Intermolecular photo-redox reaction between nitroarenes and phenylmethylamines under continuous photo-flow conditions

Abstract

Intermolecular photo-redox reactions can promote a wide range of chemical transformations. It is challenging to predict the photo-induced redox reaction for the intermolecular transfer of hydrogen and oxygen atoms. Herein, we have reported a strategy for transferring the intermolecular redox activity between nitroarenes and phenylmethylamines to make aromatic amines, aldehydes, and imines. The approach successfully synthesizes useful drug intermediates as well as final drug compounds, showcasing its efficiency and versatility. This protocol highlights significant practical utility in advanced organic synthesis.

---

Aromatic amines and various aldehydes are key building blocks in the synthesis of bioactive motifs (Scheme 1a)^1–4 and are widely present in numerous agrochemicals, paints,⁵ drugs, and organic molecules.^6–10 To prepare aromatic amines or aldehydes, mostly expensive metal catalysts are utilized,^11–13 which gives a lot of by-product wastage.^14–17 On the other hand, photo-redox reactions in organic chemistry have a long history, involving the simultaneous oxidation of one reactant and the reduction of another through electron transfer.^18,19 Intermolecular photo-redox reactions offer significant advantages, including versatility, efficiency, green chemistry principles, simplicity, chemoselectivity, mild reaction conditions, scalability, low operational costs, and broad substrate scope.^20–26

Scheme 1 Intermolecular photo-redox reaction for mutually exchanging oxygen and hydrogen.

Recent literature studies noted that the conventional nitrobenzene approach uses a photo-oxidizing agent to transfer an oxygen atom to an aromatic alkene for producing aldehydes, ketones, or alcohols.^25,27 However, in batch processes, nitrobenzene itself ends up as a waste by-product.¹⁵ On the other hand, when the reaction is carried out in a batch photolysis system, the reaction encounters multiple challenges, such as inadequate light exposure, inefficient mixing, reduced quantum efficiency, a limited surface-to-volume ratio, and extended reaction durations.^28–32 To address these challenges, Maiti et al. have transitioned the batch process to a continuous-flow technique for the oxidation of aromatic alkenes to the corresponding aldehydes or ketones.^33,34

Furthermore, since the reduction of nitroarenes generates multiple intermediates, it presents an opportunity to precisely intercept nitrosoarene³⁵ and hydroxylamine³⁶in situ. In continuation of the work, Baumann et al. have reported that photochemical intramolecular nitro reduction forms the intermediate nitroso species (Scheme 1b).³⁷ However, achieving the complete transformation of nitro to amine has remained challenging until 2024. In the course of our experiments, we observed that Jin et al. had further advanced the photochemical reduction of nitro to amine by incorporating EDA complex-based intermolecular hydrogen and electron transfer from an aliphatic amine.³⁸ However, the electron donor substrate is unused to synthesize any useful molecule. Under these batch reaction conditions, water was claimed to be a green solvent. However, since nitrobenzene is insoluble in water, a glycine mediator is required to facilitate the reaction. In some cases, the use of an organic solvent such as ethyl acetate becomes necessary.³⁸ Another significant limitation was that insoluble reagents or starting materials were not suitable under photo-flow conditions. Considering these challenges, we aimed to leverage intermolecular hydrogen and oxygen transfer to generate valuable compounds for organic synthesis. Thus, the development of photo-flow conditions for the reduction of nitro compounds to their corresponding amines has emerged as a significant research priority. Herein, we have reported a strategy for transferring the intermolecular photo-redox activity between nitroarenes and various active methylene groups to successfully synthesize arylamine compounds, aldehydes, and imines, showcasing its efficiency and versatility. This protocol highlights significant practical utility in advanced organic synthesis. Notably, this system exhibits a broad substrate scope, enabling the synthesis of both simple and complex arylamines and aldehydes. Furthermore, it facilitates the production of commercially significant drugs and intermediates, including benzocaine, butamben, and lenalidomide as well as essential drug intermediates such as lenalidomide and apixaban, highlighting its synthetic importance.

We began our investigation as a proof of concept for the photo-redox reaction, utilizing readily available 4-nitrobenzonitrile (1a) as a model oxidant and 1-benzylpiperidine (2a) as a model reductant. Prior to conducting the experiment, we recorded the UV spectra of 1a and 2a, which revealed a concentration-dependent n–π* absorption peak for 1a in the range of λ = 325–250 nm (Fig. S7). To access this wavelength range, we designed a system covering wavelengths from 254 to 395 nm and initiated the model reaction. A 0.05 M stock solution of the model substrate, 4-nitrobenzonitrile (1a), was prepared with 5 equivalents of 1-benzylpiperidine (2a) in DCM and passed through an in-house fabricated micro-photo flow reactor (μPFR). The solution was passed through a 10 mL perfluoroalkoxy (PFA) tubular reactor assembled with 310 nm light at a 5 bar pressure and 50 μL min⁻¹ flow rate, with a residence time of 200 min, which led to the formation of compound 3a with 90% yield (Table 1, entry 1). A control experiment without light was conducted, confirming the necessity of continuous light irradiation (Table 1, entry 2). We then investigated the effect of different light-emitting diode (LED) sources on the reaction (Fig. S1–S6); however, no formation of product 3a was observed (Table 1, entry 3). When the energy source was shifted from a 310 nm to a 395 nm LED (100 W), adjusting the pressure to 5 bar improved yields significantly, ranging from 78% to 88% (Table 1, entries 4 and 5).

Table 1 Optimization of the photo-redox reaction

Entry	Deviation from standard conditions	Yield (%)
		3a	4a	5a
Reaction conditions: 0.05 M solution of 1a in dichloromethane (DCM) +5 eq. of 2a, temp: 25 ± 5 °C; flow rate 50 μL min^−1.a Flow rate 25 μL min^−1.b Represents light intensity ≤50 000 lux; 5 bar.c Represents the corresponding aldehyde. NR means yield is less than 5%. Yields are based on isolated yield.
1	None	90	52	NR
2	Without light	NR	NR	NR
3	460, white, 530, and 620 nm instead of 310 nm light	NR	NR	NR
4	395 nm instead of 310 nm light	78	43	NR
5^a	395 nm instead of 310 nm light	88	50	NR
6^a^,^b	Sunlight instead of 310 nm light	88	42	NR
7	2 eq. of 2a instead of 5 eq. of 2a	46	23	NR
8	4 eq. of 2a instead of 5 eq. of 2a	79	57	NR
9	1 bar instead of 5 bar	NR	NR	NR
10	3 bar instead of 5 bar	77	31	NR
11	ACN instead of DCM	60	24	NR
12	EtOH instead of DCM	54	23	NR
13	THF instead of DCM	40	15	NR
14	DCE instead of DCM	25	NR	NR
15	DMSO, DMF instead of DCM	NR	NR	NR
16	2b instead of 2a	60	30^c	NR
17	2c instead of 2a	85	35^c	NR
18	2d, 2e, 2f, 2g, and 2h instead of 2a	NR	NR	NR
19	400 min instead of 200 min	50	15	25

---

Considering the varying intensities, we also tested the reaction under natural sunlight using a previously reported solar tracker reactor.³⁹ We observed the minimum yield of 3a when the flow rate was 50 μL min⁻¹ (residence time 200 min). Therefore, we examined the reaction at a reduced flow rate of 25 μL min⁻¹, resulting in a reaction time of 400 min (Table 1, entry 6).

Over this extended period, the formation of imine (5a) was also observed (Table 1, entry 6). After screening various light sources, we examined the influence of proton source equivalency on the reaction yield by adjusting the amount of proton source (2a) from 2 to 4 equivalents instead of 5, which resulted in a gradual decrease in 3a yield (46–79%) (Table 1, entries 7 and 8).

We next investigated the effects of pressure and found that a pressure of 5 bar is essential for the reaction (Table 1, entries 9 and 10). To ensure the photo-induced redox reaction for practical application, we were interested in checking the solvent feasibility, including ACN, EtOH, THF, DCE, DMSO, and DMF (Table 1, entries 11–15).

Our results indicated that DCM led to complete conversion and excellent selectivity in 200 min of residence time. Additionally, we explored the intermolecular redox process of electron donor–acceptor complexes by substituting the 2a moiety with various other hydrogen atom transfer (HAT) and electron donor groups. The results demonstrated that amines with an α-hydrogen atom could significantly enhance the reaction performance (Table 1, entries 16–18). Once we have increased the reaction time from 200 min to 400 min, we observed that the amine further reacted with the aldehyde and formed the imine product (Table 1, entry 19).

Building on the success of our preliminary intermolecular photo-redox photo-flow setup, we aimed to investigate the reduction of additional nitroarenes, recognizing the importance of their corresponding amines in various applications, such as agrochemicals, paints, pharmaceuticals, and other organic compounds. It is widely recognized that nitro-reduction processes using batch methods, with either inorganic hydrogen sources or molecular hydrogen in the presence of transition metal catalysts (such as Cu, Sn, Zn, Mn, Ni, Ir, Pd, V), require external energy input.^40–46 Additionally, a major challenge in the selective reduction of nitro groups occurs when bulky substituents, metal-sensitive compounds, heat-sensitive groups, or unsaturated substituents are present on the aromatic rings, often resulting in the formation of unwanted side products.^47,48 To tackle all the above-mentioned challenges, a variety of substituted nitroarenes containing the electron-withdrawing group (3a–3m) were chosen to undergo a photo-redox reaction smoothly. Moreover, easily synthesized nitrobenzene esters were successfully converted to two specific anaesthetics,^49,50 benzocaine (3i) and butamben (3j), with good to excellent yields. Furthermore, the standard protocols have been adapted to reduce substituted nitrobenzene containing the electron-donating group (3n–3p), and these compounds displayed excellent tolerance and site selectivity towards the methyl substitution.

Next, nitro compounds, containing benzothiazole (1r), quinoline (1s), and morpholine-substituted (1u), substrates yielded their corresponding derivatives 3r, 3s, and 3u along with the potential intermediate 3q when 1q reacted with 2a, which is used for the synthesis of lenalidomide intermediate. It is worth noting that the protocol can also be applied to the synthesis of the anti-cancer drug lenalidomide (3t).⁵¹ After investigating various nitro compounds for intermolecular photo-redox reactions, we turned our attention to imine synthesis by extending the reaction time from 200 to 400 min.^52,53 The developed protocol was further tested with various other substituted groups, and we successfully obtained imine products in good yields (Fig. 1, 5b–5e). It is worth mentioning that the batch process for the same reaction showed a previously reported poor space–time yield (STY) of 1.565 mg per day mL⁻¹, even with the addition of 1 equivalent of glycine as an additive. In contrast, our photo-flow process demonstrates significantly higher productivity and an STY of 54.9 mg per day per mL⁻¹ (SI, Section S6, Table S1).³⁸ Next, when comparing the E-factors from previously reported studies³⁸ with those of our current work, the values are found to be approximately the same (SI, Section S6, Table S2).

Fig. 1 Photo-flow process for the synthesis of aryl amines. Reaction conditions: 0.05 M 1a to 1u, reactor volume 10 mL, residence time 200 min and 400 min. For 5b to 5e, yields are based on isolated yield.

To illustrate the synthetic potential of this protocol, we conducted a gram-scale reaction for the nitro reduction of 1u to synthesize apixaban intermediate (3u) in 52% yield (SI, Section S8), which is used for the synthesis of an anticoagulant drug to prevent stroke and systemic embolism in non-valvular atrial fibrillation as well as to treat deep vein thrombosis (DVT) and pulmonary embolism (PE) in patients recovering from hip or knee replacement surgery⁵⁴ (Fig. 2).

Fig. 2 Gram-scale synthesis of apixaban intermediate 3u.

To further assess the feasibility under sunlight, we employed our previously developed litre-scale photo-flow reactor to carry out the reaction.⁵⁵ Finally, we obtained (5.6 g, Fig. S17, SI, Section S10) an isolated 36% yield under fluctuating sunlight (20 000–200 000 lux).

To elucidate the mechanistic study, we conducted several control experiments under standard conditions. Initially, radical quenching experiments were conducted to explore the possibility of the EDA complex further converting to the radical.^27,33 For this purpose, we added radical scavengers such as tetramethylpiperidine-N-oxide (TEMPO) to the reaction mixture under identical reaction conditions (SI, Section S9) but we could not obtain the desired product. This implies that the current photo-redox reactions may proceed via a free radical pathway. For this, we experimented by synthesizing the possible intermediates (1a′, 1a′′, and 1a′′′) and examined them under standard conditions to convert the nitroso (1a′) and hydroxylamine (1a′′) intermediates to the product. However, with the azobenzene intermediate (1a′′′), we failed to obtain 3a as a product (Fig. 3, SI, Section S9). Further, to optimize the reaction, we analysed samples at specific intervals using NMR and GC, allowing us to track real-time progress. NMR data showed steady nitroarene consumption and increasing amine product yield. Notably, hydroxylamine intermediates (1a′′) appeared at around 60 min, suggesting a stepwise reduction mechanism where nitroarenes are reduced first to hydroxylamine, then to the amine (3a). Additionally, the yield of the amine, which increased as the nitroarene was consumed, was assessed using both GC and NMR analysis (Fig. 3, SI, Section S9). We then plotted a graph that correlates these analytical results for a clear representation of the reaction progress.

Fig. 3 Mechanistic investigation of photo-redox reaction. (I) Control experiments. (II) NMR spectral analysis of the time-based study. (III) NMR and GC spectral analysis for the mechanistic studies in time-based screening.

Based on experimental findings and a previous report, we proposed the mechanism as shown in Fig. 4.

Fig. 4 Plausible mechanism of the photo-redox reaction.

At first, nitroarene 1a forms the EDA complex I with 2a, which engages in hydrogen atom transfer (HAT) of the benzylic α-C(sp³)–H bond,²⁷ absorbs 310 nm light, and enters the excited state. This induces the formation of a radical ion pair via an intermolecular single-electron transfer (SET) process. 1a′, is subsequently generated through hydrogen atom abstraction from 2a and the elimination of a water molecule. 2a was converted to an iminium ion, and hydrolysis in an aqueous medium may lead to aldehyde formation (4a, isolated yield 52% under standard conditions) and piperidine. Next, 1a′, rapidly combines with 2a to form EDA complex II, which undergoes another SET and HAT cycle, generating a 1a′′, intermediate. Finally, EDA complex III is formed, and a third round of SET and HAT processes leads to the formation of target 3a.

Conclusions

In summary, we have developed a cascade EDA complex-mediated strategy for the photochemical reduction of nitroarenes to aromatic amino compounds through intermolecular SET and HAT processes. This method effectively facilitates the formation of photoexcited EDA complexes between electron-deficient compounds (such as nitroarene and nitrosoarene) or neutral hydroxylamine adducts and primary, secondary, and tertiary amines. The protocol offers numerous benefits, including ease of operation, low cost, catalyst- and metal-free conditions, compatibility with various easily reducible functional groups (–CN, –COR, –CHCHR, and –halo), successful application in pharmaceutical intermediates and commercial drug synthesis, with scalability to gram quantities.

Conflicts of interest

The authors declare no competing financial interest.

Data availability

Supplementary information is available. See DOI: https://doi.org/10.1039/D5RE00131E.

The data that support the findings of this study are available in the communication or its SI file.

Acknowledgements

Mounika Kukudala and A. K. S. thank the CSIR (Govt. of India), New Delhi, for the FTT project (file no. 31-2(281)/2023-24). CSIR-IICT manuscript communication no. IICT/Pubs./2024/144.

Notes and references

1. A. Cabré, X. Verdaguer and A. Riera, Chem. Rev., 2022, 122, 269–339.
2. M. Li, Y. Jin, Y. Chen, W. Wu and H. Jiang, J. Am. Chem. Soc., 2023, 145, 9448–9453.
3. S. S. Shivatare, V. S. Shivatare and C.-H. Wong, Chem. Rev., 2022, 122, 15603–15671.
4. D. Sokolova, T. C. Lurshay, J. S. Rowbotham, G. Stonadge, H. A. Reeve, S. E. Cleary, T. Sudmeier and K. A. Vincent, Nat. Commun., 2024, 15, 7297.
5. S. Biswas and C. Bolm, Org. Lett., 2024, 26, 1511–1516.
6. A. Rana, R. Chauhan and A. K. Singh, React. Chem. Eng., 2024, 9, 1313–1319.
7. A. Rana, R. Chauhan, A. Mottafegh, D. P. Kim and A. K. Singh, Commun. Chem., 2024, 7, 251.
8. P. Zhou, L. Jiang, F. Wang, K. Deng, K. Lv and Z. Zhang, Sci. Adv., 2017, 3, e1601945.
9. R. V. Jagadeesh, A.-E. Surkus, H. Junge, M.-M. Pohl, J. Radnik, J. Rabeah, H. Huan, V. Schünemann, A. Brückner and M. Beller, Science, 2013, 342, 1073–1076.
10. S. Kumar, D. Aand, S. Pabbaraja, D.-P. Kim and A. K. Singh, ACS Sustainable Chem. Eng., 2019, 7, 19605–19611.
11. R. Chauhan, A. Rana, S. Ghosh, P. Srihari and A. K. Singh, React. Chem. Eng., 2023, 8, 908–916.
12. G. Zhang, F. Tang, X. Wang, L. Wang and Y.-N. Liu, ACS Catal., 2022, 12, 5786–5794.
13. V. Chugh, B. Chatterjee, W.-C. Chang, H. H. Cramer, C. Hindemith, H. Randel, T. Weyhermüller, C. Farès and C. Werlé, Angew. Chem., Int. Ed., 2022, 61, e202205515.
14. J. K. Mitchell, W. A. Hussain, A. H. Bansode, R. M. O'Connor and M. Parasram, J. Am. Chem. Soc., 2024, 146, 9499–9505.
15. J. K. Mitchell, W. A. Hussain, A. H. Bansode, R. M. O'Connor, D. E. Wise, M. H. Choe and M. Parasram, Org. Lett., 2023, 25, 6517–6521.
16. R. Luo, J. Tong, L. Ouyang, L. Liu and J. Liao, RSC Adv., 2023, 13, 29607–29612.
17. Y. Yamane, X. Liu, A. Hamasaki, T. Ishida, M. Haruta, T. Yokoyama and M. Tokunaga, Org. Lett., 2009, 11, 5162–5165.
18. Y. Tan, M. Pei, K. Yang, T. Zhou, A. Hu and J.-J. Guo, Org. Lett., 2024, 26, 8084–8089.
19. Z.-M. Lai, Y. Xie, L.-L. Huang, J. Guo and G. Lu, Chem. Sci., 2025, 16, 4352–4359.
20. V. Srinivasu, K. Pal, S. Giri and D. Sureshkumar, Org. Lett., 2024, 26, 10328–10333.
21. M. Capone, G. Dell'Orletta, C. G. Page, T. K. Hyster, G. D. Scholes and I. Daidone, ACS Catal., 2024, 14, 16488–16496.
22. Y.-Z. Liu, Y. Chen, A. Wang, Z. Shen, X. Zhou, J. Zhang, Y. Jian and X. Ma, Green Chem., 2024, 26, 2280–2289.
23. D. Barik, N. Chakraborty, A. K. Sahoo, H. N. Dhara and B. K. Patel, Chem. Commun., 2024, 60, 12577–12580.
24. F. Meng, Y. Cui, W. Xu and W.-C. Yang, Org. Lett., 2024, 26, 6884–6888.
25. J. D. Lasso, D. J. Castillo-Pazos, J. M. Salgado, C. Ruchlin, L. Lefebvre, D. Farajat, D. F. Perepichka and C.-J. Li, J. Am. Chem. Soc., 2024, 146, 2583–2592.
26. X. Huang and C. Cai, Chem. Commun., 2025, 61, 5126–5129.
27. D. E. Wise, E. S. Gogarnoiu, A. D. Duke, J. M. Paolillo, T. L. Vacala, W. A. Hussain and M. Parasram, J. Am. Chem. Soc., 2022, 144, 15437–15442.
28. J. M. Paolillo, A. D. Duke, E. S. Gogarnoiu, D. E. Wise and M. Parasram, J. Am. Chem. Soc., 2023, 145, 2794–2799.
29. M. Purwa, A. Rana and A. K. Singh, React. Chem. Eng., 2022, 7, 2084–2092.
30. L. Capaldo, Z. Wen and T. Noël, Chem. Sci., 2023, 14, 4230–4247.
31. A. Laybourn, K. Robertson and A. G. Slater, J. Am. Chem. Soc., 2023, 145, 4355–4365.
32. S. Cañellas, M. Nuño and E. Speckmeier, Nat. Commun., 2024, 15, 307.
33. A. Ruffoni, C. Hampton, M. Simonetti and D. Leonori, Nature, 2022, 610, 81–86.
34. G. Prakash, J. Grover, P. Pathak, A. K. Mittal, P. Balasubramaniam and D. Maiti, React. Chem. Eng., 2024, 9, 1032–1035.
35. H. Zhu, J. N. Powell, V. A. Geldchen, A. S. Drumheller and T. G. Driver, Angew. Chem., Int. Ed., 2025, 64, e202416126.
36. T. Kaneko, R. Ito, T. Okamura and T. Sato, Org. Lett., 2025, 27, 2042–2048.
37. J. García-Lacuna and M. Baumann, J. Org. Chem., 2024, 89, 617–623.
38. X. Zhuang, H. Song, J. Wang, Z. Zhang, J. Wang, B. Sun, W. Su and C. Jin, Green Chem., 2024, 26, 9682–9689.
39. A. Rana, B. K. Malviya, D. K. Jaiswal, P. Srihari and A. K. Singh, Green Chem., 2022, 24, 4794–4799.
40. D. Formenti, F. Ferretti, F. K. Scharnagl and M. Beller, Chem. Rev., 2018, 119, 2611–2680.
41. B. Lei, F. Cao, M. Chen, X. Wang and Z. Mo, J. Am. Chem. Soc., 2024, 146, 17817–17826.
42. S. C. Cosgrove, G. J. Miller, A. Bornadel and B. Dominguez, ACS Sustainable Chem. Eng., 2023, 11, 8556–8561.
43. M. Bal, W. Van Hoey, R. Cleirbaut, F. Lemiere, S. Van Doorslaer, P. Cool and B. U. Maes, ACS Catal., 2025, 15, 4726–4738.
44. J. Gao, L. Zhang, L. Luo and N. Wang, Green Chem., 2025, 27, 1723–1728.
45. Y. Pan, X. Zhang, G. Sun, Y. Li and B. Liu, ACS Catal., 2025, 15, 3674–3698.
46. P. Biswal, G. Sai Kumar and V. Chandrasekhar, J. Org. Chem., 2025, 90, 3194–3201.
47. H. An, Y. Ding, P. Sautet, G. Sun and N. Yan, J. Am. Chem. Soc., 2024, 146, 29315–29324.
48. R. Jana and K. Pradhan, Chem. Commun., 2024, 60, 8806–8823.
49. S. Khair-ul-Bariyah, M. Arshad, M. Ali, M. I. Din, A. Sharif and E. Ahmed, Mini-Rev. Med. Chem., 2020, 20, 3–11.
50. R. Manikandan, R. Shanmugam and A. Pratheepkumar, Chem. Commun., 2025, 61, 9043–9046.
51. Ç. Ağtaş and E. B. Atici, J. Pharm. Biomed. Anal., 2025, 258, 116757.
52. M. Yus, J. C. Gonzalez-Gomez and F. Foubelo, Chem. Rev., 2013, 113, 5595–5698.
53. K. Acharyya and P. S. Mukherjee, Am. Ethnol., 2019, 131, 8732–8745.
54. R. Chauhan, A. Rana, A. Mottafegh, D. P. Kim and A. K. Singh, Org. Process Res. Dev., 2025, 29, 881–888.
55. D. Aand, A. Rana, A. Mottafegh, D. P. Kim and A. K. Singh, React. Chem. Eng., 2024, 9, 2427–2435.

---