Visible-Light-Driven Decarboxylative Coupling of 2H-Indazoles with α-Keto Acids without Photocatalysts and Oxidants

An efficient synthesis of functionalized 3-acyl-2H-indazoles via visible-light-induced self-catalyzed energy transfer was developed. This method utilized a self-catalyzed energy transfer process between 2H-indazoles and α-keto acids, offering advantages like absence of photosensitizers, metal catalysts, and strong oxidants, broad substrate compatibility, and operational simplicity under mild conditions.


■ INTRODUCTION
Nitrogen-containing heterocycles play an essential role in the molecular structures of natural compounds, 1 pharmaceuticals, 2 and various biologically active substances. 32H-indazole derivatives are known for their diverse range of biological activities, including antitumor, 4 antibacterial, 5 anti-inflammatory, 6 and anticancer 7 properties.This makes them valuable components in the development of pharmaceuticals and other biologically active compounds.Thus, the pursuit of environmentally friendly and highly efficient methods for synthesizing functionalized 2H-indazoles has become a focal point in the field of synthetic methodology. 8n particular, C3-acylated 2H-indazoles have been used as polymerase inhibitors 9 and demonstrated notable value in pharmacology. 10While several methods for the synthesis of 3acyl-2H-indazoles have been developed, these methods often involve the use of transition metals, environmentally unfriendly stoichiometric oxidants, or high temperature conditions. 11hese conditions lead to the generation of chemical wastes and significantly raise the overall manufacturing cost.For example, Liu and co-workers introduced a Ag-catalyzed [3 + 2]  cycloaddition for the synthesis of 3-acyl-2H-indazoles utilizing benzynes and diazocarbonyl compounds (Figure 1Aa). 12Kim and co-workers reported a Rh(III)-catalyzed C−H functionalization of azobenzene with keto aldehydes (Figure 1Ab). 13ou and co-workers developed a tandem Rh(III)-catalyzed C− H alkylation/intramolecular decarboxylative cyclization method using azoxy compounds and diazoesters (Figure 1Ac). 14In addition to the above-mentioned cycloaddition methods, alternative approaches to the synthesis of 3-acyl-2H-indazoles involve the direct addition of acyl radicals to 2H-indazoles.In these cases, acyl radicals could be generated from various acyl precursors, such as α-keto acids 15 and aldehydes. 16For example, in the presence of strong oxidants (e.g., Na 2 S 2 O 8 and TBHP), silver nitrate 17 and NiCl 2 18 have been employed for the C-3 substituted acyl 2H-indazoles using α-keto acids and aldehydes, respectively (Figure 1Bd,Be).Recently, a metal free method was developed by using aldehyde and di-tertbutylperoxide (DTBP) at a high temperature of 120 °C (Figure 1Bf). 19To the best of our knowledge, a mild C-3 acylation of 2H-indazoles at room temperature without the use of transition metals, oxidants, and photosensitizers has never been reported.
Building on our continued efforts to develop environmentally friendly and efficient methods for the synthesis of functionalized N-heterocycles, 20 herein, we reported a visiblelight-induced self-catalyzed decarboxylative coupling reaction of 2H-indazoles and α-keto acids (Figure 1C).Notably, this method did not require the use of photocatalysts or oxidants under mild conditions.
Furthermore, the influence of the ratio between 1a and 2a on the reaction was explored.As demonstrated in Table 1 (entries 5 −7), increasing the ratio between 1a and 2a from 1:2 to 1:5 led to an enhanced yield of 3aa from 43 to 62%.However, when the ratio was further increased to 1:6 (Table 1, entry 8), there was no further improvement in the yield of 3aa.Therefore, the optimal ratio between 1a and 2a was identified as 1:5.It was known that α-keto acids (e.g., 2a) could undergo decarboxylation under light irradiation, leading to the formation of acyl radicals. 21Indeed, during the reaction optimization, benzil was identified as one of the byproducts (details see Figure 4a), which could be generated from acyl radicals.It was found that the concentration of starting materials could also affect the efficiency of reaction.As indicated in Table 1 (entries 13−15), the amount of solvent utilized in the reaction (2−5 mL) also had an impact on the performance of the reaction, with the optimal solvent amount of 4 mL for 0.2 mmol scale.
It has been reported that the addition of oxidants (e.g., K 2 S 2 O 8 , organic peroxides) and photoinitiators could help the generation of free acyl radicals from 2a, facilitating subsequent free radical coupling reactions. 22Thus, the addition of various additives such as photoinitiators (4CzIPN), oxidants (e.g., K 2 S 2 O 8 , DTBP), bases ( t BuOK, K 2 CO 3 ), and acids (trifluoroacetic acid, triflic acid) were examined under our optimized conditions (Table 1, entries 9−12, 16−17, 26−27).However, no improvement in the yield of product 3aa was found.During the optimization process, different wavelengths of the light source, ranging from 360 to 455 nm, were tested.The results indicated that the wavelength (420−425 nm) was the most effective light source for the reaction (Table 1, entries 18−22).To further improve the yield of 3aa, degassed hexafluoroisopropanol (HFIP) was added as cosolvent.Pleasingly, the solvent mixture of MeCN and HFIP (V MeCN :V HFIP = 3:1) led to the highest yield of 3aa at 71% (Table 1, entries 23−25).This improvement could be attributed to the previously reported role of HFIP as a radical stabilizer. 23Optimizing the reaction conditions revealed that the optimal reaction time is   d Solvent (5.0 mL).e Reaction time was 6 h.f Reaction time was 16 h.g Reaction conditions: 1a (0.2 mmol), 2a (amount as specified), solvent as specified (2.0 mL), room temperature, nitrogen atmosphere, 10 h. 10 h, as indicated in Table 1 (Table 1, entries 24, 28, and 29).Interestingly, extending the reaction time beyond 10 to 16 h did not result in an improvement in the yield of 3aa.
With the optimized reaction condition in hand, the applicability of the synthetic approach to a range of functionalized 2H-indazoles was explored (Figure 2).The functional group tolerance of 2H-indazoles at C-5 and C-6 position was also examined.The introduction of both halogen atoms (−F, −Cl, −Br) (3ba−3da) or electron-donating groups (−OMe, −Me, −OMeO−) (3ea−3ha) at the C-5 and C-6 position of 2H-indazoles resulted in the target products with the yields ranging from 48 to 64%.The synthesis of product 3ia showcased the feasibility of conducting the synthesis using 2H-indazoles containing multiple functional groups on various sites.Various indazoles bearing aromatic rings that were functionalized with diverse functional groups (e.g., −Me, −OMe, −SMe, −F, −Cl, −Br, −CF 3 , −COOEt, −phenyl) along with heteroaryl substituents (e.g., pyridyl) at the 2N position were effectively converted into their corresponding 3-acyl-2H-indazoles in fair to good yields (3ja−3za, 33−78%).It was noted that substrates with orthomethyl substituted phenyl groups at 2N position led to lower yields of the products (e.g., 3qa) compared to the products with meta-and para-substituted phenyl groups (e.g., 3oa, 3pa).This suggested that steric hindrance at the ortho position affected the reaction efficiency.Additionally, when aromatic functional group at 2N position was switched to aliphatic substituent, as in the case of 2-(cyclohexyl)-2H-indazole (1w), a poor isolated yield of 3wa was obtained (25%), even after extending the reaction time to 20 h.It was proposed that the broader light absorption range of C3-acylated 2H-indazole products (e.g., 3aa and 3wa) could inhibit the continuous formation of desired product (for the product inhabitation effect and UV−vis spectrum, see Figures S2−S4 in Supporting Information).
After the exploration of the scope of 2H-indazole derivatives, a variety of aryl and hetaryl α-keto acids were synthesized (see Experimental Section).These α-keto acids were then investigated for the photoinitiated decarboxylative radical acylation reaction with 1a.It was found that our reaction system exhibited versatility and accommodated a wide range of α-keto acids, allowing them to smoothly undergo the acylation process and yield the desired acylated 2H-indazole derivatives (3ab−3ak) with moderate to good yields.Aromatic α-keto acids bearing electron-withdrawing groups (e.g., −F, −Cl, −Br, −CF 3 ) at the ortho-and para-positions of the aromatic rings provided the desired products (e.g., 3ab−3af) with yields ranging from 52 to 76%.Moreover, α-keto acids with electrondonating groups (−OMe, −Me, −OMeO−) were also effective acylating reagents, although they resulted in slightly lower yields (40−59%) of acylated 2H-indazoles (3ag−3aj).Pleasingly, a good yield (62%) of 3ak was obtained when naphthyl α-keto acid (2k) was used.Unfortunately, aliphatic αketo acids, such as 2-oxopropanoic acid (2l), could not lead to the desired products 3al using our method.To further establish the scalability of our method, we performed largescale reactions using different quantities of 1a as starting materials, ranging from 3 (0.58 g) to 6 mmol (1.17 g).The isolated yields of 3aa were 64 and 56%, respectively (for details, please refer to the SI).
To gain preliminary insight into the reaction mechanism, a series of control experiments were conducted.When the free radical inhibitor TEMPO (2,2,6,6-tetramethylpiperidinooxyl) was introduced (Figure 3a), the transformation was completely suppressed.The HRMS analysis of the reaction mixture confirmed the formation of acyl radical during the photoreaction, as evidenced by the detection of compound 4, an adduct of TEMPO and acyl radical ([M + H] + found 262.1800, calculated 262.1802, detailed HRMS see Figure S1).This finding strongly suggested that the reaction likely proceeded through a free radical pathway.
As depicted in Figure 3b, when the light source was removed from the reaction, product 3aa was not detected, confirming that the reaction was indeed light driven.Interestingly, when the reaction was conducted in an air atmosphere, the reaction was quenched and did not proceed.This observation diverged from a method developed by He and co-workers, 24 where oxygen played a crucial role in visible-light-induced aerobic acylation of quinoxalin-2(1H)-ones with α-keto acids.In this process, an acyl radical and byproduct H 2 O 2 were generated through a hydrogen-atom-transfer (HAT) process via excitedstate singlet oxygen ( 1 O 2 ).Jin and co-workers reported the generation of hydrogen gas during the synthesis of The Journal of Organic Chemistry quinazolinone derivative under photoinduced conditions using α-keto acids. 25To investigate the possibility of formation of hydrogen gas from the H• radical, the gas mixture from the headspace of the reaction mixture was analyzed by GC with a TCD detector (Figure S7).However, only the formation of CO 2 and minor CO were confirmed, and no hydrogen gas was generated in our system.In addition, we carried out fluorescence quenching experiments involving 1a and 2a.Upon the addition of 2a to the visible light irradiated 1a, the fluorescence intensity of 1a decreased (Figure 3c, see SI for a detailed procedure).The straight Stern−Volmer plot of the quenching experiment in Figure 3c indicated that the excited state of 1a could be quenched by 2a.Further fluorescence quenching study demonstrated that no energy transfer between 3aa and 2a (Figure S6 in Supporting Information).Moreover, GC-MS analysis of reaction mixture at 10 h (reaction condition as listed in entry 24, Table 1) revealed the generation of byproducts such as benzaldehyde, benzoic acid and benzil during the reaction (Figure 4a).−26 Based on the above obtained results and relevant literature studies, 26,27,28 a plausible reaction mechanism was proposed (Figure 4b).Initially, 1a absorbs visible light, transitioning to an exciting state (1a*).Subsequently, the ground state 2a undergoes energy transfer from 1a*, leading to the excited state 2a*, which could homolyze to form an acyl radical (A) and carboxyl radical (B). 26,27he acyl radical (A) could then form benzil (C) or further decompose to CO and phenyl radical (Ph•), with the later react with B to form benzoic acid (D).In the meantime, B could react with another B to form CO 2 and formic acid (E). 26he generated acyl radical A could attack the C-3 position of 1a, forming radical intermediate F.

■ CONCLUSIONS
In conclusion, we have presented a novel and efficient method for visible-light-induced synthesis of functionalized 3-acyl-2Hindazoles.Notably, the utilization of a self-catalyzed energy transfer process between 2H-indazoles and α-keto acids offered this method several advantages, induced organic synthesis without photocatalyst conditions.This method demonstrated a broad substrate scope and operational simplicity under mild conditions.Further investigation on visible-light-induced organic synthesis under without photocatalyst conditions is currently underway in our laboratory.
■ EXPERIMENTAL SECTION General Information. 1 H NMR spectra and 13 C NMR spectra were recorded on a Bruker AVANCE NEO (400 MHz/500 MHz/ 600 MHz) spectrometer and Bruker AVANCE NEO (100 MHz/125 MHz/150 MHz) spectrometer at 25 °C, respectively.High resolution mass spectra (HRMS) were measured with an Agilent 6230 TOF instrument.
Safety Statement.Caution!Aldehydes and anilines have irritating odors, and inhalation can cause damage to the body.Therefore, weighing and transferring these chemicals should be carried out within a fume hood.Additionally, exposure to light sources can be harmful to the eyes, necessitating the use of protective goggles.
General Procedure for the Synthesis of Functionalized 3-Acyl-2H-indazoles.In a clean Schlenk flask, 2H-indazoles (0.2 mmol) and α-keto acids (5 equiv., 1.0 mmol) were placed.The Schlenk flask was evacuated and purged with nitrogen three times using a Schlenk line.Subsequently, 3 mL of degassed MeCN and 1 mL of degassed HFIP were added under nitrogen charging conditions, and the flask was tightly sealed.The reaction was conducted under 420−425 nm light for 10−20 h.After completion of the reaction, the solution was concentrated, and flash chromatography was performed using petroleum ether (PE)/ethyl acetate (EA) = 100:1−15:1 as eluent.
2-(p-tolyl)-2H-indazole (1o). 311o was prepared by the general procedure with p-toluidine.White solid; 1  General Procedure for the Synthesis of α-Keto Acids. 38Aryl ketones (32 mmol) and SeO 2 (2 equiv., 7.10 g, 64 mmol) were added to a round-bottom flask under a nitrogen atmosphere.Then, degassed pyridine (30 mL) was added as the solvent.The reaction mixture was heated to 120 °C using an oil bath for 15 h.When reaction was finished, the reaction mixture was filtered and washed with EA, and the filtrate was collected and evaporated using a rotavapor.Saturated aqueous NaOH solution was added to this concentrated reaction mixture to adjust the pH to 12.Then, the aqueous phase was collected and extracted with EA and saturated NaCl solution.After the aqueous phase was collected and adjusted to pH < 2 using 2 M hydrochloric acid, the aqueous phase was extracted with EA.Then, the organic layer was collected, dried with anhydrous NaSO 4 , and filtered before it was concentrated and purified by flash chromatography (EA and PE).

a
Yield of 3aa was determined by GC using docosane as internal standard.N.R.: no reaction.b Solvent (3.0 mL).c Solvent (4.0 mL).
Finally, the extraction of H• from intermediate F took place with radical R• (e.g., A, B, or other radicals generated during the decomposition of α-keto acids), leading to the formation of the final target product 3aa.

Figure 4 .
Figure 4. Organic intermediates determined by GC-MS and the proposed mechanism.