Synthesis of benzoyl cyanide through aerobic photooxidation of benzyl cyanide using carbon tetrabromide as a catalyst

Abstract

We developed a synthetic method toward benzoyl cyanide through aerobic photooxidation of benzyl cyanide in the presence of carbon tetrabromide under visible light irradiation with fluorescent lamps.

---

Acyl cyanide is a useful synthetic intermediate in organic synthesis as it can be utilized in a variety of transformations as both CO and CN functionalities. For instance, effective cycloaddition of azides with benzoyl cyanides affords benzoyltetrazols.¹ Furthermore, it can be used in the synthesis of oxazoles^2–4 and tetrazolinones.⁵ In general, benzoyl cyanides are synthesized by cyanation of benzoyl chloride using AgI⁶ or oxidation of mandelonitrile by various methods.⁷ In addition, aroyl cyanides are synthesized via the formation of cyanohydrin nitrate intermediates using AgNO₃.⁸ Furthermore, oxidation of benzyl cyanide has been used for the synthesis of benzoyl cyanide; however, most of these reactions require expensive and environmentally hazardous transition metal catalysts such as ruthenium or vanadium.^9–12

Because of the increasing demand for more environmentally benign syntheses, molecular oxygen has received much attention because it is more atom-efficient than other oxidants such as toxic heavy metals or complex organic reagents. In addition, molecular oxygen is inexpensive and theoretically produces only water as the end product of oxidation.¹³ However, to the best of our knowledge, there has been no investigation of the catalytic oxidation of benzyl cyanide to benzoyl cyanide under metal-free conditions using molecular oxygen as the terminal oxidant.

In contrast, the effective use of visible light is important for environmental load-reducing reactions as it exudes no residue and has neither shape nor weight. As such, it is an important component in the examination of environmentally benign processes.¹⁴ As a result, we have been studying the aerobic photooxidation of various substrates using molecular oxygen and visible light from fluorescent lamps.¹⁵ For instance, we previously reported the photooxidation of a benzyl amide to the corresponding imide using catalytic amounts of a bromine source with photoirradiation.¹⁶ However, benzyl cyanide is less stable than benzyl amide; therefore, we considered that the oxidation of benzyl cyanide would be possible under these mild conditions. After investigating the ideal reaction conditions, we found that the photooxidation of benzyl cyanides occurred in the presence of carbon tetrabromide (CBr₄) and water with irradiation using fluorescent lamps (Scheme 1). Here we report a detailed study of the photooxidation of benzyl cyanides to benzoyl cyanides.

Scheme 1 Photo-oxidation of benzyl cyanide.

To explore this approach, we selected benzyl cyanide (1a) as the test substrate to optimize the reaction conditions (Table 1). We first compared the effect of additives with K₂CO₃ as the base, TFA as the acid, or no additives. From this, we found that TFA dramatically improved the yield of 2a, while oxidation in the presence of K₂CO₃ did not proceed at all (entries 1–3). Thus, we screened solvents and bromine sources in the presence of TFA and molecular oxygen under photoirradiation from fluorescent lamps (entries 4–21). From this, we found that benzoyl cyanide (2a) was produced most efficiently and in the best yield using 0.2 equivalents of CBr₄ as the catalyst and ethyl acetate as the solvent (entry 2). Next, we examined other acids as additives, which produced benzoyl cyanides in low to moderate yields (entries 22–24). We also found that H₂O was the best additive as a proton source (entry 25–28). Furthermore, the best yield of benzoyl cyanide was obtained when 1.5 equivalents of water was used (entry 27).

Table 1 Study of reaction conditions^a

Entry	Bromo source (equiv.)	Solvent	Additive (equiv.)	Conversion^b (%)
a Reaction conditions: 1a (0.3 mmol), bromo source (0.2 equiv.) and additive in solvent was stirred and irradiated externally with fluorescent lamps for 20 h.b Conversion yields, determined by ^1H-NMR with 1,1,2,2-tetrachloroethane as an internal standard.
1	CBr4 (0.2)	EtOAc	—	6
2	CBr4 (0.2)	EtOAc	TFA (0.1)	69
3	CBr4 (0.2)	EtOAc	K2CO3 (0.1)	0
4	CBr4 (0.2)	Hexane	TFA (0.1)	3
5	CBr4 (0.2)	Cyclohexane	TFA (0.1)	7
6	CBr4 (0.2)	THF	TFA (0.1)	43
7	CBr4 (0.2)	CHCl3	TFA (0.1)	0
8	CBr4 (0.2)	CH2Cl2	TFA (0.1)	Trace
9	CBr4 (0.2)	Acetone	TFA (0.1)	Trace
10	CBr4 (0.2)	i-PrOH	TFA (0.1)	0
11	CBr4 (0.2)	MeOH	TFA (0.1)	Trace
12	CBr4 (0.2)	Neat	TFA (0.1)	Trace
13	48% aq. HBr (0.2)	EtOAc	TFA (0.1)	0
14	NBS (0.2)	EtOAc	TFA (0.1)	7
15	KBr (0.2)	EtOAc	TFA (0.1)	4
16	NaBr (0.2)	EtOAc	TFA (0.1)	4
17	AgBr (0.2)	EtOAc	TFA (0.1)	Trace
18	MgBr2 (0.2)	EtOAc	TFA (0.1)	Trace
19	CBr4 (0.1)	EtOAc	TFA (0.1)	Trace
20	CBr4 (0.3)	EtOAc	TFA (0.1)	55
21	CBr4 (0.4)	EtOAc	TFA (0.1)	16
22	CBr4 (0.2)	EtOAc	AlCl3 (0.1)	25
23	CBr4 (0.2)	EtOAc	AcOH (0.1)	40
24	CBr4 (0.2)	EtOAc	HBF4 (0.1)	61
25	CBr4 (0.2)	EtOAc	H2O (0.5)	70
26	CBr4 (0.2)	EtOAc	H2O (1.0)	68
27	CBr4 (0.2)	EtOAc	H2O (1.5)	74
28	CBr4 (0.2)	EtOAc	H2O (2.0)	71

---

Table 2 shows the scope and limitations of the oxidation of benzyl cyanide under the optimized reaction conditions (entry 27).¹⁷ As the crude products are labile and we couldn't purify them by flush column chromatography or Kugelrohr distillation,^18,19 they were characterized by ¹H NMR (Table 2, yield 1) and isolated as the corresponding carboxylic acids after hydrolysis of in the presence of sulfuric acid (Table 2, yield 2). As a result, benzoic acid was obtained in 73% isolated yield (entry 1). The corresponding benzoyl cyanides were detected when electron withdrawing groups were present on the benzene ring (entries 2–9). In particular, 4-chlorobenzyl cyanide and 4-bromobenzyl cyanide gave the corresponding benzoyl cyanides in good yields (entries 2 and 3). In contrast, oxidation of 4-fluorobenzyl cyanide was difficult; thus, an increase in the time and amount of CBr₄ used was necessary (entry 4). 3-Chlorobenzoyl cyanide and 3-bromobenzoyl cyanide were obtained in moderate yields (entries 5 and 6). Both 2-chlorobenzyl cyanide and 2,4-dichlorobenzyl cyanide provided the corresponding products in modest yields because of steric reasons (entries 7 and 8). 3,4-Dichlorobenzoyl cyanide resulted in a low yield (entry 9). Substrates with an electron donating group, 4-methoxy, or 1-naphthaleneacetonitrile were converted to the desired products in low yields when CBr₄ was increased and reaction time was extended (entries 11 and 12). Unfortunately, aliphatic cyanide was not oxidized (entry 13).

Table 2 Scope and limitation^a

Entry	Time (h)	Product 1	Conversion^b	Product 2	Yield^c
a Reaction conditions: substrate (0.3 mmol), bromo source and additive (1.5 equiv.) in solvent was stirred and irradiated externally with fluorescent lamps for 20 h. The solvent was removed by rotary evaporation, then we took ^1H NMR. After dryness by rotary evaporation again, it was added H2O (140 mL) and conc. H2SO4 (150 mL). The mixture was stirred under reflux for 30 min.b Conversion yields, determined by ^1H-NMR with 1,1,2,2-tetrachloroethane as an internal standard.c Isolated yields.d 0.4 equiv. of CBr4.e 0.8 equiv. of CBr4.
1	20		74		73
2	20		73		72
3	20		88		71
4^d	40		45		49
5^d	65		68		60
6	20		64		67
7	20		53		46
8	20		34		40
9	20		25		19
10^d	65		71		21
11^e	65		37		34
12^e	110		28		16
13	20		0

---

To determine the reaction mechanism, we performed several control experiments (Scheme 2). Under air, the oxidation proceeded slightly with 89% of the starting material and 8% of the intermediate being recovered (eqn (1)). Under an argon atmosphere, the oxidation did not proceed at all (eqn (2)). These results suggest the necessity of sufficient molecular oxygen for this oxidation. The oxidation also did not proceed at all in the dark (eqn (3)); thus, photoirradiation is required for this oxidation. Without CBr₄, the reaction did not proceed at all; thus, CBr₄ is essential for the oxidation (eqn (4)). These results suggest that CBr₄, photoirradiation, and molecular oxygen are critical for this reaction. A catalytic amount of BHT, which is a radical inhibitor, completely suppressed the reaction (eqn (5)), suggesting that this reaction most likely proceeds via a radical chain mechanism.

Scheme 2 Study of reaction mechanism.

We also performed a time course experiment for the oxidation of benzyl cyanide (1a) under the optimized reaction conditions (Fig. 1). Mandelonitrile (4a) was detected by ¹H NMR in the early stages of the reaction, as well as small amounts of benzoyl cyanide (2a). In the middle stages, benzoyl cyanide (2a) was synthesized quickly, while the amount of mandelonitrile (4a) decreased. After 20 h, the starting material and mandelonitrile (4a) completely disappeared. Thus, it was confirmed that mandelonitrile was synthesized first, while benzoyl cyanide was synthesized gradually. This suggests that mandelonitrile is the intermediate of this reaction.

Fig. 1 Time course. The conversions were determined by ¹H-NMR with 1,1,2,2-tetrachloroethane as an internal standard.

Scheme 3 shows a further study on the availability of benzoyl cyanide (2a). The synthesis of α-cyanobenzyl benzoate (5a) from benzoyl cyanide (2a) was previously reported.²⁰ Thus, we attempted to synthesize α-cyanobenzyl benzoate (5a) in a one-pot reaction using the aerobic photooxidation from benzyl cyanide (1a) through benzoyl cyanide (2a). As a result, the product was obtained in moderate yields.

Scheme 3 Synthesis of α-cyanobenzyl benzoate.

Conclusions

In conclusion, we developed a synthetic method for the aerobic photooxidation of benzyl cyanide to benzoyl cyanide using a catalytic amount of carbon tetrabromide in the presence of water under photoirradiation from fluorescent lamps. Although the detailed mechanism of this reaction was not determined, it was proven that it most likely proceeds via a radical chain reaction with H₂O as the proton source. This novel reaction is valuable as it uses inexpensive and harmless molecular oxygen and visible light from general purpose fluorescent lamps. The application of this photooxidation to other reactions is now in progress in our laboratory.

Notes and references

1. Z. P. Demko and K. B. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2113.
2. R. Lakhan and R. L. Singh, J. Heterocycl. Chem., 1988, 25, 1413.
3. A. Radspieler and J. Liebscher, Synthesis, 2001, 745.
4. M. Davis, R. Lakhan and B. Ternai, J. Heterocycl. Chem., 1977, 14, 317.
5. J. P. Horwitz, B. E. Fisher and A. J. Tomasewski, J. Am. Chem. Soc., 1959, 81, 3076.
6. Z. Li, S. Shi and J. Yang, Synlett, 2006, 2495.
7. (a) J. G. Lee, J. A. Lee and S. Y. Sohn, Synth. Commun., 1996, 26(3), 543; (b) J. C. Lee and S. B. Lee, Repub. Korean Kongkae Taeho Kongb (South Korean patent) KR 2009030384 A; (c) S. Murahashi, et al., Tetrahedron Lett., 1985, 26(7), 925.
8. T. Sueda, M. Shoji and K. Nishide, Tetrahedron Lett., 2008, 49, 5070.
9. S. Murahashi, T. Naota and T. Kuwabara, Synlett, 1989, 62.
10. H. Harle and H. C. Jochims, Chem. Ber., 1986, 119, 1400.
11. B. M. Choudary and P. N. Reddy, J. Org. Chem., 1996, 112, 385.
12. X. Chen, T. Chen, Q. Li, Y. Zhou, L. Han and S. Yin, Chem.–Eur. J., 2014, 20, 12234.
13. For a recent review see: (a) T. Punniyamurthy, S. Velusamy and J. Iqbal, Chem. Rev., 2005, 105, 2329; (b) T. Mallat and A. Baiker, Chem. Rev., 2004, 104, 3037; (c) S. S. Stahl, Angew. Chem., Int. Ed., 2004, 43, 3400; (d) J. Piera and J. E. Backvall, Angew. Chem., Int. Ed., 2008, 47, 3506; (e) G. Urgoitia, R. SanMartin, M. T. Herrero and E. Dominguez, Chem. Commun., 2015, 51, 4799.
14. (a) Photochemistry of Organic Compounds: From Concepts to Practice, ed. P. Klán and J. Wirz, John Wiley & Sons, Ltd, 2009; (b) N. Hoffmann, Chem. Rev., 2008, 108, 1052.
15. (a) K. Matsui, T. Ishigami, T. Yamaguchi, E. Yamaguchi, N. Tada, T. Miura and A. Itoh, Synlett, 2014, 25, 2613; (b) I. Itoh, Y. Matsuzaki, A. Fujiya, N. Tada, T. Miura and A. Itoh, Tetrahedron Lett., 2014, 55, 3160; (c) T. Yamaguchi, T. Nobuta, N. Tada, T. Miura, T. Nakayama, B. Uno and A. Itoh, Synlett, 2014, 25, 1453; (d) Y. Nagasawa, Y. Matsusaki, T. Hotta, T. Nobuta, N. Tada, T. Miura and A. Itoh, Tetrahedron Lett., 2014, 55, 6543; (e) A. Fujiya, A. Kariya, T. Nobuta, N. Tada, T. Miura and A. Itoh, Synlett, 2014, 25, 884.
16. N. Tada, K. Ban, M. Yoshida, S. Hirashima, T. Miura and A. Itoh, Tetrahedron Lett., 2010, 51, 6098.
17. General Procedure: a solution of benzyl cyanide (1a, 0.3 mmol), carbon tetrabromide (0.2 equiv.), and H₂O (1.5 equiv.) in dry EtOAc (5 mL) in a pyrex test tube, purged with an O₂ balloon, was stirred and irradiated externally with fluorescent lamps for 20 h. The reaction mixture was concentrated in vacuo. Then 140 μL of water and 150 μL of conc. H₂SO₄ were added. The mixture was stirred under reflux for 30 min. After that the mixture was basified by 10% NaOH and extracted with ether. Water layer was acidified by 2 N HCl and extracted with ether. Organic layer was evaporated and product was obtained. When the product is not pure, it was purified by preparative TLC.
18. K. E. Koenig and W. P. Weber, Tetrahedron Lett., 1974, 26, 2275.
19. T. Ando, T. Kawate, J. Yamawaki and T. Hanafusa, Synthesis, 1983, 637.
20. B. Alejandro, N. Carmen, G. R. Ma and S. Jose, Synthesis, 2005, 2787.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra13269j

---