^1i,7 The most common homogeneous reductant for XEC is tetrakis(dimethylamino)ethylene (TDAE; E° = −1.11 V vs ferrocenium/ferrocene (Fc⁺/Fc), Figure 1a). TDAE has been used to facilitate several reductive transformations,⁸ but it is an expensive and air-sensitive liquid that is difficult to handle and store. Recently, we developed a series of reductants based on the tetraaminoethylene framework, TME (E° = −0.85 V), TPiE (E° = −1.06 V), and TPyE (E° = −1.32 V), that can facilitate similar XEC reactions to TDAE (Figure 1a).⁹ The new reductants are solids under ambient conditions, and the weakest reductant TME is air-stable. Additionally, the different reduction potentials of the new reductants can be used to control the rate of radical generation in XEC, enabling rare examples of couplings between aryl halides and benzylic Katritzky salts. Nevertheless, these reductants, like TDAE, are challenging to synthesize, and for this reason there is interest in preparing reductants that are not based on the tetraaminoethylene scaffold.

Figure 1:

a) Previous examples of homogenous reductants used in XEC. Potentials are reported relative to ferrocenium/ferrocene (Fc⁺/Fc) in DMF. b) New homogeneous reductant for XEC developed in this work.

In contrast to the tetraaminoethylene based reductants, a homogeneous reductant based on 4,4’-NMe₂-2,2’-bipyridine, DMAP-OED₃ (Figure 1a), is more straightforward to synthesize.¹⁰ Unfortunately, DMAP-OED₃ has found limited use in XEC,¹¹ likely because it is a relatively strong reductant (E° = −1.69 V) and as such, can undergo deleterious side reactions with substrates such as aryl and alkyl halides. One of the advantages of homogeneous reductants is that their reduction potentials are tunable. Research from both Murphy and co-workers and our group has demonstrated that by removing the NMe₂ substituents in DMAP-OED₃ and changing the length of the linker connecting the nitrogen atoms of the 2,2’-bipyridine scaffold, it is possible to generate weaker reductants,¹² but these reductants have not been used in XEC. In this work, we describe the synthesis of a new reductant based on the 2,2’-bipyridine scaffold, ^tBu-OED₄ (E° = −1.33 V), which has a similar reduction potential to TPyE, the strongest of the tetraaminoethylene reductants (Figure 1b). We demonstrate that ^tBu-OED₄ can be readily synthesized in two steps from simple and abundant precursors and is compatible with a Ni/Co dual catalyzed XEC reaction between aryl and alkyl halides. Overall, our work adds a new homogeneous organic reductant to the limited set available for facilitating XEC reactions and suggests that the 2,2’-bipyridine scaffold is a candidate for developing other reductants with varying reduction potentials.

Results and Discussion

Synthesis of ^tBu-OED₄

We synthesized ^tBu-OED₄ in two steps starting from commercially available 4,4’-^tBu₂-2,2’-bipyridine (~$6.50/gram from a search on SciFinder-n) (Figure 2). In the first step, 4,4’-^tBu₂-2,2’-bipyridine was alkylated with 1,4-diiodobutane (~$0.60/gram) to generate the dicationic salt, I, in 82% yield. Compound I was characterized by ¹H and ¹³C NMR spectroscopy, as well as HRMS and elemental analysis. The reduction of I with Mg⁰ powder in acetonitrile generates ^tBu-OED₄, which is a solid under ambient conditions, in 93% yield. Using this two-step procedure, we were able to synthesize approximately 7 grams of ^tBu-OED₄ in one reaction. Importantly, no chromatography is required to purify either I or ^tBu-OED₄ and only washing, extraction, and filtration are needed to generate both I and ^tBu-OED₄ in high purity (>95%). Although I is air stable, ^tBu-OED₄ is air-sensitive and needs to be prepared, handled, and stored under an inert atmosphere.

Figure 2.

Synthetic scheme and cyclic voltammogram of ^tBu-OED₄. Cyclic voltammogram was recorded in DMF with 0.1 M TBAPF₆ as the supporting electrolyte at 100 mV/s. E° = −1.33 V (vs Fc⁺/Fc). Arrow indicates the scan direction of the cyclic voltammogram.

^tBu-OED₄ was characterized by ¹H and ¹³C NMR spectroscopy, UV-Vis spectroscopy, HRMS, and elemental analysis. The cyclic voltammogram of ^tBu-OED₄ in DMF gives a single two electron reductive feature, which is characteristic of systems where the loss of the second electron occurs more readily than the loss of the first electron (Figure 2). Other reductants based on a 2,2’-bipyridine scaffold, such as DMAP-OED₃ display the same single two electron reduction.¹³ The two-electron reduction potential of ^tBu-OED₄ is −1.33 V, which indicates that ^tBu-OED₄ is a significantly weaker reductant than DMAP-OED₃ (E° = −1.69 V) and will be compatible with a wider range of organic substrates. Interestingly, the cyclic voltammograms of related reductants such as ^tBu-OED₃ and Me-OED₃, which contain three-carbon linkers, display two single reductions.^12c The potential for the first reduction is similar to ^tBu-OED₄ but it is harder to remove the second electron. This suggests that the use of ^tBu-OED₃ or Me-OED₃ in XEC may result in a buildup in the concentration of the singly reduced reductant, which could sequester alkyl radicals unproductively. Another advantage of ^tBu-OED₄ is that it has good solubility in a range of common organic solvents including pentane, benzene, THF, and DMF (see SI Section D) indicating that it should be compatible with a broad range of reaction conditions. This is likely benefit of synthesizing a reductant based on 4,4’-^tBu₂-2,2’-bipyridine compared with either 2,2’-bipyridine or 4,4’-Me₂-2,2’-bipyridine.

Application of ^tBu-OED₄ in Ni/Co Dual-Catalyzed XEC

Our previous studies indicate that homogeneous reductants can often not be used as direct replacements for heterogeneous reductants in XEC.^8e,9,11 Instead, to account for subtle mechanistic differences, either the reaction conditions need to be changed or additives used in order to maximize product yield. To promote XEC reactions between aryl and alkyl halides using homogeneous reductants, we recently reported a Ni/Co dual catalytic method for C(sp²)–C(sp³) XEC using TDAE as the reductant.^8e Initially, we demonstrated that ^tBu-OED₄ can be used as a direct replacement for TDAE in a Ni/Co co-catalyzed reaction involving 2-bromo-toluene and 1-bromo-3-phenylpropane without modification to the optimal catalyst loadings used in the studies involving TDAE (Eq 1). The NMR yield is 87%, which is comparable to our yields with TDAE as the reductant. This result indicates that ^tBu-OED₄ is suitable for XEC reactions and confirms our previous observation that this Ni/Co XEC reaction is tolerant of reductants spanning a wide range of reduction potentials (up to 0.6 V range in E°).⁹

A feature of the Ni/Co dual catalytic XEC method is its wide substrate scope.^8e To test if the substrate scope was retained with the new ^tBu-OED₄ reductant, we performed a series of XEC reactions with different substrates (Figure 3). For each substrate, the NMR yield was optimized to ~70% yield or greater by varying the catalyst loadings of Ni and Co. The optimization can be completed rationally rather than empirically because the reaction by-products guide changes to the catalyst loadings.^8e Specifically, if alkyl bromide remains after all the aryl halide substrate is consumed, then this indicates that the rate of alkyl radical generation from the alkyl bromide by Co^II(Pc) is too slow and the Co loading needs to be increased (or the Ni loading decreased). Conversely, if aryl halide remains after all the alkyl bromide is consumed, this indicates that the rate of alkyl radical generation is too fast and not matched with the rate of the production of the Ni^II aryl halide intermediate, which is responsible for radical trapping. Consequently, the loading of Ni needs to be increased (or the loading of Co decreased).

Figure 3:

Substrate scope for dual-catalyzed cross-electrophile coupling between aryl halides and 1-bromo-3-phenylpropane using ^tBu-OED₄ as the reductant. Values outside of parentheses are isolated yields, and values inside of parentheses are NMR yields, which were determined by integration of selected peaks in ¹H NMR spectra against a hexamethylbenzene internal standard.

Using ^tBu-OED₄ as the reductant, we can successfully couple bromobenzene (3a), as well as aryl halides containing electron donating (3b) and electron withdrawing substituents (3c). A major advantage of homogeneous reductants in XEC is that they are generally more tolerant to steric bulk and the successful coupling of substrates 3d, 3e, and 3f indicate that this is also the case with ^tBu-OED₄. In particular, the coupling of 2,6-dimethoxyiodobenzene is notable because there are relatively few systems for XEC that can couple di-ortho substituted aryl halides.¹⁴ Lower product yield was observed with less activated 2,6-dimethoxybromobenzene. The new reductant is also compatible with functional groups that can lead to further functionalization, such as pinacolborane (3g), alcohols (3h), aryl chlorides (3i), and esters (3j). In some cases, relatively low isolated yields were obtained for these substrates due to separation problems, but the reactions themselves typically proceed in high yield. In medicinal chemistry, structures containing heteroaromatic groups are ubiquitous, and therefore coupling methods for heteroaryl halides are important.¹⁵ Using ^tBu-OED₄ as the reductant, both a bromopyridine (3k) and a benzothiophene (3l) could be coupled. To further demonstrate the compatibility of ^tBu-OED₄ with heteroaryl substrates, we performed a reaction between 1-bromo-3-phenylpropane and a complex aryl halide that was identified as an intermediate in a drug discovery program from the Aryl Halide Informer Library (Eq. 2).¹⁶ We were able to isolate product II in 71% yield demonstrating that ^tBu-OED₄ can be used with complex molecules. Overall, these results demonstrate that ^tBu-OED₄ can be used as a reductant for Ni/Co catalyzed XEC. The yields are generally comparable to those obtained using TDAE, but as discussed earlier there are some practical benefits associated with ^tBu-OED₄. At this stage we have not determined the fate of ^tBu-OED₄ in catalysis, but likely it is forming an oxidized salt such as compound I (Figure 2). Given that our synthesis of ^tBu-OED₄ involves reduction of I, this suggests that if an analogue of I can be isolated from the reaction mixture it can be reduced back to ^tBu-OED₄. Thus, in principle it should be possible to recycle ^tBu-OED₄.

Conclusions

In this work, we developed a new homogeneous reductant, ^tBu-OED₄, based on 4,4’-^tBu₂-2,2’-bipyridine. The new reductant can be synthesized on a multigram scale in two steps from inexpensive and commercially available reagents. ^tBu-OED₄ has a much lower reduction potential, −1.33 V, compared to DMAP-OED₃, an analogous reductant based on 4,4’-NMe₂-2,2’-bipyridine. As a result, ^tBu-OED₄ can be used as the reductant in a Ni/Co dual-catalyzed XEC reaction between aryl and alkyl halides, where it displays high functional group tolerance. We anticipate that modification of the 2,2’-bipyridine scaffold will lead to the development of additional homogeneous organic reductants that span a wider range of reduction potentials, and this is the subject of ongoing research in our laboratory.