Wei Liu‡
a,
Sheng Liu‡a,
Hongqi Xieab,
Zhixing Qinga,
Jianguo Zengab and
Pi Cheng*ab
aCo-Innovation Center of Hunan Province for Utilization of Functional Ingredients from Botanicals, Hunan Agricultural University, Changsha, Hunan 410128, China. E-mail: picheng55@126.com; Fax: +86 (731) 84686560
bPre-State Key Laboratory for Germplasm Innovation and Utilization of Crop, Hunan Agricultural University, Changsha, Hunan 410128, China
First published on 3rd February 2015
A visible light promoted and TBHP mediated oxidative reaction of N-2-alkynylphenyl α-amino carbonyl compounds to N-2-alkynylphenyl oxalic amides was developed. In the presence of CuBr and photocatalyst Ru(bpy)3Cl2·6H2O, the reaction proceeded smoothly to afford the corresponding oxalic amides under the irradiation of a 26 W compact fluorescence bulb at room temperature. Furthermore, N-2-alkynylphenyl oxalic amides could be subsequently transferred to 2-aryl indoles without an additional deacylation step through a favored 5-endo-dig N-cyclization process using AgNO3 as catalyst.
However, the CDC α-functionalization is not the predominant transformation of α-amino carbonyl during the radical-mediated protein damaging process in biological systems. In fact, the formation of α-carbon radicals of α-amino carbonyl units generally leads to the peptide backbone oxidations and fragmentations via oxalic acid derivatives and imine intermediates.3 Recently, Huo reported an aerobic auto-oxidative of glycine derivatives to obtained oxalic acid derivatives using dioxygen.7 Li and coworkers also demonstrated that visible light could promote the oxidation of N-aryl glycine ester to oxalic amide under dioxygen atmosphere with Ru(bpy)3Cl2 as photocatalyst.6g In the last half decade, visible light photoredox catalysis with transition metal complexes such as polypyridyl complexes of ruthenium has received much attention.8 Although that molecular oxygen has also been used as oxidant in the photoredox-mediated oxidative organic reactions,9 these reactions were often catalyzed by photocatalysts in combination with stoichiometric amounts of chemical oxidants such as BrCCl3,10 K2S2O8,11 Co(acac)3,12 m-dinitrobenzene (m-DNB)13 and aryl diazonium salts.14 Following our exploring process for the visible light promoted oxidative reaction,15 we found that visible light could promote the chemoselective oxidation of N-2-alkynylphenyl glycine ester 1a to the oxalic amide derivatives 2a with stoichiometric oxidant tert-butyl hydroperoxide (TBHP) (Scheme 1).
Initially, we hoped to access 2,3-diacylated indole 3a (Scheme 1) with N-2-alkynylphenyl glycine ester 1a as substrate via activated imine intermediate 4a followed by an intramolecular nucleophilic attack with the alkyne as reported by Patel.16 Thus, the reaction was firstly carried out in DMSO with aqueous TBHP as oxidant. Disappointingly, no desired indole derivative 3a was obtained. The oxalic derivative 2a, an isomer of 3a, was isolated in 3% yield along with the imine 4a and 2-ethynylaniline 5a as major products. To gain an insight into this type of oxidative reaction, we were encouraged to optimize the reaction parameters to promote the yield of oxalic acid derivative 2a under more mild conditions using photocatalytic strategy.
We opted to investigate the photoredox transformation of module substrates N-2-alkynylphenyl glycine ester 1a to oxalic amide 2a (Table 1). The ruthium complex Ru(bpy)3Cl2·6H2O and aqueous TBHP were chosen as the photocatalyst and oxidant respectively. When a solution of 1a, aqueous TBHP and another 20 mol% of CuBr in MeCN was irradiated with a household 26 W compact fluorescent bulb in the presence of 2 mol% of Ru(bpy)3Cl2·6H2O for 24 hours, the desired oxalic derivative 3a was obtained only in 4% isolated yield (Table 1, entry 1). When DMSO was used as solvent instead of MeCN in the reaction, the yield of 3a slightly increased to 8% (Table 1, entry 2). Under the above conditions, 2-ethynylaniline (5a) was isolated as major product. Thus, we considered that water from aqueous TBHP solution had a significant influence on the reaction outcome because of the side hydrolysis reaction. Next, a 6 M solution of TBHP in decane was used as oxidant instead of aqueous TBHP. Fortunately, the change of oxidant resulted in significantly increased yield of 2a to 35% and 42% in MeCN (Table 1, entry 3) and DMSO (Table 1, entry 4) respectively. Based on the above results, the reaction was then carried out in super dried solvent and led to slightly increased yield of 2a in MeCN (40% yield, entry 5) and DMSO (50% yield, entry 6) respectively as we expected. In contrast, replacement of Ru(bpy)3Cl2·6H2O with the complex Ru(phen)3Cl2 or Ir(ppy)3 resulted in decreased yield of 3a (entries 7 and 8). Notably, replacement of CuBr with FeCl3 or CuI provided 3a in decreased yields respectively (27% and 31% yield, entries 9 and 10). Furthermore, it should be noted that when CuBr was reduced to 10% mol or increased to 40% mol, only 41% and 35% yield of 3a were achieved respectively (entries 11 and 12). In contrast, no desired product was detected when the oxidation reaction was carried out under open air condition in MeCN with or without Cu salts (entries 13 and 14). Furthermore, when 1.1 eq. of K2CO3 was added to reaction system under open air condition, no transformation of the substrate was observed (entry 15). Interestingly, however, we found that a mixed solvents of MeCN and DMSO (VMeCN:
VDMSO = 4
:
1) could slightly improve the yield of 3a (57% yield, entry 16). Furthermore, when 4.0 eq. of K2S2O8 was used as oxidant instead of TBHP, 100% conversion of substrate was observed, but only 7% yield of desired compound 3a could be isolated with unidentified side products (entry 17). When 3.0 eq. of TBHP was used as oxidant, the desired compound 3a was still obtained in 55% yield (entry 18). Finally, to verify the role of photocatalyst and visible light in this type of oxidation reaction, control experiments were conducted. As outlined in Table 1, the requirement of a photocatalyst and visible light was verified, as significantly decreased outcome of 3a was observed without photocatalyst or visible light irradiation respectively (entries 19 and 20).
Entry | Oxidants | Solvent | Photocatalyst 2% mol | Metal salts 20% mol | Yieldb % |
---|---|---|---|---|---|
a Unless stated otherwise, the reaction was carried out with 1a (0.5 mmol), oxidants (4.0 eq.), metal salts (20 mol%), and photocatalyst (2 mol%) in the indicated solvent (2 mL) and irradiated with 26 W compact fluorescent lamp for 24 h at room temperature.b Isolated yields.c Super dried solvents.d 10 mol% CuBr was used.e 40 mol% CuBr was used.f 1.1 eq. of K2CO3 was added to the reaction system.g Mixed solvents of super dried MeCN and DMSO (VMeCN![]() ![]() ![]() ![]() |
|||||
1 | aq. TBHP | MeCN | Ru(bpy)3Cl2·6H2O | CuBr | 4 |
2 | aq. TBHP | DMSO | Ru(bpy)3Cl2·6H2O | CuBr | 8 |
3 | TBHP in decane | MeCN | Ru(bpy)3Cl2·6H2O | CuBr | 35 |
4 | TBHP in decane | DMSO | Ru(bpy)3Cl2·6H2O | CuBr | 42 |
5 | TBHP in decane | MeCNc | Ru(bpy)3Cl2·6H2O | CuBr | 40 |
6 | TBHP in decane | DMSOc | Ru(bpy)3Cl2·6H2O | CuBr | 50 |
7 | TBHP in decane | DMSOc | Ru(phen)3Cl2 | CuBr | 28 |
8 | TBHP in decane | DMSOc | Ir(ppy)3 | CuBr | 23 |
9 | TBHP in decane | DMSOc | Ru(bpy)3Cl2·6H2O | FeCl3 | 27 |
10 | TBHP in decane | DMSOc | Ru(bpy)3Cl2·6H2O | CuI | 31 |
11d | TBHP in decane | DMSOc | Ru(bpy)3Cl2·6H2O | CuBr | 41 |
12e | TBHP in decane | DMSOc | Ru(bpy)3Cl2·6H2O | CuBr | 35 |
13 | Open air | MeCN | Ru(bpy)3Cl2·6H2O | CuBr | 0 |
14 | Open air | MeCN | Ru(bpy)3Cl2·6H2O | — | 0 |
15f | Open air | MeCN | Ru(bpy)3Cl2·6H2O | — | 0 |
16 | TBHP in decane | DMSO/MeCNg | Ru(bpy)3Cl2·6H2O | CuBr | 57 |
17 | K2S2O8 | DMSO/MeCNg | Ru(bpy)3Cl2·6H2O | CuBr | 7 |
18h | TBHP in decane | DMSO/MeCNg | Ru(bpy)3Cl2·6H2O | CuBr | 55 |
19 | TBHP in decane | DMSO/MeCNg | — | CuBr | 16 |
20i | TBHP in decane | DMSO/MeCNg | Ru(bpy)3Cl2·6H2O | CuBr | 5 |
With the optimal oxidative conditions in hand, we next sought to determine the scope of N-2-alkynylphenyl glycine esters that can be employed in this photocatalytic oxidation protocol. As shown in Table 2, the scope of alkyne motif adjacent to the nitrogen atom was examined firstly. Generally, the electron-donating group on the phenyl ring of phenylacetylene motif decreased the oxidation efficiency. For example, when group R2 was (4-methylphenyl)ethynyl (2b, 52% yield, Table 2, entry 2), (4-ethylphenyl)ethynyl (2c, 49% yield, entry 3) or (4-methoxylphenyl)ethynyl (2d, 46% yield, entry 4), yields of the desired compounds were slightly lower than the model substrate (1a, 57% yield, entry 1). When an electron-withdrawing fluorine atom was substituted on the phenyl ring of phenylethynyl motif, moderate yields of desired compounds (2e, 50% yield, entry 5; 2f, 54% yield, entry 6) were achieved. Meanwhile, the optimized conditions could be applied to synthesize oxalic derivatives 2g (43% yield, entry 7) and 2h (41% yield, entry 8) that possessed tert-butylethynyl and 1-pentyl adjacent to the nitrogen atom respectively. At this point, we thought that substituent group R1 that directly connected on the N-phenyl ring possibly had more obvious effect on the yields of target compounds. Thus, we next examined the scope of R1. As outlined in Table 2, when the substituent at the N-phenyl ring was an electron-donating methyl group, a deleterious effect on overall reaction yield was observed (entries 9–13). Meanwhile, 2-ethynyl anilines 5, the byproduct from imine hydrolysis reaction, could be isolated with increased yields. To further verify the influence of R1, fluoro (R1 = F, entries 14–16) and chloro (R1 = Cl, entries 17 and 18) substituted substrates were chosen for the oxidation experiments. As we expected, the yields of target compounds 2n–2p (R1 = F, entries 14–16) increased obviously compared with corresponding methyl substituted substrates (R1 = Me) due to the strong electron-withdrawing ability of fluorine atom. With respect to the chloro substrates (R1 = Cl, entries 17 and 18), target compounds 2q–2r were provided in 49% and 38% yields which slightly decreased from the corresponding fluoro substrates (2j–2k). In most cases as shown in Table 2, 2-ethynyl anilines 5 could be isolated as the main byproducts. Furthermore, the structure of compound 2q was further established by X-ray diffraction study (Fig. 1).
Entry | R1 | R2 | Compounds 2 | Yield of 2b (%) | Compounds 5 | Yield of 5b (%) |
---|---|---|---|---|---|---|
a Conditions: compounds 2 (1.0 mmol), Ru(bpy)3Cl2·6H2O (2 mol%), CuBr (20 mol%), TBHP (4.0 eq.) in solvent (4 mL).b Isolated yield.c Hydrolysis products 5 were not isolated. | ||||||
1 | H | Phenyl | 2a | 57 | 5a | 11 |
2 | H | 4-Me phenyl | 2b | 52 | 5b | 19 |
3 | H | 4-Et phenyl | 2c | 49 | 5c | 18 |
4 | H | 4-MeO phenyl | 2d | 46 | 5d | 7 |
5 | H | 4-F phenyl | 2e | 50 | 5e | 15 |
6 | H | 2-F phenyl | 2f | 54 | 5f | 14 |
7c | H | t-Butyl | 2g | 43 | 5g | — |
8c | H | n-Propyl | 2h | 41 | 5h | — |
9 | 4-Me | Phenyl | 2i | 33 | 5i | 35 |
10 | 4-Me | 4-Me phenyl | 2j | 37 | 5j | 31 |
11 | 4-Me | 4-MeO phenyl | 2k | 42 | 5k | 28 |
12 | 4-Me | 4-F phenyl | 2l | 39 | 5l | 14 |
13 | 4-Me | 2-F phenyl | 2m | 32 | 5m | 23 |
14 | 4-F | Phenyl | 2n | 69 | 5n | 5 |
15 | 4-F | 4-Me phenyl | 2o | 55 | 5o | 8 |
16 | 4-F | 2-F phenyl | 2p | 56 | 5p | 9 |
17 | 4-Cl | 4-F phenyl | 2q | 49 | 5q | 14 |
18 | 4-Cl | 2-F phenyl | 2r | 38 | 5r | 12 |
To the best of our knowledge, the mechanism of TBHP mediated oxidation of α-amino carbonyl to oxalic derivatives was not clear. Based on our experimental results and previous available literature,8,9 a plausible catalytic mechanism for the visible light mediated oxidation process was proposed. As shown in Scheme 2, visible light excited Ru complex (*Ru(bpy)32+) was a strong reductant (E*II/III1/2 = − 0.81 V vs. SCE),8a which is oxidized by TBHP to generate Ru(bpy)33+ through a single electron transfer (SET) process. Meanwhile, Ru(bpy)33+, a powerful oxidant (EIII/II1/2 = 1.29 V vs. SCE),8a is able to oxidize substrate 1 to generate the corresponding amino radical cation 6 and Ru(bpy)32+ to start a new photocatalytic cycle. The relatively acidic tertiary amine radical cation 6 is deprotonated by a strongly basic species –OH and then further oxidized by an SET process resulting in the reactive imine intermediate, which coordinates with Cu+ to provide a more reactive intermediate 7. At this point, the activated imine carbon of intermediate 7 accepts the nucleophilic attack from TBHP or t-BuOH and H2O generated from the reaction system to give N-hybrid acetal peroxide 8 and byproduct aniline respectively. Intermediate 8 attends the second photocatalytic cycle and is then oxidized by Ru(bpy)33+ to give amino radical cation 9, which is deprotonated by basic species t-BuO− or OH− followed by another SET process to give imine peroxide 10. Intermediate 10 decomposes through a SET process to provide enolated anion 11, which is finally protonated to afford the target compound 2.
According to previous literature, the protocol for transition metal such as gold, palladium or copper catalyzed cycloisomerization of 2-alkynylaniline to indole derivatives was well established.17 However, when the nitrogen atom of 2-alkynylaniline was acylated, the acylated 2-alkynylaniline carried two kind of nucleophiles, the N atom of amide and O atom of enolated amide. Thus, the heterocyclization could take place either through O- or N-ring closure processes.18 With compounds 2 in hand, we next attempted to study the subsequent transformation of these oxalic amides. As shown in Scheme 3, when a solution of compound 2, 20 mol% of AgNO3 and 2.0 eq. of K2CO3 in MeCN was heated at 70 °C for 2 hours, the reaction proceeded through the favored 5-endo-dig N-cyclization process to provide 2-aryl indoles 12 in excellent isolated yield. Notably, no further step was needed to deacylate the N atom of indole. This transformation could be applied to synthesis of 3-aryl indole derivatives efficiently using cheap AgNO3 as catalyst.
In conclusion, we have demonstrated a visible light promoted oxidation of N-2-alkynylphenyl glycine ester to oxalic amides with Ru(bpy)3Cl2·6H2O as photocatalyst and TBHP as oxidant respectively. The possible oxidation mechanism was also proposed. Furthermore, the N-2-alkynylphenyl oxalic amides can be used to synthesize 2-aryl indoles efficiently using AgNO3 as catalyst without additional deacylation step.
Footnotes |
† Electronic supplementary information (ESI) available. CCDC 1041450. For ESI and crystallographic data in CIF or other electronic format. See DOI: 10.1039/c4ra17232a |
‡ These authors contributed equally to this work and should be considered joint first authors. |
This journal is © The Royal Society of Chemistry 2015 |