Acrylonitriles, especially substituted acrylonitriles, are versatile building blocks widely occurring in the pharmaceutical industry, natural products and synthetic organic chemistry (Fringuelli et al., 1994; Fleming, 1999; Fleming et al., 2010; Carta et al., 2011; Shen et al., 2015; Baker et al., 2020; Sirim et al., 2020; Solangi et al., 2020; Baker et al., 2021). Among acrylonitrile-containing molecules, aryl acrylonitriles are an important subclass in the medicinal chemistry and pharmaceutical industry (ANI-7 (Tarleton et al., 2011), CDCPA (Baker et al., 2018), TPAT-AN-XF (Niu et al., 2019), CC-5079 (Zhang et al., 2006), Entacapone (Seeberger and Hauser, 2009), and Rilpivirine (Clercq, 2005) Figure 1). Therefore, the development of efficient and practical approaches for the synthesis of aryl acrylonitriles remains in demand.

Classical synthetic routes to acrylonitrile derivatives include the Wittig/Horner−Wadsworth−Emmons reaction (Zhang et al., 1998; Kojima et al., 2002; Fang et al., 2011; Ando et al., 2013) and Peterson type reactions (Kojima et al., 2004; Pabmo’ et al., 1990; Palomo et al., 1990). However, these procedures suffer from limitations such as a poor substrate scope, low efficiency for the synthesis of polysubstituted acrylonitriles. During the past decade, organic chemists keep searching new and efficient reactions, including oxidative Heck-type reactions (Zou et al., 2003; Zhang and Liebeskind, 2006), cyanation of alkenyl halides (Stuhl, 1985; Alterman and Hallberg, 2000; Pradal and Evano, 2014; Ahuja and Sudalai, 2015; Chaitanya and Anbarasan, 2015; Yang et al., 2018), alcohols (Oishi et al., 2009; Rokade et al., 2012; Thiyagarajan and Gunanathan, 2018; Yadav et al., 2020), aldehydes (Tomioka et al., 2011; Laulhe et al., 2012; Del Fiandra et al., 2016; Wu et al., 2016), acrylamide/oxime dehydration (Yamaguchi et al., 2007; Zhou et al., 2009), carbocyanation of alkynes (Nakao et al., 2007; Cheng et al., 2008; Minami et al., 2013; Yang et al., 2013; He et al., 2016; Qi et al., 2017), cross-metathesis (Crowe and Goldberg, 1995; Randl et al., 2001; Mu et al., 2019), and direct conversion of allylic carbon to nitrile (Qin and Jiao, 2010; Zhou et al., 2010) have been developed and could be applied for the synthesis of acrylonitriles. For example, Jiao developed a series of powerful synthesis of substituted acrylonitriles, which used allyl esters or halides and NaN₃ or TMSN₃ by a tandem Pd-catalyzed azidation and the subsequent oxidative rearrangement process (Scheme 1A) (Qin and Jiao, 2010; Zhou et al., 2010; Jiao et al., 2011; Wang and Jiao, 2014). Engle reported a direct oxidative cyanation of terminal and internal alkenes to prepare substituted acrylonitriles using a homogeneous copper catalyst and a bystanding N–F oxidant (Scheme 1B) (Gao et al., 2018). Recently, Liu reported an elegant synthesis of aryl substituted terminal acrylonitriles through Ni/Mn-catalyzed hydrocyanation of terminal alkynes with Zn(CN)₂ (Scheme 1C) (Zhang et al., 2018). Milstein reported an effective synthesis of aryl acrylonitriles through dehydrogenative coupling of alcohols with nitriles catalyzed by a pincer complex of manganese at 135°C for 43–60 h (Scheme 1D) (Chakraborty et al., 2017).

Despite these advances, these motheds are generally restricted by the addition of dangerous reagents (cyanide reagents, azide reagents) and stoichiometric amount of oxidants (DDQ, Selectfluor), high catalyst loading, tedious synthetic procedures, low yielding and high reaction temperatures. Therefore, an optional method for the efficient synthesis of aryl acrylonitrile derivatives under mild reaction conditions using simple, easily available substrates are very necessary. Herein, we report an efficient synthesis of aryl acrylonitrile derivatives using a Palladium/NIXANTPHOS-based catalyst system. This approach furnishes efficient access to a variety of substituted and functionalized aryl acrylonitriles (21 examples, up to 95%). The scalability of the transformation was demonstrated and the derivatizations of the aryl acrylonitrile were conducted.

We initiated our reaction optimization by using phenylacetonitrile 1a and 2-bromoprop-1-ene 2a as the model substrates. At the outset, based on our experience with deprotonative cross-coupling processes of weakly acidic substrates (Yang et al., 2016; Duan et al., 2018; Liu et al., 2018), we have found that NIXANTPHOS can effectively implement these conversions. The high reactivity of the Pd/NIXANTPHOS-based system may be due to the presence of the main group metal and the deprotonation of the ligand N−H moiety under basic reaction conditions (Zhang et al., 2014). A variety of palladium source including different Pd⁰ and Pd^II precursors, phosphine ligands and six bases (LiN(SiMe₃)₂, NaN(SiMe₃)₂, KN(SiMe₃)₂, LiO ^t Bu, NaO ^t Bu and KO ^t Bu) were examined the coupling of phenylacetonitrile 1a and 2-bromoprop-1-ene 2a in DME at 65°C for 1 h (Table 1, entries 1–10) (see the optimization of reaction conditions on page S2 in Supplementary Material). The top Pd/L/base combination from this screen was Pd(OAc)₂/NIXANTPHOS/NaO ^t Bu resulted in 20% assay yield (AY, determined by ¹H NMR analysis). Other four solvents (dioxane, CPME, THF and Toluene) were tested, which only afforded trace amount of product (0%–13%) (entries 11–14). Raising the reaction temperature to 80 and 100°C led to increases to 57 and 28% AY, respectively (entries 15 and 16). Changing the equivalent of 2a from 2 to 4 led to increases of AY (entries 17–19). When a 3 equivalent was employed, the AY increased to 77% (75% isolated yield, entry 18). Reducing the Pd/ligand ratio to 5:10, AY dropped to 57% (entry 20).

L1: NIXANTPHOS, L2: XANTPHOS, L3: PPh₃, L4: P (o-TOL)₃, L5: P (1-NAP)₃, L6: rac-BINAP, L7: JOHNPHOS, L8: PCy₃

B1: LiN(SiMe₃)₂, B2: NaN(SiMe₃)₂, B3: KN(SiMe₃)₂, B4: LiO ^t Bu, B5: NaO ^t Bu, B6: KO ^t Bu

With the optimized reaction conditions (Table 1, entry 18), we explored the structural diversity of vinyl halides/triflates using phenylacetonitrile 1a as the model substrate. As shown in Table 2, 2-bromo-1-ene 2a delivered aryl acrylonitrile 3aa in 75% yield, while 2-chloro-1-ene 2a’ gave 55% yield. Vinyl chloride 1-chloro-2-methylprop-1-ene 2b led to product 3ab in 54% yield. (E)-(1-bromoprop-1-en-2-yl)benzene 2c provided product 3ac in 81% yield (67% yield for 5% Pd/10% L). Sterically hindered bromomethylenecyclohexane 2d rendered product 3ad with excellent yield of 84% yield (70% yield for 5% Pd/10% L). Trans- and cis-2-bromobut-2-ene (2e and 2f) furnished products 3ae and 3af in overall 57% and 51% yields. 2-Bromo-3-methylbut-2-ene 2g afforded product 3ag in overall 65% yield. Cycloolefin halides/triflates were all suitable reaction partners in this transformation and provided a series of cycloalkane-functionalized aryl acrylonitriles in moderate yields. 1-Chlorocyclopent-1-ene 2h led to product 3ah in 50% yield. Six/seven/eight-membered cycloolefin triflates proceeded the corresponding products 3ai, 3aj, and 3ak in 65%, 61% and 55% yields, respectively.

We next explored the scope of arylacetonitriles using sterically hindered bromomethylenecyclohexane 2d as the model substrate. As shown in Table 3, in general, arylacetonitriles bearing electron-donating and electron-withdrawing Ar groups or heterocyclic rendered good to excellent yields under the standard conditions (Table 3). Arylacetonitriles possessing alkyl 4-Me (1b) and 2-Me (1c) reacted with bromomethylenecyclohexane 2d to give aryl acrylonitriles 3bd and 3cd in 84% and 81% yields (69% and 63% yields for 5% Pd). Arylacetonitrile with electro-donating (4-OMe, 1d) substituents provided product 3dd in 67% yield. Arylacetonitriles bearing electron-withdrawing 4-F (1e), 4-Cl (1f) and 4-Br (1g) generated the products 3ed, 3fd and 3gd in 83% (78% yield for 5% Pd), 95% (79% yield for 5% Pd) and 50% yields, respectively. The sterically demanding 2-naphthyl acetonitrile (1h) was well tolerated, led to product 3hd in 67% yield. Interesting, medicinally important heterocyclic-containing acetonitriles were suitable reaction partners. 2-(1-Methyl-1H-indol-3-yl)acetonitrile (1i) reacted with 2d to generate the aryl acrylonitrile 3id with excellent yield of 93% (79% yield for 5% Pd). 2-(Thiophen-2-yl)acetonitrile (1j) provided product 3jd in 47% yield.

To evaluate the scalability of our transformation, we next carried out the reaction of phenylacetonitrile 1a and 2-bromo-1-ene 2a on a gram-scale under the optimal conditions (Scheme 2). The desired aryl acrylonitrile 3aa was isolated in 1.03 g (70% yield), demonstrating the scalability of our method.

Finally, to illustrate further the synthetic versatility of the resulting aryl acrylonitrile, a series of derivatizations were performed on 3aa (Scheme 3). Thus, the selective reduction of the carbon-carbon double bond of aryl acrylonitrile 3aa using Pd/C and hydrogen led to the substituted saturated phenylacetonitrile 4aa in 95% yield. Then, the selective reduction of the nitrile group of 3aa employing DIBAL-H in toluene at 0°C generated the corresponding α,β-unsaturated aldehyde 4ab in 39% yield (Chen et al., 2019). Meanwhile, the hydrolysis of the nitrile group of 3aa using 30% H₂O₂ and NaOH in MeOH rendered the corresponding α,β-unsaturated amide 4ac in 78% yield. Furthermore, the epoxidation of the carbon-carbon double bond and hydrolysis of the nitrile group of 3aa using 30% H₂O₂ and K₂CO₃ in DMSO afforded the corresponding α,β-epoxy amide 4ad in 57% yield.

A possible catalytic cycle is shown in Scheme 4 based on Walsh’s work on the palladium-catalyzed deprotonative cross-coupling processes (Hussain et al., 2014; Jia et al., 2014; Mao et al., 2014; Jia et al., 2015). The deprotonation of aryl acetonitrile by NaO ^t Bu gives benzyl anions. After oxidative addition of the vinyl bromide to Pd (0), the vinyl palladium intermediate is proposed to bind the benzyl anions to form the palladium complex. Then, reductive elimination occurs to afford the enenitrile and regenerates Pd (0). Finally, enenitrile isomerizes to obtain aryl acrylonitrile.

In conclusion, we have successfully synthesized a series of aryl acrylonitrile derivatives employing a Pd/NIXANTPHOS-based catalyst system for the first time. In this protocol, commercially available arylacetonitriles and vinyl bromides/chlorides/triflates underwent palladium-catalyzed α-alkenylation to furnish efficient access to a variety of substituted and functionalized aryl acrylonitriles. The scalability of the mothed was demonstrated by the gram-scale reaction. A series of derivatization of aryl acrylonitrile were performed, including the selective reduction of the double bond or nitrile group, the hydrolysis of the nitrile group, and the epoxidation of the double bond, which demonstrated the synthetic versatility of aryl acrylonitrile. It is noteworthy that this approach does not require dangerous reagents and stoichiometric amount of oxidants, which enables the synthesis of a range of aryl acrylonitriles in an effective and straightforward means.