Tuning Benzylic C−H Functionalization of (Thio)xanthenes with Electrochemistry

Here, we report a tunable electrochemical benzylic C−H functionalization of (thio)xanthenes with terminal alkynes and nitriles in the absence of any catalyst or external chemical oxidant. The benzylic C−H functionalization can be well controlled by varying the electrochemical conditions, affording the specific coupling products via C−C and C−N bond formation.

Molecules 2023, 28, 6139 2 with terminal alkynes and nitriles has been less explored. Herein, we will describe a approach for the direct benzylic C(sp 3 )−H functionalization of (thio)xanthenes that ca well tuned by varying the reaction temperature and reaction medium, leading to the mation of carbon−carbon and carbon−nitrogen bonds in an efficient manner, respect (Scheme 1f).

Results and Discussion
Initially, we chose xanthene 1a and phenylacetylene 2a as coupling partners for the optimization of reaction conditions, and the results are listed in Table 1. Based on our recent findings regarding electrochemical benzylic C−H functionalization [32][33][34][35], acetonitrile was preferentially selected as a reaction medium for this investigation. With n Bu4NPF6 as an electrolyte, the reaction of 1a with 2a was performed under constant-current conditions (8 mA) in an undivided cell equipped with a carbon anode and a platinum cathode. After proceeding at room temperature under an air atmosphere for 3 h, the product 3a was isolated in 32% yield. Meanwhile, the formation of 4a generated from the reaction of 1a with CH3CN was obtained with a 40% yield (entry 1). The ratio of 3a/4a was effectively improved by varying the reaction temperature from 30 to 60 °C (entries 2−4). Remarkably, reaction temperature could control the formation of 3a as a major product, Scheme 1. Electrochemical methods for the benzylic C(sp 3 )−H functionalization of (thio) xanthenes. In addition, it is well known that tuning chemoselectivity involving unsaturated alkenes and alkynes is always a big challenge in the field of organic synthesis [41][42][43]. For alkyne addition, achieving high regioselectivity is very challenging, mainly due to critical conditions such as high temperature and transition metals. Additionally, most of the reported reactions involving nitrile often proceed in a sequence of distinct steps upon treatment with high temperatures, high pressure or strong inorganic acid, and overhydrolysis is not absolutely controlled. Although much advancement in benzylic C−H functionalization with electrochemistry has been achieved, investigation with an electrochemical method for tuning the electrochemical benzylic C−H functionalization of (thio)xanthenes with terminal alkynes and nitriles has been less explored. Herein, we will describe a rare approach for the direct benzylic C(sp 3 )−H functionalization of (thio)xanthenes that can be well tuned by varying the reaction temperature and reaction medium, leading to the formation of carbon−carbon and carbon−nitrogen bonds in an efficient manner, respectively (Scheme 1f).

Results and Discussion
Initially, we chose xanthene 1a and phenylacetylene 2a as coupling partners for the optimization of reaction conditions, and the results are listed in Table 1. Based on our recent findings regarding electrochemical benzylic C−H functionalization [32][33][34][35], acetonitrile was preferentially selected as a reaction medium for this investigation. With n Bu 4 NPF 6 as an electrolyte, the reaction of 1a with 2a was performed under constant-current conditions (8 mA) in an undivided cell equipped with a carbon anode and a platinum cathode. After proceeding at room temperature under an air atmosphere for 3 h, the product 3a was isolated in 32% yield. Meanwhile, the formation of 4a generated from the reaction of 1a with CH 3 CN was obtained with a 40% yield (entry 1). The ratio of 3a/4a was effectively improved by varying the reaction temperature from 30 to 60 • C (entries 2-4). Remarkably, reaction temperature could control the formation of 3a as a major product, but higher temperature (60-70 • C) led to an inferior yield of 3a (entries 5 and 6). Furthermore, the use of C(+)|Ni(−) or Pt(+)|C(−) still resulted in the mixture of 3a/4a (entries 7 and 8). Additionally, the amount of water extremely affects the formation of 3a (for details, see Table S1 in Supporting Information). Interestingly, we then found that the model reaction with GF(+)|GF(−) as electrode generated 4a as a sole product, albeit with 43% isolated yield (entry 9). Screening of both electrolytes and constant current failed to enhance the yield of 3a (entries 10-13). We clearly observed that the reaction temperature and electrode could affect the formation of 3a and 4a (entries 4, 7-9). Subsequently, with GF(+)|GF(−) as an electrode, 4a was formed at rt to give a 35% yield (entry 14). The use of dry CH 3 CN could enhance the yield of 4a (entries 15-16). Replacing n-Bu 4 NBF 4 with n-Bu 4 NPF 6 did not improve the benzylic C−H amination of xanthene electrolytes (entry 17). Our attempt to improve the yield of 4a by varying the constant current from 5 to 10 mA was unsuccessful (entries [18][19]. The above model reaction did not proceed without an electric current (entry 20). Notably, Faraday efficiency values for 3a and 4a were determined as 45.6% and 28.5%, respectively (for details, see the Supporting Information). but higher temperature (60−70 °C) led to an inferior yield of 3a (entries 5 and 6). Furthermore, the use of C(+)|Ni(−) or Pt(+)|C(−) still resulted in the mixture of 3a/4a (entries 7 and 8). Additionally, the amount of water extremely affects the formation of 3a (for details, see Table S1 in Supporting Information). Interestingly, we then found that the model reaction with GF(+)|GF(−) as electrode generated 4a as a sole product, albeit with 43% isolated yield (entry 9). Screening of both electrolytes and constant current failed to enhance the yield of 3a (entries 10-13). We clearly observed that the reaction temperature and electrode could affect the formation of 3a and 4a (entries 4, 7-9). Subsequently, with GF(+)|GF(−) as an electrode, 4a was formed at rt to give a 35% yield (entry 14). The use of dry CH3CN could enhance the yield of 4a (entries 15-16). Replacing n-Bu4NBF4 with n-Bu4NPF6 did not improve the benzylic C−H amination of xanthene electrolytes (entry 17). Our attempt to improve the yield of 4a by varying the constant current from 5 to 10 mA was unsuccessful (entries 18-19). The above model reaction did not proceed without an electric current (entry 20). Notably, Faraday efficiency values for 3a and 4a were determined as 45.6% and 28.5%, respectively (for details, see the Supporting Information). With the optimized reaction conditions in hand, we next focused on the scope of this tunable benzylic C−H functionalization of xanthenes (Scheme 2). We first examined the electrochemical reaction of the aromatic terminal alkynes with xanthenes under the optimized conditions described in entry 4 of Table 1. Several typical terminal alkynes were used to react with xanthene 1a under standard conditions and produced the crosscoupling products in good yields. On the terminal alkyne substrates listed in Scheme 2, the ones having electron-rich groups, such as Me, n-amyl and 4-Et-Phenyl, incorporated at para-position reacted with 1a to generate the asymmetric A-alkylated aryl ketones 3b-d in 63-75% isolated yields. However, the use of a methyl group at the ortho-, meta-position in alkynes led to decreasing yields of the target product 3e and 3f at 53% and 68%, respectively. The results indicate that there might be a detrimental steric effect in this electrochemical benzylic C−H functionalization. We found that this reaction was also applicable to 1ethynylnaphthalene and produced 3g in 61% yield. Several typical substituted xanthenes were then examined under the above system, showing that yield of the product was insensitive to the position of the alkyl group on the aromatic ring (3h-j). Unfortunately, it was found that 9H-thioxanthene and 9,10-dihydroacridines did not react with 2a under the standard reaction conditions.  Table 1. Several typical terminal alkynes were used to react with xanthene 1a under standard conditions and produced the cross-coupling products in good yields. On the terminal alkyne substrates listed in Scheme 2, the ones having electron-rich groups, such as Me, n-amyl and 4-Et-Phenyl, incorporated at para-position reacted with 1a to generate the asymmetric ɑ-alkylated aryl ketones 3b-d in 63−75% isolated yields. However, the use of a methyl group at the ortho-, meta-position in alkynes led to decreasing yields of the target product 3e and 3f at 53% and 68%, respectively. The results indicate that there might be a detrimental steric effect in this electrochemical benzylic C−H functionalization. We found that this reaction was also applicable to 1-ethynylnaphthalene and produced 3g in 61% yield. Several typical substituted xanthenes were then examined under the above system, showing that yield of the product was insensitive to the position of the alkyl group on the aromatic ring (3h-j). Unfortunately, it was found that 9H-thioxanthene and 9,10-dihydroacridines did not react with 2a under the standard reaction conditions. We next explored the Ritter-type amination of xanthenes with nitriles under the optimal electrochemical conditions described in Table 1, and the results are listed in Scheme 3. With acetonitrile as a solvent and substrate, a variety of 2-substituted xanthenes including Me, Et, Ph and CF 3 were all viable substrates and afforded amination products in 56-75% yields (4b-e). This result indicates that the introduction of an electron-withdrawing group does not favor this electrochemical process. Then, we examined the xanthene derivatives having Me or Ph substituents at the 4-position and found that the reaction proceeded smoothly to form the corresponding products in acceptable yields (4f and 4g). Next, the reactions with dimethylsubstituted xanthenes generated the corresponding products in 67-72% yields (4h-j). Similarly, the reaction with asymmetric xanthenes having a naphthyl group also proceeded smoothly, affording the amination products in moderate yields (4k-m). Additionally, the common solvent n-butyronitrile was employed as a substrate to react with 1a under optimal conditions, generating the compound 4n in 50% yield. Finally, in the case of thioxanthene, the desired product 4o was obtained in 64% yield.
We next explored the Ritter-type amination of xanthenes with nitriles under the optimal electrochemical conditions described in Table 1, and the results are listed in Scheme 3. With acetonitrile as a solvent and substrate, a variety of 2-substituted xanthenes including Me, Et, Ph and CF3 were all viable substrates and afforded amination products in 56−75% yields (4b-e). This result indicates that the introduction of an electron-withdrawing group does not favor this electrochemical process. Then, we examined the xanthene derivatives having Me or Ph substituents at the 4-position and found that the reaction proceeded smoothly to form the corresponding products in acceptable yields (4f and 4g). Next, the reactions with dimethylsubstituted xanthenes generated the corresponding products in 67−72% yields (4h-j). Similarly, the reaction with asymmetric xanthenes having a naphthyl group also proceeded smoothly, affording the amination products in moderate yields (4k-m). Additionally, the common solvent n-butyronitrile was employed as a substrate to react with 1a under optimal conditions, generating the compound 4n in 50% yield. Finally, in the case of thioxanthene, the desired product 4o was obtained in 64% yield. To gain insight into the reaction mechanism, some experiments were specially conducted under given conditions. We observed that the oxidation potential of xanthene 2a (first peak at Ep 1.74 V) is lower than that of 1a (Ep 2.23 V) with the method of cyclic voltammograms (CVs), as shown in Figure 2, indicating that preferential oxidation of 2a occurred and enabled the following electrophilic addition to alkyne (Scheme 4) [35,44]. Then, an experiment involving the mixture of 1a/1a-D 2 and 2a revealed a KIE value showing that cleavage of benzylic C(sp 3 )−H of xanthene may not be involved in the rate-determining step (Scheme 4a). Importantly, replacing 2a with halogenated alkynes 6a-c as substrates did not yield any product, thus showing the critical role of terminal alkynes (Scheme 4b).
To gain insight into the reaction mechanism, some experiments were specially con ducted under given conditions. We observed that the oxidation potential of xanthene 2 (first peak at Ep 1.74 V) is lower than that of 1a (Ep 2.23 V) with the method of cycli voltammograms (CVs), as shown in Figure 2, indicating that preferential oxidation of 2 occurred and enabled the following electrophilic addition to alkyne (Scheme 4) [35,44 Then, an experiment involving the mixture of 1a/1a-D2 and 2a revealed a KIE value show ing that cleavage of benzylic C(sp 3 )−H of xanthene may not be involved in the rate-deter mining step (Scheme 4a). Importantly, replacing 2a with halogenated alkynes 6a-c as sub strates did not yield any product, thus showing the critical role of terminal alkyne (Scheme 4b).  According to the experiment results and previous reports, Refs. [32][33][34][35][36][37][38][39][40], a possibl mechanism for the direct electrochemical reaction of xanthenes with terminal alkynes i proposed (Scheme 5). Initially, xanthene 1a was oxidized into I followed by losing a pro ton to yield II. Then, intermediate II underwent anodic oxidation to produce a key cati onic intermediate III. On one hand, III could be attacked by terminal alkyne 1a to generat To gain insight into the reaction mechanism, some experiments were specially con ducted under given conditions. We observed that the oxidation potential of xanthene 2 (first peak at Ep 1.74 V) is lower than that of 1a (Ep 2.23 V) with the method of cycli voltammograms (CVs), as shown in Figure 2, indicating that preferential oxidation of 2 occurred and enabled the following electrophilic addition to alkyne (Scheme 4) [35,44 Then, an experiment involving the mixture of 1a/1a-D2 and 2a revealed a KIE value show ing that cleavage of benzylic C(sp 3 )−H of xanthene may not be involved in the rate-deter mining step (Scheme 4a). Importantly, replacing 2a with halogenated alkynes 6a-c as sub strates did not yield any product, thus showing the critical role of terminal alkyne (Scheme 4b).  According to the experiment results and previous reports, Refs. [32][33][34][35][36][37][38][39][40], a possibl mechanism for the direct electrochemical reaction of xanthenes with terminal alkynes i proposed (Scheme 5). Initially, xanthene 1a was oxidized into I followed by losing a pro ton to yield II. Then, intermediate II underwent anodic oxidation to produce a key cat onic intermediate III. On one hand, III could be attacked by terminal alkyne 1a to generat According to the experiment results and previous reports, Refs. [32][33][34][35][36][37][38][39][40], a possible mechanism for the direct electrochemical reaction of xanthenes with terminal alkynes is proposed (Scheme 5). Initially, xanthene 1a was oxidized into I followed by losing a proton to yield II. Then IV, and IV would be trapped by H2O to give the V. Subsequently, V underwent deprotonation and isomerization to afford 2a (Path a) [35,44]. On the other hand, intermediate III would be trapped by an acetonitrile molecule (Ritter-type reaction) to generate 3a (Path b) [45].

General Considerations
NMR spectra were recorded on a Bruker-600 (Bruker, Germany (600 MHz for 1 H; 151 MHz for 13 C). 1 H NMR spectra were referenced relative to internal Si(Me)4 (TMS) at δ 0.00 ppm or CDCl3 at δ 7.26 ppm. 13 C NMR spectra were recorded at ambient temperature on Bruker-600 (151 MHz) spectrometers and are referenced relative to CDCl3 at δ 77.16 ppm. Data for 1 H, 13 C NMR are recorded as follows: chemical shift (δ, ppm), multiplicity (s = singlet, d = doublet, t = triplet, m = multiplet, q = quartet, quint = quintet, br = broad), integration and coupling constant (Hz). High-resolution mass spectra were recorded on a P-SIMS-Gly produced by Bruker Daltonics Inc. (Bruker, Germany) using electrospray-ionization time of flight (ESI-TOF) and an Agilent Technologies 7250 GCQTOF using EI-TOF (Agilent Technologies, CA, USA). n-Bu4NBF4, phenylacetylene and CH3CN were purchased from Energy Chemical Company and Taitan Chemical Company in China. Other substituted xanthenes and thioxanthenes were synthesized according to the known methods [32,34].

Typical Procedure for the Synthesis of 3a
To an undivided cell (10 mL columnar round-bottom flask with a 24# mouth) fitted with a carbon rod (Φ 6 mm) anode and a platinum cathode (10 mm×10 mm×0.3 mm), the solid reagents xanthene (0.45 mmol) and n-Bu4NPF6 (0.36 mmol) were added. Then, the liquid reagents phenylacetylene (0.3 mmol), H2O (1.2 mmol) and CH3CN (5 mL) were added in sequence via syringe. Electrolysis was carried out with a constant current (5 mA) at 50 °C for 5 h. Then, the solvent was evaporated to dryness under reduced pressure, and the residue was purified by column chromatography on silica gel to give product 3a as a white solid (63.8 mg, 71% yield).

General Considerations
NMR spectra were recorded on a Bruker-600 (Bruker, Germany (600 MHz for 1 H; 151 MHz for 13 C). 1 H NMR spectra were referenced relative to internal Si(Me) 4 (TMS) at δ 0.00 ppm or CDCl 3 at δ 7.26 ppm. 13 C NMR spectra were recorded at ambient temperature on Bruker-600 (151 MHz) spectrometers and are referenced relative to CDCl 3 at δ 77.16 ppm. Data for 1 H, 13 C NMR are recorded as follows: chemical shift (δ, ppm), multiplicity (s = singlet, d = doublet, t = triplet, m = multiplet, q = quartet, quint = quintet, br = broad), integration and coupling constant (Hz). High-resolution mass spectra were recorded on a P-SIMS-Gly produced by Bruker Daltonics Inc. (Bruker, Germany) using electrospray-ionization time of flight (ESI-TOF) and an Agilent Technologies 7250 GCQTOF using EI-TOF (Agilent Technologies, CA, USA). n-Bu 4 NBF 4 , phenylacetylene and CH 3 CN were purchased from Energy Chemical Company and Taitan Chemical Company in China. Other substituted xanthenes and thioxanthenes were synthesized according to the known methods [32,34].

Typical Procedure for the Synthesis of 3a
To an undivided cell (10 mL columnar round-bottom flask with a 24# mouth) fitted with a carbon rod (Φ 6 mm) anode and a platinum cathode (10 mm × 10 mm × 0.3 mm), the solid reagents xanthene (0.45 mmol) and n-Bu 4 NPF 6 (0.36 mmol) were added. Then, the liquid reagents phenylacetylene (0.3 mmol), H 2 O (1.2 mmol) and CH 3 CN (5 mL) were added in sequence via syringe. Electrolysis was carried out with a constant current (5 mA) at 50 • C for 5 h. Then, the solvent was evaporated to dryness under reduced pressure, and the residue was purified by column chromatography on silica gel to give product 3a as a white solid (63.8 mg, 71% yield).

Typical Procedure for the Synthesis of 4a
To an undivided cell (10 mL columnar round-bottom flask with a 24# mouth) fitted with a graphite felt anode (10 mm × 10 mm × 3 cm) and a graphite felt cathode (10 mm × 10 mm × 3 cm), the solid reagents xanthene (0.3 mmol) and n-Bu 4 NPF 6 (0.36 mmol) were added. Then, the liquid CH 3 CN (5 mL) was added in sequence via syringe. The electrolysis was carried out with a constant current (5 mA) at room temperature for 5 h. Then, the solvent was evaporated to dryness under reduced pressure, and the residue was purified by column chromatography on silica gel to give product 4a as a white solid (48.7 mg, 68% yield).

Conclusions
In summary, we have developed an electrochemical benzylic C−H functionalization of (thio)xanthenes with terminal alkynes and nitriles in the absence of any catalyst or external chemical oxidant. The method enables selective benzylic C−H bond functionalization by varying electrochemical conditions, providing an efficient approach to synthesizing xanthene derivatives under mild conditions. Efforts in our laboratory are ongoing to explore other challenging inert C−H functionalizations.