Sustainable synthesis of heteroaryl ethers from azine N-oxides via phosphoramide catalysis

Abstract

A rapid and eco-friendly approach has been devised for synthesizing heteroaryl ethers. This methodology involves the reaction between nitrogen heteroaromatic N-oxides and phenol derivatives or alcohols, catalyzed by an in situ generated phosphonium salt. This salt acts as the activating agent and is formed through the reaction of the byproduct phosphoramide with phosphoryl tribromide. Compared to previously reported methods, this method stands out due to its excellent atom economy (92%), the obviation of C1- or C2-prefunctionalized heteroaromatics, the use of an eco-friendly solvent, mild reaction conditions, the incorporation of feedstock materials as catalysts, a short reaction time, and a broad substrate scope, yielding a diverse array of heteroaryl ethers.

---

Green foundation

1. This methodology involves the reaction between nitrogen heteroaromatic N-oxides and phenol derivatives or alcohols, catalyzed by an in situ generated phosphonium salt. This salt acts as the activating agent and is formed through the reaction of the byproduct phosphoramide with phosphoryl tribromide. In this instance, we not only successfully recycle the reaction byproduct but also synthesize the activating agent necessary for the reaction.

2. The developed method adheres to six of the twelve green chemistry principles. Compared to previous reports, this method stands out due to its excellent atom economy, the obviation of C1- or C2-prefunctionalized heteroaromatics, the use of an eco-friendly solvent, mild reaction conditions, the incorporation of feedstock materials as catalysts, a short reaction time, and a broad substrate scope.

3. Future research will focus on developing more effective catalysts to enhance the efficiency of C–H bond etherification reactions of nitrogen heteroaromatic N-oxides under mild conditions.

Introduction

Diaryl ether scaffolds are commonly encountered structures in natural products,¹ agrochemicals,² and compounds exhibiting biological activity.³ They are also widely used as key intermediates in synthetic chemistry.⁴ Despite their prevalence, the de novo synthesis of diaryl ethers through C–H bond functionalization, particularly heteroaryl ethers, remains challenging (Scheme 1). Classically, heteroaryl halides can undergo coupling reactions with phenols in the presence of transition metal catalysts, resulting in the formation of diaryl ethers⁵ (Scheme 1A). Yamaguchi and coworkers developed a catalytic decarbonylative etherification of aromatic esters using a palladium or nickel catalyst with a diphosphine ligand, yielding the desired diaryl ethers.⁶ However, these phenoxylation reactions do not involve direct C–H bond functionalization and typically require harsh reaction conditions with transition-metal catalysts, owing to the intrinsically low nucleophilicity of phenolic compounds. In 2024, a photochemical approach for the reaction between aryl halides and phenols was documented.⁷ While this method offers mild reaction conditions, heteroaryl halides are often not readily available for purchase and typically need to be synthesized from the corresponding heteroaryl N-oxides.⁸ Furthermore, comprehensive investigations into the substrate scope are imperative, as only three specific examples have been successfully synthesized. An alternative method for the synthesis of diaryl ethers involves the use of easily accessible isoquinoline N-oxides. While the ortho C–H functionalization of heteroaryl N-oxides has been extensively developed,⁹ there are only two reports in the literature that address the selective formation of C–O bonds. Two synthetic methods use diisopropyl phosphite¹⁰ or PyBroP¹¹ as the activating agent to synthesize diaryl ethers, respectively. Although both are efficient, only one example has been reported for each, and the substrate scope needs to be further investigated. Moreover, there are some material-related issues, such as the need to use an excess of activating agent, the high cost of the material (PyBroP), highly dilute conditions, long reaction times and the use of highly toxic carbon tetrachloride. For the above reasons, general, practical, fast and green synthetic methods for the preparation of 2-aroxyisoquinolines are in demand.

Scheme 1 Challenges and approaches to the synthesis of 2-phenoxyisoquinolines.

Due to the essential role of activating agents in the functionalization of N-oxides, a variety of such agents have been identified. Among them, PyBroP, a phosphonium salt, stands out as an exceptional choice for facilitating the reaction of N-oxides with multiple nucleophiles.^9a,d,11,12 Nevertheless, the use of this reagent comes with certain drawbacks, including its high cost and the formation of phosphoramide as a byproduct (2, as shown in Scheme 1C). Additionally, the substantial molecular weight of PyBroP (MW = 466.19) and its associated phosphoramide byproduct (MW = 257.32) contribute to their reduced atom efficiency and overall reaction mass efficiency. In order to solve the problem, we propose to prepare PyBroP in situ, so that only a catalytic amount of PyBroP is required in the reaction. We were inspired by a report that 2 can react with phosphoryl tribromide to produce PyBroP.¹³ In this instance, we not only successfully recycled the reaction byproduct but also synthesized the activating agent necessary for the reaction. This design adheres to six of the twelve green chemistry principles:¹⁴ prevent waste, maximize atom economy, design less hazardous chemical syntheses, use safer solvents and reaction conditions, increase energy efficiency, and use catalysts, not stoichiometric reagents.

Results and discussion

The reaction between isoquinoline N-oxides and phenol under the catalysis of either PyBroP or phosphoramide (2) was set as the pilot reaction (Scheme 2). Commercially available and inexpensive phosphoramides were evaluated under the initial reaction conditions, and HMPA (2a) was the most efficient catalyst for this reaction. The reaction intriguingly yielded 4-bromoisoquinoline as a byproduct, which arises from an excess of phosphorus tribromide. We attempted to suppress its formation by adding silver salts or reducing the equivalent of phosphorus tribromide. However, both approaches resulted in a decrease in the yield of the target diaryl ether. Remarkably, this contrasts with the usual outcome when reacting with N-oxides, wherein 1-bromoisoquinoline is typically formed.^8b,8c,8f In addition to the selection of catalysts, the reaction conditions have been further optimized by an extensive exploration of other reaction parameters, including solvent, base, temperature, equivalent, and concentration (35 conditions screened, see Table S1†).

Scheme 2 Screening of an appropriate catalyst for the synthesis of 2-phenoxyisoquinolines.

The generalization of this 2-aryloxylation reaction is shown in Scheme 3 using a variety of phenols, naphthol and isoquinoline N-oxides. We successfully produced a series of 2-aroxyisoquinolines in yields ranging from moderate to good. Additionally, the reaction is remarkably swift, consistently achieving completion within a mere 5 minutes for all substrates examined. The substrate scope is broad, and multiple functional groups, including benzylic methyl, halogen, cyano, ester, methoxy and vinyl groups, are all well tolerated under the developed reaction conditions. Those products bearing bromo (3g, 3j–k, 3n, 3p–r, 3u, and 3w) and methyl (3b, 3o, 3s, and 3v) substituents are compatible with the reaction, thus providing additional handles for further functionalization at the halogenated or benzylic positions using cross-coupling reactions. The electronic properties of phenols are crucial for these reactions. It can be inferred that neutral phenols are superior reactants compared to those with electron-donating groups (3a > 3b > 3d) or electron-withdrawing groups (3a > 3h or 3i). Furthermore, the steric bulk of the phenol had little impact on the reaction outcome, as 2,6-dimethyl phenol performed well (3l). Regarding the scope of isoquinoline N-oxides, substituents at various positions, including at 3, 4, 5, 6, 7, and 8 positions, are all applicable. Notably, the brominated isoquinoline byproduct is suppressed in most substituted isoquinoline substrates. Generally, halogen-substituted examples tend to offer appreciable yields, such as compounds 3n, 3q and 3x. Most of these products are novel compounds, probably because these products are not accessible through conventional transition metal catalysis, as aromatic halides would react with transition metals. The versatility and mildness of the developed method were further illustrated through its application in the late-stage functionalization of biologically relevant, structurally complex targets. The reaction with carvacrol proceeded smoothly, yielding 3z in good quantity. Additionally, substrates containing labile functional groups, such as vinyl (3y, 3aa) or methoxy (3y, 3ab) groups, were also found to be compatible under the established reaction conditions.

Scheme 3 Substrate scope for the 2-aryloxylation of isoquinoline N-oxides. All reactions were conducted at 0.5 M concentration with isoquinoline N-oxide (100 mg, 1.0 equiv.), ArOH (3.0 equiv.), HMPA (0.4 equiv.), DIEA (2.0 equiv.) and POBr₃ (2.0 equiv.) in ethyl acetate (EA) at rt.

Next, we investigated the application of this reaction system to other azine N-oxides. After fine-tuning the reaction conditions for the isoquinoline N-oxide reaction (see Tables S2 and S3†), we discovered that the optimized system is also highly effective for other nitrogen heteroaromatics, such as quinoline, pyridine, pyrimidine, and 1,10-phenanthroline (Scheme 4). Notably, when using quinoline N-oxides, we observed the absence of the bromide by-product for most substrates and obtained yields ranging from moderate to excellent (up to 87%). The reaction is exceptionally swift for quinoline and pyrimidine substrates (within 5 minutes) and is compatible with various functional groups, such as benzylic methyl, halogen, ester, methoxy, vinyl and alkynyl groups. Besides, the reaction system is adaptable to different substitution positions on the quinoline ring. Electron-neutral and electron-rich phenols demonstrate exceptional performance, whereas electron-deficient phenols often underperform (6b > 6a > 6f). Pyridine substrates have shown effective performance, with 3-cyano pyridine emerging as the most beneficial. Nonetheless, the formation of regioisomeric products was observed, predominantly featuring substitution at the 6-position (7g–1). This preference is likely due to steric factors. In contrast to quinoline and isoquinoline substrates, pyridine substrates displayed reduced reactivity. This diminished reactivity necessitated longer reaction times and frequently resulted in the retention of unreacted starting materials. As a result, these factors collectively contributed to the moderate reaction yields observed.

Scheme 4 Substrate scope for the 2-aryloxylation or 2-alkoxylation of azine N-oxides. All reactions were conducted at 0.5 M concentration. Conditions with phenols: N-oxide substrate (1.0 equiv.; pyridine N-oxide: 1.05 mmol, other azine N-oxides: 0.69 mmol), ArOH (1.5–2.0 equiv.), HMPA (0.1–0.2 equiv.), DIEA (2.0 equiv.) and POBr₃ (1.0–2.0 equiv.) in EA at rt. ^a3.0 equiv. of phenol was employed. Conditions with alcohols: azine N-oxide (0.69 mmol, 1.0 equiv.), ROH (3.0 equiv.), PyBroP (0.4 equiv.), DIEA (2.0 equiv.) and POBr₃ (2.0 equiv.) in EA at rt.

The reactions of these azine N-oxides with alcohols were subsequently investigated (Scheme 4). After optimizing the reaction conditions, it was determined that HMPA is not an effective catalyst, whereas PyBroP proved to be effective (see Table S4†). Although the reaction proceeded to completion, the yields of isoquinoline and quinoline N-oxides (10a–10d) were low. The mass balance was primarily attributed to the formation of reduced products (isoquinoline or quinoline) and brominated byproducts. Notably, pyridine N-oxide (10e) remained unreactive toward alcohols under the developed reaction conditions.

Another remarkable feature of our developed method is scalability. The phenoxylation reactions were conducted with either isoquinoline or quinoline N-oxides on a 1 g scale. Both reactions were completed within 30 minutes, affording yields comparable to those achieved on the milligram-scale (Scheme 5A). Although these diaryl ethers appear simple, 24 out of the 57 are new compounds. This indicates the underdevelopment of simple, inexpensive, and practical synthetic methods for this series. It is instructive to comment on the means by which the known compounds were prepared, as shown in Scheme 5B. In nearly all cases, the current method offers a more feasible and economically viable alternative. Isoquinoline phenyl ethers, for instance, cannot be accessed under the classic S_NAr conditions, which involve using a strong base in a polar solvent under high temperatures. This is because very low yields were observed for 3a, 3d, 3f and 3h when using this approach (Process B, Scheme 5B).¹⁵ Compounds 3a^3a,5,16 and 3e¹⁷ were accessed from 1-haloisoquinoline and phenol derivatives by transition metal catalysis at high temperatures. Additionally, compound 3a can be produced from isoquinoline N-oxides by following established procedures,^10,11 as shown in Scheme 1B. However, these methods necessitate the use of excess activating agents. The key advantages of the developed method include the obviation of C1-prefunctionalized isoquinoline, the elimination of costly transition metals, and the reduction in the stoichiometric number of activating agents required.

Scheme 5 (A) Gram scale synthesis. (B) Practical alternatives to conventional synthesis.

Finally, mechanistic experiments were conducted to gain deeper insights into the reaction mechanism (Scheme 6A). Although 4-bromoisoquinoline (Cpd. 4) is generated during the reaction, and 1-bromoisoquinoline formed as a potential intermediate could theoretically react with phenol via an S_NAr pathway to yield product 3a, no reaction was observed when 1-bromoisoquinoline was employed as the substrate. This result effectively excludes the possibility of an S_NAr mechanism. Instead, a direct reaction between the N-oxide and phenol is proposed, as shown in Scheme 6B. First, phosphoramide (2) reacts with phosphoryl tribromide to produce the PyBroP analogue (A–Y).¹³ Coordination of 1a with A–Y forms an activated complex 11, whose electrophilicity is enhanced. Subsequent nucleophilic attack of phenol affords the desired diaryl ether 3a and phosphoramide, which proceeds into the next catalytic cycle.

Scheme 6 (A) Control experiments. (B) Proposed mechanism.

Conclusions

In conclusion, we have successfully developed an environmentally friendly synthetic method for producing heteroaryl ethers. This innovative process involves the reaction of azine N-oxides with phenol derivatives or alcohols, utilizing an in situ prepared PyBroP analogue as the activating agent. This analogue is formed through the reaction of the byproduct phosphoramide with phosphoryl tribromide. The standout features of this reaction include its exceptional atom economy (92% for 3a), the use of an eco-friendly solvent, mild reaction conditions, a short reaction time, and a broad substrate scope.

Notably, this is the first report utilizing the feedstock material HMPA as a catalyst to facilitate the reaction between nitrogen heteroaromatic N-oxides and phenol derivatives. Research into the reaction of N-oxides with other nucleophiles is ongoing, and the findings will be reported in due course.

Author contributions

D. W. designed the experiments and directed the project. D. L. optimized the reaction conditions and expanded the substrate scope. F. X. and T. H. expanded the substrate scope. D. L., F. X., T. H. and D. W. analyzed the data and wrote the ESI.† D. W. wrote the manuscript.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support from the Science Foundation for Distinguished Young Scholars of Xinjiang (No. 2022D01E34), the Foundation of Tianchi Innovation Leading Talent Project (No. 51052300410) and the National Natural Science Foundation of China (No. 22261051) is greatly acknowledged.

References

1. E. N. Pitsinos, V. P. Vidali and E. A. Couladouros, Eur. J. Org. Chem., 2011, 1207–1222.
2. (a) T. Chen, H. Xiong, J.-F. Yang, X.-L. Zhu, R.-Y. Qu and G.-F. Yang, J. Agric. Food Chem., 2020, 68, 9839–9877; (b) Y. Liu, B. Fu, Y. Xu, B. Ren and Z. Qin, Molecules, 2022, 27, 6803.
3. (a) R. Saari, J.-C. Törmä and T. Nevalainen, Bioorg. Med. Chem., 2011, 19, 939–950; (b) F. Bedos-Belval, A. Rouch, C. Vanucci-Bacque and M. Baltas, MedChemComm, 2012, 3, 1356–1372; (c) X. Chen, S. Ding, P. Zhan and X. Liu, Curr. Pharm. Des., 2013, 19, 2829–2838.
4. (a) G. Kervefors, A. Becker, C. Dey and B. Olofsson, Beilstein J. Org. Chem., 2018, 14, 1491–1497; (b) T. Kim, H. Kwon, D.-Y. Lee, D.-J. Kim, Y. Jeon, H. Shin, H. S. Kim, J. Hur, C. Lim, E.-H. Kim, D. Shin and S.-H. Kim, Bioorg. Chem., 2021, 116, 105398; (c) Y. Zhou, G. E. Klinger, E. L. Hegg, C. M. Saffron and J. E. Jackson, Nat. Commun., 2022, 13, 2050.
5. J.-H. Chu, P.-S. Lin and M.-J. Wu, Organometallics, 2010, 29, 4058–4065.
6. R. Takise, R. Isshiki, K. Muto, K. Itami and J. Yamaguchi, J. Am. Chem. Soc., 2017, 139, 3340–3343.
7. Y.-M. Tian, W. Silva, R. M. Gschwind and B. König, Science, 2024, 383, 750–756.
8. (a) D. Wang, H. Jia, W. Wang and Z. Wang, Tetrahedron Lett., 2014, 55, 7130–7132; (b) D. Wang, Y. Wang, J. Zhao, L. Li, L. Miao, D. Wang, H. Sun and P. Yu, Tetrahedron, 2016, 72, 5762–5768; (c) S. E. Wengryniuk, A. Weickgenannt, C. Reiher, N. A. Strotman, K. Chen, M. D. Eastgate and P. S. Baran, Org. Lett., 2013, 15, 792–795; (d) Y. Chen, J. Huang, T.-L. Hwang, M. J. Chen, J. S. Tedrow, R. P. Farrell, M. M. Bio and S. Cui, Org. Lett., 2015, 17, 2948–2951; (e) K. Qiao, L. Wan, X. Sun, K. Zhang, N. Zhu, X. Li and K. Guo, Eur. J. Org. Chem., 2016, 1606–1611; (f) R. S. Malykhin and A. Y. Sukhorukov, Adv. Synth. Catal., 2021, 363, 3170–3188.
9. (a) A. T. Londregan, S. Jennings and L. Wei, Org. Lett., 2010, 12, 5254–5257; (b) T. Nishida, H. Ida, Y. Kuninobu and M. Kanai, Nat. Commun., 2014, 5, 4387; (c) V. A. Rassadin, D. P. Zimin, G. Z. Raskil'dina, A. Y. Ivanov, V. P. Boyarskiy, S. S. Zlotskii and V. Y. Kukushkin, Green Chem., 2016, 18, 6630–6636; (d) Y. Lian, S. B. Coffey, Q. Li and A. T. Londregan, Org. Lett., 2016, 18, 1362–1365; (e) R. Muta, T. Torigoe and Y. Kuninobu, Org. Lett., 2019, 21, 4289–4292; (f) N. Singh, S. Sharma and A. Sharma, Adv. Synth. Catal., 2023, 365, 3505–3511; (g) Q. Ma, Y. Shi and D. Wang, Org. Lett., 2023, 25, 9181–9185; (h) C. Bu, K. Wang, C. Gong and D. Wang, Green Chem., 2024, 26, 4659–4664; (i) D. Wang, L. Désaubry, G. Li, M. Huang and S. Zheng, Adv. Synth. Catal., 2021, 363, 2–39; (j) J. Singh, R. I. Patel and A. Sharma, Adv. Synth. Catal., 2022, 364, 2289–2306; (k) K. Wang, L. Xu and D. Wang, Sci. Sin. Chim., 2023, 53, 1369–1376; (l) D. Zhang, K. Qiao, J. Hua, Z. Liu, H. Qi, Z. Yang, N. Zhu, Z. Fang and K. Guo, Org. Chem. Front., 2018, 5, 2340–2344; (m) Y. Zhang, P. K. Sahoo, P. Ren, Y. Qin, R. Cauwenbergh, P. Nimmegeers, G. SivaRaman, S. Van Passel, A. Guidetti and S. Das, Chem. Commun., 2022, 58, 11454–11457; (n) Y. Zhang, T. Zhang and S. Das, Chem, 2022, 8, 3175–3201.
10. W.-Z. Bi, C. Qu, X.-L. Chen, L.-B. Qu, Z.-D. Liu, K. Sun, X. Li and Y.-F. Zhao, Eur. J. Org. Chem., 2017, 5125–5130.
11. A. T. Londregan, S. Jennings and L. Wei, Org. Lett., 2011, 13, 1840–1843.
12. (a) A. T. Londregan, K. Burford, E. L. Conn and K. D. Hesp, Org. Lett., 2014, 16, 3336–3339; (b) D. Wang, J. Zhao, Y. Wang, J. Hu, L. Li, L. Miao, H. Feng, L. Desaubry and P. Yu, Asian J. Org. Chem., 2016, 5, 1442–1446; (c) D. Wang, H. Feng, L. Li, Z. Liu, Z. Yan and P. Yu, J. Org. Chem., 2017, 82, 11275–11287; (d) L. Zhao, L. Hao, Y. Fu, Y. Cheng, G. Pan, L. Désaubry, P. Yu and D. Wang, Adv. Synth. Catal., 2020, 362, 3841–3845.
13. J. Coste, E. Frerot and P. Jouin, J. Org. Chem., 1994, 59, 2437–2446.
14. P. T. Anastas and J. C. Warner, Green Chemistry: Theory and Practice, Oxford University Press, 2000.
15. E. S. Kim, S. E. Yoo, K. Y. Yi, S. Lee, J. S. Noh, Y. S. Jung, E. Kim and N. Jeong, Bull. Korean Chem. Soc., 2002, 23, 1003–1010.
16. C. H. Burgos, T. E. Barder, X. Huang and S. L. Buchwald, Angew. Chem., Int. Ed., 2006, 45, 4321–4326.
17. S. H. Ahn, K.-S. Hwang, D.-S. Shin, S. S. Kim, J. Y. Yang, B. H. Lee, E. J. Bae, B. W. Choi, M. A. Bae and J. H. Ahn, Bioorg. Med. Chem. Lett., 2020, 30, 127201.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5gc00249d

---