Introduction

The development of mild and practical strategies to produce high-value pharmaceuticals and fine chemicals from inexpensive and readily available commodity chemicals is a central goal in organic synthesis1,2. DCM, a fundamental chemical in the chemical industry and laboratory organic synthesis, is commonly used as an organic solvent to mediate organic reactions, purification processes, or sample analysis, however, is rarely applied as a C1 building block in organic reactions in the past decades (Scheme 1a)3,4,5,6,7,8. This is because that compared with alkyl bromides [C(sp3)-Br = 285 kJmol-1] and alkyl iodides [C(sp3)-I = 213 kJmol−1], the activation of C(sp3)-Cl bond requires high energy (327 kJmol−1)9,10. Hence, the development of novel organic transformations to synthesize value-added chemicals from stable and abundant feedstock chemical DCM is highly desired. Indeed, such transformations would potentially open up new vistas in organic synthesis.

Scheme 1: Chemoselective multicomponent synthesis of 1,4,2-dioxazoles from DCM.
scheme 1

aSynthesis of value-added chemicals from DCM. bRepresentative biologically active 1,4,2-dioxazoles. cMethods for the synthesis of 1,4,2-dioxazoles. dThis work: DCM as both a C1 source and solvent.

Full size image

1,4,2-Dioxazoles widely existed in bioactive molecules such as I (antiamoebic activity)11 and II (modest central nervous system depressant activity)12 (Scheme 1b). Furthermore, 1,4,2-dioxazoles were also important building blocks which could be utilized as nitrene transfer reagents13, nitrene precursors14, amidating reagents15, and protecting agents16. Recently, some methods to access 1,4,2-dioxazoles have been reported (Scheme 1c). Velo and Polat-Cakir achieved the synthesis of 1,4,2-dioxazoles via the 1,3-dipolar cycloaddition reaction of nitrile oxides with acyl derivatives17,18,19. 1,4,2-Dioxazoles were also prepared by the reaction of acyl azides with ketones reported by Abraham20. Besides, 1,4,2-dioxazoles synthesis could be accomplished via the addition of benzhydroxamic acids to alkynes, diethylketals or vinyl aryl ethers12,21,22,23,24,25. Despite these advances, these developed methods suffered from inconveniently available starting materials, limited substrate scope and harsh reaction conditions such as high temperatures or noble metal catalysts. Furthermore, some of them involved multistep reactions and required tedious operation and complex purification processes. The aforementioned aspects would significantly thwart their wide practical application. Given the impressive pharmacological and synthetic characters of 1,4,2-dioxazoles, and in continuation of our efforts in developing novel organic transformations26,27, we wondered if 1,4,2-dioxazoles could be synthesized from readily available feedstock chemical DCM as the C1 source in a simple, economical and metal-free fashion with a broader substrate scope (Scheme 1d). One of the key challenges in this multicomponent reaction is chemoselective control since several side reactions can occur to produce undesired products, such as those from the direct nucleophilic attack on DCM by different nucleophiles28,29. The success of this approach would provide not only a new platform for DCM application, but also an excellent complementary strategy to the established 1,4,2-dioxazoles synthesis.

Results and discussion

To realize this transformation, we initially chose 1,3-dioxoisoindolin-2-yl cyclohexanecarboxylate 1a, DCM, and various nucleophiles as starting materials (Table 1). Unfortunately, the multicomponent reaction did not occur when using aniline as the nucleophile in the presence of DBU. In addition, the sole side product 3′ was obtained in 41 and 72% yields employing 4-methylbenzenethiol and phenylmethanethiol, respectively. Excitingly, 2-methoxyphenol as the nucleophile exclusively led to the formation of the desired product 1,4,2-dioxazole 3 in 63% yield (Table 1, entry 1). The possible side product 3′′ was not observed in the process. Inspired by the above results, we further screened various additives and found that DBU was superior. Surprisingly, neither organic bases (entries 3–7) nor inorganic bases (entries 9 and 10) could yield 1,4,2-dioxazole 3. The reaction using Et3N or 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD) as the additive gave 3 in poor yield (entries 2 and 8). It was worth noting that the yield of 3 was improved to 72% when 2.5 equivalents of 2 and DBU was used (entry 11). Control experiments indicated that DBU played a vital role in this reaction (entry 12, for more optimization details, see Supplementary Tables S1–S5).

Table 1 Optimization of reaction conditionsaView full size image
Full size table

With the optimal reaction conditions established, the generality of this method was then explored (Scheme 2). As shown in Scheme 2a, a range of esters, including those bearing halides (4-6), methyl (7 and 9), or methoxyl (8) groups at different positions (ortho or meta) of phenyl rings and a naphthalene ring (10), reacted smoothly to yield the desired 1,4,2-dioxazoles in moderate to good yields with excellent chemoselectivity. The structure of 3 was further confirmed by X-ray crystallography (Supplementary Data 1). Notably, the aliphatic (11) and heterocyclic (12) succinimide derivatives were also compatible with the optimized conditions to afford 1,4,2-dioxazoles in moderate yields. Next, we investigated various nucleophiles. Phenols with diverse functional groups including methoxyl (14 and 15), halides (16-19), and some potentially reactive groups such as nitrile (20) and terminal alkene (21) worked well to exclusively furnish the desired products in 61-80% yields. Moreover, 1,4,2-dioxazoles 22 and 23 could be successfully obtained with 5,6,7,8-tetrahydronaphthalen-2-ol and naphthol. The reactions using heterocyclic phenols also gave the corresponding products 24 and 25. In addition, acyclic and cyclic alcohols with different functional groups as the nucleophiles were well tolerated to exclusively afford 1,4,2-dioxazoles 26-41 in 48-83% yields. More importantly, the multicomponent reaction could smoothly proceed with two types of simple commodity chemicals DCM and methanol or ethanol to synthesize value-added 1,4,2-dioxazoles 36 and 37. Subsequently, we turned our attention to assess amines as nucleophiles and found that secondary aliphatic amines, including sulfamides, exclusively led to the formation of 1,4,2-dioxazoles 42-47. When a series of thiols were subjected to the standard conditions, the desired 1,4,2-dioxazoles 48-53 were obtained in good to high yields. Noteworthy is that this transformation has excellent chemoselectivity and the potential side products 3′ was not observed. Pharmaceutically important motifs, including 1H-indazoles (54-56), 1H-pyrazoles (57-59) and 2H-indazole (60), were also successfully incorporated into the 1,4,2-dioxazole skeleton. The structure of 54 was also confirmed by X-ray crystallography (Supplementary Data 2). Unfortunately, the reaction of 1a with other chloralkanes, including 1,2-dichloroethane, 1,1-dichloroethane, chloroform and tetrachloromethane, did not produce the desired product. The pharmacokinetic properties of small-molecule drugs can be positively influenced by deuterium incorporation, resulting in improved safety, tolerability or efficacy30. As such, a variety of deuterated 1,4,2-dioxazoles 61-65 were smoothly synthesized by using readily available CD3OD and CD2Cl2. This newly developed approach using DCM as a C1 source could offer a facile and inexpensive means for late-stage deuterium incorporation into pharmaceutically active compounds.

Scheme 2: Substrate scope.
scheme 2

aReaction conditions: 1 (0.2 mmol), 2 (0.5 mmol) and DBU (0.5 mmol) in DCM (2 mL) at 60 °C for 24 h under argon. Isolated yields are reported. b4, 5: r.r. = 1:1.3; 6: r.r. = 1:1; 7, 8: r.r. = 1:1.1; cAt 80 °C in DCM (1 mL) and CH3CN (3 mL). dAt 80 °C in DCM (1 mL) and dioxane (3 mL).

Full size image

To demonstrate the robustness of this novel method, the late-stage functionalization of biologically active molecules, such as estrone (66), 7-hydroxyflavone (67), (R)-spirodichol (68), diacetone-D-galactose (69), paracetamol (70), thymol (71), and paeonol (72), was evaluated under the optimized conditions and the desired products were obtained in 50-72% yields with excellent chemoselectivity (Scheme 3a). Furthermore, a large-scale reaction was conducted to yield 3 in 66% yield (Scheme 3b). 1,4,2-Dioxazoles were also versatile synthetic blocks, and their synthetic transformation was explored (Scheme 3c). A new ring-opening reaction of 13 by PCl5 smoothly occurs to form phenyl 2-cyanobenzoate 73 in 71% yield. 13 could also be reduced by NaBH4 to afford 74 in 75% yield. In addition, the yield of 36 was significantly improved by the transesterification of 13 with KOMe. Upon treatment of 13 with NaBH4 and PCl5, 75 was generated in a 71% yield, and more importantly, the Cl atom in 75 could serve as a handle for the construction of more complex molecules. Excitingly, 1,2-phenylenedimethanol (76), serving as the versatile synthetic blocks and ligand was obtained in 69% yield through the reduction of 1331,32.

Scheme 3: Late-stage functionalization and synthetic applications.
scheme 3

aLate-stage functionalization of biologically active molecules. bLarge-scale synthesis of 3. cSynthetic transformation of 13.

Full size image

To better elucidate the reaction mechanism, we carried out a series of control experiments. The yield of 3 was not significantly influenced when the radical scavenger TEMPO or ethene-1,1-diyldibenzene was added to the model reaction (Scheme 4a), indicating that a radical process might not be involved in this reaction. The model reaction after 2.5 h was analysed by HRMS, and the possible intermediate I was detected (Scheme 4b). When I was subjected to the standard conditions, 3 was obtained in 55% yield, which further verified the intermediacy of I (Scheme 4c). Intriguingly, II, a DBU-containing intermediate, was also detected by HRMS in the model reaction (Scheme 4b). DBU has been demonstrated to serve as a superior nucleophilic catalyst in Baylis-Hillman reactions33 and for the synthesis of methyl esters34 and amides35,36. In order to further verify the role of DBU in the multicomponent reaction, the reaction of DBU with DCM was performed and the cationic part of III was detected by HRMS. Furthermore, when I was added to the reaction mixture of DBU and DCM, we observed the cationic part of II and III by HRMS (Scheme 4c). The aforementioned experimental results showed that DBU might serve as a nucleophilic catalyst in 1,4,2-dioxazoles synthesis using DCM as a C1 source.

Scheme 4: Mechanistic studies.
scheme 4

aRadical inhibition experiment. bActive species trapping experiment. cControl experiment.

Full size image

On the basis of experimental results and previous reports33,34,35,36, we proposed a plausible reaction mechanism in which DBU functioned as not only a proton scavenger, but also a nucleophilic catalyst (Scheme 5). First, the reaction of 1a with the nucleophile gave the intermediate I and the acyl derivative V in the presence of DBU. V has been isolated in >95% yield when using 2 as the nucleophile. The nucleophilic attack on DCM by DBU generated DBU adduct IV. With a quaternary ammonium ion and a chlorine atom attached, IV is extremely electrophilic, making the intermediate much more reactive than DCM37. Then, the disubstituted intermediate II was formed via the reaction of IV with I. Finally, the nucleophilic attack-induced skeletal editing of II was achieved to yield the target product 1,4,2-dioxazoles.

Scheme 5: Proposed reaction mechanism.
scheme 5

DBU functioned as not only a proton scavenger, but also a nucleophilic catalyst.

Full size image

In conclusion, we have developed a novel multicomponent strategy for the chemoselective synthesis of valuable 1,4,2-dioxazoles by using stable and abundant feedstock chemical DCM as a C1 source. Notably, the easily-occurring side reactions are successfully inhibited by tuning the type of nucleophiles. The reaction operates under mild and simple conditions, exhibiting excellent chemoselectivity, metal free, and broad substrate scope covering different types of nucleophiles. Furthermore, the potential of this method is further demonstrated by the preparation of deuterated 1,4,2-dioxazoles, the late-stage functionalization of biologically active molecules, and large-scale synthesis. Preliminary mechanistic studies suggest the dual roles of DBU as both a proton scavenger and a nucleophilic catalyst. Further endeavors toward the development of novel strategies for expanding the application of DCM as a C1 source in organic synthesis are ongoing in our laboratory.

Methods

Reagents were used as received from commercial suppliers without further purification, unless otherwise stated. TLC was carried out on SiO2 (silica gel 60 F254, Merck). Flash chromatography was carried out on SiO2 (type: specifications 3, 300– mesh). NMR spectra were recorded for 1H NMR (400 and 600 MHz), 13C NMR (100 and 150 MHz) and 19F NMR (565 MHz) using TMS as an internal standard and Bruker AV 400 and 600 as an instrument. The following abbreviations were used to describe peak patterns where appropriate: singlet (s), doublet (d), triplet (t), and multiplet (m). High-resolution mass spectroscopy (HRMS) were obtained using Bruker Apex IV RTMS. The X-ray diffraction data were collected on an XtaLab PRO MM007HF-DW diffractometer.

Procedure for the synthesis of 1,4,2-dioxazoles

To a 10 mL Schlenk tube equipped with a stir bar, starting material 1 (0.2 mmol, 1.0 equiv.) was added. The Schlenk tube was purged with argon, and then CH2Cl2 (2.0 mL), 2 (55.0 µL, 0.5 mmol, 2.5 equiv.) and DBU (74.7 µL, 0.5 mmol, 2.5 equiv.) were added. The reaction mixture was stirred at 60  °C for 24 h. Subsequently, the mixture was evaporated under reduced pressure and the crude residue was purified by column chromatography on silica gel to afford the desired 1,4,2-dioxazoles.