English

• Amide functional groups are significant structural motifs that are highly prevalent in biologically active compounds and pharmaceuticals [1-5]. One of the most important properties of amides is that the C═O and N—H groups allow to form hydrogen bonding interaction, such as with proteins, to increase the binding specificity [6]. On the other hand, when alkyl or aryl groups are introduced at the α-position of acetamide, these hydrophobic groups are likely crucial elements in determining the binding affinity [5]. Some approved α-alkyl or α-aryl amide drugs are representatively shown in Scheme 1, such as zofenopril [7], levomilnacipran [8], netupitant [9], disopyramide [10], which contain α-alkyl or α-aryl amide groups as a part of pharmacophoric or auxophoric molecular fragments (Scheme 1). Besides these approved drugs, there are still emerging numerous α-substituted amides in drug discovery. Although amide synthesis is well-developed in the past decades and becomes one of the most frequently utilized organic reactions in the pharmaceutical sector [11-14], yet the amide formation avoiding poor atom economy reagents was voted as the top challenge for green chemistry by the American Chemical Society Green Chemistry Institute (ACSGCI) in 2007 (Scheme 2A) [15]. Therefore, given the significance of the α-substituted amide structure, developing facile, highly effective (to bring the product with both high yield and purity), and scalable approaches with high atom economies and less waste generation toward synthesis of α-alkyl or α-substituted amides on demand is synthetically useful and desired [16,17].

  Scheme 1

  

  Scheme 1.  α-Alkyl and α-aryl amide drugs.

  DownLoad: Full-Size Img PowerPoint

  Scheme 2

  

  Scheme 2.  (A) Top challenge for green chemistry voted by ACSGCI. (B) Decomposition of ketene under irradiation. (C) Fast Wolff-rearrangement in flow toward synthesis of α-substituted amides.

  DownLoad: Full-Size Img PowerPoint

  Wolff rearrangement has been developed for over a century during which it has triggered an avalanche of applications, and it is still one of the most important transformations hitherto in organic synthesis [18-24]. α-Diazoketones involved Wolff rearrangement are featured with 1,2-shift of an alkyl or an aryl group to generate disubstituted ketenes. Due to their electrophilic nature, they are highly reactive acylating reagents to undergo addition reactions with a wide range of nucleophiles, including amines, to give α-substituted amides or other carboxylic acid derivatives [25]. Furthermore, with the use of asymmetrically substituted diazoketones, a chiral center is able to be created in the α-substituted carboxylic acid derivative products via the nucleophilic substitution to the prochiral ketenes [26,27].

  Typically, Wolff rearrangements are conducted under thermal [28-33], metal-catalyzed [34-41], or photochemical conditions [42-51]. The photochemical Wolff rearrangement, which can be carried out at ambient temperatures without any other additives, thus is relatively milder, greener, less costly and more atom-economic than that under any other conditions. Nevertheless, photochemical reactions commonly suffer from a limited penetration depth in batch photochemical reactions. Therefore, photochemical Wolff rearrangement in batch usually shows unsatisfied efficiency, especially for scaling up reactions. To settle these issues, microreactors featured with large surface-to-volume ratio extremely enhance the mass and heat transfer properties, as well as improve the light penetration, thus is considered ideal for photochemical transformations [52-59]. Konopelski et al. reported the generation of ketenes by photochemical Wolff rearrangement in continuous flow for the first time in 2010, allowing the synthesis of β-lactams from α-dizao-β-ketoamides [60]. Thereafter, there are only isolated examples to date for the generation of ketenes via photochemical Wolff rearrangement in flow to take part in transformations, including cycloaddition, nucleophilic addition or Arndt–Eistert homologation [61-66]. However, a relatively long retention time (usually in minutes or even longer) is usually required for intermolecular reactions, which increases the risk of photodecomposition of ketenes [26,63], accompanying the generation of byproducts at the same time (Scheme 2B). Therefore, an excess amount of one of the reactants is required to improve the yields, and a procedure of column chromatography for product purification become necessary. Recently, Yoshida et al. developed a concept of flash chemistry, which allowed the reaction to proceed extremely fast (typically within a reaction time in the range from milliseconds to seconds), avoiding the decomposition of the active intermediate, so that realized some challenging reactions in micro reactors instead of flasks [67,68]. Enlighten by the concept of flash chemistry, we envisioned that fast photochemical Wolff rearrangement followed by fast nucleophilic addition, which occurs very rapidly, may avoid the photochemical decomposition of the short-live ketene intermediate and concurrently improve the utilization of the reactants. Therefore, an equimolar ratio or near equimolar ratio of reactants is possibly enough for such an efficient transformation so that the resulting product is produced with enough purity and can be easily separated from the mixture without the tedious chromatography procedure (Scheme 2C). To reach this goal, herein we showcase a fast photochemical Wolff rearrangement toward synthesis of a series of α-substituted amides.

  At the outset, 1-diazo-1-phenylpropan-2-one (1a) and aniline (2a, 2 equiv.) were used as the model substrates in batch reactions to investigate the preliminary reaction conditions. Dichloromethane (DCM) was found to be better than any other tested solvents, including 1,2-dichloroethane, acetonitrile, toluene, tetrahydrofuran, tert‑butyl methyl ether, and 1,4-dioxane (Table S1 in Supporting information), affording the desired product N,2-diphenyl-propanamide (3aa) in 94% yield. With the batch conditions in hand, next an initial flow experiment was conducted with a homemade micro-tubing photochemical reactor, but showed reaction efficiencies far behind our expectations. After careful investigation of the micro-structured reactors, a commercially available HANU™ flow photochemical reactor (from Creaflow, for detail see Supporting information), which had a visualized internal flow channel equipped with a series of cubic static mixing elements and could be used in combination with an extra auxiliary oscillatory diaphragm pulsator to vibrate the flowing reaction mixtures for enhancing the split-and-recombine mixing [69-71], was found to be effective to promote the transformation under the irradiation of blue LEDs. When a solution of 1a (0.04 mol/L) and 2a (0.1 mol/L) in dichloromethane was pumped into the photochemical oscillatory flow reactor (POFR) at a flow rate of 1.5 mL/min (t_R = 600 s), the desired product, N,2-diphenyl-propanamide (3aa) was furnished in 96% yield with 50% maximum light intensity (LI) of the blue LEDs (Table 1, entry 1). Increasing the flow rate to 15 mL/min (t_R = 60 s) made an adverse effect on the yield albeit under the irradiation with maximum light intensity (entries 2 and 3). Nevertheless, embedding a back pressure regulator (BPR) (75 psi) into the downstream system to ensure a steady flow rate and to prevent cavitation facilitated the conversion of the reaction within a residence time of 60 s (entry 4), whilst no further improvement was observed by increasing the back pressure (entry 5). With the help of the back pressure regulator, the solution was allowed to be superheated and resulted in a better yield at 40 ℃ than that at either an elevated temperature or an ambient temperature (entries 6 and 7). Notably, the flow protocol was found to be robust even when the reactants were fed into the POFR in an equimolar ratio, yet without compromising the yield through the dilution of the reaction (entries 8–10). At this point, without much superfluous reactants remaining in the crude mixture after reaction, the product was easily purified from the mixture by recrystallization to give 3aa in 96% yield under the optimal conditions (entry 10). Further shortening the residence time to 50 s brought a deleterious impact on the yield (entry 11).

  Table 1

  

  Table 1.  Reaction optimization with the use of POFR in flow.^a

  DownLoad: CSV

  With the optimal conditions in hand, we next turned our attention to investigating the substrate scope (Scheme 3). Electron-donating groups (Me, OMe) and electron-withdrawing groups (halides, CN, CF₃) are well tolerated on the phenyl ring of anilines (2b-2n). α- and β-naphthylamines (2o, 2p) were employed and afforded the corresponding amides in 92% and 97% yield, respectively. Pyridin-2-amine (2q) was also a suitable substrate in our protocol albeit producing the product 3aq in a moderate yield. To our delight, the substrate scope could be expanded to aliphatic amines, including benzylamine, and other primary and secondary amines that gave their corresponding amides 3ar-3av in excellent yields. Additionally, a series of 1-diazo-1-phenylpropan-2-ones bearing halides were tested and produced the desired α-disubstituted amides in excellent yields (3ba-3ha), whilst the introduction of halides in these products offered opportunities for further derivatization. Besides the methyl group, ethyl (3ia) and aryl groups (3ja-3na) also migrated readily in the Wolff rearrangement, but a bromo atom on the phenyl ring had a negative impact on its migratory aptitude (3ma). Notably, all the above-mentioned products were isolated and purified by recrystallization.

  Scheme 3

  

  Scheme 3.  Substrate scope.

  DownLoad: Full-Size Img PowerPoint

  Control experiments were performed to understand the mechanism of the reaction. α-Diazoketone 1a pumped to the reactor alone under the standard reaction conditions led to complete conversion to phenyl methyl ketene. Notably, the retention time was able to be reduced to less than 40 s without any deleterious effect on the conversion. Then the ketene intermediate was thoroughly converted to 3aa after that the collected ketene solution was mixed with aniline and once again pumped into the POFR within a retention time of 60 s (Supporting information). It indicates that a fast Wolff rearrangement is involved and the rate of the whole reaction is limited by the rate of nucleophilic addition of amine to the ketene intermediate. Based on the control experiments and previous studies [22], a proposed mechanism is shown in Scheme 4. An excited singlet state of α-diazo ketone (¹1*) resulting from irradiation undergoes decomposition to generate a ketene intermediate via two pathways. Ketene is possibly formed by a concerted nitrogen extrusion and 1,2-shift of the methyl group (path a). Another pathway involves a stepwise process in which a singlet state α-oxo-carbene intermediate is produced via nitrogen extrusion followed by the methyl group migration (path b). Finally, the amide 3 is furnished by the addition of aniline to the ketene intermediate.

  Scheme 4

  

  Scheme 4.  Proposed mechanism.

  DownLoad: Full-Size Img PowerPoint

  To demonstrate the synthetic value of this protocol, the reaction was performed on a 5 mmol scale to produce 1.02 g 3aa with 92% yield after recrystallization (Scheme 5a). Additionally, the present protocol may provide an alternative for the preparation of bioactive α-alkyl-α-arylcarboxylic acid derivatives. For instance, 2-(4-isobutylphenyl)-N-phenylpropanamide 5 was readily prepared under the standard conditions from the diazo ketone 4 and aniline, which could be further hydrolyzed into (±)-ibuprofen, a popular anti-inflammatory drug (Scheme 5b).

  Scheme 5

  

  Scheme 5.  Synthetic application.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, we have developed a fast photochemical Wolff rearrangement toward synthesis of a range of α-substituted amides with the use of a POFR under visible-light irradiation. Control experiment indicates that a fast process of the Wolff rearrangement (<40 s) is involved. The present protocol is easily scalable, and does not require the excess use of any reactants, and the α-substituted amides could be isolated by recrystallization in good to excellent yields.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Huashan Huang: Investigation. Jingze Chen: Investigation. Luyun Zhang: Investigation. Hong Yan: Writing – review & editing, Writing – original draft. Siqi Li: Investigation. Fen-Er Chen: Supervision.

  Acknowledgments

  We acknowledge financial support from the National Natural Science Foundation of China (No. 22208279) and Financial support from the Fuzhou University (No. 0041/511095)

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109992.

  
  
• 1. [1]

     Y. Fang, L. Xiao, Z. Chen, et al., Electrochem. Energy Rev. 1 (2018) 294–323. doi: 10.1007/s41918-018-0008-x

  2. [2]

     W. Luo, F. Shen, C. Bommier, et al., Acc. Chem. Res. 49 (2016) 231–240. doi: 10.1021/acs.accounts.5b00482

  3. [3]

     F. Li, Z. Wei, A. Manthiram, et al., J. Mater. Chem. A 7 (2019) 9406–9431. doi: 10.1039/c8ta11999f

  4. [4]

     H.J. Li, S.C. Han, B.W. Lyu, et al., Chin. Chem. Lett. 32 (2021) 2887–2892.

  5. [5]

     Y.J. Ke, J.X. Zhang, X. Peng, et al., Nano Res. (2024) 1998-0124.

  6. [6]

     H. Ying, W.Q. Han, Adv. Sci. 4 (2017) 1700298.

  7. [7]

     M. Sha, H. Zhang, Y. Nie, et al., J. Mater. Chem. A 5 (2017) 6277–6283.

  8. [8]

     Y. Liu, N. Zhang, L. Jiao, et al., Adv. Funct. Mater. 25 (2014) 214–220.

  9. [9]

     H. Wang, E. Matios, C. Wang, et al., J. Mater. Chem. A 7 (2019) 23747–23755. doi: 10.1039/c9ta05176g

  10. [10]

      C. Kim, K.Y. Lee, I. Kim, et al., J. Power Sources 317 (2016) 153–158. doi: 10.17995/kjgs.2016.11.33.153

  11. [11]

      Y. Xu, Y. Zhu, Y. Liu, et al., Adv. Energy Mater. 3 (2012) 128–133. doi: 10.1080/03602559.2011.578294

  12. [12]

      H. Bian, Z. Li, X. Xiao, et al., Adv. Funct. Mater. 29 (2019) 1901000.

  13. [13]

      F. He, Q. Xu, B. Zheng, et al., RSC Adv. 10 (2020) 6035–6042. doi: 10.1039/c9ra08897k

  14. [14]

      X. Xiao, W. Yao, T. Yan, et al., Nanoscale 15 (2023) 15405–15414. doi: 10.1039/d3nr02414h

  15. [15]

      W. Zhao, X. Ma, L. Gao, et al., Carbon 167 (2020) 736–745.

  16. [16]

      L. Yu, X.Y. Yu, X.W. Lou, Adv. Mater. 30 (2018) 1800939.

  17. [17]

      L. Yu, H. Hu, H.B. Wu, et al., Adv. Mater. 29 (2017) 1604563.

  18. [18]

      H. Fu, Q. Wen, P.Y. Li, et al., Small Methods 6 (2022) 2201025.

  19. [19]

      J.Y. Cheng, Z.L. Niu, Z.P. Zhao, et al., Adv. Energy Mater. 13 (2022) 2203248.

  20. [20]

      M. Chen, P. Xiao, K. Yang, et al., Angew. Chem. Int. Ed. 62 (2023) e202219177.

  21. [21]

      N. Ali, B. Singh, V.A.R.S. Lal, et al., J. Phys. Chem. C. 125 (2021) 7734–7745. doi: 10.1021/acs.jpcc.0c08407

  22. [22]

      W. Zeng, W. Shu, J. Zhu, et al., ACS Energy Lett. 9 (2024) 346–354. doi: 10.1021/acsenergylett.3c02502

  23. [23]

      M. Han, Z. Liu, X. Shen, et al., Adv. Energy Mater. 10 (2020) 346–354. doi: 10.1080/24701394.2020.1830075

  24. [24]

      M.C. Biesinger, B.P. Payne, A.P. Grosvenor, et al., Appl. Surf. Sci. 257 (2011) 2717–2730.

  25. [25]

      Y. Zhu, Z. Qian, J. Song, et al., Nano Lett. 21 (2021) 3588–3595. doi: 10.1021/acs.nanolett.1c00489

  26. [26]

      X.S. Liu, W.H. Zhao, B.Z. Zheng, et al., Angew. Chem. Int. Ed. 134 (2022) e202200475.

  27. [27]

      Z.Y. Zou, N. Ma, A.P. Wang, et al., Adv. Energy Mater. 10 (2020) 2001486.

  28. [28]

      N. Kuganathan, N. Kelaidis, A. Chroneos, Materials 12 (2019) 3243. doi: 10.3390/ma12193243

  29. [29]

      M. Song, C. Wang, D. Du, et al., Sci. China Chem. 62 (2019) 616–621. doi: 10.1007/s11426-018-9422-y

  30. [30]

      Y. Zhu, Q. Yao, R. Shao, et al., Nano Lett. 22 (2022) 7976–7983. doi: 10.1021/acs.nanolett.2c03334

• 

  Figure 1  DOS values of (a) ZnO, (b) Sn-ZnO, (d) Zn_0.9Mn_0.1O, and (e) Sn-Zn_0.9Mn_0.1O. Electrostatic potentials of (c) Sn and (f) Sn···Mn. (g) Na⁺ diffusion energy barrier of Sn and Sn···Mn. (h) Diffusion path of Na⁺ in Sn···Mn.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Schematic illustration of the construction of Sn-Zn_0.9Mn_0.1O/CNT. SEM images of (b) Zn_0.9Mn_0.1Sn(OH)₆/CNT and (c, d) Sn-Zn_0.9Mn_0.1O/CNT. (e) HRTEM images of Sn-Zn0_.9Mn_0.1O/CNT and (f) the TEM morphologies of Sn/Zn_0.9Mn_0.1O interfacial zone. (g) Fast Fourier transform (FFT) from the marked zone of (f).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  TEM images of (a) Sn-Zn_0.9Mn_0.1O/CNT and (b) overall element distribution. (c-f) Elemental mapping of Sn-Zn_0.9Mn_0.1O/CNT (Zn, O, Sn, and Mn).

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) TGA curves of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT from 30 ℃ to 900 ℃ (air atmosphere). (b) XRD of patterns of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT. Comparison of XRD diffraction angles of (c) Sn phases and (d) ZnO phases in Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT. (e) ICP results of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT. (f) XPS survey spectrums of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT. High-resolution (g) Sn 3d and (h) Mn K-edge XANES and reference samples (Mn foil, Mn₃O₄, MnO, and Mn₂O₃). (i) Zn K-edge XANES (inset: a zoomed-in view) of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  CV curves of the (a) Sn-Zn_0.9Mn_0.1O/CNT and (d) Sn-ZnO/CNT at a scan rate of 0.1 mV/s. Charge/discharge profiles of (b) Sn-Zn_0.9Mn_0.1O/CNT and (e) Sn-ZnO/CNT electrode at 1 A/g. Charge/discharge profiles of (c) Sn-Zn_0.9Mn_0.1O/CNT and (f) Sn-ZnO/CNT electrodes at different current rates (0.1, 0.5, 1, 2, 3, and 5 A/g). (g) Rate performance of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT. Cycle performance of Sn-Zn_0.9Mn_0.1O/CNT and Sn-ZnO/CNT electrodes at (h) 2, (i) 1, and (j) 5 A/g.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  CV curves of the (a) Sn-ZnO/CNT and (d) Sn-Zn_0.9Mn_0.1O/CNT electrode at various scanning rates from 0.1 mV/s to 1.0 mV/s. Peak current versus scan rate for (b) Sn-ZnO/CNT and (e) Sn-Zn_0.9Mn_0.1O/CNT electrodes (logarithmic format). Normalized contribution ratio of capacitive- and diffusion-controlled capacities of (c) Sn-ZnO/CNT and (f) Sn-Zn_0.9Mn_0.1O/CNT at different scanning rates from 0.1 mV/s to 1.0 mV/s. (g) GITT curves and Na⁺ diffusion coefficients of Sn-ZnO/CNT and Sn-Zn_0.9Mn_0.1O/CNT during (h) discharging and (i) charging process.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  (a) GCD curve of Sn-Zn_0.9Mn_0.1O/CNT. (b, c) Half-in-situ XRD patterns of Sn-Zn_0.9Mn_0.1O/CNT during (b) discharging (0.21, 0.03, and 0.01 V) and (c) charging processes (0.22, 0.54, and 0.62 V) respectively. Partial magnification of XRD patterns of Sn-Zn_0.9Mn_0.1O/CNT during the (d) discharging and (e) charging processes.

  下载: 全尺寸图片 幻灯片
  

  Figure 8  Electrochemical performance of Sn-Zn_0.9Mn_0.1O/CNT//NVP full cell. (a) Working principle diagram of the full cell. (b) Charge and discharge curves of NVP//Na, Sn-Zn_0.9Mn_0.1O/CNT//Na, and NVP//Sn-Zn_0.9Mn_0.1O/CNT full cells. (c) CV curve of NVP//Zn_0.9Mn_0.1O/CNT full cell at a scan rate of 0.1 mV/s. (d) Charge and discharge curve, (e) cycle performance, and (f) rate performance of NVP//Zn_0.9Mn_0.1O/CNT full cell.

  下载: 全尺寸图片 幻灯片
•