A Paternò–Büchi Reaction of Aromatics with Quinones under Visible Light Irradiation

Reported herein is a Paternò–Büchi reaction of aromatic double bonds with quinones under visible light irradiation. The reactions of aromatics with quinones exposed to blue LED irradiation yielded oxetanes at −78 °C, which was attributed to both the activation of double bonds in aromatics and the stabilization of oxetanes by thiadiazole, oxadiazole, or selenadiazole groups. The addition of Cu(OTf)2 to the reaction system at room temperature resulted in the formation of diaryl ethers via the copper-catalyzed ring opening of oxetanes in situ. Notably, the substrate scope was extended to general aromatics.

The vast majority of PB reactions require the use of UV light.A few visible-light PB reactions have recently been reported.In 2019, functionalized azetidines were synthesized via intramolecular aza PB reactions between imines and alkenes under visible light irradiation catalyzed by precious Ir-based complexes, as reported by Schindler and coworkers [30].In 2020, Dell'Amico and coworkers developed a visible-light PB reaction of substituted indoles and aromatic ketones [27].
Under the optimized reaction conditions, the substituent effect was significant and similar to many reported PB reactions [10].The presence of methyl groups in 1 at position 5 or 6 facilitated the formation of oxetane products under the optimized conditions.As X in 1 was selenium or oxygen, oxetanes 3b and 3c were formed in 64% and 43% yields, respectively (Scheme 2).The presence of only a single methyl in 5-methylbenzo[c][1,2,5]thiadiazole (1d) at position 5 resulted in the selective synthesis of compound 3d in 38% yield.The PB reaction only involved the double bonds with the methyl group in 1d.The substrate 2-methylcyclohexa-2,5-diene-1,4-dione (2b) was also used in the PB reactions involving 1a, [1,2,5]oxadiazole (1c) to synthesize oxetanes.Only the carbonyl at position 4 in 2b participated in the PB reaction.The two isomers (3e and 3e ′ , 3f and 3f ′ ), whose configurations (3e and 3e ′ ) were confirmed via NOESY spectra (pages S16 and S18 in Supplementary Materials), were formed via the reaction of 2b with 1a and 1b, respectively.However, only isomer 3g was detected in the reaction between 2b and 1c.Naphthalene-1,4-dione (1e) was also used as a substrate in this PB reaction to obtain oxetane 3h in 8% yield.at -40 °C, the yield was increased to 48% (entry 9).At -78 °C, no oxetane 3a was detected when Et2O or THF was used as a solvent (Table 1, entries 10 and 11).In dichloromethane (DCM), the optimal yield of 3a was 74%.And, when the reaction was carried out in the dark at -78 °C, there was no 3a detected.Thus, the use of 1a (0.9 mmol) and 2a (0.3 mmol) in DCM (3 mL) at -78 °C under blue LED irradiations represents optimized reaction conditions.  e]  3:1 -DCM -78 np a Reaction conditions: 1a and 2a (0.30 mmol) in the solvent (3 mL) under blue LED irradiation for 48 h.b Isolated yields. [c] Light source: 254 nm. [d] np = no product. [e] No blue LED irradiation.
At room temperature, the oxetane products in solution can be translated into the starting materials, e.g., a portion of 3a was transformed into 1a and 2a in deuterated chloroform in 2 h.Thus, an equilibrium may exist between the oxetane products and the starting materials in the reactions under the optimized reaction conditions.In the case of the other substrates, the equilibrium shifted in favor of the starting materials, resulting in a low yield of oxetanes, which could not be separated.To expand the application of this reaction, Cu(OTf)2 was added to translate the oxetane to shift the equilibrium to the right At room temperature, the oxetane products in solution can be translated into the starting materials, e.g., a portion of 3a was transformed into 1a and 2a in deuterated chloroform in 2 h.Thus, an equilibrium may exist between the oxetane products and the starting materials in the reactions under the optimized reaction conditions.In the case of the other substrates, the equilibrium shifted in favor of the starting materials, resulting in a low yield of oxetanes, which could not be separated.To expand the application of this reaction, Cu(OTf) 2 was added to translate the oxetane to shift the equilibrium to the right by the ring-opening reaction of the oxetane catalyzed by Lewis acid [33,34].Compound 4a was obtained when the reaction of 1a, 2a, and Cu(OTf) 2 (0.1 equiv.) in CH 3 CN was carried out under blue LED irradiation, which prompted us to develop a method for the synthesis of diaryl ethers (Scheme 3).
Molecules 2024, 29, x FOR PEER REVIEW 4 of 16 by the ring-opening reaction of the oxetane catalyzed by Lewis acid [33,34].Compound 4a was obtained when the reaction of 1a, 2a, and Cu(OTf)2 (0.1 equiv.) in CH3CN was carried out under blue LED irradiation, which prompted us to develop a method for the synthesis of diaryl ethers (Scheme 3).
Diaryl ethers are important structural units, which are widely present in natural product molecules and functional materials [35,36].Therefore, the development of effective methods for synthesizing diaryl ether has great research significance and widespread application.The general method to produce diaryl ethers involves the reaction of aryl halides with phenoxides via substitution reactions at high temperature in the presence of electron-withdrawing groups in aryl halides [37,38], and with phenols or phenoxides via transition metal-catalyzed reactions [39][40][41].Methods for the synthesis of diaryl ethers also were developed without the use of aryl halides.The reactions involving phenols with aryl boronic acids in the presence of copper acetate were developed to synthesize diaryl ethers [42][43][44][45].Recently, Ritter and coworkers developed a two-step process of diaryl ether synthesis from electron-rich aromatics and phenols using aryl thianthrenium salts by photoredox chemistry [46].In this work, diaryl ethers were synthesized via a reaction between quinones and aromatics, which included electron-rich, -poor, and -neutral aromatics under blue LED in the presence of Cu(OTf)2.
To ensure optimized conditions, benzo[c][1,2,5]thiadiazole (1b) was used as the model substrate in the reaction with 2a for conditional screening (Table 2).To our delight, the diaryl ether product 4b was afforded in 50% yield using Cu(OTf)2 (10% of 2a) as the catalyst in CH3CN at 40 °C under blue LED for 48 h (entry 1).Photosensitizers (Ir(bpy)3, Ru(bpy)3Cl2•6H2O, and thioxanthen-9-one) were also added to the reactions to increase the product yield, but there was no effect on the yield increase in 4a (entries 2-4).The Lewis acids were selected as catalysts to optimize this reaction.When the Lewis acids, Fe(OTf)3, Zn(OTf)2, Yb(OTf)3, Nd(OTf)3, and Sc(OTf)3 (entries 5-9), were used to replace Cu(OTf)2 in the reactions, product 4a could be obtained, but the yields were lower than that of entry 1.The using of Pd(OAc)2, AgOAc, and BF3•Et2O did not give 4a.The product yield was also affected by reaction temperature.When the reaction was carried out at 25 °C, the yield of 4a was increased to 53% (entry 13).The decrease in the ratio of 1b and 2a did not increase the yield of 4a (entries [15][16][17][18].Under the 3:1 ratio of 1b and 2a, the yield of 4a increased to 65% when the amount of Cu(OTf)2 increased to 20% of 2a (entry 19).Then, the solvent effect was explored to increase the product yield.Product 4a was not detected when the reactions were carried out in CH3OH, acetone, Et2O, or THF (entries 20-23).When DCM was used as a solvent, 4a was obtained in 45% yield.Thus, a mixed solvent of DCM and CH3CN was used in the reaction, which resulted in the highest yield of 4a (67%, entry 25).In addition, without blue LED irradiation, no 4a was obtained (entry 26), which indicated that the excitation of the starting material was a decisive factor in the reaction rather than the copper-catalyzed coupling reaction.Finally, after the optimization of Lewis acids, reaction temperature, and solvents, 1b (0.9 mmol), 2a (0.3 mmol), and Cu(OTf)2 (0.06 mmol) in CH3CN/DCM (1:1, 3 mL) at room temperature for 48 h under blue LED were determined as the optimized conditions.Diaryl ethers are important structural units, which are widely present in natural product molecules and functional materials [35,36].Therefore, the development of effective methods for synthesizing diaryl ether has great research significance and widespread application.The general method to produce diaryl ethers involves the reaction of aryl halides with phenoxides via substitution reactions at high temperature in the presence of electron-withdrawing groups in aryl halides [37,38], and with phenols or phenoxides via transition metal-catalyzed reactions [39][40][41].Methods for the synthesis of diaryl ethers also were developed without the use of aryl halides.The reactions involving phenols with aryl boronic acids in the presence of copper acetate were developed to synthesize diaryl ethers [42][43][44][45].Recently, Ritter and coworkers developed a two-step process of diaryl ether synthesis from electron-rich aromatics and phenols using aryl thianthrenium salts by photoredox chemistry [46].In this work, diaryl ethers were synthesized via a reaction between quinones and aromatics, which included electron-rich, -poor, and -neutral aromatics under blue LED in the presence of Cu(OTf) 2 .
To ensure optimized conditions, benzo[c][1,2,5]thiadiazole (1b) was used as the model substrate in the reaction with 2a for conditional screening (Table 2).To our delight, the diaryl ether product 4b was afforded in 50% yield using Cu(OTf) 2 (10% of 2a) as the catalyst in CH 3 CN at 40 • C under blue LED for 48 h (entry 1).Photosensitizers (Ir(bpy) 3 , Ru(bpy) 3 Cl 2 •6H 2 O, and thioxanthen-9-one) were also added to the reactions to increase the product yield, but there was no effect on the yield increase in 4a (entries 2-4).The Lewis acids were selected as catalysts to optimize this reaction.When the Lewis acids, Fe(OTf) 3 , Zn(OTf) 2 , Yb(OTf) 3 , Nd(OTf) 3 , and Sc(OTf) 3 (entries 5-9), were used to replace Cu(OTf) 2 in the reactions, product 4a could be obtained, but the yields were lower than that of entry 1.The using of Pd(OAc) 2 , AgOAc, and BF 3 •Et 2 O did not give 4a.The product yield was also affected by reaction temperature.When the reaction was carried out at 25 • C, the yield of 4a was increased to 53% (entry 13).The decrease in the ratio of 1b and 2a did not increase the yield of 4a (entries [15][16][17][18].Under the 3:1 ratio of 1b and 2a, the yield of 4a increased to 65% when the amount of Cu(OTf) 2 increased to 20% of 2a (entry 19).Then, the solvent effect was explored to increase the product yield.Product 4a was not detected when the reactions were carried out in CH 3 OH, acetone, Et 2 O, or THF (entries 20-23).When DCM was used as a solvent, 4a was obtained in 45% yield.Thus, a mixed solvent of DCM and CH 3 CN was used in the reaction, which resulted in the highest yield of 4a (67%, entry 25).In addition, without blue LED irradiation, no 4a was obtained (entry 26), which indicated that the excitation of the starting material was a decisive factor in the reaction rather than the copper-catalyzed coupling reaction.Finally, after the optimization of Lewis acids, reaction temperature, and solvents, 1b (0.9 mmol), 2a (0.3 mmol), and Cu(OTf) 2 (0.06 mmol) in CH 3 CN/DCM (1:1, 3 mL) at room temperature for 48 h under blue LED were determined as the optimized conditions.We then explored the scope of diaryl ether synthesis (Scheme 4).As substrate 1 contains thiadiazole, etherification always occurs at the active positions 4 or 7 in 1.Under the optimized conditions, 4b was obtained in 67% yield.The introduction of the same functional groups at positions 5 and 6 in 1, such as CH3, F, Cl, and Br, resulted in diaryl ethers 4a and 4c-4e in good yield (besides 4c).In the presence of a single substituent group in benzothiadiazole, the regioselectivity of the etherification was controlled by the electronic effects of the substituent.The presence of an electron-rich group (CH3, CH3O) at position 5 of 1 resulted in the synthesis of compounds 4f and 4g in 67% and 55% yield, respectively, via etherification at position 4 of 1.With halogen at position 5 in 1, two isomers of etherification at positions 4 and 7 were obtained, but the main products were 4h, 4i, and 4j.When position 5 of 1 was substituted by the formyl group (electron-poor group), its etherification at position 4 produced 4k in 49% yield.The presence of a CH3 group at position 4 in 1 led to its etherification at position 7 to generate 4l in 58% yield.The reaction of 6-methylbenzo[c][1,2,5]thiadiazole-5-carbaldehyde (1m, R 1  We then explored the scope of diaryl ether synthesis (Scheme 4).As substrate 1 contains thiadiazole, etherification always occurs at the active positions 4 or 7 in 1.Under the optimized conditions, 4b was obtained in 67% yield.The introduction of the same functional groups at positions 5 and 6 in 1, such as CH 3 , F, Cl, and Br, resulted in diaryl ethers 4a and 4c-4e in good yield (besides 4c).In the presence of a single substituent group in benzothiadiazole, the regioselectivity of the etherification was controlled by the electronic effects of the substituent.The presence of an electron-rich group (CH 3 , CH 3 O) at position 5 of 1 resulted in the synthesis of compounds 4f and 4g in 67% and 55% yield, respectively, via etherification at position 4 of 1.With halogen at position 5 in 1, two isomers of etherification at positions 4 and 7 were obtained, but the main products were 4h, 4i, and 4j.When position 5 of 1 was substituted by the formyl group (electron-poor group), its etherification at position 4 produced 4k in 49% yield.The presence of a CH 3 group at position 4 in 1 led to its etherification at position 7 to generate 4l in 58% yield.The reaction of 6-methylbenzo[c][1,2,5]thiadiazole-5-carbaldehyde (1m, R 1 = 5-CHO, R 2 = 6-CH 3 ) and 2a afforded 4m in low yield.However, 5-bromo-6-methylbenzo[c][1,2,5]thiadiazole (1n, R 1 = 5-Br, R 2 = 6-CH 3 ) reacted smoothly with 2a under standard conditions to generate two isomers 4n and 4n' in good yields.General aromatics without the activating group (thiadiazole) were also used in this reaction to afford the corresponding diaryl ethers.
Benzene as a substrate reacted with benzoquinone under standard conditions to yield 4o (46%).Reducing the reaction temperature to −40 • C resulted in a 95% yield of 4o (Table S1).The synthesis of 4p, 4q, 4s, and 4u was carried out in the presence of Cu(OTf) 2 in CH 3 CN/DCM (1:1) at −40 • C for 48 h.When methyl 2,4,6-trimethylbenzoate was used as the substrate, the corresponding product 4p was obtained in 64% yield, although it had steric hindrance.Etherification at the α position of naphthalene afforded 4q in 87% yield.The substituent quinones also were tolerated in this reaction.The reactions of 2,3dimethylbenzoquinone (2c, R 3 = 2,3-dimethyl) with 1b and benzene were carried out to obtain 4r and 4s with a yield of 34% and 44%, respectively.2-Chlorobenzoquinone (2d, R 3 = 2-Cl) reacted with 1b to afford 4t in 24% yield but with benzene to yield 4u in good yield (89%).General aromatics without the activating group (thiadiazole) were also used in this reaction to afford the corresponding diaryl ethers.Benzene as a substrate reacted with benzoquinone under standard conditions to yield 4o (46%).Reducing the reaction temperature to −40 °C resulted in a 95% yield of 4o (Table S1).The synthesis of 4p, 4q, 4s, and 4u was carried out in the presence of Cu(OTf)2 in CH3CN/DCM (1:1) at −40 °C for 48 h.When methyl 2,4,6-trimethylbenzoate was used as the substrate, the corresponding product 4p was obtained in 64% yield, although it had steric hindrance.Etherification at the α position of naphthalene afforded 4q in 87% yield.The substituent quinones also were tolerated in this reaction.The reactions of 2,3-dimethylbenzoquinone (2c, R 3 = 2,3-dimethyl) with 1b and benzene were carried out to obtain 4r and 4s with a yield of 34% and 44%, respectively.2-Chlorobenzoquinone (2d, R 3 = 2-Cl) reacted with 1b to afford 4t in 24% yield but with benzene to yield 4u in good yield (89%).The products all were characterized by 1 H NMR, 13 C NMR, and HRMS.The configurations of products were confirmed by X-ray analysis of the single crystals of 3a, 4b, 4k, 4n, and 4t.The products all were characterized by 1 H NMR, 13 C NMR, and HRMS.The configurations of products were confirmed by X-ray analysis of the single crystals of 3a, 4b, 4k, 4n, and 4t.
To elucidate the reaction mechanisms, the control experiments, UV-vis absorption, and fluorescence involved 1a and 2a were investigated.In UV-vis absorption spectra (Figure S1), only benzoquinone (2a) had an absorption peak at >400 nm, which was not affected by 1a or Cu(OTf) 2 .As shown in Scheme 5, no products were detected when 2,2,6,6-tetramethylpiperidinooxy (TEMPO) was added to the reactions during the synthesis of oxetane 3a and diaryl ether 4a under standard conditions (I and II in Scheme 5).The results showed that 2a was excited under blue LED and the processes involved free radicals.Product 3a could be smoothly obtained in the reaction with triplet quencher DABCO (III in Scheme 5), or the reaction under atmosphere (IV in Scheme 5) [47], which indicated that the triplet state was not favored in the reaction system.The result of the energy transfer calculations from the Gibbs energy of the photoinduced electron transfer (PET) equation showed ∆G > 0 (page S51 in Supplementary Materials), forbidding the possibility of the PET mechanism, which also needs to involve the triplet state [47].Oxetane 3a was treated with Cu(OTf) 2 (0.2 equiv.) in DCM/CH 3 CN (1:1) at room temperature to obtain compound 5, whose structure was confirmed via X-ray analysis of its single crystal (V in Scheme 5).
To elucidate the reaction mechanisms, the control experiments, UV-vis absorption, and fluorescence involved 1a and 2a were investigated.In UV-vis absorption spectra (Figure S1), only benzoquinone (2a) had an absorption peak at >400 nm, which was not affected by 1a or Cu(OTf)2.As shown in Scheme 5, no products were detected when 2,2,6,6-tetramethylpiperidinooxy (TEMPO) was added to the reactions during the synthesis of oxetane 3a and diaryl ether 4a under standard conditions (I and II in Scheme 5).The results showed that 2a was excited under blue LED and the processes involved free radicals.Product 3a could be smoothly obtained in the reaction with triplet quencher DABCO (III in Scheme 5), or the reaction under atmosphere (IV in Scheme 5) [47], which indicated that the triplet state was not favored in the reaction system.The result of the energy transfer calculations from the Gibbs energy of the photoinduced electron transfer (PET) equation showed ΔG > 0 (page S51 in Supplementary Materials), forbidding the possibility of the PET mechanism, which also needs to involve the triplet state [47].Oxetane 3a was treated with Cu(OTf)2 (0.2 equiv.) in DCM/CH3CN (1:1) at room temperature to obtain compound 5, whose structure was confirmed via X-ray analysis of its single crystal (V in Scheme 5).

Scheme 5. Control experiments.
Based on the literature [13,48] and the experimental evidence, a plausible mechanism of the reactions involving 1a and 2a was proposed (Scheme 6).Under blue LED irradiation, 2a was excited to the singlet state and reacted with 1a to mainly form the singlet exciplex (A) [49].The two oxetane products (3a and 3a′) may be produced via singlet biradical intermediates, which could go back to 1a and 2a.Oxetane 3a was more stable than 3a′ (the internal energy difference was 8 kcal/mol, which was calculated by DFT at the B3LYP/6-31G* level), which can be separated and characterized.Treatment of 3a with Cu(OTf)2 resulted in the cleavage of the carbon-carbon bond in oxetane to form intermediate C, followed by methyl migration and aromatization resulting in 5.When the starting materials were treated with Cu(OTf)2 under blue LED irradiation at room temperature, the equilibrium was biased toward raw materials.However, the highly reactive intermediate 3a′ was transformed into intermediate E, followed by aromatization to form 4a, which eventually promoted the transformation of starting materials to 4a.Accordingly, only 4a was obtained instead of 5 because of the high energy barrier from C to D. Based on the literature [13,48] and the experimental evidence, a plausible mechanism of the reactions involving 1a and 2a was proposed (Scheme 6).Under blue LED irradiation, 2a was excited to the singlet state and reacted with 1a to mainly form the singlet exciplex (A) [49].The two oxetane products (3a and 3a ′ ) may be produced via singlet biradical intermediates, which could go back to 1a and 2a.Oxetane 3a was more stable than 3a ′ (the internal energy difference was 8 kcal/mol, which was calculated by DFT at the B3LYP/6-31G* level), which can be separated and characterized.Treatment of 3a with Cu(OTf) 2 resulted in the cleavage of the carbon-carbon bond in oxetane to form intermediate C, followed by methyl migration and aromatization resulting in 5.When the starting materials were treated with Cu(OTf) 2 under blue LED irradiation at room temperature, the equilibrium was biased toward raw materials.However, the highly reactive intermediate 3a ′ was transformed into intermediate E, followed by aromatization to form 4a, which eventually promoted the transformation of starting materials to 4a.Accordingly, only 4a was obtained instead of 5 because of the high energy barrier from C to D. Scheme 6. Speculated mechanism.

Conclusions
In summary, a PB reaction between aromatics and quinones under visible light irradiation was developed.The PB reactions between aromatics and quinones were accomplished via activation of the benzene ring double bond using the combination of thiadiazole, oxadiazole, or selenadiazole under blue LED irradiation, resulting in the synthesis of oxetanes with an obvious substituent effect.The addition of Cu(OTf) 2 opened the generated oxetane rings in situ to transform into diaryl ethers, which expanded the range of substrate adaptation.

Scheme 2 .
Scheme 2. Substrate scope of the Paternò-Büchi reaction of aromatic with quinones.

Scheme 4 .
Scheme 4. Substrate scope for the synthesis of diaryl ethers.

Scheme 4 .
Scheme 4. Substrate scope for the synthesis of diaryl ethers.

Table 1 .
The Paternò-Büchi reaction of aromatic with quinones a .Selected examples of the Paternò-Büchi reaction.

Table 1 .
The Paternò-Büchi reaction of aromatic with quinones a .

Table 1 .
The Paternò-Büchi reaction of aromatic with quinones a .

Table 2 .
Optimization of the reaction conditions for the synthesis of diaryl ethers a .

Table 2 .
Optimization of the reaction conditions for the synthesis of diaryl ethers a .