Synthesis of (Z)-Cinnamate Derivatives via Visible-Light-Driven E-to-Z Isomerization

Key words

visible light - isomerization - Ir₂(ppy)₄Cl₂ - cinnamate - coumarins

(Z)-Cinnamate derivatives are commonly encountered in natural products[1] and have been widely used in organic synthesis.[2] The synthesis of (Z)-cinnamate derivatives has attracted interest since the early era of organic chemistry and it remains a focus of interest. Over the past ten years, many methods have been established to obtain the thermodynamically less stable (Z)-cinnamate derivatives,[3] of which the most widely used have been semihydrogenation of alkynes with Pd,[4] Ru,[5] Au,[6] Rh,[7] Cu,[8] or Ni[9] complexes. Wittig olefination,[10] stereoselective cross-coupling,[11] and olefin metathesis[12] are also commonly employed methods for the efficient synthesis of Z-olefins. Photochemical reactions, as mild, atom-economic and environmentally benign methodologies, have been widely used in organic synthesis,[13] especially in the synthesis of diazo compounds,[14] heterocycles,[15] helicene-like compounds,[16] natural products with complex structures,[17] and thermodynamically unstable (Z)-cinnamates.[18]

During our investigation of the synthetic utility of visible-light photoredox catalysts for [2+2] photodimerization,[19] it was observed that the thermodynamically stable (E)-cinnamamide (1a) could be transformed into its Z-isomer (2a) effectively (in 65% isolated yield), rather than forming the expected dimer, when treated with 1 mol% Ir₂(ppy)₄Cl₂ (Scheme [1]). Further experimentation revealed that air had no clear influence on the reaction efficiency, which made the reaction operation very convenient (Table [1], entry 1). Inspired by these interesting results, we considered developing a practical method for the synthesis of (Z)-cinnamate derivatives by visible-light-driven isomerization.

Scheme 1 Photochemical reaction of 1a catalyzed by Ir₂(ppy)₄Cl₂

Table 1 Optimization of Isomerization Conditions^a

Entry	Catalyst (mol%)	Solvent	Yield (%)b	Z/E c
1	Ir2(ppy)4Cl2 (1)	1,4-dioxane	68	70:30
2	Ir2(ppy)4Cl2 (1)	CH3CN	71	72:28
3	Ir2(ppy)4Cl2 (1)	DMF	67	71:29
4	Ir2(ppy)4Cl2 (1)	DMA	65	70:30
5	Ir2(ppy)4Cl2 (1)	DCE	67	69:31
6	Ir2(ppy)4Cl2 (1)	THF	61	64:36
7	Ir2(ppy)4Cl2 (1)	PhMe	12	14:86
8	Ir2(ppy)4Cl2 (1)	EtOAc	34	36:64
9	–	CH3CN	NRd	0:100
10e	Ir2(ppy)4Cl2 (1)	CH3CN	NRd	0:100
11	Ir2(ppy)4Cl2 (3)	CH3CN	67	69:31
12	Ir2(ppy)4Cl2 (5)	CH3CN	62	63:37
13f	Ir2(ppy)4Cl2 (1)	CH3CN	70	72:28
14g	Ir2(ppy)4Cl2 (1)	CH3CN	65	69:31
15h	Ir2(ppy)4Cl2 (1)	CH3CN	73	76:24
16i	Ir2(ppy)4Cl2 (1)	CH3CN	8	13:87
17	Ir(ppy)3 (1)	CH3CN	74	77:23
18	Ir(ppy)2(dtbbpy)PF6 (1)	CH3CN	72	75:25

^a Reaction conditions: 5 W LED (460–470 nm), 1a (0.4 mmol), catalyst (0.004–0.02 mmol), solvent (4.0 mL), air, r.t., 24 h, unless otherwise noted.

^b Isolated yield of Z-isomer (2a).

^c Determined by ¹H NMR analysis of the crude product.

^d Not detected.

^e Performed in the dark.

^f 10 mol% N,N-diisopropylethylamine (DIPEA) was added.

^g Reaction duration 12 h.

^h Reaction duration 36 h.

ⁱ 40 °C.

Initially, we investigated the photoisomerization of (E)-cinnamamide (1a) catalyzed by 1 mol% Ir₂(ppy)₄Cl₂ in different solvents.[20] The results showed that the isomerization occurred smoothly in a wide range of solvents such as CH₃CN, DMF, DMA, DCE, and THF, which provided the application of this strategy with more options (Table [1], entries 2–6). In reactions conducted in toluene and EtOAc, only a small proportion of 1a isomerized; no other by-products were observed (entries 7 and 8). Finally, CH₃CN was chosen as the solvent of choice. Further experiments showed that the photocatalyst and visible light were both essential for the success of this reaction; no reaction took place in the absence of Ir₂(ppy)₄Cl₂ (entry 9) or visible light (entry 10). Increasing the amount of Ir₂(ppy)₄Cl₂ to 3 or 5 mol% produced lower yields (entries 11 and 12). Feasibly, the higher loadings of photocatalyst increased self-quenching and reduced light penetration into the reaction solution. The influence of external base on this isomerization was also studied, with addition of 10 mol% N,N-diisopropylethylamine (DIPEA) (entry 13). However, no clear influence on the reaction efficiency was observed. To demonstrate the equilibrium point in this E-to-Z isomerization, the reaction times were set as 12 h and 36 h, respectively (entries 14 and 15, vs. entry 1). This indicated that 24 h was required to obtaining efficient conversion. To test the effect of temperature on the yield of Z-isomer, the reaction temperature was set at 40 °C (entry 16 vs. entry 1). However, only 8% of 2a was obtained after 24 h. This might be due to the fact that Ir₂(ppy)₄Cl₂ is temperature sensitive and loses its catalytic capability at the higher temperature. When compared to other commercial available photoredox catalysts, such as Ir(ppy)₃ [18b] (entry 17) and Ir(ppy)₂(dtbbpy)PF₆ [17d] (entry 18), which have been reported to be effective catalysts for obtaining Z-alkenes, Ir₂(ppy)₄Cl₂ afforded a similar yield. However, it should be noted that, among the three studied photoredox catalysts, Ir₂(ppy)₄Cl₂ was the most economical.

An optimization of light sources with different wavelengths (λ) and powers (W) was also carried out (Table [2]). This study showed that 5 W blue light with wavelengths between 460 and 470 nm afforded the best selectivity (entry 4). When the reaction was simply exposed to sunlight, little conversion to the Z-isomer was observed. Based on these results, we finally established optimal reaction conditions for the E-to-Z isomerization to be 1 mol% Ir₂(ppy)₄Cl₂ in CH₃CN, irradiated with a 5 W blue LED lamp (λ ≈ 460–470 nm) for 24 h at room temperature. The reaction apparatus was set up as shown in Figure [1].

Table 2 Optimization of Light Source^a

Entry	Light source	λ (nm)	Yield (%)b	Z/E c
1	5 W LED	400–415	42	43:57
2	5 W LED	420–430	48	50:50
3	5 W LED	440–445	69	72:28
4	5 W LED	460–470	71	73:27
5	5 W LED	480–485	27	28:72
6	5 W LED	490–495	11	13:87
7	7 W LED	460–470	70	72:28
8	10 W LED	460–470	71	73:27
9	sunlight	–	trace	2:98

^a Reaction conditions: Light source, 1a (0.4 mmol), catalyst (0.004 mmol), CH₃CN (4.0 mL), air, r.t., 24 h.

^b Isolated yield of the Z-isomer (2a).

^c Determined by ¹H NMR analysis of the crude product.

Figure 1 Reaction apparatus for photochemical isomerization

With these optimized reaction conditions in hand, we next investigated the scope of (E)-cinnamate derivatives that could be applicable to the transformation (Table [3]).

Table 3 Scope of Photocatalytic Isomerization^a

2b (59%, 71:29)	2c (63%, 70:30)	2d (58%, 63:37)
2e (57%, 64:36)	2f (54%, 60:40)	2g (47%, 54:46)
2h (62%, 64:36)	2i (67%, 71:29)	2j (80%, 85:15)
2k (37%, 48:52)	2l (56%, 60:40) (53%, 57:43)b	2m (58%, 63:37)
2n (60%, 68:32)	2o (66%, 70:30)	2p (87%, 92:8)
2q (70%, 75:25)	2r (60%, 66:34)	2s (76%, 81:19)
2t (60%, 67:33)

^a Reaction conditions: 5 W LED (460~470 nm), 1b–t (0.4 mmol), catalyst (0.004 mmol), CH₃CN (4.0 mL), air, r.t., 24 h; isolated yield of the Z-isomer; Z/E ratio was determined by ¹H NMR analysis of the crude product.

^b The reaction was conducted on a 2-gram scale.

Various (E)-cinnamate derivatives with electron-withdrawing, electron-donating and neutral substituents at the ortho-, meta-, or para-positions of the benzene ring underwent E-to-Z isomerization smoothly, giving the corresponding Z-isomers 2b–t in modest to high yields. From the results shown in Table [3], it can be concluded that electron-withdrawing substituents promoted this transformation, with yields of Z-isomers increasing notably when F, CF₃, or Cl groups are present (2i, 2j, 2p). However, a range of substituents were well tolerated, and no other products could be detected other than the two isomeric alkenes. In some cases, the efficiency of this isomerization could be enhanced by introducing a methyl substituent onto the β-position of the double bond (eg. 2q vs. 2e), which is consistent with previous reports.[18b] The isomerization of substituted heterocycles also occurred smoothly to give the Z-isomers (2s, 2t). To test whether the reaction could proceed efficiently at a larger scale, we carried out the isomerization of (E)-methyl-4-hydroxycinnamate (1l) on a 2-gram scale, which gave 2l in 53% yield. The E-isomer (1l) could be readily removed by repeated crystallization from methanol.

A plausible pathway for the E-to-Z transformation is proposed based on the experimental data and literature.[18b] [21] The addition of DIPEA, which is a common strategy to promote photoredox reaction,[22] did not facilitate this E-to-Z transformation (Table [1], entry 13). Therefore, we propose that an energy-transfer (ET) process occurs in this isomerization, rather than an outer-sphere single-electron-transfer (SET) process (Scheme [2]). After absorbing a photon, the photocatalyst in its singlet ground state PC(S₀) is excited to the first singlet excited state *PC(S₁), which relaxes to the triplet excited state *PC(T1) through internal conversion and successive fast intersystem crossing (ISC). The energy transfer between *PC(T₁) and *Q(S₀) regenerates the PC(S₀) and provides *Q(T₁) to engage in the photochemical isomerization reaction. The selectivity of the E-to-Z isomerization is determined by the different rates of the photo-quenching of the *PC(T₁) by the two isomers. Generally, the E-isomer quenches the *PC(T₁) at a faster speed than its Z-isomer, thus resulting in an increase in the amount of Z-isomer in this process.[17c] [21b]

Scheme 2 Plausible pathway for the E-to-Z isomerization

Encouraged by the range of successful isomerization examples, we speculated that this facile and practical E-to-Z isomerization of cinnamate derivatives under mild conditions could be used as an attractive synthetic methodology in organic synthesis. Coumarins, a class of oxygen-containing heterocycles bearing a typical benzopyrone framework, are widely distributed in plants and they exhibit broad biological activities.[23] Among the various methods developed for the synthesis of coumarin derivatives,[24] we considered that cyclization of ortho-hydroxycinnamates via double-bond isomerization might prove to be convenient and atom-economic. Thus, we tested our optimized reaction conditions for the synthesis of coumarins and were delighted to obtain the target compounds efficiently (Table [4]).[25]

Table 4 Synthesis of Coumarins via Photochemical Isomerization^a

Entry	Substrate		Product	Isolated yield (%)
1		1u		2u	94
2		1v		2u	95b
3		1w		2v	71c
4		1x		2w	96

^a Reaction conditions: 5 W LED (460–470 nm), 1u–x (0.4 mmol), catalyst (0.004 mmol), CH₃CN (4.0 mL), air, r.t., 72 h.

^b Reaction duration 36 h.

^c 20% of 1w was recovered.

In summary, we have developed a practical and efficient strategy for the synthesis of (Z)-cinnamate derivatives by visible-light-driven isomerization with Ir₂(ppy)₄Cl₂ as the photocatalyst. Z-Isomers of cinnamate derivatives could be readily obtained in good to excellent yields. These mild, convenient reaction conditions could be used as an effective methodology for the synthesis of coumarins.

• References and Notes

• 1a Ibrahim AM, Mansoor AA, Gross A, Ashfaq MK, Jacob M, Khan SI, Hamann MT. J. Nat. Prod. 2009; 72: 2141
• 1b Li T, Kongstad KT, Staerk D. J. Nat. Prod. 2019; 82: 249
• 1c Shuab R, Lone R, Koul KK. Acta Physiol. Plant. 2016; 38: 64
• 2a Fukuda H, Nishikawa K, Fukunaga Y, Okuda K, Kodama K, Matsumoto K, Kano A, Shindo M. Tetrahedron 2016; 72: 6492
• 2b Shing TK. M, Luk T, Lee CM. Tetrahedron 2006; 62: 6621
• 3 Siau WY, Zhang Y, Zhao Y. Top. Curr. Chem. 2012; 327: 33

For representative reports, see:
• 4a Kuwahara Y, Kango H, Yamashita H. ACS Catal. 2019; 9: 1993
• 4b Iwasaki R, Tanaka E, Ichihashi T, Idemoto Y, Endo K. J. Org. Chem. 2018; 83: 13574
• 4c Prabusankar G, Sathyanarayana A, Raju G, Nagababu C. Asian J. Org. Chem. 2017; 6: 1451
• 4d Surmiak SK, Doerenkamp C, Selter P, Peterlechner M, Schaefer AH, Eckert H, Studer A. Chem. Eur. J. 2017; 23: 6019
• 4e Maesing F, Nuesse H, Klingauf J, Studer A. Org. Lett. 2017; 19: 2658
• 4f Maesing F, Wang X, Nuesse H, Klingauf J, Studer A. Chem. Eur. J. 2017; 23: 6014
• 4g Du W, Gu Q, Li Y, Lin Z, Yang D. Org. lett. 2017; 19: 316
• 4h Jagtap SA, Sasaki T, Bhanage BM. J. Mol. Catal. A: Chem. 2016; 414: 78
• 4i Slack ED, Gabriel CM, Lipshutz BH. Angew. Chem. Int. Ed. 2014; 53: 14051
• 5 Jagtap SA, Bhanage BM. Mol. Catal. 2018; 460: 1

For representative reports, see:
• 6a Wissing M, Niehues M, Ravoo BJ, Studer A. Eur. J. Org. Chem. 2018; 3403
• 6b Li S.-S, Tao L, Wang F.-Z.-R, Liu Y.-M, Cao Y. Adv. Synth. Catal. 2016; 358: 1410
• 6c Liang S, Hammond GB, Xu B. Chem. Commun. 2016; 52: 6013
• 6d Mitsudome T, Yamamoto M, Maeno Z, Mizugaki T, Jitsukawa K, Kaneda K. J. Am. Chem. Soc. 2015; 137: 13452
• 6e Vasilikogiannaki E, Titilas I, Vassilikogiannakis G, Stratakis M. Chem. Commun. 2015; 51: 2384
• 7a Jagtap SA, Bhanage BM. ChemistrySelect 2018; 3: 713
• 7b Kiryutin AS, Yurkovskaya AV, Lukzen NN, Vieth H.-M, Ivanov KL. J. Chem. Phys. 2015; 143: 234203-1
• 8 Fedorov A, Liu H.-J, Lo H.-K, Coperet C. J. Am. Chem. Soc. 2016; 138: 16502
• 9 Richmond E, Moran J. J. Org. Chem. 2015; 80: 6922

For representative reports, see:
• 10a Kumar P, Dey R, Banerjee P. Org. Lett. 2018; 20: 5163
• 10b Dey R, Kumar P, Banerjee P. J. Org. Chem. 2018; 83: 5438
• 10c Weissenborn MJ, Loew SA, Borlinghaus N, Kuhn M, Kummer S, Rami F, Plietker B, Hauer B. ChemCatChem 2016; 8: 1636
• 10d Abascal NC, Lichtor PA, Giuliano MW, Miller SJ. Chem. Sci. 2014; 5: 4504
• 10e Zhang B, Lv C, Li W, Cui Z, Chen D, Cao F, Miao F, Zhou L. Chem. Pharm. Bull. 2015; 63: 255
• 11a Cheung CW, Zhurkin FE, Hu X. J. Am. Chem. Soc. 2015; 137: 4932
• 11b Lee H, Mane MV, Ryu H, Sahu D, Baik MH, Yi CS. J. Am. Chem. Soc. 2018; 140: 10289
• 12a Ahmed TS, Grubbs RH. Angew. Chem. Int. Ed. 2017; 56: 11213
• 12b Montgomery TP, Johns AM, Grubbs RH. Catalysts 2017; 7: 87
• 12c Herbert MB, Grubbs RH. Angew. Chem. Int. Ed. 2015; 54: 5018
• 12d Mann TJ, Speed AW, Schrock RR, Hoveyda AH. Angew. Chem. Int. Ed. 2013; 52: 8395
• 13 Hoffmann N. Chem. Rev. 2008; 108: 1052
• 14 Ciszewski LW, Rybicka-Jasinska K, Gryko D. Org. Biomol. Chem. 2019; 17: 432

For reviews, see:
• 15a Kaur N. J. Heterocycl. Chem. 2019; 56: 1141
• 15b Kaur N. Synth. Commun. 2018; 48: 1259
• 15c Kaur N. Curr. Org. Synth. 2018; 15: 298
• 15d Kaur N. Curr. Org. Synth. 2017; 14: 972
• 16 Hoffmann N. J. Photochem. Photobiol., C 2014; 19: 1
• 17a Bach T, Hehn JP. Angew. Chem. Int. Ed. 2011; 50: 1000
• 17b Ando Y. Yuki Gosei Kagaku Kyokaishi 2010; 68: 1067
• 17c Weaver JD, Singh K, Staig S. J. Am. Chem. Soc. 2014; 136: 5275
• 17d Fabry DC, Ronge MA, Rueping M. Chem. Eur. J. 2015; 21: 5350
• 18a Bhadra M, Kandambeth S, Sahoo MK, Addicoat M, Balaraman E, Banerjee R. J. Am. Chem. Soc. 2019; 141: 6152
• 18b Zhan K, Li Y. Catalysts 2017; 7: 337
• 18c Metternich JB, Gilmour R. Synlett 2016; 27: 2541
• 19a Lei T, Zhou C, Huang M.-Y, Zhao L.-M, Yang B, Ye C, Xiao H, Meng Q.-Y, Ramamurthy V, Tung C.-H, Wu L.-Z. Angew. Chem. Int. Ed. 2017; 56: 15407
• 19b Pagire SK, Hossain A, Traub L, Kerres S, Reiser O. Chem. Commun. 2017; 53: 12072
• 20 Typical Synthetic Procedure: A solution of 1a (0.4 mmol) and Ir₂(ppy)₄Cl₂ (1 mol%) in CH₃CN (4.0 mL) was irradiated at room temperature with 5 W blue LED for 24 h. The solvent was then removed under reduced pressure and the residue was purified by flash chromatography on silica gel to afford 2a (71%) as a white solid; mp 85–86.5 °C; R_f = 0.50 (petroleum/EtOAc, 1:2). ¹H NMR (400 MHz, CD₃OD): δ = 7.53 (d, J = 7.2 Hz, 2 H), 7.34–7.26 (m, 3 H), 6.75 (d, J = 12.8 Hz, 1 H), 6.02 (d, J = 12.8 Hz, 1 H). ¹³C NMR (100 MHz, CD₃OD): δ = 172.7, 138.3, 136.8, 130.4, 129.7, 129.4, 124.3.
• 21a Schultz DM, Yoon TP. Science 2014; 343: 1239176
• 21b Metternich JB, Gilmour R. J. Am. Chem. Soc. 2015; 137: 11254
• 22 Shaw MH, Twilton J, MacMillan DW. C. J. Org. Chem. 2016; 81: 6898
• 23a Srikrishna D, Dubey PK, Godugu C. Mini-Rev. Med. Chem. 2018; 18: 113
• 23b Pereira TM, Franco DP, Vitorio F, Kummerle AE. Curr. Trends Med. Chem. 2018; 18: 124

For representative reports, see:
• 24a Metternich JB, Gilmour R. J. Am. Chem. Soc. 2016; 138: 1040
• 24b Salem MA, Helal MH, Gouda MA, Ammar YA, El-Gaby MS. A, Abbas SY. Synth. Commun. 2018; 48: 1534
• 24c Moskvina VS, Khilya VP. Chem. Heterocycl. Compd. 2019; 55: 300
• 25 Typical Synthetic Procedure: A solution of 1u (0.4 mmol) and Ir₂(ppy)₄Cl₂ (1 mol%) in CH₃CN (4 mL) was irradiated at room temperature with a 5 W blue LED light for 72 h. The solvent was then removed under reduced pressure and the residue was purified by flash chromatography on silica gel to afford 2u (94%) as a white solid; mp 70–71.4 °C; R_f = 0.35 (petroleum–EtOAc, 3:1). ¹H NMR (400 MHz, CDCl₃): δ = 7.65 (d, J = 9.6 Hz, 1 H), 7.48 (m, 1 H), 7.42 (d, J = 7.6 Hz, 1 H), 7.27 (d, J = 8.4 Hz, 1 H), 7.21 (d, J = 7.2 Hz, 1 H), 6.39 (d, J = 9.6 Hz, 1 H). ¹³C NMR (100 MHz, CDCl₃): δ = 161.0, 154.3, 143.6, 132.0, 128.1, 124.6, 119.0, 117.1, 116.9.