English

• 2-Arylpropionic acids are prevalent in bioactive molecules and marketed pharmaceuticals [1]. For example, they represent an important class of non-steroidal anti-inflammatory drugs (NSAIDs), including ibuprofen [2] and loxoprofen [3]. Hence, considerable effort has been dedicated to developing methodologies that enable efficient and sustainable synthesis of such structures [4]. Although several approaches are available, such as methylation of aryl acetic acid derivatives [5], hydrolysis of corresponding nitriles [6], and transition metal-catalyzed carbonylation reactions [7], they are often plagued with the low availability of substrates, the handling of toxic reagents, and the harsh reaction conditions.

  The 1,2-aryl migration of α-substituted alkyl aryl ketones is a practical strategy for preparing 2-arylpropionic acids [8,9]. The ZnO-promoted migration of α–chloro ketals is adapted as a key step for the manufacture of ibuprofen in China at > 8000 tons of annual production capacity (Scheme 1B, route 1) [10,11]. On the other hand, the need to acetalize the carbonyl group before the migration and the subsequent hydrolysis procedures has stimulated further advances in rearrangement protocols that can bypass this tedious sequence. Representative examples include the oxidative migration of propiophenone derivatives (route 2) [12–15] and the photo-initiated Favorskii rearrangement of α-chloroketones (route 3). In particular, the photorearrangement process has attracted continuous attention for its cost efficiency and adhere to the principles of green chemistry [16,17]. To this end, Sonawane and others made significant contributions to the photochemical rearrangement of 2–chloro-1-arylpropanones [18–21]. Baxendale recently applied this chemistry to the flow synthesis of ibuprofen using a continuous photochemistry platform for enhanced photon delivery [22]. However, all the existing reports employed (over)stoichiometric propylene oxide as an acid scavenger, which is acutely toxic and carcinogenic with high vapor pressure (vp = 445 mmHg at 20 ℃).

  Scheme 1

  

  Scheme 1.  1,2-Aryl migration strategy to synthesize 2-arylpropinoic acids.

  DownLoad: Full-Size Img PowerPoint

  To design a more efficient photo-rearrangement protocol, we pursued a non-hazardous alternative to propylene oxide. Discussions based on the reaction mechanism suggest that an ideal acid scavenger should be a mild base and poor electron donor (Scheme 2) [23–25]. Mechanistic investigations of photo-Favorskii rearrangements of the closely related, p-hydroxyphenacyl derived 3 suggested the generation of triplet ³3* under light irradiation [26,27]. Adiabatic formation of the triplet biradical ³4 followed by ³I → ¹I ISC furnished the Dewar zwitterion 5 [28,29], where the deprotonation of ³3* is likely assisted by general acid and general base catalysis of water [30,31]. For 1a without a p-OH group, the absence of a phenolic proton should shut down such a reaction channel. Instead, deprotonation could occur with strong bases at other acidic sites, such as the α-position of the carbonyl group substituted with a Cl atom (Scheme 2B and Scheme S1 in Supporting information) and the acidic proton in product 2a [32–34]. Moreover, an additive could act as a hydrogen atom donor or an electron donor, facilitating the photoreduction pathways to generate 8a [35–39]. Bearing these considerations, we report the photo-Favorskii rearrangement of 1 using HCOONa as an acid scavenger. The current protocol was further transferred to a continuous-flow platform, leading to improved photon delivery, reproducibility, and scalability [40–46].

  Scheme 2

  

  Scheme 2.  Mechanistic considerations of photo-Favorskii rearrangement. ^a The pK_a values were obtained by DFT calculations.

  DownLoad: Full-Size Img PowerPoint

  We embarked on initial optimization efforts using the photo-induced rearrangement of 2–chloro-1-(p-tolyl)propan-1-one (1a) as a model reaction (Table 1). The desired acid 2a was obtained in 50% yield at 55 ℃ under irradiation of 365 nm in a mixed solution of acetone-water (w/w = 9:1) (entry 1), and the yield of 2a increased in the presence of propylene oxide (entry 2). The addition of mild bases led to decreased yields of 2a (Table S1 in Supporting information). For example, when triethylamine was used, the desired 2a was obtained in a marginal 13% yield (Table 1, entry 3), with the dehalogenated 7a detected as the major byproduct (62% yield). The reaction employing pyridine delivered 2a in a 79% yield (entry 4), which is comparable to the conditions with propylene oxide (entry 2). These results suggested that the reaction yields were inversely related to the basicity and reduction ability of additives, which encouraged us to examine the effect of carboxylate salts. While sodium acetate was less satisfying (entry 5), sodium formate [47] proved to be an effective replacement for the hazardous propylene oxide (entry 6).

  Table 1

  

  Table 1.  Screen of acid scavengers. ^a

  DownLoad: CSV

  Further evaluations of other reaction conditions were performed with HCOONa (Table 2 and Table S1). The reaction operated under the wavelength range of 335–385 nm (entries 2 and 3, see Figs. S1 and S2 in Supporting information for details of the UV–vis absorbance spectra). Increasing the equivalence of HCOONa showed little improvement in the reaction yield in batch (entry 4). Acetonitrile (ε = 37.5) can also be used as a solvent (entry 5), but other solvents, such as THF (ε = 7.6) and DMF (ε = 2.3), gave a mixture of 2a and 7a (entry 6) [48]. Lowering the reaction temperature to 45 ℃ led to a slightly compromised yield (80%, entry 7). The rapid kinetics of the photo-induced migration is well illustrated in entry 8. Irradiation of 1a for 5 min yielded 2a, and a trace amount of 1a was recovered (89% conversion).

  Table 2

  

  Table 2.  Optimization of reaction conditions in batch. ^a

  DownLoad: CSV

  Next, we sought to address the scalability issue inherent to photochemical transformations under batch conditions. Transferring a photochemical reaction to a continuous-flow platform offers many advantages, such as ease of scaling up, enhanced mass transfer, and more homogeneous irradiation. We began our investigations of the continuous-flow reaction using a homemade photoreactor with a reaction unit featuring coiled FEP tubing (0.8 mm ID, 4.0 mL). The coil was placed in a water bath to control the reaction temperature (Fig. S2). The limited solubility of HCOONa and NaCl (as a byproduct) resulted in solid formation upon extended irradiation in batch reactions. In flow, mixing a solution of 1a in neat acetone (0.086 mol/L) and HCOONa in water (1.5 mol/L) prevented precipitation (Table 3). A quick screen of flow rates (entries 1–4) identified the optimal conditions (entry 3), and 2a was successfully obtained in 85%−87% yields within 4 min using the flow protocol (entry 3). Further reducing the residence time led to decreased yield (entry 5). Similar to batch reactions, acetonitrile could serve as an optional solvent (entry 6). We also examined the effect of pyridine as a basic additive in flow, and 2a was obtained in 80% yield (entry 7).

  Table 3

  

  Table 3.  Optimization of reaction conditions in flow. ^a

  DownLoad: CSV

  With optimized conditions in hand, we explored the generality of the photochemical aryl migration module in batch and flow (Scheme 3A). A collection of para-alkyl substituted derivatives underwent the rearrangement with good yields of the desired products (2b-2f). Notably, both the batch and flow reactions of 1d proceeded smoothly to yield ibuprofen 2d without acetalization-hydrolysis steps. To our delight, the reaction tolerated the ketone functional group at a remote position (2e and 2f), and loxoprofen 2f was isolated in 69% and 79% yields in batch and flow under the established conditions. Also, m-Me substituted 1g was an effective substrate (2g, 81% yield in flow), but o-Me substitution led to undesired Norrish type Ⅱ reactions (vide infra).

  Scheme 3

  

  Scheme 3.  Substrate scope (A) and scale-up synthesis of loxoprofen (B). Batch reactions were performed with 0.4 mmol 1 at 55 ℃ for 1 h. Flow reactions were performed with 2.0 mmol of 1 at 55 ℃ with t_R = 4 min. All reported yields are based on isolated 2. ^a MeCN was used as solvent.

  DownLoad: Full-Size Img PowerPoint

  In general, substrates without electron-donating alkyl substituents reacted with better yields in polar MeCN (ε = 37.5) than in acetone (ε = 20.5). As such, 2-phenylpropanoic acid (2h) was obtained in satisfying yields in MeCN (84% in flow). Similarly, halogen substitution was accommodated to provide 2i-2k with additional functional handles. In addition, products with p-OMe (2l) and p-CF₃ (2m) were obtained in poor yields along with photolysis products. Trifluoromethoxy substituted 2n was obtained in a moderate 66% yield, indicating the window for the electronic properties of substituents. Structural variations of the substrates delivered 2o and 2p with different α-alkyl substituents.

  The reaction setup is straightforward and requires no precautions for dry solvents and air. These features, coupled with the environmental benefits of photochemical transformations, demonstrate the suitability of this method for the decagram synthesis of NSAIDs. Specifically, multigram quantities of 1f reacted well under standard conditions, and loxoprofen (2f, 76%) was isolated by recrystallization (Scheme 3B).

  Additional observations concerning the detailed reaction pathway under the current conditions are shown in Scheme 4. We performed the reactions with triethylamine and sodium formate in D₂O and observed no evidence of deuterium incorporation in 2a. On the other hand, deuterium exchange was feasible for 1a in the dark (Scheme S1). Accordingly, deprotonation of 1a and related intermediates is not involved during the desired photochemical pathway. The attenuated yields of 2l and 2m were attributed to the presence of α-cleavage pathways [49], where photolysis of 2l and 2m competed with the desired 1,2-aryl migration to yield the Norrish Type Ⅰ products 12l and 12m. Similarly, a hydrogen atom abstraction-cyclization mechanism was operative for o-Me substituted 2q [50,51], as evidenced by the formation of 13q. Overall, these results are consistent with the reaction manifold [20] depicted in Scheme 2C.

  Scheme 4

  

  Scheme 4.  Experimental results of detailed reaction pathways.

  DownLoad: Full-Size Img PowerPoint

  In summary, the photo-Favorskii rearrangement of α-chloroketones to produce 2-arylpropionic acids was achieved by using HCOONa as an acid scavenger. Incorporating the continuous-flow setup avoided the issues associated with batch-mode photochemistry, showing more reproducible and scalable results. This method was compatible with the multigram synthesis of NSAIDs, and we expect that the scalability and the experimental ease of this photo-induced transformation will lead to its extensive use.

  Declaration of competing interests

  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.

  Acknowledgment

  Financial support from the National Natural Science Foundation of China (No. 22077018) is gratefully acknowledged.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108576.

  
  
• 1. [1]

     A.M. Gouda, E.A. Beshr, F.A. Almalki, et al., Bioorg. Chem. 92 (2019) 103224. doi: 10.1016/j.bioorg.2019.103224

  2. [2]

     G.M. Halford, M. Lordkipanidzé, S.P. Watson, Platelets 23 (2012) 415–422. doi: 10.3109/09537104.2011.632032

  3. [3]

     A. Terada, S. Naruto, K. Wachi, et al., J. Med. Chem. 27 (1984) 212–216. doi: 10.1021/jm00368a019

  4. [4]

     M.W. Ha, S.M. Paek, Molecules 26 (2021) 4792. doi: 10.3390/molecules26164792

  5. [5]

     F. Gualtieri, G. Conti, S. Dei, et al., J. Med. Chem. 37 (1994) 1704–1711. doi: 10.1021/jm00037a022

  6. [6]

     N.J. Stuart, A.S. Sanders, Patent, US3385886A, 1968.

  7. [7]

     V. Elango, M.A. Murphy, B.L. Smith, et al., US4981995A, 1991.

  8. [8]

     G. Castaldi, A. Belli, F. Uggeri, C. Giordano, J. Org. Chem. 48 (1983) 4658–4661. doi: 10.1021/jo00172a041

  9. [9]

     R.C. Larock, R. Rozhkov, Formation of nitriles, carboxylic acids, and derivatives by rearrangements, in: R.C. Larock (Ed.), Comprehensive Organic Transformations, John Wiley & Sons, Inc., 2018, pp. 1–28.

  10. [10]

      H. Wu, J. Zhai, C. Zhao, et al., Patent, CN110563694A, 2019.

  11. [11]

      J. Wang, X. Li, X. Liu, G. Dou, Patent, CN102701949A, 2012.

  12. [12]

      Y. Tamura, T. Yakura, Y. Shirouchi, J. Haruta, Chem. Pharm. Bull. 33 (1985) 1097–1103. doi: 10.1248/cpb.33.1097

  13. [13]

      T. Yamauchi, K. Hattori, K. Nakao, K. Tamaki, J. Org. Chem. 53 (1988) 4858–4859. doi: 10.1021/jo00255a037

  14. [14]

      A.R. Bogdan, S.L. Poe, D.C. Kubis, S.J. Broadwater, D.T. McQuade, Angew. Chem. Int. Ed. 48 (2009) 8547–8550. doi: 10.1002/anie.200903055

  15. [15]

      D.R. Snead, T.F. Jamison, Angew. Chem. Int. Ed. 54 (2015) 983–987. doi: 10.1002/anie.201409093

  16. [16]

      A. Albini, M. Fagnoni, Green Chem. 6 (2004) 1–6. doi: 10.1039/b309592d

  17. [17]

      Y. Kim, C.J. Li, Green Synth. Catal. 1 (2020) 1–11. doi: 10.1016/j.gresc.2020.06.002

  18. [18]

      H.R. Sonawane, B.S. Nanjundiah, D.G. Kulkarni, J.R. Ahuja, Tetrahedron: Asymmet. 2 (1991) 251–252. doi: 10.1016/S0957-4166(00)80045-2

  19. [19]

      H.R. Sonawane, B.S. Nanjundiah, D.G. Kulkarni, J.R. Ahuja, Tetrahedron 44 (1988) 7319–7324. doi: 10.1016/S0040-4020(01)86104-4

  20. [20]

      H.R. Sonawane, N.S. Bellur, G.M. Nazeruddin, Tetrahedron 51 (1995) 11281–11294. doi: 10.1016/0040-4020(95)00690-A

  21. [21]

      D.D. Dhavale, V.P. Mali, S.G. Sudrik, H.R. Sonawane, Tetrahedron 53 (1997) 16789–16794. doi: 10.1016/S0040-4020(97)10081-3

  22. [22]

      M. Baumann, I.R. Baxendale, React. Chem. Eng. 1 (2016) 147–150. doi: 10.1039/C5RE00037H

  23. [23]

      R.S. Givens, L. Strekowski, J. Am. Chem. Soc. 97 (1975) 5867–5873. doi: 10.1021/ja00853a036

  24. [24]

      G. Jones Ⅱ, L.P. McDonnell, J. Am. Chem. Soc. 98 (1976) 6203–6210. doi: 10.1021/ja00436a023

  25. [25]

      Y. Yagci, Y.Y. Durmaz, B. Aydogan, Chem. Rec. 7 (2007) 78–90. doi: 10.1002/tcr.20110

  26. [26]

      P. Klán, T. Šolomek, C.G. Bochet, et al., Chem. Rev. 113 (2013) 119–191. doi: 10.1021/cr300177k

  27. [27]

      T. Field, J. Peterson, C. Ma, et al., Photochem. Photobiol. Sci. 19 (2020) 1364–1372. doi: 10.1039/d0pp00067a

  28. [28]

      R.S. Givens, M. Rubina, K.F. Stensrud, J. Org. Chem. 78 (2013) 1709–1717. doi: 10.1021/jo301640q

  29. [29]

      T. Šolomek, D. Heger, B.P. Ngoy, R.S. Givens, P. Klán, J. Am. Chem. Soc. 135 (2013) 15209–15215. doi: 10.1021/ja407588p

  30. [30]

      P.G. Conrad, R.S. Givens, B. Hellrung, et al., J. Am. Chem. Soc. 122 (2000) 9346–9347. doi: 10.1021/ja001134f

  31. [31]

      R.S. Givens, D. Heger, B. Hellrung, et al., J. Am. Chem. Soc. 130 (2008) 3307–3309. doi: 10.1021/ja7109579

  32. [32]

      H. Yang, C. Tian, D. Qiu, et al., Org. Chem. Front. 6 (2019) 2365–2370. doi: 10.1039/c9qo00495e

  33. [33]

      Y. He, H. Yang, D. Gao, et al., Chin. J. Org. Chem. 41 (2021) 4725–4731. doi: 10.6023/cjoc202111014

  34. [34]

      B.G. Cai, W.Z. Yao, L. Li, J. Xuan, Org. Lett. 24 (2022) 6647–6652. doi: 10.1021/acs.orglett.2c02671

  35. [35]

      L. Kammari, L. Plíštil, J. Wirz, P. Klán, Photochem. Photobiol. Sci. 6 (2007) 50–56. doi: 10.1039/b612233g

  36. [36]

      D.E. Falvey, C. Sundararajan, Photochem. Photobiol. Sci. 3 (2004) 831–838. doi: 10.1039/b406866a

  37. [37]

      A. Banerjee, D.E. Falvey, J. Org. Chem. 62 (1997) 6245–6251. doi: 10.1021/jo970495j

  38. [38]

      A. Banerjee, K. Lee, D.E. Falvey, Tetrahedron 55 (1999) 12699–12710. doi: 10.1016/S0040-4020(99)00754-1

  39. [39]

      A. Banerjee, D.E. Falvey, J. Am. Chem. Soc. 120 (1998) 2965–2966. doi: 10.1021/ja971431t

  40. [40]

      L. Buglioni, F. Raymenants, A. Slattery, S.D.A. Zondag, T. Noël, Chem. Rev. 122 (2022) 2752–2906. doi: 10.1021/acs.chemrev.1c00332

  41. [41]

      L. Wan, G. Kong, M. Liu, et al., Green Synth. Catal. 3 (2022) 243–258. doi: 10.1016/j.gresc.2022.07.007

  42. [42]

      C. Sambiagio, T. Noël, Trend. Chem. 2 (2020) 92–106. doi: 10.1016/j.trechm.2019.09.003

  43. [43]

      L. Rogers, K.F. Jensen, Green Chem. 21 (2019) 3481–3498. doi: 10.1039/c9gc00773c

  44. [44]

      J. Xie, D. Zhao, Chin. Chem. Lett. 31 (2020) 2395–2400. doi: 10.1016/j.cclet.2020.03.022

  45. [45]

      Y. Gao, J. Wang, Chin. J. Org. Chem. 38 (2018) 1275–1291. doi: 10.6023/cjoc201712029

  46. [46]

      D. Liu, Y. Zhu, S. Gu, F. Chen, Chin. J. Org. Chem. 41 (2021) 1002–1011. doi: 10.6023/cjoc202007051

  47. [47]

      I. Bownik, P. Šebej, J. Literák, et al., J. Org. Chem. 80 (2015) 9713–9721. doi: 10.1021/acs.joc.5b01770

  48. [48]

      P. Sebej, B.H. Lim, B.S. Park, R.S. Givens, P. Klan, Org. Lett. 13 (2011) 644–647. doi: 10.1021/ol102887f

  49. [49]

      N.C. Yang, D.S. McClure, S. Murov, J.J. Houser, R. Dusenbery, J. Am. Chem. Soc. 89 (1967) 5466–5468. doi: 10.1021/ja00997a037

  50. [50]

      P. Klán, M. Zabadal, D. Heger, Org. Lett. 2 (2000) 1569–1571. doi: 10.1021/ol005789x

  51. [51]

      A.P. Pelliccioli, P. Klán, M. Zabadal, J. Wirz, J. Am. Chem. Soc. 123 (2001) 7931–7932. doi: 10.1021/ja016088d

• 

  Scheme 1  1,2-Aryl migration strategy to synthesize 2-arylpropinoic acids.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Mechanistic considerations of photo-Favorskii rearrangement. ^a The pK_a values were obtained by DFT calculations.

  下载: 全尺寸图片 幻灯片
  

  Scheme 3  Substrate scope (A) and scale-up synthesis of loxoprofen (B). Batch reactions were performed with 0.4 mmol 1 at 55 ℃ for 1 h. Flow reactions were performed with 2.0 mmol of 1 at 55 ℃ with t_R = 4 min. All reported yields are based on isolated 2. ^a MeCN was used as solvent.

  下载: 全尺寸图片 幻灯片
  

  Scheme 4  Experimental results of detailed reaction pathways.

  下载: 全尺寸图片 幻灯片

  Table 1.  Screen of acid scavengers. ^a

  下载: 导出CSV

  Table 2.  Optimization of reaction conditions in batch. ^a

  下载: 导出CSV

  Table 3.  Optimization of reaction conditions in flow. ^a

  下载: 导出CSV
•