2022 Volume 70 Issue 11 Pages 765-768
We developed the addition reaction of α-silyl amines with benzalmalononitriles catalyzed by a Mg2+-conjugated pyrene catalyst under visible light irradiation. The catalytic activity of this complex was higher than pyrene alone, a Mg2+ Lewis acid alone, and the sum of these two independent catalytic elements. The observed enhancement in catalytic activity was likely due to electrostatic interactions of the Mg2+ Lewis acid with the pyrene radical anion, which was generated through photoinduced single electron transfer from α-silyl amines to the catalyst’s pyrene moiety.
Photoredox catalysts (PCs) enable conversions of various organic compounds through photoinduced single electron transfer (SET) between PC and substrates1) (Chart 1a). Since the activity of PC depends on the efficacy of SET in many cases, tuning electronic and photophysical properties of PC is important for the reactivity.2) Lewis acids can accelerate SET through complexation with PC’s heteroatoms.3) In the cases of polycyclic aromatic hydrocarbon (PAH) PCs without coordinating heteroatoms, however, Lewis acids barely affect their activity. Pyrene is one of the most attractive PAH-PCs due to its unique properties, such as high fluorescence quantum yield and the relatively long lifetime of its singlet excited states.4) In 2004, Fukuzumi reported that photocatalytic oxygenation of hexamethylbenzene was promoted by the combination of Sc(OTf)3 and pyrene5) (Chart 1b). The electron acceptor ability of the singlet excited state of pyrene was dramatically enhanced in the presence of Sc(OTf)3, resulting in the efficient SET from hexamethylbenzene to photoexcited pyrene to produce the radical cation from hexamethylbenzene. Inspired by this report, we expected to improve the activity of pyrene PC by conjugating with a Lewis acid. In this study, we found that a Mg2+-conjugated pyrene catalyst promoted the addition reaction of α-silyl amines to benzalmalononitriles (Chart 1c).
Since pyrene is known to undergo SET with dimethylaniline under light irradiation,6–10) we assessed if the addition of a Lewis acid would enhance the efficacy of SET between pyrene catalysts and α-silyl amine 1. The reaction between N-trimethylsilylmethylaniline (1a) and benzalmalononitrile (2a) was chosen as a model reaction11) (Table 1). The reaction of 1a and 2a in dichloromethane (DCM) under blue light irradiation afforded product 3a only in 8% in the absence of PC (entry 1). In the presence of pyrene, 3a was obtained in 26% yield (entry 2). Addition of Lewis acid Mg(OTf)2 in the absence of pyrene was not effective for improving the yield (20% yield, entry 3). A combination of pyrene and Mg(OTf)2, however, afforded 3a in 54% yield (entry 4). To further improve the reactivity, PC1 was designed containing a pyrene moiety linked to a bisoxazoline moiety. Though yield using PC1 and Mg(OTf)2 (55% yield, entry 5) was almost unchanged from entry 4, it was clear that the addition of the Lewis acid had an acceleration effect (entry 6). After examining other Lewis acids, Mg(NTf2)2 was selected as the best additive (81% yield, entry 7). When the solvent was changed to acetone, product 3a was obtained in 90% (entry 8). When pyrene was used instead of PC1 in the presence of Mg(NTf2)2 with or without L1, yields were lower than that of entry 8, confirming positive effects of the pyrene-Mg2+ conjugation (entries 9 and 10).
![]() |
Under the optimized conditions, we studied the scope of α-silyl amines (Chart 2). N-Trimethylsilylmethylanilines bearing a methyl (3b) or a bromide substituent (3c) afforded the corresponding product in high yield. Aliphatic tertiary amine provided product 3d in 32% yield. Then, the scope of radical acceptors was investigated. Methyl (3e), bromine (3f), or methoxy (3g) substituted benzalmalononitrile provided the products in good yield. When pyrene was used instead of PC1, yields were also lower for 3b and 3c (82 and 66%). These results suggested that PC1 is superior to pyrene.
The proposed reaction mechanism is shown in Chart 3. First, pyrene moiety of PC1 was excited under blue light irradiation. Then, SET from aniline 1 to the pyrene moiety provides radical ion pairs (Eox (1a·+/1a))=+0.6 V vs. SCE, E(*Py/Py·−)=+1.2 V vs. SCE).12,13) This step is promoted by electrostatic interaction with pyrene radical anion and Mg2+ ion 4. The silyl group is easily removed from radical cation species 5 to give α-aminomethyl radical 6. Radical 6 reacts with 2 to afford 7. Radical 7 is reduced by radical anion species 4 to give 8, which is subsequently trapped by a silyl cation to afford 9. The silyl group is removed in workup to give product 3.
Here we found that a Mg2+-conjugated pyrene catalyst promoted the addition reaction of α-silyl amines to benzalmalononitriles under visible light irradiation. The Mg2+-conjugated pyrene catalyst exhibited high reactivity and was superior to those using Mg2+ salt and pyrene as independent catalysts. This acceleration effect is likely due to the electrostatic stabilization of the pyrene radical anion by the Mg2+ Lewis acid. Currently, the effect of linking pyrene and Mg2+ source is not that great (Table 1 entry 8 vs. entries 9 and 10), but we believe that more efficient catalysts can be developed by further investigating linkers and Lewis acid sources.
1H-NMR spectra were recorded on JEOL ECX500 (500 MHz for 1H-NMR), and JEOL ECS400 (400 MHz for 1H-NMR) spectrometer. For 1H-NMR, chemical shifts are reported in ppm with the solvent resonance as the internal standard (1H: CDCl3, δ 7.26, 13C: CDCl3, δ 77.0). Data are reported as follows: s = singlet, d = doublet, t = triplet, m = multiplet; coupling constants in Hz; integration. Deuterated solvent was purchased from Kanto chemical (Tokyo, Japan). Column chromatographies were performed with silica gel Merck 60 (230–400 mesh ASTM), Biotage Isolera One and Biotage SNAP Ultra or Yamazen Universal Column. Super dehydrated acetone was purchased from Wako Pure Chemical Corporation (Osaka, Japan), and Mg(NTf2)2 and benzalmalononitrile were purchased from Tokyo Chemical Industry Co., Ltd. (TCI, Tokyo, Japan). They were used as received. A Valore VBP-L24-C2 with 38W LED lamp (VBL-SE150-BBB (430)) was used as the 430 nm light source.
SynthesisMagnesium bistriflimide (11.7 mg, 0.020 mmol, 0.2 equivalent (equiv.)) was dissolved in super dehydrated acetone (1.0 mL) in a screw-capped vial under argon atmosphere. Then, PC1 (7.6 mg, 0.020 mmol, 0.2 equiv.) was added into the vial, and the solution was stirred at room temperature under argon for an hour. Benzalmalononitrile 2 (0.10 mmol, 1.0 equiv.) and α-silyl amine 1 (0.10 mmol, 1.0 equiv.) were added to the solution. The vial was subjected to blue LED irradiation for 16 h at room temperature. Then, the reaction mixture was evaporated and 1,1,2,2-tetrachloroethane was added as an internal standard. The crude mixture was purified by flash column chromatography on silica gel to give product 3.
2-(2-(Methyl(phenyl)amino)-1-phenylethyl)malononitrile (3a)14)1H-NMR (391.78 MHz, CDCl3): δ = 7.46–7.40 (5H, m), 7.31 (2H, dd, J = 8.2, 7.5 Hz), 6.86 (1H, t, J = 7.5 Hz), 6.81 (2H, d, J = 8.2 Hz), 4.25 (1H, d, J = 4.5 Hz), 3.90 (1H, dd, J = 14.8, 10.3 Hz), 3.71 (1H, dd, J = 14.8, 4.9 Hz), 3.65–3.60 (1H, m), 2.98 (3H, s). 13C-NMR (125.77 MHz, CDCl3): δ = 148.6, 135.0, 129.7, 129.5, 129.5, 128.2, 118.9, 113.9, 112.3, 111.9, 55.5, 44.9, 41.0, 27.1.
2-(2-(Methyl(p-tolyl)amino)-1-phenylethyl)malononitrile (3b)14)1H-NMR (500.16 MHz, CDCl3): δ = 7.45–7.40 (5H, m), 7.12 (2H, d, J = 8.6 Hz), 6.75 (2H, d, J = 8.6 Hz), 4.30 (1H, d, J = 4.0 Hz), 3.82 (1H, dd, J = 14.6, 10.6 Hz), 3.66–3.57 (2H, m), 2.95 (3H, s), 2.29 (3H, s). 13C-NMR (125.77 MHz, CDCl3): δ = 146.7, 135.0, 130.2, 129.5, 129.4, 128.7, 128.3, 114.8, 112.4, 111.9, 55.7, 44.9, 41.5, 27.1, 20.4.
2-(2-((4-Bromophenyl)(methyl)amino)-1-phenylethyl)malononitrile (3c)1H-NMR (500.16 MHz, CDCl3): δ = 7.47–7.42 (3H, m), 7.39–7.36 (4H, m), 6.65 (2H, d, J = 8.6 Hz), 4.16 (1H, d, J = 4.6 Hz), 3.92 (1H, dd, J = 14.9, 9.7 Hz), 3.69 (1H, dd, J = 14.9, 5.7 Hz), 3.61–3.57 (1H, m), 2.93 (3H, s). 13C-NMR (125.77 MHz, CDCl3): δ = 147.5, 134.9, 132.4, 129.6, 128.1, 115.3, 112.1, 111.8, 110.9, 55.4, 44.7, 40.9, 27.2. IR(neat): 3005, 2956, 2922, 2852, 1715, 1362, 1078, 904, cm−1. HRMS (ESI+): Calcd for [C18H16BrN3 + Na]+ 376.0420. Found 376.0420.
2-(1-Phenyl-2-(4-phenylpiperidin-1-yl)ethyl)malononitrile (3d)1H-NMR (500.16 MHz, CDCl3): δ = 7.44–7.37 (5H, m), 7.32 (2H, dd, J = 7.4 Hz), 7.25–7.20 (3H, m), 4.71 (1H, d, J = 4.1 Hz), 3.42 (1H, ddd, J = 8.0, 4.1, 4.1 Hz), 3.13 (1H, d, J = 11.5 Hz), 3.06–3.00 (2H, m), 2.67 (1H, dd, J = 13.2, 4.1 Hz), 2.56–2.45 (2H, m), 2.14 (1H, t, J = 9.7 Hz), 1.90–1.74 (4H, m). 13C-NMR (125.77 MHz, CDCl3): δ = 146.0, 135.6, 129.4, 129.2, 128.6, 128.2, 127.0, 126.5, 113.0, 112.1, 59.5, 56.5, 53.3, 44.5, 42.5, 33.8, 33.4, 27.2. IR(neat): 3005, 2918, 2847, 1715, 1362, 1083 cm−1. HRMS (ESI+): Calcd for [C22H23N3 + Na]+ 352.1784. Found 352.1783.
2-(2-(Methyl(phenyl)amino)-1-(p-tolyl)ethyl)malononitrile (3e)15)1H-NMR (500.16 MHz, CDCl3): δ = 7.32–7.24 (6H, m), 6.86 (1H, t, J = 7.2 Hz), 6.81 (2H, d, J = 8.6 Hz), 4.22 (1H, d, J = 4.6 Hz), 3.88 (1H, dd, J = 15.1, 10.6 Hz), 3.68 (1H, dd, J = 15.1, 5.2 Hz), 3.62–3.58 (1H, m), 2.98 (3H, s), 2.39 (3H, s). 13C-NMR (98.52 MHz, CDCl3): δ = 148.7, 139.4, 131.9, 130.2, 129.7, 128.1, 118.8, 113.9, 112.4, 112.0, 55.5, 44.6, 40.9, 27.2, 21.3.
2-(1-(4-Bromophenyl)-2-(methyl(phenyl)amino)ethyl)malononitrile (3f)1H-NMR (500.16 MHz, CDCl3): δ = 7.59 (2H, d, J = 8.6 Hz), 7.33–7.28 (4H, m), 6.87 (1H, t, J = 7.2 Hz), 6.80 (2H, d, J = 8.0 Hz), 4.24 (1H, d, J = 4.6 Hz), 3.85 (1H, dd, J = 14.9, 10.3 Hz), 3.68 (1H, dd, J = 14.9, 5.2 Hz), 3.62-3.58 (1H, m), 2.98 (3H, s). 13C-NMR (125.77 MHz, CDCl3): δ = 148.5, 133.8, 132.7, 129.9, 129.8, 123.8, 119.2, 114.1, 112.1, 111.6, 55.3, 44.4, 41.1, 26.9. IR(neat): 2961, 2913, 2847, 1716, 1541, 1507, 1361, 1078 cm−1. HRMS (ESI+): Calcd for [C18H16BrN3 + Na]+ 376.0420. Found 376.0420.
2-(1-(4-Methoxyphenyl)-2-(methyl(phenyl)amino)ethyl)malononitrile (3g)15)1H-NMR (500.16 MHz, CDCl3): δ = 7.34–7.29 (4H, m), 6.97 (2H, d, J = 8.6 Hz), 6.86 (1H, t, J = 7.2 Hz), 6.80 (2H, d, J = 8.6 Hz), 4.21 (1H, d, J = 4.6 Hz), 3.89–3.84 (4H, m), 3.67 (1H, dd, J = 14.9, 5.2 Hz), 3.61–3.57 (1H, m), 2.98 (3H, s). 13C-NMR (125.77 MHz, CDCl3): δ = 160.3, 148.7, 129.7, 129.4, 126.8, 118.8, 114.8, 113.9, 112.4, 112.0, 55.5, 55.5, 44.3, 40.9, 27.4.
This work was supported in part by JSPS KAKENHI Grant Numbers JP17H06442 (M.K.) (Hybrid Catalysis), JP20H05843 (Dynamic Exciton) and JP21K15220 (H.M.).
The authors declare no conflict of interest.
This article contains supplementary materials.