Oxidative ortho-amino-methylation of phenols via C–H and C–C bond cleavage

Abstract

Initiated by CCl₃Br, phenols undergo efficient ortho-selective oxidative cross dehydrogenative coupling (CDC) with trimethylamine. When tetramethylethylenediamine (TMEDA) is used instead of trimethylamine, oxidative carbon–carbon activation coupling (CAC) could occur to give the same salicylamines together with CDC by-products. These reactions are accelerated by a gold salt.

---

Phenols are important building blocks in pharmaceutics, lubricants and fine chemicals.¹ Selective alkylation of phenols has been of interest for many decades. In some related examples, Pettus et al. reported a constructing method using protected phenols bearing an ortho-carbonyl group with organo-metal reagents; Kuninobu and Takai et al. reported a rhenium-catalyzed ortho-alkylation of phenols with alkenes; in addition to early reports of ortho alkylation of phenols with allyl alcohols, Yi et al. recently reported an intriguing dehydrating coupling between phenols and alcohols to afford ortho-alkylated phenols; Capdevielle et al. reported a copper mediated oxidation of amines into amino N-oxide to couple with electron deficient phenol; Hansen et al. utilized Eschenmoser's salt and its derivatives to react with phenols via a Mannich process.² Despite these efforts, lack of chemo-selectivity and phenol functional group tolerance are still the main drawbacks of the existing methodologies.

Pioneered by Li and others, the concept of cross dehydrogenative coupling (CDC) is employed in synthesizing various valuable building blocks, which generally involves a metal mediated intermolecular oxidative process.³ Hansen's method is efficient in constructing salicylamines, however, it is an extended classic Friedel–Crafts reaction and the CDC process with a phenol substrate is rare and needs further exploration and elucidation.^2g,i

Compared with C–H bonds, C–C bonds are weaker in bond energy but much less accessible as the steric hindrance tremendously limits their synthetic applications.⁴ Intramolecular C–C bond cleavage and reformation are well documented in various rearrangement reactions from the early stages of organic synthesis,⁵ however, C–C activation coupling (CAC) via an intermolecular manner is less developed.⁶ Recently, we reported a visible light induced Henry type reaction as shown in Scheme 1.⁷ In this process, TMEDA could be viewed as a masked Eschenmoser's salt⁸ under oxidative conditions, which may participate in further transformations such as the aminomethylation of phenols. To the best of our knowledge, there are no previous reports on salicylamine synthesis via a CAC pathway. Herein, we report a gold accelerated oxidative CAC process between phenols and TMEDA to afford salicylamines. Similar conditions could also be applied to CDC reactions between phenols and trimethylamine to yield the same products.

Scheme 1 Photo-induced CAC between TMEDA and nitro compounds.

The visible light-induced conditions shown in Scheme 1 were initially tested (entry 1, Table 1). To our disappointment, only a 28% NMR yield is achieved with O₂ as the oxidant. The conversion increases almost quantitatively when the oxidant is changed to BrCCl₃; however, BrCCl₃ without a catalyst could also mediate this transformation with a prolonged reaction time (entry 2, Table 1). Various Lewis acids/Brønsted acids were tested for their accelerating efficiency and HAuCl₄ was superior for this reaction, showing an improved conversion. In this reaction, we also identified a CDC coupling product formed between TMEDA and phenol. The combined conversion is reported in Table 1. Control reactions with or without light gave similar results when a gold catalyst was present and brand new reaction vessels and stir bars were used to eliminate possible traces of low-valent transition metal contamination.

Table 1 Screening of conditions

Entry	Catalyst	Solvent	Oxidant	Conversion (%)^b	Reaction time (h)
a Reactions were performed on a 0.1 mmol scale based on phenol with 3 equiv. oxidant and 10 equiv. TMEDA in 2 mL solvent at room temperature, with 5% catalyst loading unless indicated otherwise.b Conversion based on NMR with CH2Br2 as the internal standard.c O2 balloon.d Reaction was performed on a 1 mmol scale in 5 mL solvent with 5% catalyst, 3 equiv. BrCCl3 and 10 equiv. TMEDA at room temperature, which are set as the standard conditions. Combined isolated yields are given.
1^c	Ru(bpy)3Cl2	CH3CN	O2	28	36
2	None	DCM	BrCCl3	70	36
3	AgI	DCM	BrCCl3	50	24
4	CuCl2	DCM	BrCCl3	52	24
5	AgI	DCM	CBr4	28	24
6	CF3COOH	DCM	BrCCl3	25	14
7	HAuCl4	DCM	BrCCl3	76	14^d

---

Gold salts may play roles other than as optimal Lewis/Brønsted acids to mediate electrophilic aromatic substitution of in situ generated iminiums by phenol addition to the in situ generated Eschenmoser's salt, since phenols bearing electron withdrawing groups react as well as or better than phenols bearing electron donating groups. The exact role of the gold catalyst in this reaction still needs further elucidation.⁹

Inspired by the CDC product formed between TMEDA and phenol, we tested the reactivity between trimethylamine and phenols. The results are summarized in Table 2. Para- or ortho-substituted phenols generally give good to excellent yields with high ortho-selectivity. Dual additions to both the ortho- and para- positions are also found in small amounts in some cases (entries 4 and 5, Table 2). meta-Iodo-phenol shows no selectivity between its two ortho-positions, giving a low yield (entry 7, Table 2). Phenol gives both mono- and dual-addition products which could be separated as 63% and 23% respectively (entry 8, Table 2). Also, phenols bearing a heterocycle or with poly substitutions react well under the standard conditions (entries 9 and 10, Table 2). Phenols with either electron withdrawing groups or electron donating groups are all suitable substrates for this reaction. To our disappointment, ethyldimethylamine and butyldimethylamine could only give limited conversion (∼10%), partly due to the difficulty of generating the corresponding iminiums.

Table 2 CDC reactions between trimethylamine and phenols

Entry^a	Phenol	Product	Yield (%)^b
a Reaction conditions: phenol (1.0 mmol), TMA (gas), HAuCl4·3H2O (0.05 mmol), DCM (5 mL), BrCCl3 (3 mmol), r.t.b Isolated yields based on phenol.c Combined with 8% for entry 4 and 11% for entry 5 of the ortho, para-dual-addition product.d Combined yields. (10% 2-ortho coupling, 12% 5-ortho coupling.)e 63% mono-addition product plus 23% dual-addition product.
1			95
2			61
3			50
4			78^c
5			98^c
6			52
7			22^d
8			86^e
9			91
10			88
11			80

---

Interestingly, when naphthalene-2-ol is tested, together with 37% of the expected product, we identified naphthalene-fused oxazine in 8% isolated yield, which indicates an intriguing and novel method for fused heterocycle synthesis. Aminomethylated naphthalene-2-ol could not undergo this reaction (Scheme 2).

Scheme 2 Formation of naphthalene-fused oxazine, (a) standard conditions, (b) isolated yields.

The results of the reactions between TMEDA and phenols are summarized in Table 3. The reactions give good yields with moderate to good selectivity between the CAC and CDC products, which varies from 1 : 1 to 3 : 1. ortho-Substituted phenols show better reactivity (entries 1–9) than simple phenol (entry 14). Highly substituted phenols also give reasonable yields (entries 10 and 13). Interestingly, comparing entry 11 with Scheme 2, no oxazine product is detected. For entries 12 and 15, only CAC products are isolated.

Table 3 CAC reactions between TMEDA and phenols

Entry^a	Substrate	R	Yield (%)^b	Ratio^c
a Standard conditions.b Combined isolated yields.c CAC : CDC product ratio.
1		F	66	33 : 30
2		Cl	81	52 : 29
3		Br	88	53 : 35
4		I	76	43 : 33
5		Ph	80	35 : 45
6		CF3	68	29 : 39
7		OMe	81	50 : 31
8		Me	62	31 : 31
9		Allyl	61	31 : 30
10			72	49 : 23
11			78	57 : 21
12			60	60 : 0
13			56	23 : 33
14			59	45 : 14
15			38	38 : 0

---

meta-Substituted phenols are unfavourable substrates in previous reports of phenol alkylation, which is a major limitation.³ In our case, despite their lower reactivities, meta-iodo and meta-ethyl phenol react to give moderate yields as shown in Scheme 3. This process has excellent ortho-selectivity towards phenols. If both ortho-positions are blocked by methyl groups, the aminomethylation would occur at the para-position with only the CAC product in 23% isolated yield. Various diamines with similar structures to TMEDA were tested, however they could only afford very low conversions (ESI Table 1†).

Scheme 3 Reactions of substituted phenols under standard conditions.

When TEMPO is added, the yield dramatically decreases and a TEMPO·CCl₃ adduct is identified by MS, which hints that a radical process may be involved. However, phenol bearing a radical sensitive cyclopropane is not affected in either products or recovered substrates, which suggests an iminium pathway, shown in Scheme 4.

Scheme 4 The possible pathway of this reaction.

A tentative mechanism is proposed in Scheme 5: BrCCl₃ initiates the reaction by generating two radical species, acquiring an electron from the N atom of TMEDA to form the radical cation. TEMPO addition could inhibit the reactivity during the first two steps. Then there could be two competitive pathways: iminiums form by C–C cleavage or by C–H cleavage, which leads to the CAC and CDC products respectively. The steps involving phenols are not radical in nature and won’t disturb the radical sensitive cyclopropane, and the gold catalyst could accelerate the last step(s). The role of the gold catalyst is elusive; gold salt could significantly increase the reaction rate, however, the electron density of phenols doesn’t play a decisive role in these reactions. This fact weakens the assumption of gold acting as a simple Lewis acid/Brønsted acid in this reaction. However, from the gold salts we screened, HAuCl₄ behaves better than Ph₃PAuCl, Ph₃PAuCl/AgOTf and AuCl, which indicates the importance of the catalyst's acidity. Further mechanistic study is in process to reveal the effect of the gold salt in this reaction.

Scheme 5 Tentative mechanism.

Conclusions

In summary, we report a novel phenol aminomethylation via either C–C or C–H bond cleavage initiated by BrCCl₃ with good ortho-selectivity. This reaction provides important clues for further study of C–C bond cleavage and it is the first case of CAC alkylation of phenols. We propose a radical initiated iminium intermediate. Further applications and mechanism elucidation of this methodology are under current investigation.

Experimental

General procedure: an oven-dried round bottom flask (25 mL) was equipped with magnetic stir bar and charged with phenol (1 mmol, 1.0 equiv.), TMEDA (10 mmol, 10.0 equiv.), HAuCl₄·3H₂O (0.05 mmol, 0.05 equiv.), BrCCl₃ (3 mmol, 3.0 equiv.) and DCM (10.0 mL). The mixture was then stirred under a nitrogen atmosphere at room temperature until the starting material disappeared from the TLC. After that the solvent was removed under reduced pressure, and the residue was purified by silica gel column chromatography to afford the desired pure product.

Acknowledgements

We thank the National Natural Science Foundation of China Grant 21102007, the Shenzhen Science and Technology Innovation Committee SW201110060 and SW201110018, and the Shenzhen Peacock Program (KQTD201103) to Z. G. L.

Notes and references

1. L. R. C. Barclay and M. R. Vinqvist. The Chemistry of Phenols, ed. Z. Rappoport, Wiley-VCH, Weinheim, 2003, p. 840.
2. (a) R. W. Van De Water, D. J. Magdziak, J. N. Chau and T. R. R. Pettus, J. Am. Chem. Soc., 2000, 122, 6502; (b) Y. Kuninobu, T. Matsuki and K. Takai, J. Am. Chem. Soc., 2009, 131, 9914; (c) A. V. Malkov, S. L. Davis, I. R. Baxendale, W. L. Mitchell and P. Kočovský, J. Org. Chem., 1999, 64, 2751; (d) A. V. Malkov, P. Spoor, V. Vinader and P. Kočovský, J. Org. Chem., 1999, 64, 5308; (e) Y. Yamamoto and K. Itonaga, Org. Lett., 2009, 11, 717; (f) D. H. Lee, K. H. Kwon and C. S. Yi, J. Am. Chem. Soc., 2012, 134, 7325; (g) H. F. Anwar, L. Skattebol and T. V. Hansen, Tetrahedron, 2007, 63, 9997; (h) D. R. Hwang and B. Uang, Org. Lett., 2002, 4, 463; (i) G. Rousselet, P. Capdevielle and M. Maumy, Tetrahedron Lett., 1995, 36, 4999.
3. C.-J. Li, Acc. Chem. Res., 2009, 42, 335 , and references therein.
4. J. Halpern, Acc. Chem. Res., 1982, 15, 238.
5. B. Rybtchinski and D. Milstein, Angew. Chem., Int. Ed., 1999, 38, 870 , and references therein.
6. For recent reviews of the C–C cleavage, see: (a) C.-H. Jun, Chem. Soc. Rev., 2004, 33, 610; (b) T. Tobisu and N. Chatani, Chem. Soc. Rev., 2008, 37, 300; (c) T. Satoh and M. Miura, Top. Organomet. Chem., 2005, 14, 1; (d) C.-H. Jun and J. W. Park, Top. Organomet. Chem., 2007, 24, 117; (e) Z. Xi, Acc. Chem. Res., 2010, 43, 1342.
7. S. Y. Cai, X. Y. Zhao, X. B. Wang, Q. S. Liu, Z. G. Li and D. Z.-G. Wang, Angew. Chem., Int. Ed., 2012, 51, 8050.
8. J. Schreiber, H. Maag, N. Hashimoto and A. Eschenmoser, Angew. Chem., Int. Ed., 1971, 10, 330.
9. Gold-accelerated C–C cleavage processes are rare, except for some decarboxylation processes: (a) K. Vikse, G. N. Khairallah, J. S. McIndoe and R. A. J. O'Hair, Dalton Trans., 2013, 42, 6440; (b) J. P. Fackler, M. N. I. J. Khan, C. King, R. J. Staples and R. E. P. Winpenny, Organometallics, 1991, 10, 2178; (c) S. Y. Park, D. Kwon, H. Mok and B. H. Chung, Analyst, 2013, 138, 3146; (d) D. Xing and D. Yang, Org. Lett., 2010, 12, 1068; (e) J. Zou, Z. J. Guo, J. A. Parkinson, Y. Chen and P. J. Sadler, Chem. Commun., 1999, 1359; (f) P. García-García, M. A. Fernández-Rodríguez and E. Aguilar, Angew. Chem., Int. Ed., 2009, 48, 5534; (g) Y. F. Chen, M. Chen and Y. H. Liu, Angew. Chem., Int. Ed., 2012, 51, 6493.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3ra46373g

---