HFIP mediated transition metal-free synthesis of C4-aryl-substituted quinolines

Abstract

A transition metal-free method for C4-arylated quinolines synthesis from propargylic chloride and aniline was developed using 1,1,1,3,3,3,-hexafluoroisopropanol (HFIP). During the N-alkylation process, overalkylation was effectively resolved by hydrogen bonding interaction in HFIP. The cyclized intermediates were oxidized to quinolines under ambient air without additional oxidants. A cosolvent system was found to expand the substrate scope to include unstable electron-rich propargyl substrates by preventing acid-induced decomposition.

---

Quinolines and their derivatives are widely found in natural products and biologically active drug candidates. In particular, 4-aryl quinolines have been identified in compounds exhibiting antiproliferative, anti-HBV, and anti-tumor activities (Scheme 1).¹

Scheme 1 Bioactive natural products with 4-aryl quinolines.

Due to the significant therapeutic potential of 4-aryl quinolines, various synthetic methods have been developed. Among those, one of the most well-established and widely employed strategies for synthesizing 4-aryl quinolines involves transition metal-catalyzed reactions.² However, these transition metals are costly and pose toxicity concerns that limit their use in pharmaceutical applications.³ Consequently, alternative strategies have been developed to synthesize 4-aryl quinolines without the use of transition metals, for example, by reacting alkynes and anilines in DMSO with various oxidants.⁴ Our group has also developed a strategy to access 4-arylated tetrahydroquinolines (THQs) and quinolones directly from propargylic chlorides and N-substituted anilines (Scheme 2a).⁵ In this process, HFIP was found to be crucial for the activation of the C–Cl bond, facilitating C–N bond formation with poorly nucleophilic anilines. After the cyclization, the resulting disproportionated mixture is then converted into either THQs or quinolones, depending on the use of a proper reductant or oxidant.

Scheme 2 (a) HFIP-mediated syntheses of THQs and quinolones. (b) Selective monomethylation of amines with HFIP. (c) HFIP-mediated synthesis of quinolines.

In previous studies, the use of free aniline was avoided due to concerns on overalkylation. HFIP is known not only to activate C–X,⁶ C–O,⁷ C–H,⁸ and C=N⁹ bonds, but also to suppress overalkylation through hydrogen bonding with primary amines.¹⁰ This was exemplified by the Legros group, where HFIP enabled the selective monomethylation of anilines under mild conditions (Scheme 2b).¹¹ Building upon these insights, it was hypothesized that HFIP could effectively mitigate the overalkylation observed in prior studies. Herein, we demonstrate the synthesis of C4-aryl-substituted quinolines from propargylic chlorides and free anilines, confirming the ability of HFIP to resolve overalkylation.

In the preliminary investigation, aniline (1a) and propargylic chloride 2a were selected as model substrates to evaluate the effects of different solvents and temperatures. Initially, the HFIP system was tested at 80 °C with 1 equiv. of aniline. Under these conditions, the desired quinoline product, 5aa, was obtained along with an N-alkylated intermediate and a trace amount of overalkylated byproduct 4a (Table 1, entry 1). When the reaction was conducted in MeCN, formation of 5aa was not observed, while the production of the overalkylated byproduct increased (Table 1, entry 2), suggesting that HFIP suppresses the overalkylation of the nitrogen. Increasing the amount of 1a to 3 equiv. slightly improved the yield of 5aa, although compounds 3a and 4a remained present (Table 1, entry 3). Subjecting 3 equiv. of 1a in MeCN, only resulted in further accumulation of 3a (Table 1, entry 4), indicating that HFIP plays a critical role in the cyclization step. In contrast to the results at 80 °C, full conversion to 5aa was observed with no overalkylated byproducts at 120 °C (Table 1, entry 5).

Table 1 Overalkylation test with anilines^a

Entry	1a (equiv.)	Solvent	Temp (°C)	Ratio of 3a : 4a : 5aa^b
a Reaction conditions: 1a (1 or 3 equiv.), 2a (0.10 mmol), TfOH (0.1 equiv.), and solvent (0.20 mL) at designated temperature for 20 h. b NMR yield determined by using CH2Br2 as an internal standard.
1	1	HFIP	80	25 : 2 : 28
2	1	MeCN	80	19 : 9 : 0
3	3	HFIP	80	28 : 2 : 33
4	3	MeCN	80	62 : 6 : 0
5	3	HFIP	120	0 : 0 : 60

---

To investigate the ability of HFIP to suppress overalkylation, additional experiments were conducted using the overalkylated byproduct 4a (Table 2). Upon reaction of separately prepared 4a with aniline in HFIP at 120 °C, quinoline 5aa was obtained in 36% yield (Table 2, entry 1). In contrast, no quinoline formation was observed in the absence of aniline or at 80 °C (Table 2, entries 2 and 3). To further elucidate the conversion pathway of 4a to 5aa, a kinetic study was performed. The quinolinium intermediate 5aa′ grew rapidly, followed by the formation of quinoline 5aa and N-alkylated intermediate 3a (see SI). This transformation was only observed under the conditions of 120 °C in the presence of aniline, suggesting that 4a undergoes cyclization to form 5aa′ at elevated temperature, which then reacts with aniline via nucleophilic attack to produce both quinoline 5aa and N-alkylated intermediate 3a. These findings suggest that HFIP exerts a dual function by first suppressing overalkylation via hydrogen bonding and then enabling the conversion of the overalkylated byproduct into quinoline.

Table 2 Conversion of 4a to quinoline^a

Entry	1a (equiv)	Temp (°C)	Yield (%) of 5aa^b
a Reaction conditions: 1a (each equiv.), 4a (0.10 mmol), TfOH (0.1 equiv.), and HFIP (0.20 mL) at designated temperature for 20 h. b NMR yield determined by using CH2Br2 as an internal standard.
1	3	120	36
2	0	120	N.D.
3	3	80	N.D.

---

To further optimize the reaction conditions, a screening was performed to evaluate the effects of temperature and the amounts of aniline, Brønsted acids, and oxidants (see SI). A key factor in the optimization process was the variation in reactivity of propargylic chlorides depending on their electron density. In the previous studies, electron-deficient propargyl chlorides degraded slowly under the acidic reaction conditions, while the analogs with increased electron density were found to undergo rapid decomposition. Thus, the reaction conditions were optimized accordingly depending on the electron densities of the propargylic chlorides. For electron-deficient substrates, elevated temperatures (120 °C) and catalytic amounts (1.0 or 0.1 equiv.) of TfOH were critical for accelerating the reaction. In contrast, for electron-rich substrates, decomposition was effectively suppressed by employing toluene as a cosolvent with HFIP. These results highlight the necessity of tuning reaction conditions based on the electronic nature of the propargylic chloride to achieve both efficient conversion and minimal decomposition.

The optimized conditions were applied to a variety of propargylic chlorides, using p-toluidine (1b) as the nitrogen source. This choice was made based on the expectation that the electron-rich aromatic ring of p-toluidine would enhance the rate of the Friedel–Crafts reaction, potentially leading to slightly higher yields (Scheme 3). Electron-deficient propargylic chlorides 2a–o were subjected to condition A. Propargylic chlorides substituted with diverse electron-withdrawing groups at the para position were first evaluated. Methyl ester substituted propargylic chloride 2a afforded quinoline 5ba in a good yield (5ba, 64%). Halogenated propargylic chlorides 2b–2e not particularly efficient substrates, resulting in somewhat low yields (5bb–5be, 38–46%). The para-substituted trifluoromethoxy propargylic chloride 2f produced quinoline 5bf in a reduced yield of 30%, likely due to the unique properties of the trifluoromethoxy group. While the trifluoromethoxy group introduced an inductive effect through the oxygen-substituted trifluoromethyl group rendering the aromatic ring electron-deficient, the repulsion between the lone pairs on fluorine and the π-electrons of the aromatic ring increased the electron density specifically at the para position.¹² The decreased stability of propargylic chloride 2f is likely ascribed to increased electron density at the para position, which leads to rapid decomposition and consequently a lower yield compared to other electron-deficient substrates. In contrast, propargylic chloride 2g bearing a trifluoromethoxy group at the meta position was unaffected by such destabilizing interactions and behaved like other electron-deficient substrates, affording quinoline 5bg in 45% yield.

Scheme 3 Scope of propargylic chlorides. Condition A: 1b (3.0 equiv.), 2 (0.25 mmol), TfOH (1.0 equiv.), and HFIP (0.50 mL) at 120 °C for 16 h; condition B: 1b (3.0 equiv.), 2 (0.25 mmol), TfOH (0.1 equiv.), and HFIP/toluene (1 : 1 mixture, 0.50 mL) at 80 °C for 16 h. Isolated yield. ^a0.1 equiv. of TfOH was used. ^bOnly HFIP was used as the solvent. ^cReaction performed at 120 °C with HFIP (1.0 equiv.) in toluene (0.5 mL).

Electron-withdrawing propargylic chlorides, 2h–2k featuring nitrile, trifluoromethyl, nitro or ethyl ester groups, delivered the desired quinolines in acceptable yields of 44% to 53% (5bh–5bk). The electron-deficient nature of 3,5-trifluoromethyl-substituted propargylic chloride 2l required the use of 1 equiv. of TfOH, affording quinoline 5bl in 51% yield. For substrate 2m, bearing methyl ester groups at the meta position, a fair yield of 45% was obtained. Propargylic chlorides 2n and 2o, substituted with a 2-naphthyl group and a para-methyl ketone, respectively, both afforded a 29% yield. For the 2-naphthyl derivative 2n, it required high temperature to fully convert the N-alkylation intermediate, but this promoted decomposition. For 2o, while para-methyl ketone substituents have shown good yields in previous studies, the presence of free aniline in this reaction led to imine formation resulting in reduced yields.

Electron-rich propargylic chlorides 2p–2v were evaluated under condition B. Compared to other electron-rich propargylic chlorides, the lower electron density of propargylic chloride 2p led to sluggish cyclization of the N-alkylated intermediate when toluene was employed. Consequently, the reaction with 2p was carried out in HFIP without any cosolvent. Although the N-alkylated intermediate was fully consumed, decomposition of propargylic chloride resulted in a low yield (35%) of quinoline 5bp. For propargylic chloride 2q bearing a para-substituted OMe group, decomposition was effectively suppressed by employing only 1 equiv. of HFIP. Under the reaction conditions at 120 °C, quinoline 5bq was afforded in a good yield of 58%. Propargylic chlorides 2r–2t, with methyl substituents at the para-, meta-, and ortho-positions respectively, yielded the corresponding quinolines in 40–42% yields, indicating that steric effects at these positions exert minimal influence on the reactivity (5br–5bt). Quinoline 5bu was smoothly obtained in a moderate yield of 42% from 2u, which contains a thiophene substituent known for its versatility in various reactions. For 2v, substituted with a bulky tert-butyl group at the para position, quinoline 5bv was produced in 45% yield.

To further expand the substrate scope, functionalized anilines were also investigated (Scheme 4). Several anilines were tested with methyl ester substituted propargylic chloride 2a. Electron-neutral aniline 1a, along with electron-rich anilines 1c and 1d, substituted with methoxy and tert-butyl groups at the para position, provided the corresponding quinolines in good yields (5aa–5da, 50–59%). Interestingly, electron-deficient anilines 1e and 1f, substituted at the para position with fluorine or nitro groups, afforded quinolines 5ea and 5fa in good yields of 61% and 66%, respectively. Anilines with moderate electron-withdrawing groups, such as chloro 1g and methyl ester 1h substituents, delivered quinolines in fair yields (5ga–5ha, 49–54%). In contrast, 3,5-dimethyl-substituted aniline 1i exhibited reduced yield (5ia, 39%), likely due to steric hindrance at the meta positions impacting the reaction. Aniline derivative 1j, containing a 1,3-dihydrofuran moiety at the 3,4-positions, which is a structural motif commonly found in pharmaceuticals, was converted to quinoline 5ja in 29% yield, with a 6% isomer observed by NMR spectroscopy.

Scheme 4 Scope of anilines. Reaction conditions: 1 (0.75 mmol), 2a (0.25 mmol), TfOH (1.0 equiv.), and HFIP (0.50 mL) at 120 °C for 24 h. Isolated yield.

A plausible reaction mechanism has been proposed (Scheme 5). The process begins with HFIP-mediated activation of the C–Cl bond in propargylic chloride, enabling nucleophilic attack by aniline to form N-alkylated intermediate 3. This intermediate diverges into two pathways: cyclization to dihydroquinoline 6 (path A) or additional propargylation to form intermediate 4 (path B). In path A, 6 undergoes acid-mediated disproportionation to give quinoline 5 and THQ 7, and the THQ is further oxidized to quinoline 5 by molecular oxygen in air. In Path B, intermediate 4 cyclizes to form quinolinium 5′, stabilized by HFIP. Nucleophilic attack of 5′ by aniline yields quinoline 5 along with regeneration of the intermediate 3. Kinetic studies with 4 as the substrate confirmed the formation of 5′ at early stages, followed by quinoline production (see SI).

Scheme 5 Plausible mechanism.

For further demonstration, this methodology was applied to the total synthesis of C4-aryl quinoline derivative yaequinolone A2 known for its antibacterial activity (Scheme 6).¹³ In previous studies, the synthesis of structurally similar aflaquinolone F required an additional oxidation step starting from THQ 5aq.⁵ In contrast, the current method facilitated the direct one-pot conversion of quinoline 5aq to quinolone 8,¹⁴ followed by vicinal dihydroxylation to furnish yaequinolone A2 (9) in only three steps.

Scheme 6 Total synthesis of yaequinolone A2.

Conclusions

In summary, A method for synthesizing C4-aryl quinolines from propargylic chlorides and anilines was developed, minimizing the use of excess oxidants and effectively suppressing overalkylation. The reaction was facilitated by HFIP, which played a pivotal role in activating the C–Cl bond, stabilizing cationic intermediates, and promoting air-mediated oxidation of tetrahydroquinolines without the need for additional oxidants. Moderate to good yields of C4-aryl quinolines were obtained through the use of a cosolvent system, including those derived from electron-rich substrates that were previously incompatible due to their instability. The method's utility was further demonstrated by the efficient three-step total synthesis of yaequinolone A2.

Author contributions

H. M. C. conceived and supervised the project. T. K. K. conducted the experimental analyses. The paper was written through the contributions of all authors.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included in this published article and its SI.

SI containing experimental details and characterization data of the new compounds as well as copies of ¹H and ¹³C NMR spectra with HRMS data of all new compounds are available. See DOI: https://doi.org/10.1039/d5ob01182e.

Acknowledgements

This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2020R1C1C1015194 and 2020R1A5A1019023) and the Basic Science Research Institute fund (2021R1A6A1A10042944).

References

1. (a) P. Y. Chung, Z. X. Bian, H. Y. Pun, D. Chan, A. S. Chan, C. H. Chui, J. C. Tang and K. H. Lam, Future Med. Chem., 2015, 7, 947–967; (b) B. S. Matada, R. Pattanashettar and N. G. Yernale, Bioorg. Med. Chem., 2021, 32, 115973; (c) R. H. Guo, Q. Zhang, Y. B. Ma, X. Y. Huang, J. Luo, L. J. Wang, C. A. Geng, X. M. Zhang, J. Zhou, Z. Y. Jiang and J. J. Chen, Bioorg. Med. Chem., 2011, 19, 1400–1408; (d) Y. Qu, T. Y. Zhou, F. W. Guo, M. Y. Wei, G. Y. Chen, Y. C. Gu, C. Y. Wang and C. L. Shao, Eur. J. Med. Chem., 2023, 250, 115183.
2. (a) Y. Wang, C. Chen, J. Peng and M. Li, Angew. Chem., Int. Ed., 2013, 52, 5323–5327; (b) A. Arcadi, M. Aschi, F. Marinelli and M. Verdecchia, Tetrahedron, 2008, 64, 5354–5361.
3. M. Yaman, G. Kaya and H. Yekeler, World J. Gastroenterol., 2007, 4, 612–618.
4. (a) M. Phanindrurde, S. B. Wakade, D. K. Tiwari, P. R. Likhar and D. K. Tiwari, J. Org. Chem., 2018, 83, 9137–9143; (b) D. Chandra, A. K. Dhiman, R. Kumar and U. Sharma, Eur. J. Org. Chem., 2019, 2753–2758; (c) S. Faraz and A. T. Khan, Org. Biomol. Chem., 2023, 21, 7553–7560.
5. (a) S. H. Lee and H. M. Chi, Org. Lett., 2023, 25, 1083–1087; (b) S. H. Lee, M. S. Kang and H. M. Chi, Helv. Chim. Acta, 2023, 106, e202300153.
6. P. A. Champagne, Y. Benhassine, J. Desroches and J.-F. Paquin, Angew. Chem., Int. Ed., 2014, 53, 13835–13839.
7. (a) X. Mo, J. Yakiwchuk, J. Dansereau, J. A. McCubbin and D. G. Hall, J. Am. Chem. Soc., 2015, 137, 9694–9703; (b) N. K. Mishra, A. Rakshit, K. Moon, P. Singh and I. S. Kim, Bull. Korean Chem. Soc., 2024, 45, 131–144.
8. S. K. Sinha, T. Bhattacharya and D. Maiti, React. Chem. Eng., 2019, 4, 244–253.
9. S. K. Senapati, T. Das and A. Das, Org. Chem. Front., 2025 10.1039/d5qo00519a , Advance Article.
10. (a) D. Vuluga, J. Legros, B. Crousse, A. Z. Slawin, C. Laurence, P. Nicolet and D. D. Bonnet, J. Org. Chem., 2011, 76, 1126–1133; (b) I. Colomer, ACS Catal., 2020, 10, 6023–6029.
11. T. Lebleu, X. Ma, X. Maddaluno and J. Legros, Chem. Commun., 2014, 50, 1836–1838.
12. E. Castagnetti and M. Schlosser, Chem. – Eur. J., 2002, 8, 799–804.
13. R. Uchida, R. Imasato, H. Tomoda and S. Omura, J. Antibiot., 2006, 59, 646–651.
14. L. Y. Xie, Y. Duan, L. H. Lu, Y. J. Li, S. Peng, C. Wu, K. J. Liu, Z. Wang and W. M. He, ACS Sustainable Chem. Eng., 2017, 5, 10407–10412.

---