1 Introduction

Coumarins are significant natural products, which display wide and interesting pharmacological properties such as anti-breast cancer, anti-HIV, anti-Alzheimer, vasorelaxant, and anti-aggregatory activities [1], [2], [3], [4]. Coumarin derivatives have proven to be useful skeletons in organic synthesis as valuable building blocks and in pharmaceuticals. In particular, 3-arylcoumarins have important applications in dyes [5], [6] and pharmaceutical industries [7], [8]. There are several drugs constituting functionalized 3-arylcoumarin as the basic core in them, which possess wide-ranging activities [9], [10], [11], [12], [13], [14], [15].

Despite pervasiveness of 3-arylcoumarins in several fields of science, they can be obtained by a few general methods such as using a cycloaddition ring construction approach [16], [17], [18], and by transition metal-catalyzed coupling reactions through C–H bond activation [19], [20], [21], [22], [23], [24], [25]. Of these pathways, a direct arylation approach to form 3-arylcoumarins has been realized via transition metal-mediated cross-coupling reactions [26], [27], [28], [29]. The common limitations of most of these methods are inevitability of transition metals which possess high cost and toxicity, necessity of prefunctionalization of the C–H bond at 3-position of the starting coumarins with a halogen or carboxyl group, and use of expensive arylating counterparts such as boronic acid, elevated temperature, and longer reaction time. Because a radical arylation reaction can basically offer a much more straightforward access to reaction substrates, mainly due to the fact that such arylations are comparable to C–H activation reactions and less functionalized starting materials are therefore required, there has recently been increasing interest in improving the reaction type [30], [31], [32]. Based on the traditional aryl radical sources for arylation reactions, which are arenediazonium salts, remarkable advances have recently been achieved by employing photocatalysts and optimized Gomberg–Bachmann conditions [33], [34], [35]. But arenediazonium salts are difficult to synthesize, unstable, known as explosive under certain circumstances, and give very poor yield with formation of azoproducts. Therefore, with the background of these inadequacies, developing an adaptable method for 3-arylation of coumarins, which has short reaction time under mild reaction conditions, is still a significant challenge.

Arylhydrazines are known to undergo oxidation to form aryl radicals in the presence of oxygen [36], [37], catalysts [38], [39], and stoichiometric amounts of high valent metal oxidants such as lead(IV) acetate [40], manganese(III) acetate [36], [41], ferrate salt [42], [43], or hypervalent iodine(V) reagent [44], [45]. Another aspect that will facilitate future applications is that arylhydrazines are simple and cheap starting materials. Recently, the oxidative radical 3-arylation of coumarins and quinolin-4-ones using arylhydrazines as aryl radical sources has been reported in the presence of oxygen from air [46], [47]. But the oxidative radical coupling reaction has longer time, and the scope of substrates is also narrow. Guided by the recent studies on using arylhydrazines as aryl radical sources, we decided to investigate the modular synthesis for 3-arylcoumarins by this method. Herein, we disclose an oxidant radical regioselective arylation of coumarins and quinolinones with arylhydrazines using KMnO₄ as oxidant to produce 3-arylcoumarin and 3-arylquinolinone derivatives in moderate to good yields under mild conditions (Scheme 1).

Scheme 1:

Synthesis of 3-arylcoumarins and 3-arylquinolinones.

2 Results and discussion

We began our study with the selection of coumarin (1a) and phenylhydrazine (2a) as model substrates to optimize the reaction condition (Table 1). Initially in the absence of the oxidant, the desired product 3-arylcoumarin (3a) was not obtained when the reaction was carried out in CH₃CN at 80°C for 4.0 h under air atmosphere (entry 1). Fe(NO₃)₃, MnO₂, Mn(OAc)₂, Mn(OAc)₃, KMnO₄, H₂O₂, and t-butyl hydroperoxide (TBHP) were investigated as oxidants (Table 1, entries 2–8), but only Mn(OAc)₃ and KMnO₄ gave the desired product in decent yields, and KMnO₄ proved to be the most effective. The amount of KMnO₄ was also examined (Table 1, entries 6 and 9–12). The best yield was 65% when 3.0 eq KMnO₄ was used, which indicated that the oxidant KMnO₄ and the quantity of oxidant played important roles in this reaction. Subsequently, different solvents such as CH₃CN, H₂O, dioxane, CH₃OH, dimethylsulfoxide (DMSO), 1,2-dichloroethane (DCE), and tetrahydrofuran (THF) were screened based on the solubility of the substrates (Table 1, entries 6 and 13–18) and CH₃CN was found to be the best solvent for this transformation. The ratio of substrates coumarin and phenylhydrazine was investigated, and the ratio 1:2 of coumarin and phenylhydrazine proved to be the best result, providing 80% yield of 3a (Table S1, Supplementary Information available online). With increasing amount of phenylhydrazine, the yield of 3a decreased as the side reaction occurred to lead to byproducts. When the reaction temperature was improved from 40°C to 100°C, the yield of 3a was enhanced from 40% to 85% (Table S1, entry 4, and Table 1, entries 19–22). However, the product yields dramatically dropped if the reaction temperature was higher than 80°C, and the optimal reaction temperature was 80°C. Finally, reaction times were also examined, and 3.0 h gave good yield (Table 1, entries 23–26). After screening the reaction conditions, the optimal result was identified: the reaction of coumarin (1.0 equiv.) and phenylhydrazine (2.0 equiv.) was carried out in the presence of KMnO₄ (3.0 equiv.) in CH₃CN at 80°C for 3.0 h (Table 1, entry 26), which afforded 3-arylcoumarin in 85% yield.

With optimal reaction conditions in hand, the scope and generality of this methodology was further investigated (Table 2). First, a variety of substituted arylhydrazines were allowed to react with coumarin (Table 2, 3a–3o). Initially, the direct use of commercially available arylhydrazine hydrochlorides as the substrates did not lead to the desired products. Pretreatment of these hydrochloride salts with NaOH is required. Reactions of arylhydrazines having electron-donating groups such as –CH₃, –C₂H₅, and –OCH₃ with coumarin afforded 3-arylcoumarins 3a–3f in good to excellent yields of 78%–89%. Also the reaction of arylhydrazines bearing electron-withdrawing groups such as –F, –Cl, –CN, and –OCF₃ with coumarin proceeded smoothly and afforded 3g–3j, 3m, and 3n in moderate yields of 40%–70%. Especially, arylhydrazines bearing the strong electron-withdrawing –CN and –OCF₃ groups could also provide corresponding 3-arylated products in moderate yield (3i and 3j). It was noteworthy that arylhydrazine halides were well tolerated, providing handles for further functionalization (e.g. with a Suzuki reaction). However, 4-bromine function in the phenylhydrazine afforded only 20% yield (3l); the reason remained unknown. On the other hand, no expected product was observed in the case of ortho-bromophenylhydrazine (3k). It is not suitable for this transformation likely because of steric hindrance of ortho-bromophenylhydrazine. It is worth mentioning that an alkylhydrazine, tert-butylhydrazine, was also applied to the reaction, and the expected product was obtained (3o) in 35% yield. The fact demonstrated that alkylhydrazines are suitable precursors for alkyl radicals in this process.

Table 2:

Synthesis of 3-arylcoumarins from coumarins and arylhydrazines^a.

^aReaction conditions: coumarins 1 (0.5 mmol), arylhydrazines 2 (1.0 mmol), KMnO₄ (1.5 mmol), CH₃CN (10 mL) at 80°C for 3.0 h.

^bIsolated yields in parentheses.

Further, the scope of this methodology was also investigated with respect to coumarins (Table 2, 3p–3w). Various substituted coumarins were allowed to react with phenylhydrazine. It could be seen that the substituted coumarins bearing electron-rich groups such as –OH, –CH₃, –OCH₃, –OC₂H₅, and –N(C₂H₅)₂ were transformed into the corresponding 3-arylcoumarins in good yields (3p–3v). Notably, the standard reaction conditions could also be applied to 4-substituted coumarins, affording the corresponding products (3s–3v) in 58%–70% yields, which indicated that the steric hindrance of coumarins played a weak role in this reaction. Pleasingly, even a sensitive functionality such as the hydroxyl group was also tolerated and the coupling reactions proceeded with no requisite for protection of this group (3p, 3s, and 3v). This feature is ubiquitous in hydroxycoumarin-based biologically active products, which eliminates the requirement of protection and deprotection of hydroxyl groups. Unfortunately, coumarin possessing an electron-withdrawing –NO₂ group at the C6 position did not furnish the desired product 3w.

To further illustrate the generality and efficacy of this coupling reaction, a series of quinolinone substrates were also briefly surveyed (Table 3). Pleasingly, the C3 position of quinolinone derivatives was exclusively arylated, affording 3-arylquinolinones in 30%–83% yields. It was noteworthy that 2-quinolinone derivatives with a sensitive amide group were also tolerated and the coupling reaction proceeded with no requisite for protection of this group in good yield (5a and 5b). Notably, quinolinone derivatives bearing electron-rich (–OCH₃) and electron-poor (–F, –Br) groups could proceed smoothly under the current condition. Although quinolinones with an electron-poor –F or –Br substituent had lower reactivity than that with an electron-rich group, the corresponding 3-arylated quinolinones (5band 5f) with intact halo groups could serve as good precursors for further functionalization. The reaction proved to be fairly general for other electron-poor and electron-rich heterocycles, such as quinolinone derivatives.

Table 3:

Synthesis of 3-arylquinolines from quinolinones and arylhydrazines^a.

^aReaction conditions: quinolinones 1 (0.5 mmol), arylhydrazines 2 (1.0 mmol), KMnO₄ (1.5 mmol), CH₃CN (10 mL) at 80°C for 3.0 h.

^bIsolated yields in parentheses.

On the basis of the previous studies [36], [39], a plausible reaction mechanism for all arylations described above is proposed in Scheme 2. The phenylhydrazine A can be oxidized by KMnO₄ to form a phenyldiazene B. The diazene B is then rapidly converted to a phenyl radical C by the release of N₂ in the presence of KMnO₄. The phenyl radical C can selectively add to the C3 position of coumarin to form a radical intermediate D, allowing the carbocation intermediate E to be formed by a single-electron transfer. Finally, the intermediate Eloses a proton to produce the C3 arylated coumarin F.

Scheme 2:

Plausible reaction mechanism for the formation of 3-arylcoumarin.

3 Conclusions

In summary, we have successfully developed an improved and regioselective C-3 arylation on coumarins and quinolinones with arylhydrazines using KMnO₄ as an oxidant. This protocol provides a valuable approach to synthesize biologically interesting 3-arylcoumarins and 3-arylquinolinones. The advantages of this reaction are high efficiency, moderate to good yield, a broad functional group tolerance, and the use of cheap and readily available starting materials.

4 Experimental section

5 Supplementary information

Further experimental details and NMR spectra of compounds 3a–3v and 5a–5fare given as Supplementary Information available online (http://dx.doi.org/10.1515/znb-2016-0109).