A computational study for the reaction mechanism of metal-free cyanomethylation of aryl alkynoates with acetonitrile.

A computational study of metal-free cyanomethylation and cyclization of aryl alkynoates with acetonitrile is carried out employing density functional theory and high-level coupled-cluster methods, such as coupled-cluster singles and doubles with perturbative triples [CCSD(T)]. Our results indicate that the reaction of aryl alkynoates with acetonitrile in the presence of tert-butyl peroxybenzoate (TBPB) under metal-free conditions tends to proceed through cyanomethylation, spirocyclization and ester migration of the kinetically favoured coumarin derivatives. 1,2-Ester migration in the spiro-radical intermediate 10 does not proceed via the formation of the carboxyl radical 11 suggested by Sun and co-workers. Our results also demonstrate that the t-butoxy radical is substantially responsible the formation of the cyanomethyl radical by the abstraction of a hydrogen atom from acetonitrile. This journal is © The Royal Society of Chemistry.

Introduction

Cyanomethylation is a very useful reaction to synthesize cyano-containing compounds in medicinal and synthetic organic chemistry due to easy conversion of the cyano group into primary amines, ketones, carboxylic acids, esters, amides and even tetrazoles.[1-3] In addition, cyanomethyl moieties obtained via cyanomethylation are found as versatile motifs in some important bioactive natural products and drugs.[4,5] A variety of protocols have been reported, including transition-metal-catalyzed methods[6-16] and metal-free catalyzed methods[17-22] for the synthesis of cyanomethyl-containing compounds. Compared to the toxic metal used, metal-free catalyzed methods used acetonitrile[9-12,17,18] and its analogs such as bromoacetonitrile,[13-16] cyanoacetic acid[19,20] and ethyl cyanoacetate[5] have drawn considerable attention due to its inexpensive and environmental-friendly properties. In recent years, metal-free cyanomethylation using of acetonitrile as a cyanomethyl source in the synthesis of the cyano-containing compounds was reported.[17,18,21] In 2016, Sun and co-workers[18] demonstrated a convenient and efficient method for the synthesis of 3-cyanomethylated coumarins, a class of very important heterocyclic compounds in pharmaceuticals and dyes, via cyanomethylation and cyclization of aryl alkynoates using acetonitrile as the cyanomethyl source in the presence of tert-butyl peroxybenzoate (TBPB) under metal-free conditions. A control experiment carried out in the presence of 2,6-di-tert-butyl-4-methylphenol (BHT) or 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) showed that the thermal cyanomethylation and cyclization of aryl alkynoates proceeds via a radicalic process as described in Scheme 1.

Scheme 1

Proposed mechanism by Sun and co-workers.[18]

According to the mechanism proposed by Sun and co-workers,[18] at first, TBPB (1) decomposed into the benzoyloxy (2) and tert-butoxy radical (3) at high temperature. One of these radicals abstracted a hydrogen atom from acetonitrile (4) to generate the cyanomethyl radical (5). An electrophilic attack of cyanomethyl radical (5) to alkynoate (8) yields the alkenyl radical 9. Then, an intramolecular spirocyclization of the alkenyl radical 9 gives the spiro-radical intermediate 10. Next, the ester migration took place via a carboxyl radical 11 to yield the radical 12. Finally, the radical 12 was oxidized to a carbocation or a radical followed by deprotonation or abstraction of a hydrogen radical to yield the cyano-containing coumarin derivatives 13. Additionally, in these reactions, benzoic acid (6) and tert-butanol (7) are also formed as by-product.

Result and discussion

Our results for the mechanism proposed by Sun and co-workers are reported in Fig. 1 and 2. For the initiation of the radicalic process, the 1 → 2 + 3 conversion, the computed reaction free energy is 30.8 kcal mol−1 (Fig. 1). Zhou[23] was reported that the bond dissociation energy (BDE) (Gibbs free energy at 383.15 K) of TBPB is 11.2 kcal mol−1 at M06-L-D3/6-31+G(d,p) level and no any transition state (TS) for this step. We did not also find any TS for this step. The decomposing of TBPB (1) into benzoyloxy (2) and t-butoxy radical (3) was carried out at high temperature, which provide enough energy for this dissociation process. To generating the cyanomethyl radical (5), a hydrogen atom from acetonitrile (4) is abstracted by one of these radicals. For the abstraction of a hydrogen atom from acetonitrile (4) by benzoyloxy radical (2), 2 → 5, the reaction free energy and barrier are −14.1 and 14.4 kcal mol−1, respectively. For the other path for the formation of cyanomethyl radical (5), 3 → 5, the reaction free energy and barrier are −9.4 and 8.9 kcal mol−1, respectively. The barrier of the abstraction of a hydrogen atom from acetonitrile (4) by t-butoxy radical (3) is 5.5 kcal mol−1 lower than that of benzoyloxy radical (2) (Fig. 1). These results show that t-butoxy radical (3) is substantially responsible the abstraction of hydrogen atom.

Fig. 1

Relative free energy profile (at 403.15 K) for the formation of cyanomethyl radical (5) shown in Scheme 1 at the DLPNO-CCSD(T)/cc-PVTZ//B3LYP-6-311G(d,p) level.

Fig. 2

Relative free energy profile (at 403.15 K) for reaction mechanism of 8a shown in Schemes 1 and 2 at the DLPNO-CCSD(T)/cc-PVTZ//B3LYP-6-311G(d,p) level.

In the literature, there are many studies for the reaction of aryl alkynotes with a variety of radicals. As a result of these studies, coumarin derivatives (13 (ref. 18 and [24-35]) or 16 (ref. 36–45)) or styrene derivatives (21 (ref. 46–48) or 22 (ref. 49)) were obtained under different reaction conditions. All plausible mechanism for aryl alkanoates 8 with cyanomethyl radical (5) is proceed as described in Scheme 2. There are two possible electrophilic attack of cyanomethyl radical (5) to the C–C triple bond of aryl alkynoate derivatives 8. In the first one, the cyanomethyl radical (5) is reacted with alkynes, 8, to form the intermediate 9. The alternative electrophilic attack of cyanomethyl radical (5) to alkynes, 8, is formed the intermediates 14. For the electrophilic attack of cyanomethyl radical (5) to the C–C triple bond of aryl alkynoate derivative 8a, 8a → 9a, the reaction free energy and barrier are −24.2 and 6.9 kcal mol−1, respectively. For the alternative electrophilic attack of cyanomethyl radical (5) to the alkyne 8a, 8a → 14a, the reaction free energy and barrier are −12.6 and 10.5 kcal mol−1, respectively. Both the intermediate 9a is more stable than the intermediate 14a (the relative energy of 9a is lower 11.6 kcal mol−1 than that of 14a) and the barrier for the formation of 9a is lower 3.6 kcal mol−1 than the barrier for the formation of 14a. This result clarifies why the electrophilic attack of cyanomethyl radical (5) to alkynes, 8, is formed the formation of the intermediates 9. The intramolecular spirocyclization of the alkenyl radical 9 gives the spiro-radical intermediate 10, while the intramolecular cyclization of the alkenyl radical 9 gives the intermediate 15. For the intramolecular spirocyclization of the alkenyl radical 9, 9a → 10a, the reaction free energy and barrier are −15.3 and 6.0 kcal mol−1, respectively. For the intramolecular cyclization of the alkenyl radical 9, 9a → 12a, the reaction free energy and barrier are −15.7 and 11.8 kcal mol−1, respectively. Also, decarboxylation of the intermediate 17 forming with release acetylene derivative 18 generates phenyl radical 19. For the formation of the intermediate 17, 9a → 17a, the reaction free energy and barrier are 30.7 and 36.2 kcal mol−1, respectively. For the decarboxylation of the intermediate 17, 17a → 19a, the reaction free energy and barrier are −8.0 and 22.9 kcal mol−1, respectively. While the aryl migration via cleavage of C–O bond yields the carboxyl radical 11, the ester migration process without the formation of the carboxyl radical 11 affords the intermediate 12. For 1,2-aryl migration, 10a → 11a, the reaction free energy and barrier are 14.4 and 14.9 kcal mol−1, respectively. For the 10a → 12a conversion, the reaction energy is −0.4 kcal mol−1 and 1,2-ester migration proceeds barrierless. The cyclization of carboxyl radical 11 yields the intermediate 12 oxidized by abstraction of a hydrogen radical to yield the desire product, coumarin derivatives 13. For the cyclization of carboxyl radical 11, 11a → 12a, the reaction free energy and barrier are −14.8 and 2.9 kcal mol−1, respectively. For the oxidation by generating a hydrogen radical of the intermediate 12, the reaction free energy and barrier are 17.1 and 26.0 kcal mol−1, respectively. However, the oxidation of the intermediate 12 with single electron oxidation followed by deprotonation, or abstraction of the hydrogen atom by benzoyloxy radical (2) or t-butoxy radical (3) proceeds barrierless. The decarboxylation with release of carbon dioxide of the carboxyl radical 11 yields alkenyl radical 20, which is ready to afford the formation of styrene derivatives 21 and 22. For the decarboxylation of the carboxyl radical 11, 11a → 20a, the reaction free energy and barrier are −11.6 and 4.4 kcal mol−1, respectively.

Scheme 2

Plausible mechanism for aryl alkanoates with cyanomethyl radical.

Conclusion

In this study, all plausible mechanism for metal-free cyanomethylation of aryl alkynoates with acetonitrile has been investigated with high-level coupled-cluster methods, such as DLPNO-CCSD(T), along with the cc-pVTZ basis set. According to our computations, the decomposition of TBPB, which is the initial step of the radicalic reaction, include the highest reaction energy step and show that the rate-determining step is the formation of the benzoyloxy (2) and tert-butoxy radical (3) for the reaction scheme. 1,2-Ester migration in the spiro-radical intermediate 10 does not proceed via the formation of the carboxyl radical 11 suggested by Sun and co-workers because the relative energy of carboxyl radical 11 is nearly same that of TS 10/11 and activation free energy for the 11a → 20a conversion is very low for the proposed reaction condition. Also, solvent corrected free energies demonstrated that carboxyl radical 11 is not formed in the reaction conditions (see ESI†). Additionally, no observed styrene derivatives support that the intermediates 11 are not formed during the reaction. Our results demonstrate that the reaction of aryl alkynoates using acetonitrile as the cyanomethyl source in the presence of TBPB tends to yield the kinetically favoured product. Our results also indicate that t-butoxy radical (3) is substantially responsible in the formation of cyanomethyl radical (5). The computations are compatible with experimental results. Moreover, our computations provide useful insight into the reaction mechanism of aryl alkynotes with a variety of radicals.

Computational methods

Geometry optimizations and harmonic vibrational frequency computations for the structures considered were carried out with the density functional theory (DFT), the B3LYP functional.[50-53] For this purpose, a Pople-type polarized triple-ζ split-valence basis sets, 6-311G(d,p), was used.[54-56] Single point energies were computed at the optimized geometries with the coupled-cluster singles and doubles with perturbative triples method employing the domain based local pair natural orbitals approach [DLPNO-CCSD(T)].[57,58] In DLPNO-CCSD(T) computations Dunning's correlation consistent triple-ζ split valence basis set, cc-pVTZ, is employed.[59,60] DFT computations were carried out by using the Gaussian 09 program,[61] while the ORCA package[62,63] is employed for DLPNO-CCSD(T) computations. The solvation effects were considered employing the polarizable continuum model (PCM) in acetonitrile.[64] For the transition state (TS) between species A and B, the A/B notation is used throughout the article.

Conflicts of interest

There are no conflicts to declare.