Reactivity and mechanistic insight into the cross coupling reaction between isochromans and β-keto esters through C–H bond activation under visible light irradiation

Abstract

The visible light catalytic cross coupling reaction from two different C–H bonds provides an efficient protocol for C–H bond activation and C–C bond construction. The application of the oxidative photoredox strategy to couple C–H bonds next to an oxygen atom with other C–H bonds is more challenging because of the more positive oxidation potential of ethers than that of amines. Here, we take advantage of organic dye 9-mesityl-10-methylacridinium perchlorate (Acr⁺-Mes ClO₄⁻) as the photosensitizer and BrCCl₃ as the terminal oxidant to achieve the addition of β-keto esters to oxonium species, directly generated from isochromans, leading to the formation of alkylation products in the α-position of an oxygen atom under visible light irradiation.

---

Introduction

The dramatic development of visible light catalysis in the past several years is very closely related to the single-electron redox abilities of photosensitizers which absorb visible light.¹ Due to their excellent photophysical properties, transition metal complexes, like ruthenium, platinum or iridium polypyridyl complexes,² have so far been frequently utilized as photosensitizers. However, the high cost, potential toxicity and limited availability of noble metals have often limited their further utilization on a larger scale.³ As an alternative, organic dyes (Fig. 1) have advantages over the metal complexes with regard to lower cost, less toxicity and more convenient modifications to tune redox potentials,⁴ which have been recently used in visible light photoredox catalysis.^3–5 In 2011, for example, König and Zeitler employed eosin Y as an effective photosensitizer to successfully realize the asymmetric intermolecular α-alkylation of aldehydes.⁶ Our group exerted eosin Y or bodipy to accomplish the cross coupling reactions of N-aryltetrahydroisoquinolines with various nucleophiles.⁷ In the present work, we make use of an electron donor–acceptor-linked dyad, 9-mesityl-10-methylacridinium ion (Acr⁺-Mes),⁸ to active the C–H bond next to an oxygen atom for a new C–C bond formation.

Fig. 1 Molecular structures of common organic dyes.

The construction of a C–C bond is a very essential and vital part of organic synthesis.⁹ The cross coupling from two different C–H bonds provides one of the ideal protocols to this end, because it not only forms new C–C bonds and reduces synthetic steps by avoiding the prefunctionalization of starting materials, but also displays atom-economical and environmentally benign characteristics.¹⁰ The formation of C–C bonds via the functionalization of C–H bonds adjacent to a nitrogen atom has been extensively studied under photocatalytic conditions.¹¹ However, the visible light catalytic cross coupling of C–H bonds next to an oxygen atom with other C–H bonds is very limited due to the higher oxidation potential of ethers compared with amines.¹² Oxonium species generated by the two-electron oxidation of ethers were expected to work as reactive intermediates for ether functionalization,¹³ and several examples involving oxocarbenium ions formed by oxidative photoredox catalysis have been reported.¹⁴ For instance, Stephenson et al. first disclosed a highly efficient approach to the removal of the para-methoxybenzyl group by hydrolysis of oxonium species employing an iridium complex as the photosensitizer and BrCCl₃ as the terminal oxidant (Scheme 1, eqn (1)).^14a Later, Cheng et al. achieved a similar result by the combination of eosin Y and H₂O₂,^14b and Li et al. transformed alkyl benzyl ethers to alkyl esters via a tandem hydrolysis and esterification procedure.^14c Very recently, our group demonstrated that the formation of a C–C bond next to an O atom could be achieved under photocatalytic oxidant-free conditions via an oxonium intermediate.¹⁵ Based on these results, we questioned whether oxonium species could be captured by nucleophiles to generate coupling products under oxidative photocatalytic conditions. Considering that the oxidation byproducts (CHCl₃ and HBr) of BrCCl₃ are easily separated from the reaction system, we herein design a new combination, Acr⁺-Mes cooperated with oxidant BrCCl₃ (Scheme 1, eqn (2)), which successfully executes our plan.

Scheme 1 Utilization of oxonium species by oxidative photocatalysis.

Results and discussion

Initially to explore the above transformation, we selected isochroman (1a) and ethyl acetoacetate (2a) as substrates for our experiment with various bases under different solvent conditions (Table 1). The reaction of 1a with 2a in the presence of Acr⁺-Mes ClO₄⁻ (5 mol%), 2 equivalents of 2,6-dimethylpyridine and BrCCl₃ in anhydrous MeCN under an ambient argon atmosphere was performed, and after 24 h irradiation by blue LEDs (λ = 450 ± 10 nm), the desired cross coupling product 3a was obtained in 45% yield with 65% conversion of 2a (entry 1). Among the bases screened (entries 1–3), Na₂HPO₄ (entry 2) provided the optimal efficiency, and Na₂CO₃ (entry 3) could also facilitate this reaction despite a slightly lower yield. The solvent screening suggested that replacing MeCN with DCM (entry 4) or THF (entry 5) did not improve the reaction yield. As transitional metal salts may have the ability to activate a nucleophile and benzyl C–H bond,^9,10 several readily accessible and inexpensive copper salts were then added. To our delight, a catalytic amount of Cu(OTf)₂ proved to be the best choice for the model cross coupling reaction, which afforded 3a in an excellent yield of 87% with an almost complete conversion of 2a (96%) (entry 8). CuCl₂ (entry 6) or CuBr₂ (entry 7) as an alternative copper salt could also give 3a in a moderate yield. Notably, control experiments revealed that a very low yield was obtained in the absence of the base Na₂HPO₄ (entry 9). The selective removal of light, BrCCl₃ or photosensitizer (Acr⁺-Mes ClO₄⁻) showed that each of them was necessary for the successful oxidative coupling between 1a and 2a (entry 10).

Table 1 Optimization of reaction conditions^a

Entry	Base	Solvent	Metal salt.	Conversion 2a ^b (%)	Yield 3a ^b (%)
a Isochroman (1.0 mmol), ethyl acetoacetate (0.5 mmol), Acr^+-Mes (0.025 mmol, 5 mol%), BrCCl3 (1.0 mmol), base (1.0 mmol) and metal salt. (0.05 mmol, 10 mol%) in the solvent (2 mL) were irradiated by blue LEDs for 24 hours under an argon atmosphere. b Determined by ^1H NMR spectroscopy using an internal standard. c General conditions, but with no Acr^+-Mes, BrCCl3 or light. n.d. = not determined.
1	2,6-Dimethylpyridine	MeCN	None	65	45
2	Na2HPO4	MeCN	None	78	76
3	Na2CO3	MeCN	None	68	57
4	Na2HPO4	DCM	None	64	20
5	Na2HPO4	THF	None	74	49
6	Na2HPO4	MeCN	CuCl2	98	79
7	Na2HPO4	MeCN	CuBr2	98	76
8	Na2HPO4	MeCN	Cu(OTf)2	96	87
9	None	MeCN	Cu(OTf)2	74	27
10^c	Na2HPO4	MeCN	Cu(OTf)2	n.d.	<10

---

With the optimized reaction conditions in hand, we first sought to define the scope of β-keto esters, and the corresponding results are listed in Scheme 2. A variety of aliphatic and electronically varied aromatic β-keto esters were found to be good substrates for the reaction with a diastereomeric ratio of 1 : 1–1 : 1.9 (Scheme 2, 3a–3h). Due to the chloro-functionality of the product 3g, obtained in a relatively high yield, it could serve as a potential intermediate for further organic synthesis. tert-Butyl acetoacetate, however, showed little difference in the reactivity, giving rise to a cross coupling product in 28% yield, probably due to the decomposition of product 3e. More importantly, other nucleophiles, such as the 1,3-dicarbonyl compound (2,4-pentanedione) and benzoylnitromethane, reacted smoothly under the standard conditions, and the desired products were achieved in moderate to good yields (Scheme 2, 3i and 3j).

Scheme 2 Scope of substrates: variants of isochroman or β-keto esters. ^aIsochromans (1.0 mmol), β-keto esters (0.5 mmol), Acr⁺-Mes (0.025 mmol, 5 mol%), BrCCl₃ (1.0 mmol), Na₂HPO₄ (1.0 mmol), Cu(OTf)₂ (0.05 mmol, 10 mol%) in the MeCN (2 mL) were irradiated by blue LEDs for 24 hours under an argon atmosphere. The ratios of two diastereomers were determined by ¹H NMR spectra and indicated in brackets.

Subsequently, the substituent effect of isochroman, including methyl, tert-butyl, methoxy or chloro, on the cross coupling reaction was investigated. Alkylsubstituents on the aromatic ring of isochroman participated in the reaction with reduced yields, probably due to the increased steric bulk (Scheme 2, 3k–3o). However, no reaction took place when a methoxy or chloro substituent at C7 of isochromans was applied to the photocatalytic reaction. Then, 1,4-dihydro-2H-benzo[f]isochromene was found to be an alternative substrate, although the yield of product 3p dropped to 48% with part of the substrate remaining. Finally, when the acylic substrate (1-(methoxymethyl)-4-methylbenzene) was treated with 2a, 4-methylbenzaldehyde was isolated as the main product, and no cross coupling product was observed with most of the substrate being consumed. This observation was consistent with previous reports.¹⁴ Since an acylic oxonium ion was less stable, trace water in the reaction system may quench it and block the route of nucleophilic addition.

It should be pointed out that the reaction cannot tolerate strong electron donating or withdrawing groups toward the isochroman coupling partner. The former observation could be explained by Floreancig's mechanistic studies of DDQ-mediated ether functionalizations.¹⁶ The substrate with a lower oxidation potential was found to possess a higher bond dissociation energy of the scissile C–H bond, so the methoxy substituent isochroman variant almost remained unchanged after the reaction. The latter substrate cannot be oxidated by an acridinium photosensitizer because of a higher oxidant potential of the electron-poor aromatic group.

To shed light on the primary process for the visible light-driven reaction, the oxidation potential (E_ox) of isochroman (1a) and the reduction potential (E_red) of BrCCl₃ were determined to be 2.03 V and −0.84 V vs. SCE in MeCN (Fig. S1a and S1b†). The E_ox value of 1a is lower than the one-electron reduction potential of the electron-transfer state Acr˙-Mes˙⁺ (2.06 V vs. SCE in MeCN),¹⁷ so the electron transfer from 1a to Acr˙-Mes˙⁺ is thermodynamically feasible, whereas the electron transfer from Acr˙-Mes˙⁺ (vs. SCE in MeCN)¹⁷ to BrCCl₃ is energetically unfavourable. However, this unfavourable step is irreversible due to the decomposition of BrCCl₃,¹⁸ and may be the rate-determining step of the reaction.

Furthermore, electron paramagnetic resonance (EPR) was conducted to detect the change of reaction intermediates in the photocatalytic cross coupling reaction. The measurement was carried out in an argon-saturated MeCN solution of Acr⁺-Mes with or without 1a (or BrCCl₃). The sample cavity was first irradiated by a high-pressure mercury lamp for 2 min at 233 K. After the irradiation, the sample cell was immediately cooled to 123 K and the ESR spectra were recorded. The signal at g = 2.0036 (Fig. 2 solid line) is assigned to Acr˙-Mes⁺˙ due to the superposition of the ESR signals of the radical cation of the mesitylene moiety and the acridinyl radical moiety.¹⁹ Because the electron transfer from 1a to Acr˙-Mes⁺˙ is energetically favorable, the change of EPR signals (Fig. 2a) might result from two aspects: (1) the signal of Acr˙-Mes⁺˙ is quenched by the addition of 1a, resulting in a reduction in the intensity of the original signal; (2) Acr˙-Mes˙⁺ accepts an electron from 1a to generate Acr˙-Mes, whose signal is at g = 2.0035 (Fig. 2b).²⁰ On the other hand, Acr˙-Mes⁺˙ delivers an electron to BrCCl₃ to afford ˙CCl₃ and Br⁻, so the obvious new signal at g = 2.0092 is from ˙CCl₃, which is consistent with the reported value (Fig. 2d).²¹ These observations again demonstrate that the single electron transfer process proceeds not only between Acr˙-Mes˙⁺ and 1a, but also between Acr˙-Mes˙⁺ and BrCCl₃.

Fig. 2 EPR spectra observed after irradiation of an argon-saturated MeCN solution of Acr⁺-Mes (0.0125 M) at 233 K and then measured at 123 K. (a) In the absence and presence of isochroman (0.5 M); (b) enlarged graph in the situation of adding isochroman (c) in the absence and presence of BrCCl₃ (0.5 M); (d) enlarged graph in the situation of adding BrCCl₃.

A likely mechanism of this photocatalytic cross coupling reaction is outlined in Fig. 3. The first step involves photoinduced intramolecular electron transfer from the Mes moiety to the singlet excited state of the Acr⁺ moiety to generate Acr˙-Mes˙⁺, an extremely long-lived (e.g., 2 h at 203 K) electron-transfer state.^8a The Mes˙⁺ moiety can oxidize isochroman (1a) to generate the arene radical cation (4), whereas Acr˙ can reduce BrCCl₃ to afford Br⁻ and ˙CCl₃. The photosensitizer (Acr⁺-Mes) is then regenerated. The trichloromethyl radical ˙CCl₃ (5) subsequently abstracts a hydrogen atom from the intermediate (4) to produce the oxocarbenium ion (6) along with CHCl₃ generation. Notably, the trichloromethane CHCl₃ in the reaction system was observed by ¹H NMR and GC-MS analysis (Fig. S2†), and the kinetic isotope effect value was determined from intermolecular competition experiments to be k_H/k_D = 3.6 (Fig. 3) on the basis of analysing the ¹H NMR spectra of the isolated products (Fig. S3†). The obtained results indicated that benzylic C–H bond cleavage is probably involved in the reaction process. Finally, the resulting oxonium species react with the nucleophile (2a), activated by Cu(OTf)₂, to provide the desired C–C bond coupling product (3a).

Fig. 3 Proposed mechanism.

Conclusions

In summary, a novel acridinium catalyzed utilization of oxocarbenium ions, directly obtained from isochromans, is developed. This photocatalytic method is a practical synthetic tool for the functionalization of a C(sp³)–H bond adjacent to the O atom of ethers and α-alkylation products were obtained in moderate to good yields. Mechanistic studies give rich information about the reaction route. Electrochemistry experiments demonstrate that reductive quenching is thermodynamically feasible, whereas the unfavourable oxidation of the electron-transfer state Acr˙-Mes˙⁺ is probably the rate-determining step in this transformation. Furthermore, EPR and KIE studies reveal that BrCCl₃, after accepting an electron from Acr˙-Mes˙⁺, can generate ˙CCl₃, which would abstract a hydrogen atom from the benzylic position. Additional investigation to explore the synthetic applications of the oxonium species by this photocatalytic protocol is underway in our group.

Acknowledgements

Financial support for this research from the Ministry of Science and Technology of China (2013CB834804, 2013CB834505 and 2014CB239402), the National Natural Science Foundation of China (21390404, 91427303 and 21402217), the Key Research Programme of the Chinese Academy of Sciences (KGZD-EW-T05), and the Chinese Academy of Sciences is gratefully acknowledged.

Notes and references

1. For recent reviews on photoredox catalysis, see: (a) K. Zeitler, Angew. Chem., Int. Ed., 2009, 48, 9785; (b) T. P. Yoon, M. A. Ischay and J. Du, Nat. Chem., 2010, 2, 527; (c) J. M. R. Narayanam and C. R. J. Stephenson, Chem. Soc. Rev., 2011, 40, 102; (d) J. Xuan and W.-J. Xiao, Angew. Chem., Int. Ed., 2012, 51, 6828; (e) C. K. Prier, D. A. Rankic and D. W. C. MacMillan, Chem. Rev., 2013, 113, 5322; (f) D. M. Schultz and T. P. Yoon, Science, 2014, 343, 1239176.
2. (a) D. Zhang, L.-Z. Wu, L. Zhou, X. Han, Q.-Z. Yang, L.-P. Zhang and C.-H. Tung, J. Am. Chem. Soc., 2004, 126, 3440; (b) K. Feng, R.-Y. Zhang, L.-Z. Wu, B. Tu, M.-L. Peng, L.-P. Zhang, D. Zhao and C.-H. Tung, J. Am. Chem. Soc., 2006, 128, 14685; (c) D. A. Nicewicz and D. W. C. MacMillan, Science, 2008, 322, 77; (d) A. G. Condie, J. C. González-Gómez and C. R. J. Stephenson, J. Am. Chem. Soc., 2010, 132, 1464; (e) J. J. Zhong, Q. Y. Meng, G. X. Wang, Q. Liu, B. Chen, K. Feng, C. H. Tung and L. Z. Wu, Chem. – Eur. J., 2013, 19, 6443; (f) Q.-Y. Meng, T. Lei, L.-M. Zhao, C.-J. Wu, J.-J. Zhong, X.-W. Gao, C.-H. Tung and L.-Z. Wu, Org. Lett., 2014, 16, 5968.
3. D. P. Hari and B. Konig, Chem. Commun., 2014, 50, 6688.
4. (a) S. Fukuzumi and K. Ohkubo, Chem. Sci., 2013, 4, 561; (b) S. Fukuzumi and K. Ohkubo, Org. Biomol. Chem., 2014, 12, 6059.
5. (a) D. Ravelli and M. Fagnoni, ChemCatChem, 2012, 4, 169; (b) D. A. Nicewicz and T. M. Nguyen, ACS Catal., 2013, 4, 355.
6. M. Neumann, S. Füldner, B. König and K. Zeitler, Angew. Chem., Int. Ed., 2011, 50, 951.
7. (a) Q. Liu, Y.-N. Li, H.-H. Zhang, B. Chen, C.-H. Tung and L.-Z. Wu, Chem. – Eur. J., 2012, 18, 620; (b) J.-J. Zhong, Q.-Y. Meng, B. Liu, X.-B. Li, X.-W. Gao, T. Lei, C.-J. Wu, Z.-J. Li, C.-H. Tung and L.-Z. Wu, Org. Lett., 2014, 16, 1988; (c) X.-Z. Wang, Q.-Y. Meng, J.-J. Zhong, X.-W. Gao, T. Lei, L.-M. Zhao, Z.-J. Li, B. Chen, C.-H. Tung and L.-Z. Wu, Chem. Commun., 2015, 51, 11256.
8. (a) S. Fukuzumi, H. Kotani, K. Ohkubo, S. Ogo, N. V. Tkachenko and H. Lemmetyinen, J. Am. Chem. Soc., 2004, 126, 1600; (b) H. Kotani, K. Ohkubo and S. Fukuzumi, J. Am. Chem. Soc., 2004, 126, 15999; (c) D. S. Hamilton and D. A. Nicewicz, J. Am. Chem. Soc., 2012, 134, 18577.
9. (a) V. Ritleng, C. Sirlin and M. Pfeffer, Chem. Rev., 2002, 102, 1731; (b) C. Liu, H. Zhang, W. Shi and A. Lei, Chem. Rev., 2011, 111, 1780.
10. (a) C.-J. Li, Acc. Chem. Res., 2008, 42, 335; (b) S. A. Girard, T. Knauber and C.-J. Li, Angew. Chem., Int. Ed., 2014, 53, 74.
11. (a) A. G. Condie, J. C. González-Gómez and C. R. J. Stephenson, J. Am. Chem. Soc., 2010, 132, 1464; (b) A. McNally, C. K. Prier and D. W. C. MacMillan, Science, 2011, 334, 1114; (c) P. Kohls, D. Jadhav, G. Pandey and O. Reiser, Org. Lett., 2012, 14, 672; (d) Y. Miyake, K. Nakajima and Y. Nishibayashi, J. Am. Chem. Soc., 2012, 134, 3338.
12. J. Jin and D. W. MacMillan, Angew. Chem., Int. Ed., 2015, 54, 1565.
13. (a) Y. Zhang and C.-J. Li, Angew. Chem., Int. Ed., 2006, 45, 1949; (b) Y. Zhang and C.-J. Li, J. Am. Chem. Soc., 2006, 128, 4242; (c) Z. Li, R. Yu and H. Li, Angew. Chem., Int. Ed., 2008, 47, 7497.
14. (a) J. W. Tucker, J. M. R. Narayanam, P. S. Shah and C. R. J. Stephenson, Chem. Commun., 2011, 47, 5040; (b) Z. Liu, Y. Zhang, Z. Cai, H. Sun and X. Cheng, Adv. Synth. Catal., 2015, 357, 589; (c) P. Lu, T. Hou, X. Gu and P. Li, Org. Lett., 2015, 17, 1954.
15. M. Xiang, Q. Y. Meng, J. X. Li, Y. W. Zheng, C. Ye, Z. J. Li, B. Chen, C. H. Tung and L. Z. Wu, Chem. – Eur. J., 2015, 21, 18080.
16. H. H. Jung and P. E. Floreancig, Tetrahedron, 2009, 65, 10830.
17. K. Ohkubo, K. Mizushima, R. Iwata, K. Souma, N. Suzuki and S. Fukuzumi, Chem. Commun., 2010, 46, 601.
18. K. Ohkubo, R. Iwata, S. Miyazaki, T. Kojima and S. Fukuzumi, Org. Lett., 2006, 8, 6079.
19. S. Fukuzumi, A. Itoh, T. Suenobu and K. Ohkubo, J. Phys. Chem. C, 2014, 118, 24188.
20. S. Fukuzumi, K. Doi, A. Itoh, T. Suenobu, K. Ohkubo, Y. Yamada and K. D. Karlin, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 15572.
21. P. F. Fitzpatrick and J. J. Villafranca, J. Am. Chem. Soc., 1985, 107, 5022.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and characterization of all compounds. See DOI: 10.1039/c5qo00412h

---