Jiefeng
Hu
*a,
Xianyu
Yang
a,
Shasha
Shi
a,
Bo
Cheng
a,
Xiaoling
Luo
*c,
Yu
Lan
*d and
Teck-Peng
Loh
*ab
aInstitute of Advanced Synthesis, School of Chemistry and Molecular Engineering, Jiangsu National Synergetic Innovation Center for Advanced Materials, Nanjing Tech University, Nanjing 211816, China. E-mail: iasjfhu@njtech.edu.cn
bDivision of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore 637371, Singapore
cChongqing Key Laboratory of Inorganic Functional Materials, College of Chemistry, Chongqing Normal University, Chongqing 401331, China
dCollege of Chemistry, Institute of Green Catalysis, Zhengzhou University, Zhengzhou, Henan 450001, China
First published on 22nd January 2021
With the increasing awareness of sustainable chemistry principles, the development of an efficient and mild strategy for C(sp3)–H bond activation of nitrogen-containing compounds without the utilization of any oxidant and metal is still highly desired and challenging. Herein, we present a metal-free reaction system that enables C–H bond functionalization of aliphatic sulfonamides using DABCO as a promoter under mild conditions, affording a series of α,β-unsaturated imines in good yields with high selectivities. This protocol tolerates a broad range of functionalities and can serve as a powerful synthetic tool for the late-stage modification of complex compounds. More importantly, control experiments and detailed DFT calculations suggest that this process involves [2 + 2] cyclization/ring-cleavage reorganization, which opens up a new platform for the establishment of other related reorganization reactions.
Imines are ubiquitous and valuable intermediates in the synthesis of nitrogen-containing compounds, pharmaceuticals and natural products.16–19 Among many different imines, α,β-unsaturated imines are a unique class of difunctionalized compounds and play key roles in the synthesis of heterocycles and pharmaceuticals.20 Traditionally, these compounds are synthesized through aldol-type condensation of aldehydes, followed by dehydrative condensation with amines.21 Unfortunately, these condensation reactions usually suffer from harsh acidic conditions and high temperatures, and require dehydrating apparatus. On the other hand, some aldehydes, as reaction substrates, are not easily available and unstable, especially due to aerial oxidation. In addition, the conjugated imines obtained through this strategy are very difficult to purify because they are frequently mixed with other isomers. Thus, the widespread application of α,β-unsaturated imines is extremely hindered. Notably, in recent years significant progress has been made in the selective oxidation of amines for the synthesis of imines based on biocatalysts,22–25 complex metal catalysts26–32 or oxidizing reagents33,34. However, their substrates are limited to benzylamines and cyclic amines.35 Actually, the oxidation of amides is more difficult than that of amines because their nitrogen atoms have much lower electron density.36 Moreover, it is very important37 and challenging to directly synthesize unsaturated imines through dehydrogenation of amine derivatives. Hence, we report an efficient metal-free system for the synthesis of α,β-unsaturated imines with high geometrical selectivities through multiple C–H bond functionalization of N-fluoro sulfonamides under mild conditions (Scheme 1B). It is noteworthy that this pathway proceeded via a four-membered ring reorganization process.
Entry | Base | Solvent | Yieldb (%) |
---|---|---|---|
a Reaction conditions: 1a (0.20 mmol), 1b (0.26 mmol), 1.0 equiv. base and 150 mg 4 Å molecular sieves in solvent (1.0 mL), 24 h, at room temperature, under Ar. b Isolated yield. c Using 30 mol% DABCO. d After removing molecular sieves. DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene, DABCO = 1,4-diazabicyclo[2.2.2]octane, DMF = N,N-dimethylformamide, Ts = 4-toluenesulfonyl. | |||
1 | K2CO3 | DMF | 29 |
2 | CsOAc | DMF | 26 |
3 | K3PO4 | DMF | 31 |
4 | KOH | DMF | Trace |
5 | Et3N | DMF | 65 |
6 | DBU | DMF | 53 |
7 | DABCO | DMF | 81 |
8c | DABCO | DMF | 80 |
9 | — | DMF | N |
10d | DABCO | DMF | 72 |
11 | DABCO | MeCN | 58 |
12 | DABCO | THF | 39 |
13 | DABCO | DCE | 38 |
14 | DABCO | Toluene | N |
With this exciting initial result in hand, we proceeded to investigate the scope of the reactions using various N-fluorotosylamides, prepared from the corresponding alkyl amines (Scheme 2). As shown, N-fluorotosylamide substrates with different carbon chains were efficiently transformed into the corresponding α,β-unsaturated imines in moderate to excellent yields (2c–7c). These results revealed that the products were obtained in lower yields with sterically bulky substrates. The products with substituted alkyl ethers, phenolic ethers and silyl ethers were formed in good yields (8c–12c). It should be noted that the reaction was not affected by halo substituents such as F, CF3, Cl and Br on the substrates (13c–19c), highlighting its potential in combination with the subsequent cross-coupling transformations. Under our conditions, ester-containing substrate 20c also worked well. In particular, the substrates 21c–23c, which contained an alkenyl motif, were also found to be compatible with the reaction conditions. Excitingly, naphthalene (24c) and benzofuran (25c) were also shown to exhibit high levels of reactivities. To showcase the utility of the transformation for diversifying natural product amines, we prepared a diverse collection of N-fluorotosylamide derivatives and subjected them to the optimized reaction conditions. For example, our strategy has been applied in the modification of N-fluorotosylamide 26a with two cis-alkene groups from linolenyl alcohol, generating product 26c in 67% yield. The steroid derivative 27a was also readily transformed into the α,β-unsaturated imine 27c in 62% yield. Moreover, the N-fluorotosylamide 28a, which was synthesized from stigmasterol derivatives by a series of synthetic methods, could also be converted into the desired product 28c in 63% yield.
The scope of this transformation was further evaluated using N-fluorotosylamide (1a) in conjunction with a more diverse pool of arylimines (b) as shown in Scheme 3. A wide range of substituted imines that possess electron-neutral, electron-donating, and electron-withdrawing substituents were readily tolerated (29c–51c). Among them, the substrate bearing a CN group afforded the desired product in 67% yield (42c). Furthermore, the presence of oxygen-, sulfur- and nitrogen-containing heterocycles (44c–47c) did not interfere with the reactions. Notably, the structure of one of the products obtained, 46c, was characterized by X-ray crystallographic analysis. We were also pleased to observe that this method can be applied to some more challenging substrates. For instance, a series of π-extended systems could also participate in the transformation to afford the conjugated imines 48c–51c in moderate yields.
We next investigated the potential of α,β-unsaturated imines in the subsequent synthetic transformations. For example, allylamine compounds can be formed in the presence of DIBAL-H at −78 °C. And we also explored the metal-mediated allylation of 1c in THF, obtaining the corresponding product with a yield of 80%. Under the catalysis of FeCl3, the formed imines can undergo a cyclization reaction to provide indenamine derivatives. Interestingly, the synthetic utility of the formed unsaturated imine was further demonstrated through Prins cyclization reactions (Scheme 4).
Several experiments were conducted to provide insight into the potential mechanism of this cross-coupling reaction. (1) When using TEMPO and BHT as free radical inhibitors, this reaction was not completely suppressed. These results imply that the free radical pathway was probably not involved (Scheme 5A). (2) When the reaction was carried out in the absence of 1b under the standard reaction conditions, 52c was formed through the self-coupling of 1a, and no imine 1d was detected (Scheme 5B). Next, the progress of the experiment was monitored by GC-MS; we detected the formation of the corresponding imine 1e with 29a as the substrate (Scheme 5C). Furthermore, control experiment using N-tosylimine 1d instead of the N-fluoro sulfonamides also resulted in the formation of the desired product with Et3N·HF as a promoter (Schemes 5D and E). On the other hand, when using N-chloro sulfonamide 1f instead of the N-fluoro sulfonamides, no desired product was obtained. These observations illustrated that the imine may be a key intermediate and the fluorine atom was a necessary driving force for this C–H functionalization (Scheme 5F). (3) To rule out the possibility of the involvement of the classical aldol condensation reaction, we carried out cross-over experiments by addition of an external aldehyde. The corresponding imine with the aldehyde fragment was not detected (Schemes 5G and H). (4) To gain deeper insights into the mechanism, 30a–35a with different substituents were treated with 1b under our conditions, and the same product 1c was obtained in good yields (Scheme 5I). More excitingly, using 25b as the substrate instead of 1b, 54c was formed in 71% yield in this system (Scheme 5J). These results suggest that the sulfonamide substituent of the product (c) should be derived from the arylimine substrate (b). And a possible rearrangement process is proposed as shown in Scheme 5K.
Scheme 5 Mechanistic studies of the metal-free C–H functionalization strategy. (A–J) control experiments; (K) a possible reaction pathway. |
To further reveal the details of this transformation, we performed a detailed DFT investigation. The calculated free energy profile of the favored path is shown in Scheme 6 (the detailed information on the other paths is given in the ESI†). The reaction starts with a base-assisted HF elimination from N-fluorotosylamide 1avia transition state 55-ts with an energy barrier of 22.9 kcal mol−1, while (E)-imine 1d is generated irreversibly with an exergonic 69.2 kcal mol−1. Then an imine/enamine tautomerization38 takes place via transition state 56-ts with an energy barrier of 18.1 kcal mol−1 in the presence of DABCO to afford a hydrogen bond complex 57. Enamine 57 serves as a nucleophile, which can react with arylimine 1b through an intermolecular nucleophilic attack39via transition state 58-ts to form a new C–C covalent bond. Interestingly, the nucleophile 57 is activated by the formation of a hydrogen bond with DABCO as well as the electrophilic partner imine 1b is activated by another hydrogen bond formation with HF. Concomitantly, the generated zwitterionic intermediate 59 is stabilized by two hydrogen bonds with DABCO and HF, respectively. The overall activation free energy for this step is 26.5 kcal mol−1, which would be considered as the rate limiting step in the whole pathway. When zwitterionic intermediate 59 is formed, an intramolecular nucleophilic addition can occur via transition state 60-ts with an energy barrier of 13.4 kcal mol−1 to achieve the stepwise [2 + 2] cycloaddition. After the release of DABCO and HF, an azetidine intermediate 62 is generated reversibly. Intermediate 62 also exhibits amine characteristics, and therefore, a Brønsted acid assisted deamination can occur via transition state 64-ts with an energy barrier of 20.4 kcal mol−1 in the presence of HF. The generated iminium intermediate 65 can be further deprotonated by DABCO via transition state 66-ts to form the corresponding enamine intermediate 67. The ring strain of that intermediate results in a 4π-electrocyclic ring-opening process via transition state 68-ts to afford product 1c. To summarise, the ring-closing/ring-opening rearrangement of N-fluorotosylamide with aldimines is constituted by base-assisted HF elimination to an imine, base-assisted imine/enamine tautomerization, step-wise [2 + 2] cyclization to generate azetidine, Brønsted acid prompted deamination of the aminal moiety, and deprotonated ring-opening.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2025337. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/d0sc06603f |
This journal is © The Royal Society of Chemistry 2021 |