Optimization of the Reaction Conditions for the Synthesis of Dihydrobenzofuran Neolignans

Abstract

We have optimized the experimental conditions for the silver(I)-promoted oxidative coupling of methyl p-coumarate (I) and methyl ferulate (II), which is the most frequently used methodology to synthesize the bioactive dihydrobenzofuran neolignans 1 ((±)-trans-dehydrodicoumarate dimethyl ester) and 2 ((±)-trans-dehydrodiferulate dimethyl ester). Most of the tested conditions affected the conversion (i.e., the consumption of I and II) and the selectivity (i.e., the percentage of I and II that was converted into 1 and 2, respectively), so the optimized conditions were the ones that afforded the best balance between conversion and selectivity. Silver(I) oxide (0.5 equiv.) is the most efficient oxidant agent amongst the silver(I) reagents that were tested to convert methyl esters I and II into compounds 1 and 2, respectively. Acetonitrile, which has not yet been reported as a solvent for this reaction, provided the best balance between conversion and selectivity, besides being “greener” than other solvents that are more often employed (e.g., dichloromethane and benzene). Under the optimized conditions, the reaction time decreased from 20 to 4 h without significantly impacting the conversion and selectivity.

Keywords:
benzofuran derivatives; oxidative coupling; phenylpropanoids; radical intermediates

Introduction

Dihydrobenzofuran neolignans (DBNs) are compounds that are biosynthesized by plants as part of their secondary metabolism. DBNs result from oxidative coupling of two propenylphenols (C₆C₃), which are joined through C8–C5’ and C7–O4’ bonds.^1,2 Natural and synthetic DBNs exhibit diverse biological activities, such as anti-inflammatory,³ antioxidant,³ cytotoxic,⁴ schistosomicidal,⁵ leishmanicidal,⁶ and insecticidal actions,⁷ among others.

Despite the various synthetic methodologies that have been reported in the literature for the synthesis of DBNs (e.g., intramolecular C-H insertion⁸ and sequential cross-metathesis/isomerization/allylboration),⁹ oxidative coupling of phenylpropanoids, which mimics DBN biosynthesis in plants and affords the dihydrobenzofuran skeleton in only one synthetic step under mild conditions, is still the most commonly used.² Silver(I) oxide is the classic oxidant that is employed to promote this coupling. In the mechanism proposed in the literature¹⁰ for the silver(I)-promoted oxidative coupling of phenylpropanoids, the first step consists of 4–OH bond homolysis, to generate the phenoxy radical intermediate B (Scheme 1). In preliminary studies, mass spectrometry has been used to track changes in the reaction mixture with time. These studies suggested that the phenoxy radical originates from oxidation of a phenolate anion A and consequent silver(I) reduction to silver(0), which precipitates in the reaction vessel internal walls (data not published) (Scheme 1). Next, a C5’–C8 bond is supposedly established between two phenoxy radical units, which is followed by ring closure (resulting from O4-C7’ bond formation) and further tautomerization.

Scheme 1
Mechanism proposed for the silver(I) oxide-promoted oxidative coupling of phenylpropanoids to give dihydrobenzofuran neolignans (adapted from Daquino et al.).¹⁰

The silver(I) oxide-promoted oxidative coupling of phenylpropanoids is diastereoselective and gives DBNs as racemic mixtures of trans-enantiomers in 20-40% yield.^10,11 Many efforts have been dedicated to replacing silver(I) oxide in this reaction to obtain DBN in higher stereoselectivity and yields. To this end, iron salts,¹² Ru and Rh complexes,¹³ the horseradish peroxidase (HRP)/H₂O₂ system,¹⁴ and laccases have been employed.¹⁵ However, stereoselectivity obtained for other metals than silver(I) in the oxidative coupling of phenylpropanoids to give DBNs has not been clearly described.^12,13 On the other hand, some of these methods have provided good enantioselectivity (e.g., HRP/H₂O₂ system and laccases) and increased yields. Nevertheless, the high cost or the experimental complexity^8,9 as compared to the use of silver(I) oxide still makes the methodology based on silver(I) oxide more attractive.

In this paper, we have systematically investigated reaction conditions to optimize the silver(I)-promoted oxidative coupling of phenylpropanoids.

Experimental

General

The ¹H and ¹³C nuclear magnetic resonance (NMR) analyses were performed on a Bruker Advance DRX400 spectrometer (Karlsruhe, Germany; 400.13 MHz for ¹H and 100.61 MHz for ¹³C), as described in the literature.¹¹ The samples were dissolved in acetone-d₆ (99.8 atom% D, Sigma-Aldrich, St. Louis, USA); tetramethylsilane (TMS, 0.01%) was used as internal standard. The chemical shifts (ẟ) were expressed in parts per million (ppm) in relation to the residual solvent peak, and the multiplicity of signals was deduced according to the signals obtained in spectrum. The coupling constants (J, in Hz) were calculated in comparison to the same signal peaks, and the relative integral was deduced according to the number of hydrogens.

The infrared (IR) spectra were recorded on a spectrophotometer IR Spectrum Two (PerkinElmer, Beaconsfield, UK). The samples were macerated in potassium bromide (99.99%, Sigma-Aldrich, St. Louis, USA) and fused in a pellet under pressure.

The mass spectra were recorded on a gas chromatograph-mass spectrometer (GC-MS) Shimadzu QP2010-Plus (Kyoto, Japan) system equipped with an AOC-20i autosampler and an RTX-5MS (Restek Co., Bellefonte, PA, USA) fused silica capillary (30 m × 0.25 mm inner diameter, 0.25-μm film thickness). The electron ionization mode was used at 70 eV. Helium (99.999%) was employed as the carrier gas at a constant flow of 1.0 mL min^-1. The injection volume was 0.1 μL (split ratio of 1:10). The injector and the ion-source temperatures were set at 280 and 300 °C, respectively. The oven temperature was programmed to rise from 110 to 300 °C at 10 °C min^-1 and then held at 300 °C for 5 min. The mass spectra were taken with a scan interval of 0.5 s for masses ranging from 40 to 600 Da.

Synthesis of the dihydrobenzofuran neolignans 1 and 2

The dihydrobenzofuran neolignans (DBNs) 1 and 2, which were used as standards, were synthesized by oxidative coupling of methyl p-coumarate (I) and methyl ferulate (II), respectively, in the presence of silver(I) oxide, as oxidant, as reported previously (Scheme 2).^4,7,11,16

Scheme 2
Synthesis of the dihydrobenzofuran neolignans 1 and 2 by silver(I)-oxide oxidative coupling of methyl p-coumarate (I) and methyl ferulate (II), respectively (adapted from references 4,7,11,16).

Methyl p-coumarate (I) was obtained from coumaric acid (CA, 3.0 g, 18.3 mmol) (Sigma-Aldrich, St. Louis, USA), which was dissolved in 30 mL of methanol in a 100-mL round-bottom flask, followed by addition of 1 mL of sulfuric acid. Methyl ferulate (II) was synthesized from ferulic acid (FA, 3.0 g, 15.4 mmol) (Sigma-Aldrich, St. Louis, USA), which was dissolved in 180 mL of methanol in a 500-mL round-bottom flask, followed by addition of 3 mL of sulfuric acid. The reaction mixtures were kept under reflux at 85 °C for 4 h. After methanol was removed under reduced pressure, the resulting crude products were dissolved in ethyl acetate and extracted with water and a saturated NaHCO₃ solution (3 × 30 mL). The organic phase was dried over MgSO₄ and filtered. Ethyl acetate was removed from the samples under reduced pressure to afford compounds I and II as a yellowish powder and a brown powder in 94 and 95% yield, respectively.

Next, 3.0 g of methyl ester I (16.8 mmol) or II (14.4 mmol) were added to a two-neck 100-mL round-bottom flask covered with aluminum foil and dissolved in 40 mL of a benzene/acetone (6:4) mixture under magnetic stirring and nitrogen atmosphere. Next, silver(I) oxide (8.5 mmol) (Sigma-Aldrich, St. Louis, USA) was added, and the reaction progress was monitored by thin layer chromatography (TLC) for 4 h. The reaction mixture was filtered off, and the solvent was removed under reduced pressure. The crude products were purified by column chromatography as previously reported,¹¹ to afford compounds 1 (35% yield) and 2 (42% yield) as mixtures of trans-enantiomers.

Methyl p-coumarate (I)

¹H NMR (400 MHz, CDCl₃) ẟ 7.65 (d, J 16.1 Hz, 1H, H-1’), 7.45 (d, J 8.6 Hz, 2H, H-2 and H-6), 6.85 (d, J 8.6 Hz, 2H, H-3 and H-5), 6.31 (d, J 16.1 Hz, 1H, H-2’), 5.03 (br s, 1H, OH), 3.80 (s, 3H, H-4’); EI-MS (70 eV, m/z, relative intensity / %): 178 [M^•+] (100), 147 [M^•+ - •OCH₃] (90), 119 [M^•+ - •OCH₃ - CO] (90), 91 [M^•+ - •OCH₃ - CO - CO] (25). The NMR data are in agreement with the literature.¹⁷

Methyl ferulate (II)

¹H NMR (400 MHz, CDCl₃) ẟ 7.63 (d, J 16.0 Hz, 1H, H-1’), 7.08 (dd, J 1.9, 8.1 Hz, 1H, H-2), 7.04 (d, J 1.9 Hz, 1H, H-6), 6.93 (d, J 8.1 Hz, 1H, H-3), 6.30 (d, J 16.0 Hz, 1H, H-2’), 5.89 (br s, 1H, OH), 3.93 (s, 3H, H-7), 3.80 (s, 3H, H-4’); EI-MS (70 eV, m/z, relative intensity / %): 208 [M^•+] (100), 177 [M^•+ - •OCH₃] (60), 145 [M^•+ - •OCH₃ - CH₃OH] (40), 117 [M^•+ - •OCH₃ - CH₃OH - CO] (22). The NMR data are in agreement with the literature.¹⁷

(±)-trans-Dehydrodicoumarate dimethyl ester (1)

Yellow powder; mp 105-107 °C; IR (KBr pellet) ν_max / cm^-1 3384 (-OH), 2955 (C_sp2-H), 1715 (C=O), 1602 (C=C), 1490 (C=C), 1440 (C=C), 1240 (C-O), 1115 (C-O); ¹H NMR (400 MHz, acetone-d₆) ẟ 7.65 (1H, d, J 16.0, H7’), 7.62 (1H, br s, H6’), 7.50 (1H, dd, J 1.5 and 8.1 Hz, H2’), 7.28 (2H, dd, J 1.8 and 6.8 Hz, H2 and H6), 6.91 (1H, d, J 8.1 Hz, H3’), 6.87 (2H, dd, J 1.8 and 6.8 Hz, H5), 6.41 (1H, d, J 16.0 Hz, H8’), 6.03 (1H, d, J 7.3 Hz, H7), 4.40 (1H, d, J 7.3 Hz, H8), 3.81 (3H, s, H10), 3.73 (3H, s, H10’); ¹³C NMR (100 MHz, acetone-d₆) ẟ 172.0 (C, C9), 168.2 (C, C9’), 162.6 (C, C4), 159.1 (C, C4’), 145.5 (CH, C7’), 132.1 (C, C1’), 131.3 (C, C1), 128.9 (CH, C2 and C6), 128.0 (CH, C2’), 126.9 (C), 126.5 (CH, C6’), 116.9 (CH, C3 and C5), 116.5 (CH, C8’), 111.2 (CH, C3’), 88.2 (CH, C7), 56.0 (CH, C8), 53.4 (CH₃, C10’), 52.0 (CH₃, C10); EI-MS (70 eV, m/z, relative intensity / %): 354 [M^•+] (26), 322 [M^•+ - CH₃OH] (30), 290 [M^•+ - CH₃OH - CH₃OH] (100). The NMR data are in agreement with the literature.¹¹

(±)-trans-Dehydrodiferulate dimethyl ester (2)

Yellow oil; IR (KBr pellet) ν_max / cm^-1 3395 (-OH), 2951 (C_sp²-H), 1740 (C=O), 1723 (C=O), 1644 (C=C), 1594 (C=C), 1523 (C=C), 1432 (C=C), 1269 (C-O), 1165 (C-O), 1140 (C-O), 1095 (C-O), 1036 (C-O), 982, 941, 853; ¹H NMR (400 MHz, acetone-d₆) ẟ 7.63 (1H, d, J 16.1 Hz, H7’), 7.33 (1H, br s, H2’), 7.29 (1H, br s, H6’), 7.10 (1H, d, J 1.7 Hz, H2), 6.92 (1H, dd, J 8.1, 1.7 Hz, H6), 6.84 (1H, d, J 8.1 Hz, H5), 6.44 (1H, d, J 16.1 Hz, H8’), 6.04 (1H, d, J 7.3 Hz, H7), 4.47 (1H, d, J 7.3 Hz, H8), 3.92 (3H, s, H11’), 3.84 (3H, s, H11’), 3.81 (3H, s, H10’), 3.73 (3H, s, H10’); ¹³C NMR (100 MHz, acetone-d₆) ẟ 172.1 (C, C9), 168.2 (C, C9’), 151.5 (C, C4’), 149.0 (C, C3), 148.5 (C, C4), 146.3 (C, C3’), 145.9 (CH, C7’), 132.5 (C, C1), 129.9 (C, C1’), 127.8 (C, C5’), 120.7 (CH, C6), 119.4 (CH, C6’), 116.7 (CH, C8’), 116.3 (CH, C5), 113.9 (CH, C2’), 111.2 (CH, C2), 88.8 (CH, C7), 56.9 (CH₃, C11’), 56.8 (CH₃, C11’), 56.4 (CH, C8), 53.5 (CH₃, C10’), 52.0 (CH₃, C10’); EI-MS (70 eV, m/z, relative intensity / %): 414 [M^•+] (5), 382 [M^•+ - CH₃OH] (90), 350 [M^•+ - CH₃OH - CH₃OH] (70). The NMR data are in agreement with the literature.¹¹

GC analyses

The consumption of methyl esters I and II (i.e., conversion) and the formation of the DBNs 1 and 2 (i.e., selectivity) were monitored on a gas chromatograph GC-2010 Plus Shimadzu (Kyoto, Japan) fitted with a flame ionization detector (FID) detector and an RTX-5 fused silica capillary column (30 m × 0.25 mm i.d., 0.25-μm film thickness). For this purpose, 20-μL aliquots were sampled from the reaction mixture at pre-determined times (1, 2, 3, 4, and 20 h) and transferred to a 1-mL Eppendorf plastic vial. The samples were filtered off to remove the oxidant; added to a 1-mL vial containing 200 μL of ethyl acetate for dilution; and injected into the gas chromatograph. The temperature was programmed to increase from 70 to 310 °C at 15 °C min^-1 in 30 min. The substances of interest were identified by comparison with retention times of standard samples.

Results and Discussion

Optimization of the nature and concentration of the oxidant

Although silver(I) oxide is the most commonly employed oxidant in the synthesis of dihydrobenzofuran neolignans via oxidative coupling of phenylpropanoids, different stoichiometries and reaction conditions have been reported. For instance, Lemière et al.¹⁸ used 1 equiv. of substrate II and 0.5 equiv. of the oxidant (Ag₂O) in a dry benzene/acetone (5:3 v/v) mixture at room temperature for 20 h, which afforded DBN 2 in 31% yield ([substrate] = 0.18 mol L^-1; [oxidant] = 0.09 mol L^-1). Pieters et al.¹⁷ used 1 equiv. of substrate I and 0.7 equiv. of Ag₂O in a dry benzene/acetone (7:5 v/v) mixture ([substrate] = 0.23 mol L^-1; [oxidant] = 0.16 mol L^-1) at room temperature for 65 h, to obtain DBN 1 in 23% yield. In turn, Orlandi et al.¹⁹ employed 1 equiv. of substrate II and 1.6 equiv. of Ag₂O under inert atmosphere (Ar or N₂) in dry CH₂Cl₂ ([substrate] = 0.10 mol L^-1; [oxidant] = 0.16 mol L^-1) for 20-24 h at room temperature, which gave DBN 2 in 32% yield. Daquino et al.¹⁰ used 1 equiv. of methyl caffeate (substrate) and 1 equiv. of Ag₂O in CHCl₃ at room temperature for 2 h, to obtain the corresponding DBN in 17.6% yield ([substrate] = [oxidant] = 0.07 mol L^-1). Therefore, to date, there are no standard conditions regarding the use of silver(I) oxide as oxidant in the oxidative coupling of phenylpropanoids.

We started optimizing the oxidative coupling reaction conditions by investigating the oxidant. Sako et al.²⁰ evaluated how different silver(I) reagents (e.g., AgOAc, Ag₂O, Ag₂CO₃, and AgNO₃) affected the synthesis of the resveratrol dehydrodimer and obtained yields between 4 and 97%. Thus, we decided to examine how the use of silver(I) reagents other than Ag₂O impacted the oxidative coupling of methyl esters I and II. First, we employed 1 equiv. of methyl ester I or II (0.56 mmol) and 1 equiv. of Ag⁺ ions (0.56 mmol) dissolved in 2 mL of benzene/acetone (6:4 v/v) under magnetic stirring. Besides Ag₂O and AgNO₃ (Merck, Darmstadt, Germany), which have already been reported²⁰ as oxidants in this reaction, we also tested AgBr (Merck, Darmstadt, Germany), AgOCOCF₃ (Aldrich, St. Louis, USA), and AgOSO₂CF₃ (Merck, Darmstadt, Germany). The conversion (C) and the selectivity (S) obtained from GC-FID analyses are shown in Tables 1 (methyl ester I → DBN 1) and 2 (methyl ester II → DBN 2).

Table 1
Methyl ester I (methyl p-coumarate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 1 as a function of the silver(I) reagent, used as oxidant

0.56 mmol 1 equiv. Ag+		Ag2O 0.14 mol L-1			AgNO3 0.28 mol L-1			AgBr 0.28 mol L-1			AgOCOCF3 0.28 mol L-1			AgOSO2CF3 0.28 mol L-1
Reaction time / h		C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1		9.8	61.7		1.0	5.0		0.8	3.8		11.8	88.3		2.1	6.5
2		17.2	94.0		0.3	20.0		0.4	12.6		16.8	91.6		9.5	1.2
3		24.8	82.0		0.4	14.0		0.2	55.0		20.3	83.3		13.2	11.8
4		32.7	94.0		0.5	32.0		0.3	6.7		20.5	91.1		37.4	0.0
20		35.3	66.4		1.4	22.9		1.5	7.1		20.0	78.2		5.7	2.8

Table 2
Methyl ester II (methyl ferulate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 2 as a function of the silver(I) reagent, used as oxidant

0.56 mmol 1 equiv. Ag+		Ag2O 0.14 mol L-1			AgNO3 0.28 mol L-1			AgBr 0.28 mol L-1			AgOCOCF3 0.28 mol L-1			AgOSO2CF3 0.28 mol L-1
Reaction time / h		C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1		29.6	62.5		5.0	0.0		4.6	1.9		20.2	49.8		59.2	1.4
2		52.9	85.0		4.0	5.8		4.4	3.7		25.8	61.9		75.1	1.4
3		63.7	83.1		4.2	11.0		4.1	3.7		25.1	63.9		95.7	1.6
4		67.7	85.8		4.3	10.8		4.4	2.3		26.3	62.8		99.9	0.0
20		52.7	62.1		10.7	2.4		5.2	1.9		16.7	50.6		27.7	1.6

Under the same experimental conditions, methyl ferulate (II) conversion (i.e., the percentage of methyl ester that was consumed) was higher than methyl p-coumarate (I) conversion, regardless of the oxidant. On the other hand, selectivity (i.e., the percentage of methyl ester that was converted into the corresponding DBN) was higher for methyl p-coumarate (I) as compared to methyl ferulate (II). This difference in the conversion of methyl esters I and II and in the selectivity toward DBNs 1 and 2, respectively, could explain, at least in principle, the lack of standard conditions for the silver(I) oxide-promoted oxidative coupling reaction. However, we found that Ag₂O was the most efficient oxidant because it provided the best balance between conversion and selectivity, especially for a reaction time of 4 h: 32.7% conversion of methyl ester I and 94.0% selectivity toward DBN 1; 67.7% conversion of methyl ester II and 85.8% selectivity toward DBN 2. It is noteworthy that the use of AgCO₂CF₃ afforded good conversion and selectivity percentages even though these values were not as high as for Ag₂O.

In the literature,^2,19 Ag₂O between 0.5 and 1.6 equiv. has been employed in the oxidative coupling of phenylpropanoids (oxidant concentrations ranging between 0.07 and 0.16 mol L^-1). Here, we evaluated how different Ag₂O concentrations affected the oxidative coupling of methyl esters I and II to produce DBNs 1 and 2, respectively. To this end, we dissolved the methyl ester (1 equiv., 0.56 mmol) in a benzene/acetone (6:4 v/v, 2 mL) mixture and added the resulting solution to a round-bottom flask containing the oxidant and kept the reaction mixture under N₂ atmosphere for 20 h, covered with an aluminum foil. The oxidant impacted both the conversion and selectivity, as depicted in Tables 3 and 4. Larger Ag₂O amounts increased methyl ester conversion. On the other hand, larger Ag₂O amounts also decreased selectivity, which indicated that side reactions probably occurred when larger Ag₂O amounts were used. The best balance between conversion and selectivity was obtained for 0.5 equiv. of Ag₂O.

Table 3
Methyl ester I (methyl p-coumarate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 1 as a function of the Ag₂O amount

Reaction time / h	0.25 equiv. 0.035 mol L-1			0.5 equiv. 0.07 mol L-1			1 equiv. 0.14 mol L-1			5 equiv. 0.70 mol L-1
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	16.8	60.5		28.1	69.9		100.0	47.0		73.0	71.9
2	22.4	74.0		67.5	70.5		87.8	66.0		83.0	61.2
3	23.6	78.4		36.8	60.3		100.0	44.3		93.9	49.6
4	24.0	73.2		34.3	69.7		98.7	25.7		100.0	70.0
20	25.7	78.1		35.3	60.6		100.0	8.8		100.0	15.1

Table 4
Methyl ester II (methyl ferulate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 2 as a function of the Ag₂O amount

Reaction time / h	0.25 equiv. 0.035 mol L-1			0.5 equiv. 0.07 mol L-1			1 equiv. 0.14 mol L-1			5 equiv. 0.70 mol L-1
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	21.7	57.9		31.3	75.0		74.1	33.0		95.4	15.9
2	25.0	55.4		39.4	66.3		44.8	43.2		93.8	16.0
3	25.1	65.0		45.8	55.8		49.0	43.5		92.4	11.1
4	28.6	53.6		44.4	63.1		95.1	25.6		97.7	20.8
20	27.1	66.9		46.5	69.2		94.2	13.0		100.0	11.0

Optimization of the solvent

Several solvents have been used in oxidative coupling reactions of phenylpropanoids. Dichloromethane and benzene/acetone and dichloromethane mixtures have been reported most often. Here, besides these two solvent systems, we also tested methanol and acetonitrile. The choice of these solvents was made on the basis of substrate solubility and the fact that methanol and acetonitrile are relatively “greener” as compared to other solvents: they present decreased toxicity and can be recycled for use in other reactions more times than other solvents.²¹ In these experiments, we dissolved 1 equiv. of the substrate (methyl ester I or II) in 2 mL of the tested solvent and added the resulting solution to a 10-mL two-neck flask containing 0.5 equiv. of Ag₂O or AgCO₂CF₃, the two best oxidant agents for this reaction. We covered the reaction vessel with aluminum foil and kept the reaction mixture under stirring and N₂ atmosphere. The results are listed in Tables 5, 6, 7, and 8.

Table 5
Methyl ester I (methyl p-coumarate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 1 when Ag₂O (0.5 equiv.) is used as oxidant in different solvents

Reaction time / h	DCM			Benzene/acetone 6:4 (v/v)			MeOH			MeCN
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	5.5	34.5		5.2	41.7		5.8	28.4		26.2	60.7
2	7.1	50.4		10.0	55.5		8.6	36.6		35.0	62.7
3	15.5	37.4		14.7	59.7		13.0	38.3		36.8	60.3
4	15.3	63.8		12.7	57.2		12.7	43.7		34.3	69.6
20	27.5	73.0		35.3	66.4		30.2	51.0		35.3	60.6

• DCM: dichloromethane; MeOH: methanol; MeCN: acetonitrile.

Table 6
Methyl ester II (methyl ferulate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 2 when Ag₂O (0.5 equiv.) is used as oxidant in different solvents

Reaction time / h	DCM			Benzene/acetone 6:4 (v/v)			MeOH			MeCN
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	19.1	71.6		24.2	65.0		30.6	56.0		18.6	71.1
2	40.1	44.4		27.6	62.4		44.6	55.0		32.6	69.6
3	45.1	37.6		33.0	61.3		49.6	46.0		37.0	64.9
4	44.7	38.5		31.8	58.1		54.7	41.7		39.1	69.1
20	49.5	37.1		38.6	53.3		60.5	11.9		45.0	67.5

• DCM: dichloromethane; MeOH: methanol; MeCN: acetonitrile.

Table 7
Methyl ester I (methyl p-coumarate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 1 when AgCO₂CF₃ (0.5 equiv.) is used as oxidant in different solvents

Reaction time / h	DCM			Benzene/acetone 6:4 (v/v)			MeOH			MeCN
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	3.6	30.6		8.1	50.8		4.4	46.9		7.7	52.1
2	7.5	45.4		10.5	55.3		9.6	63.6		4.7	56.6
3	6.2	14.7		10.3	57.2		8.6	61.7		8.4	72.8
4	9.0	45.1		10.6	53.8		10.2	70.9		11.6	72.6
20	3.9	48.5		10.3	55.8		9.2	65.8		10.4	65.9

• DCM: dichloromethane; MeOH: methanol; MeCN: acetonitrile.

Table 8
Methyl ester II (methyl ferulate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignan (DBN) 2 when AgCO₂CF₃ (0.5 equiv.) is used as oxidant in different solvents

Reaction time / h	DCM			Benzene/acetone 6:4 (v/v)			MeOH			MeCN
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	1.8	13.2		13.9	51.2		15.0	56.5		18.9	70.2
2	4.5	23.2		13.6	49.4		15.8	56.2		14.7	61.4
3	6.8	27.8		22.8	42.5		20.2	46.5		17.2	59.7
4	7.0	30.0		15.3	37.6		16.2	35.3		17.2	55.6
20	24.0	43.1		15.9	41.6		30.2	29.7		7.9	48.0

• DCM: dichloromethane; MeOH: methanol; MeCN: acetonitrile.

In acetonitrile (Sigma-Aldrich, St. Louis, USA), the reactions proceeded with good yields and increased selectivity as compared to the other solvents. Silver(I) produces diverse complexes with weak coordinating ligands (e.g., halogens), besides being weakly oxophilic and forming numerous complexes with donor groups, such as S, Se, P, As, and N donor ligands.²² The increase in silver(I) valence due to formation of a complex when it is dissolved in cyanide solvents has been previously reported.²³ In this sense, the increased selectivity and good conversion rate of the substrates obtained here suggested that acetonitrile, which is a good electron donor, can form highly stable complexes with silver(I) during the oxidation. Both oxidants provided higher selectivity toward the DBNs 1 and 2. Therefore, acetonitrile, which has not been previously reported as solvent for this reaction, is the best solvent to be used with Ag₂O, which in turn provides much more expressive results as compared to AgCO₂CF₃.

Optimization of the temperature

We also evaluated how the temperature influences the oxidative coupling of methyl esters I and II. Usually, these reactions are reported^2,19 to occur at room temperature. However, Sako et al.²⁰ obtained viniferin in 40% yield when they conducted AgOAc-promoted oxidative coupling in methanol under heating at 50 °C for 2 h. Here, we investigated how three different temperatures (0 °C, room temperature, and reflux at 85 °C) affected the oxidative coupling of methyl esters I and II by using Ag₂O, as oxidant, and acetonitrile, the best solvent for these reactions (Table 9).

Table 9
Methyl ester I (methyl p-coumarate) and methyl ester II (methyl ferulate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignans (DBN) 1 and 2, respectively, when Ag₂O (0.5 equiv.) was used as oxidant in acetonitrile at different temperatures

Reaction time / h	Room temperature						0 °C						Reflux
	I			II			I			II			I			II
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	49.7	76.3		29.8	76.0		6.6	31.0		12.9	61.3		28.6	73.7		63.7	78.3
2	47.6	72.7		39.0	76.6		6.9	23.8		18.9	67.9		29.8	66.1		68.8	58.8
3	57.1	73.2		38.6	78.4		9.8	44.5		23.1	71.4		26.9	63.1		66.3	74.3
4	50.1	75.3		41.7	77.1		11.9	50.0		35.5	77.3		30.0	56.3		65.7	71.0
20	49.1	71.7		45.4	77.3		44.9	71.1		46.9	71.3		44.4	42.1		68.0	64.1

Both conversion and selectivity decreased at 0 °C for both methyl esters. On the other hand, reflux conditions provided better homogenization of the reaction mixture and gave good selectivity and conversion. However, reaction time of 20 h decreased selectivity probably because side reactions took place, to afford products other than the DBNs 1 and 2.

Optimization of the reaction time

Pieters et al.¹⁷ reported that the optimum time for the silver(I)-promoted oxidative coupling of methyl esters I and II and analogs is 20 h. Nevertheless, our results revealed that this time could be optimized to 4 h in acetonitrile and 0.5 equiv. of Ag₂O without decreasing conversion and selectivity. The disadvantages of longer reaction times include reduced selectivity due to formation of undesired products.

The reflux condition was the most efficient among the tested conditions, especially in the case of methyl ester II (Table 9). However, we selected room temperature as the most adequate temperature for this reaction because it requires less energy and provides higher selectivity than reflux conditions.

Effect of radical initiator and inhibitor

AIBN (azobisisobutyronitrile) is a radical initiator due to its decomposition into 1-cyanoprop-1-yl radical, which is driven by N₂ elimination (Scheme 3).²⁴ AIBN has been used in many reactions in which radical intermediate species are involved.²⁵ Because the mechanism proposed in the literature^10,26 for the silver(I)-oxide oxidative coupling of phenylpropanoids involves the formation of radical intermediate species (Scheme 1), we decided to investigate how AIBN affected the synthesis of DBNs 1 and 2 from methyl esters I and II, respectively (Table 10). AIBN significantly increased methyl ester conversion in all the reaction times as compared to the reaction conducted under the same experimental conditions without AIBN addition. However, the selectivity toward DBNs 1 and 2 decreased drastically. This data indicated that AIBN addition to the reaction mixture generated other radical species that resulted in the formation of products other than the DBNs 1 and 2 even when the optimized conditions for oxidant, solvent, temperature, and reaction time were employed.

Scheme 3
Mechanism of 1-cyanoisoprop-1-yl radical formation from AIBN (adapted from reference 24).

Table 10
Methyl ester I (methyl p-coumarate) and methyl ester II (methyl ferulate) conversion (C) and selectivity (S) toward dihydrobenzofuran neolignans (DBN) 1 and 2, respectively, when Ag₂O (0.5 equiv., as oxidant) in acetonitrile at room temperature was used together with AIBN (radical initiator) or isoquinoline (radical inhibitor

Reaction time / h	AIBN						Isoquinoline
	I			II			I			II
	C / %	S / %		C / %	S / %		C / %	S / %		C / %	S / %
1	80.0	5.7		70.2	7.1		57.9	49.1		38.1	35.0
2	87.6	5.8		73.8	8.2		60.1	46.5		45.6	41.2
3	85.8	6.1		76.9	7.0		60.5	47.7		43.7	41.5
4	88.8	5.6		78.3	6.4		60.2	45.7		47.8	39.0
20	86.3	5.6		78.6	10.1		62.5	47.1		48.6	40.1

• AIBN: azobisisobutyronitrile.

Finally, we decided to verify how isoquinoline impacted the synthesis of DBNs 1 and 2 from methyl esters I and II, respectively. Isoquinoline acts as a radical inhibitor in organic synthesis.²⁷ Because the silver(I) oxide-promoted oxidative coupling of methyl esters I and II to produce 1 and 2, respectively, has been suggested to involve intermediate radical species, we expected that isoquinoline addition would decrease the conversion of methyl esters I and II and the selectivity toward DBNs 1 and 2, respectively. Surprisingly, although the selectivity decreased, the conversion of methyl esters I and II increased when isoquinoline was added. These results did not clarify the involvement of radical intermediate species in this reaction.

Conclusions

Silver(I) oxide (0.5 equiv.) is the most efficient oxidant amongst the silver(I) reagents that we tested to promote the oxidative coupling of methyl p-coumarate (I) and methyl ferulate (II) to produce the dihydrobenzofuran neolignans 1 and 2. Acetonitrile, which has not been previously reported as solvent for this reaction, affords the best balance between conversion and selectivity, besides being “greener” than other more frequently employed solvents (e.g., dichloromethane and benzene). Under the optimized conditions, the reaction time can be reduced from 20 to 4 h without a significant decrease in conversion and selectivity. AIBN (a radical initiator) addition increases conversion and decreases selectivity, whereas isoquinoline (a radical inhibitor) addition slightly decreases reactivity and increases selectivity. The results of the experiments involving radical inhibitors/initiators to prove the involvement of radical intermediate species in the silver(I)-promoted oxidative coupling of the methyl esters I and II are not conclusive and must be investigated by other correlated experiments (e.g., mass spectrometry).

Supplementary Information

Supplementary information (IR, ¹H and ¹³C NMR, and EI-MS spectra) is available free of charge at http://jbcs.sbq.org.br as PDF file.

Acknowledgments

The authors thank the Brazilian Foundations FAPESP (process 2013/20094-0) and CNPq for fellowships and grants.

References

• ¹
  Moss, G. P.; Pure Appl. Chem 2000, 72, 1493.
• ²
  Dias, H. J.; Silva, E. O.; Vieira, T. M.; Crotti, A. E. M. In Benzofuran: Production and Applications; Barros, M. S., ed.; Nova Science Publisher: Hauppauge, NY, 2020.
• ³
  Liu, Q.-B.; Huang, X.-X.; Bai, M.; Chang, X.-B.; Yan, X.-J.; Zhu, T.; Zhao, W.; Peng, Y.; Song, S.-J.; J. Agric. Food Chem 2014, 62, 7796.
• ⁴
  Fukui, M. J.; Dias, H. J.; Severiano, M. E.; de Souza, M. G. M.; de Oliveira, P. F.; Ambrósio, S. R.; Martins, C. H. G.; Tavares, D. C.; Crotti, A. E. M.; ChemistrySelect 2018, 3, 1836.
• ⁵
  Dias, H. J.; Patrocínio, A. B.; Pagotti, M. C.; Fukui, M. J.; Rodrigues, V.; Magalhães, L. G.; Crotti, A. E. M.; Chem. Biodiversity 2018, 15, e1800134.
• ⁶
  Van Miert, S.; Dyck, S. V.; Schmidt, T. J.; Brun, R.; Vlietinck, A.; Lemière, G.; Pieters, L.; Bioorg. Med. Chem 2005, 13, 661.
• ⁷
  Baldin, E. L. L.; Dias, H. J.; de Souza, C. M.; Soares, M. C. E.; Grundman, C. O.; Santos, T. L. B.; Crotti, A. E. M.; J. Pest Sci 2019, 92, 861.
• ⁸
  Garcia-Muñoz, S.; Alvarez-Corral, M.; Jimenez-Gonzalez, L.; Lopez-Sanchez, C.; Rosales, A.; Muñoz-Dorado, M.; Rodriguez-Garcia, I.; Tetrahedron 2006, 62, 12182.
• ⁹
  Hemelaere, R.; Carreaux, F.; Carboni, B.; Eur. J. Org. Chem 2015, 2015, 2470.
• ¹⁰
  Daquino, C.; Rescifina, A.; Spatafora, C.; Tringali, C.; Eur. J. Org. Chem . 2009, 2009, 6289.
• ¹¹
  Medeiros, T. C. T.; Dias, H. J.; Silva, E. O.; Fukui, M. J.; Soares, A. C. F.; Kar, T.; Heleno, V. C. G.; Donate, P. M.; Parreira, R. L. T.; Crotti, A. E. M.; J. Braz. Chem. Soc 2016, 27, 136.
• ¹²
  Kuo, Y.-H.; Wu, C.-H.; J. Nat. Prod 1996, 59, 625.
• ¹³
  Blum, T. R.; Zhu, Y.; Nordeen, S. A.; Yoon, T. P.; Angew. Chem 2014, 53, 11056.
• ¹⁴
  Chioccara, F.; Poli, S.; Rindone, B.; Pilati, T.; Brunow, G.; Pietikäinen, P.; Setälä, H.; Tsuda, T.; Acta Chem. Scand. 1993, 47, 610; Yoshihara, T.; Yamaguchi, K.; Sakamura, S.; Agric. Biol. Chem 1983, 47, 217.
• ¹⁵
  Erdtman, H.; Justus Liebigs Ann. Chem 1933, 503, 283.
• ¹⁶
  Dias, H. J.; Baguenard, M.; Crevelin, E. J.; Palaretti, V.; Gates, P. J.; Vessecchi, R.; Crotti, A. E. M.; J. Mass Spectrom 2019, 54, 35.
• ¹⁷
  Pieters, L.; Van Dyck, S.; Gao, M.; Bai, R.; Hamel, E.; Vlietinck, A.; Lemiere, G.; J. Med. Chem 1999, 42, 5475.
• ¹⁸
  Lemière, G.; Gao, M.; de Groot, A.; Dommisse, R.; Lepoivre, J.; Pieters, L.; Buss, V.; J. Chem. Soc., Perkin Trans. 1 1995, 1775.
• ¹⁹
  Orlandi, M.; Rindone, B.; Molteni, G.; Rummakko, P.; Brunow, G.; Tetrahedron 2001, 57, 371.
• ²⁰
  Sako, M.; Hosokawa, H.; Ito, T.; Iinuma, M.; J. Org. Chem 2004, 69, 2598.
• ²¹
  Anastas, P. T.; Warner, J. C.; Green Chemistry: Theory and Practice; Oxford University Press: New York, NY, 1998.
• ²²
  Dance, I. J.; Polyhedron 1986, 1986, 1037; Blower, P. J.; Dilworth, J. R.; Coord. Rev. Chem. 1987, 76, 121.
• ²³
  Gimeno, M. C.; Laguna, A.; Meyer, T. J.; Comprehensive Coordination Chemistry II; Pergamon: Oxford, 2003.
• ²⁴
  Krstina, J.; Moad, G.; Willing, R. I.; Danek, S. K.; Kelly, D. P.; Jones, S. L.; Solomon, D. H.; Eur. Polym. J 1993, 29, 379.
• ²⁵
  Bloger, D. L.; Mathvink, R. J.; J. Org. Chem . 1992, 57, 1429; Chatgilialoglu, C.; Ferreri, C.; Ballestri, M.; Curran, D. P.; Tetrahedron Lett. 1996, 37, 6387; Beaufils, F.; Denes, F.; Renaud, P.; Org. Lett. 2004, 6, 2563.
• ²⁶
  Quideuau, S.; Ralph, J.; Holzforshung 1994, 48, 12; Maeda, S.; Masuda, H.; Tokoroyama, T.; Chem. Pharm. Bull. 1995, 43, 935.
• ²⁷
  Minisci, F.; Zhao, L.; Fontana, F.; Bravo, A.; Tetrahedron Lett 1995, 36, 1895; Jang, M. H.; Kim, H. Y.; Kang, K. S.; Yokozawa, T.; Park, J. H.; Arch. Pharm. Res. 2009, 32, 341; Bernotas, R. C.; Thomas, C. E.; Carr, A. A.; Nieduzak, T. R.; Adams, G.; Ohlweiler, D. F.; Hay, D. A.; Bioorg. Med. Chem . Lett. 1996, 6, 1105.

Publication Dates

• Publication in this collection
  20 Jan 2021
• Date of issue
  Jan 2021

History

• Received
  13 May 2020
• Accepted
  22 July 2020