Light-Driven Intermolecular Charge Transfer Induced Reactivity of Ethynylbenziodoxol(on)e and Phenols

Abstract

Ethynylbenziodoxol(on)es (EBXs) have been widely used in organic synthesis as electrophilic alkyne-transfer reagents involving carbon- and heteroatom-based nucleophiles. However, potential reactions of EBXs with phenols remain uninvestigated. Here, we present the formation of (Z)-2-iodovinyl phenyl ethers with excellent regio- and stereoselectivity through the reactivity between EBXs and phenols driven by visible light. We propose that this light-activated transformation proceeds through electron donor−acceptor complexes to enable new reactivity beyond existing mechanisms for alkynylation of carbon- and heteroatom-based nucleophiles. This operationally robust process was employed for the synthesis of diverse (Z)-2-iodovinyl phenyl ethers through irradiating a solution containing a phenyl-EBX, a phenol, and the base Cs₂CO₃ with a commercially available blue LED at room temperature. The (Z)-2-iodovinyl phenyl ether products can be further stereospecifically functionalized to form trisubstituted alkenes, demonstrating the potential of these products en route to chemical complexity.

Graphical Abstract

INTRODUCTION

Ethynylbenziodoxol(on)es (EBXs)^1–6 are electrophilic reagents that can be used for the alkynylation of carbon- and heteroatom-based nucleophiles.^7–19 Specifically, alkynylation of carbon,^{7−11,14–17} sulfur,^12,13 nitrogen,¹⁸ and phosphorus¹⁹ nucleophiles has been successfully developed (Figure 1a). In the process, an alkynyl group is transferred to the nucleophile and 2-iodobenzoic acid is commonly obtained as a stoichiometric byproduct. Other reactions involving EBX reagents such as copper-catalyzed oxyalkynylation of diazo compounds^20,21 (Figure 1d) and Pd(II)-catalyzed transformation of alkynylbenziodoxoles with N-aryl imines and carboxylic acids have also been reported (Figure 1b,c).^22–24 Although reactions of carbon- and heteroatom-based nucleophiles with EBX electrophiles have been extensively investigated, there remains unexplored chemistry using phenols as nucleophiles. From a synthetic standpoint, the development of strategies using EBXs and phenols is highly desirable.

Figure 1.

Development of reactions using ethynylbenziodoxol(on)e (EBX) reagents. (a) Alkynylation of carbon- and heteroatom-based nucleophiles with EBX reagents. (b) Palladium-catalyzed reaction of N-aryl imines and EBX reagents to make multisubstituted furans. (c) Upper: palladium-catalyzed reaction of EBX reagents and carboxylic acids; lower: palladium-catalyzed three-component reaction of EBX reagents, carboxylic acids, and imines. (d) Copper-catalyzed oxy-alkynylation of diazo compounds using EBX reagents. (e) This work: reactions of EBX reagents and phenols under visible light irradiation. Nu: nucleophile.

Phenol-containing compounds are frequently found in plant secondary metabolites. Notable examples include vanillin, thymol, and eugenol, which have been traditionally utilized as important food additives for color, flavor, and astringency.^25,26 Additionally, they are prevalent in a wide range of bioactive natural products such as steroid hormones, thyroid hormones, and monoamine neurotransmitters, such as estrone, estradiol, estriol,²⁷ triiodothyronine,²⁸ and serotonin.²⁹ Phenol motifs are also important in medicines. Isoproterenol is used for the treatment of heart block and levalbuterol is used for the treatment of asthma.³⁰ Industrially, phenol-containing compounds are also one of the most versatile and important organic commodity chemicals.²⁵ Therefore, direct transformation involving conversion of phenols into a variety of products with a wide range of properties is highly desirable.

Recently, we reported a visible-light-promoted C−S cross-coupling reaction of thiophenols and aryl halides in the presence of a mild base, Cs₂CO₃.³¹ The observed reactivity was enabled through visible-light-induced electron transfer within the electron donor−acceptor (EDA) complex; for example, an electron is transferred from the thiophenoxide (donor) to the aryl halide (acceptor). Inspired by this work, we hypothesized that in the presence of a suitable acceptor, phenol could also be employed as an electron donor in light-driven transformations utilizing EDA complexes.³²

Herein, we report the formation of EDA complexes^33–35 between nucleophilic phenols and electrophilic EBX reagents. Utilizing Cs₂CO₃ as a base and under blue LED irradiation, phenols and EBX reagents react to form (Z)-2-iodovinyl phenyl ethers with excellent regio- and stereoselectivity (Figure 1e). To the best of our knowledge, this represents the first example of reactivity between EBX reagents and phenols that exploits visible light irradiation at room temperature. Notably, in forming the (Z)-2-iodovinyl phenyl ether products, we discovered an unprecedented cleavage of the phenyl−I bond of EBX reagents to yield vinyl halide products. Vinyl halides are important intermediates in organic synthesis, and we demonstrate these (Z)-2-iodovinyl phenyl ethers as versatile synthetic precursors to undergo a range of transition-metal-catalyzed cross-coupling reactions, such as Buchwald− Hartwig,³⁶ Ullmann,³⁷ and Suzuki³⁸ reactions, to generate functionally diverse trisubstituted alkenes.

RESULTS AND DISCUSSION

We commenced our studies by evaluating 4-methoxyphenol 1b and phenyl-EBX 2a in similar reaction conditions to those developed for our C−S cross-coupling reaction (Table 1).³¹ Gratifyingly, reactivity between 1b and 2a was observed after 14 h of blue LED irradiation at room temperature (Table 1, entry 1), and the (Z)-2-iodovinyl phenyl ether product 3b was isolated in 46% yield. Compared to the typical alkynylation products and 2-iodobenzoic acid byproduct (Figure 1a) obtained in previous EBX reactions,^7–19 we observed a unique phenyl−I bond cleavage to synthesize 3b with benzoic acid or 14 as byproducts; this new reactivity is attributed to the photoinduced electron transfer in the EDA complex. Subsequent optimization experiments led to the selection of acetonitrile as the solvent (Table 1, entries 1−4), a loading of 1b of 1.5 equiv (entries 4−7), and a duration of 14 h irradiation to achieve good yield (entries 4, 8, 9). Among multiple bases investigated (entries 4, 11−14), the use of 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD) resulted in the highest yield (78%), while only trace product was detected in the absence of base (Table S2). Interestingly, 3b was also formed in the absence of light when the strong organic base TBD was used, albeit at lower yield (58%, Table S2). Cs₂CO₃, which is less expensive than TBD, is also an excellent base for this transformation (74%, Table 1, entry 4) and was therefore used throughout the remainder of this study. Using the inorganic Cs₂CO₃ base, light irradiation is mandatory for reactivity (Table S2), while the presence of oxygen significantly reduces product formation (Table 1, entry 10).

Table 1.

Initial Investigation of the Synthesis of (Z)-2-Iodovinyl Phenyl Ether^a

entry	t (h)	equiv 1b	solvent	base	yield (%)
1	14	1.5	DMSO	Cs2CO3	47 (46†)
2	14	1.5	DMAc	Cs2CO3	44
3	14	1.5	DMF	Cs2CO3	47
4	14	1.5	CH3CN	Cs2CO3	75 (74b)
5	14	1.0	CH3CN	Cs2CO3	46
6	14	2.0	CH3CN	Cs2CO3	61
7	14	3.0	CH3CN	Cs2CO3	51
8	2	1.5	CH3CN	Cs2CO3	35
9	4	1.5	CH3CN	Cs2CO3	48
10	14	1.5	CH3CN	Cs2CO3	24c
11	14	1.5	CH3CN	DBU	20
12	14	1.5	CH3CN	MTBD	40
13	14	1.5	CH3CN	TBD	79 (78b)
14	14	1.5	CH3CN	TMG	trace
15	14	1.5	DMSO	TBD	53
16	14	1.5	THF	TBD	37
17	14	1.5	THF	TMG	20

^aReaction conditions: 1b (0.30 mmol), 2a (0.20 mmol), base (0.30 mmol), solvent (1.5 mL), room temperature, N₂. Yields are determined by ¹H NMR using 1,3,5-trimethoxybenzene as an internal standard.

^bIsolated yield.

^cReaction carried out under air. t: time; DBU: 1,8-diazabicyclo[5.4.0]undec-7-ene; MTBD: 7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene; TBD: 1,5,7-triazabicyclo[4.4.0]dec-5-ene; TMG: 1,1,3,3-tetramethylguanidine; DMAc: N,N′-dimethylace-tamide; DMF: N,N′-dimethylformamide; DMSO: dimethyl sulfoxide; THF: tetrahydrofuran.

Using the optimized reaction conditions, a variety of phenols were transformed into the corresponding 2-iodovinyl phenyl ethers in moderate to good yields with high regio- and stereoselectivities (Scheme 1). Overall, no significant electronic effects of the phenol on the reaction was observed. Explicitly, the reaction tolerated both electron-donating and electron-withdrawing substituents on the phenol, including amine (3o, 3p, 3v,), methoxy (3b, 3c, 3x, 3y), and trimethylsilyl (3d) groups as well as halogen (3e−l, 3w, 3x), aldehyde (3q), ester (3r, 3v), and trifluoromethyl (3s) groups. A phenol bearing a boronic acid pinacol ester group also afforded the corresponding 2-iodovinyl phenyl ether derivative 3a, albeit in modest yield (45%). When an allylic-containing substrate was employed, TBD was observed to be the best base for this transformation and provided the corresponding 2-iodovinyl phenyl ether derivative 3ag (Supporting Information, SI, p S8) in 52% yield. Notably, a protected amino acid could also be transformed to the desired product 3ah (SI, p S8) in 40% yield using TBD as a base, demonstrating the high functional group tolerance of the transformation. Additionally, hydroxyl-heteroaryls, ubiquitous among biologically active molecules and pharmaceutical products, similarly reacted to provide the desired products 3ab, 3ad, 3ae, and 3af in 61, 70, 52, and 40% yields, respectively. The 2-iodovinyl phenyl ether products obtained were assigned as the Z-isomer, as supported from X-ray diffraction analysis of 3aa (Scheme 1). Density functional theory (DFT) calculations corroborated that the Z-isomer of 3b is 3.4 kcal/mol more stable than the corresponding E-isomer. Different alcohols were also explored. With CF₃CH₂OH, we were able to obtain the desired product at 31% yield, albeit with a mixture of Z- and E-isomers, while no desired products were isolated with other alcohols, such as 1,1,1,3,3,3-hexafluoro-2-propanol, benzyl alcohol, and ethanol (Table S3).

Scheme 1. Scope of Phenols.

^aTBD used as a base. ^bPerformed in the dark and obtained as ¹H NMR yield. ^cV_acetonitrile:V_benzene = 1:1, see Table S1. Reaction conditions: 1 (0.30mmol), 2 (0.20 mmol), base (0.30 mmol), solvent (1.5 mL), room temperature, N₂. Isolated yield reported. TBD: 1,5,7-triazabicyclo[4.4.0]dec-5-ene; BPin: 4,4,5,5-tetramethyl-1,3,2-dioxaborolane.

The substrate scope of EBXs was also explored (Scheme 2). A broad range of functional groups on EBXs, including nitro (4a), ester (4b), cyano (4c), trifluoromethyl (4d, 4h, 4k), methyl (4e, 4i, 4n), halogen (4f, 4g, 4j, 4o−q), and methoxy (4r−t) groups, were well tolerated in this process, affording the corresponding (Z)-2-iodovinyl phenyl ether products in moderate to good yields. EBX reagents bearing an alkyl group could also be implemented (4u). No desired product, however, could be obtained with triisopropylsilyl-EBXs (see SI, p S7. Markedly, phenyl-EBXs derived from other substituted 2-iodobenzoic acids (e.g., 2-fluoro-6-iodo and 2-iodo-5-nitrobenzoic acid, see SI) similarly reacted with 1b to produce the same product 3b in 87% and 42% yields, respectively. This protocol was also applicable to bistrifluoromethyl-substituted benziodoxole (see SI, p S7), which reacted with 1b to provide 3b in 52% yield. We note that the use of mixed solvent (V_acetonitrile:V_benzene = 1:1) in a number of cases has improved the product yield (see Schemes 1 and 2).

Scheme 2. Scope of Ethynylbenziodoxol(on)es.

^aV_acetonitrile:V_benzene = 1:1, see Table S1. Reaction conditions: 1 (0.30 mmol), 2 (0.20 mmol), base (0.30 mmol), solvent (1.5 mL), room temperature, N₂. Isolated yield reported.

To demonstrate potential synthetic value of the (Z)-2-iodovinyl phenyl ether products, we investigated their chemistry for subsequent transformations (Figure 2). The vinyl ether 3r efficiently engaged in a series of stereospecific palladium-catalyzed cross-coupling reactions, including Suzuki couplings (5, 7), a Sonogashira reaction (6), an Ullmann reaction (8), and a Buchwald−Hartwig amidation (11). Furthermore, Pd-catalyzed methoxycarbonylation of 3r generated 9 in 92% yield. Notably, Pd-catalyzed intramolecular C−H activation furnished benzo[b]furan 10 in 83% yield. Using nuclear Overhauser effect spectroscopy (NOESY), we found that compounds 3r, 5, 6, 8, and 9 are Z-isomers and that 5, 6, 8, and 9 retained their Z-conformations after metal-catalyzed transformations. On the contrary, the amidation product of 11 was determined to be an E-isomer (see SI, p S163). Overall, these results demonstrate the potential of this reaction to provide a unified approach for the preparation of trisubstituted alkenes from readily available phenols and EBX reagents.

Figure 2.

Transition-metal-catalyzed transformations of (Z)-2-iodovinyl phenyl ethers. (a) Suzuki coupling; (b) Sogonashira coupling; (c) Suzuki coupling; (d) Ullmann coupling; (e) carbonylation; (f) C−H activation; (g) Buchwald−Hartwig coupling. PPh₃: triphenylphosphine; TMHD: 2,2,6,6-tetramethyl-3,5-heptanedione; dppf: 1,1′-ferrocenediyl-bis(diphenylphosphine); Xantphos: 4,5-bis(diphenylphosphino)-9,9-dimethylxan-thene.

A series of experiments and computations were performed to gain insight into this reactivity between EBXs and phenols (Figure 3). In the absence of light, we observed that phenol nucleophilically adds across the triple bond of EBX 2a to form a vinylbenziodoxolone intermediate (12), which is isolatable (Figure 3a). This nucleophilic addition step was computed to be exergonic by 28.3 kcal/mol. We then subjected 12 to our standard reaction condition. Upon combining vinylbenziodox-olone (12), 1b, and Cs₂CO₃ in MeCN, the solution gradually turned yellow, distinctly different from the colorless parent compounds (Figure S7). In accord with the C−S coupling we examined previously,³¹ this significant red-shift in the UV−vis absorption suggests the formation of an EDA charge-transfer complex. Time-dependent (TD)-DFT calculations are consistent with the EDA hypothesis that the lowest energy excitation of the vinylbenziodoxolone 12−phenoxide complex has extensive charge-transfer character (Figure 3a, inset). Specifically, 75% and 14% of the lowest energy excitation iscomposed of ^π_HOMO−^π_LUMO and ^π_HOMO−^π_LUMO+1 transitions, respectively, where the π_HOMO stems from phenoxide (donor) while the π_LUMO and π_LUMO+1 are derived from vinyl-benziodoxolone 12 (acceptor).

Figure 3.

Mechanistic studies. (a) Proposed mechanism for the reactivity between EBXs and phenols under the influence of visible light. Thermochemistry evaluation was conducted at the M06/Lanl2dz level of theory with CPCM-described solvation in acetonitrile; CPCM: conductor-like polarizable continuum model. Time-dependent density functional theory (TD-DFT) calculations to model the lowest excited states of the EDA complex were conducted at the CAM-B3LYP/Lanl2dz/CPCM-MeCN level of theory. HOMO: highest occupied molecular orbital; LUMO: lowest unoccupied molecular orbital; f: predicted oscillator strength. λ_calc: predicted maximum wavelength of absorption of the lowest excited state. (b) ¹³C-labeling experiment. (c) Experiment involving deuterated phenol. (d) Experiment involving vinylbenziodoxolone and sodium phenoxide. (e) Light-driven reaction between EBX 2a, diboronate ester, and pyridine to yield (iodoethynyl)benzene.

To further investigate the EDA complex, we analyzed a Job plot to characterize the complexation between 12b and sodium phenoxide (Figures S8 and S9). Sodium phenoxide was used as the donor instead of (phenol + Cs₂CO₃) because the phenol is not completely deprotonated by Cs₂CO₃ and thus the concentration of the donor is unknown. Using UV−vis spectroscopy, we monitored the absorbance values at 400 nm (corresponding to the EDA complex’s absorption) and plotted them as a function of molar fraction of sodium phenoxide. We obtained a parabolic curve with a maximum absorbance value at 50% mole fraction of sodium phenoxide, indicating a 1:1 EDA complex between 12b and sodium phenoxide. Furthermore, setting [12b] at a constant value of 0.005 M, we measured the absorbance values at 400 nm as we varied [sodium phenoxide] from 0.005 M to 0.045 M. Using the Benesi−Hildebrand method,^39,40 we plotted 1/absorbance versus 1/[sodium phenoxide] and obtained a linear relationship (Figures S10 and S11). Through linear regression, the y-intercept and slope values allowed for estimation of the association constant of the EDA complex (K_EDA) as 38.9 in DMSO.

Upon blue LED irradiation of the yellow solution (containing vinylbenziodoxolone (12), phenol 1b, and Cs₂CO₃), we obtained the desired product 3b, whereas only a trace amount of product was obtained in the absence of light (Scheme S3). Thus, this observation supports the role of light in phenyl−I bond cleavage to yield product 3b. We suggest that in the EDA complex photoinduced electron transfer occurs from the phenoxide (donor) to the vinylbenziodox-olone 12 (acceptor). DFT calculations support that the one-electron reduction of vinylbenziodoxolone 12 leads to spontaneous phenyl−I bond cleavage to generate 3b and radical intermediates 13 (Figure 3a); in support of the formation of 13a, a small amount of benzoic acid could be isolated (see SI). The radical intermediate 13 can then quench to form a C−C bond, which upon rearrangement and elimination of water finally yields the 6H-benzo[c]chromen-6-one byproduct 14. The free energy change for this step was computed to be −79.7 kcal/mol, and the byproduct 14 was successfully isolated (20% yield, see SI). To further investigate the observed phenyl−I bond cleavage, we utilized an organic phenoxazine-based photoredox catalyst,^41–44 developed by our group as a strong excited state reductant, to directly reduce vinylbenziodoxolone 12 under blue LED irradiation. Consistent with our proposed mechanism, 3b was isolated in 47% yield (Scheme S3), which supports our proposition that the phenyl−I bond can be reductively cleaved.

The formation of byproduct 14 may suggest that at least 2.0 equiv of 1b is needed for reaction completion; however, we empirically determined that a 1.5 equiv loading of 1b was optimum, as higher or lower loadings led to decreased yields (Table 1, entries 5, 6, and 7). Also, at 1.5 equiv loading, in many cases we obtained a yield greater than the theoretical limit of 75% (Schemes 1 and 2). Thus, we hypothesize that the formation of byproduct 14 only contributes in part to 3b’s formation, as evident from isolation of 14 in 20% yield. It is also interesting to note that when TBD was added to a solution containing vinylbenziodoxolone and phenol 1b, a more intense yellow solution was observed (as compared to the addition of Cs₂CO₃). However, in this instance, no light was needed to promote the formation of 3b (Scheme S3) in good yield. Thus, with TBD, we propose that thermally activated electron transfer results in the phenyl−I bond cleavage.⁴⁵

In addition, we examined reactions using ¹³C-labeled EBX 2a* and phenol-OD. As shown in Figure 3b, the reaction of phenol 1b and ¹³C-labeled EBX 2a* afforded 3b-¹³C, where ¹H and ¹³C NMR spectra confirmed the bonding of the phenoxy group to the ¹³C atom and showed a vinylic hydrogen signal with a J_C−H coupling constant of 10 Hz (see SI). Furthermore, the reaction of deuterated phenol(−OD) with EBX 2a afforded 3aj (Figure 3c), where the vinylic hydrogen contains about 62% deuterium (Figure 3c). Using the structurally related vinylbenziodoxolone (15)⁴⁶ and sodium phenoxide, we demonstrated conceptually similar phenyl−I bond cleavage to generate (2-iodovinyl)benzene 16 in 50% yield under blue LED irradiation, while no reactivity was observed in the dark (Figure 3d). This experiment established that phenoxide (deprotonated phenol) is the electron donor and that photoinduced electron transfer can also promote phenyl−I cleavage in structurally related vinylbenziodoxolone 15. Lastly, by applying mechanistic concepts elucidated above, we utilized a diboronate ester and pyridine as electron donor species,⁴⁷ resulting in reductive phenyl−I cleavage of EBX 2a upon visible light irradiation to produce (iodoethynyl)benzene (17) in 55% yield (Figure 3e).

CONCLUSIONS

We report new reactivity between ethynylbenziodoxol(on)es and phenols to afford a diverse collection of (Z)-2-iodovinyl phenyl ether derivatives under irradiation with visible light. To generate the (Z)-2-iodovinyl phenyl ether products, we propose a photoinduced electron transfer step involving an intermediary vinylbenziodoxolone−phenoxide EDA complex that subsequently leads to the unprecedented phenyl−I bond cleavage. Using the (Z)-2-iodovinyl phenyl ethers as precursors, a series of transition-metal-catalyzed chemistries were performed to access unique trisubstituted alkene derivatives with significant control of stereo- and regiochemistry, underscoring the potential of this transformation in the pharmaceutical and agrochemical industries.