Copper-Catalyzed Oxidative Cleavage of the C-C Bonds of β-Alkoxy Alcohols and β-1 Compounds.

Copper-catalyzed aerobic oxidation conditions were employed to promote the C-C bond cleavage of β-alkoxy alcohols and β-1 compounds (lignin model compounds). Besides these compounds, various 1,2 and 1,3-diols were successfully converted to aldehydes. We propose the Cu(I)-catalyzed mechanism explaining the C-C cleavage of these 1,2 and 1,3-dihydroxy compounds and β-alkoxy alcohols based on XPS data. Although our reaction conditions do not include large excess of bases and elaborated ligand-modified catalysts, copper salts with/without Me-TBD show good catalytic activities for C-C bond cleavage of various lignin model compounds.

Introduction

Lignin, a key material for the formation of plant cell walls, is composed of various aromatic building blocks that make it attractive as a renewable feedstock for forming n class="Chemical">aromatic compounds via chemical processes. Compared to the chemical utilization of other biomass-based feedstocks such as cellulose and hemicellulose, the chemical conversion of lignin is not intensively conducted owing to the difficulty of depolymerizing the randomly connected chemical bonds in the complex structure. Investigating efficient and selective catalysts for the depolymerization of lignin to afford industrially useful aromatic compounds is imperative for replacing fossil fuel-based aromatic compound production.[1,2] The finite availability of petroleum along with the carbon balance achieved by using biomass waste has increased the importance of catalytic processes for converting lignin.

To develop the depolymerization process of lignin, modeling of the n class="Chemical">metal-catalyzed cleavage of the C–C or C–O bonds has been attempted by many researchers.[3−7] C–C and C–O bonds are commonly found in the β-O-4 linkage and β-1 linkage of lignin.[8−10] For C–O cleavage, Ni,[11] V,[12,13] Pd,[14,15] Co,[16] Ru,[17,18] and Zn,[19,20] as well as metal-free conditions,[21] have been employed, and for C–C cleavage, V,[22−24] Fe,[25] Ru,[26,27] Ir,[28] and Cu[29−35] catalyzed reactions, as well as metal-free[36−38] ones, have been reported. The cleavage of the C–C bonds is not as facile as that of the C–O bonds because of the robustness of the former, but the judicious choice of catalysts has been shown to promote the efficient cleavage of C–C bonds of model lignin compounds.

Our research group has been developing copper-catalyzed oxidative coupling reactions and simple oxidation for tandem asymmetric reactions.[39−41] Based on previous n class="Chemical">copper-catalyzed oxidation reactions, we decided to investigate copper-catalyzed aerobic oxidative C–C cleavage of β-alkoxy alcohols and β-1 compounds (Figure ). The β-alkoxy alcohols possessing OR1 (R1 = alkyl) and the aromatic group on the carbon backbone undergo the facile C–C bond cleavage compared to the slow reaction of β-O-4 compounds having OAr (Ar = aromatic).[13] Thus, the reactions of β-alkoxy alcohols have been examined to investigate a catalyst. Copper-catalyzed oxidative cleavage reactions of β-alkoxy alcohols and β-1 compounds were reported by Baker/Hanson and Li (Figure ). Copper catalysts modified with 2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) or saturated 1,3-dimesitylimidazol-2-ylidene were used in the presence of large excess of bases such as pyridine, 2,6-lutidine, and quinoline. Compared to previous reports, our reactions do not require large quantity of bases, and in the case of β-1 compounds, CuCl without both ligands and bases promoted the oxidative cleavage. The reaction optimization, the scope of the C–C bond cleavage including β-alkoxy alcohols, β-1 compounds, 1,2-diol, and 1,3-diols, and the proposed catalytic cycles are presented herein.

Figure 1

Copper-catalyzed oxidative cleavage reactions.

Results and Discussion

The copper-catalyzed oxidative cleavage reactions of n class="Chemical">2-methoxy-1,2-diphenylethan-1-ol (1a) were conducted as listed in Table . The combination of CuBr2 (10 mol %) and 1,5,7-triazacyclo[4.4.0]dec-5-ene (TBD) under 1 bar of oxygen promoted the desired oxidative cleavage to afford aldehyde 1b and ester 1c with 28% and 27% yields, respectively (entry 1). The yield of each product was determined by 1H NMR using an internal standard (see the Supporting Information). Higher catalyst loadings increased the yields of 1b and 1c (62 and 59%, respectively) (entry 2). In the absence of copper catalysts, the reaction did not proceed (entry 3). Instead of TBD, triethylamine (TEA), TEMPO, 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), and N-methyl TBD (Me-TBD) were examined, from which the reaction involving Me-TBD afforded products with the highest yields (entries 4–7). CuBr, Cu(OAc)2, and Cu(OTf)2 were inferior to CuBr2 for C–C bond cleavage (entries 8–10). Oxygen gas was bubbled through the solution to increase the oxygen content in the reaction media, resulting in the highest yields (entry 11). Because the yield of 1c was slightly lower than that of 1b, we re-exposed the same amount of 1b and 1c to the optimized reaction conditions. Compounds 1b and 1c were recovered quantitatively, implying that 1b and 1c were not decomposed under the reaction conditions (see Supporting Information, Scheme S1).

Table 1

Optimization of Oxidative Cleavage of 1a

entry	Cu catalyst (mol %)	additives (mol %)	yield (1b) (%)	yield (1c) (%)
1	CuBr2 (10)	TBD (15)	28	27
2	CuBr2 (20)	TBD (30)	62	59
3		TBD (30)
4	CuBr2 (20)	TEA (30)	3	2
5	CuBr2 (20)	TEMPO (30)	0.4	0.1
6	CuBr2 (20)	DBU (30)	31	23
7	CuBr2 (20)	Me-TBD (30)	88	73
8	CuBr (20)	Me-TBD (30)	24	27
9	Cu(OAc)2 (20)	Me-TBD (30)	4	11
10	Cu(OTf)2 (20)	Me-TBD (30)	5	4
11a	CuBr2 (20)	Me-TBD (30)	94	88

O2 was bubbled through the solution. Reaction condition: the mixture of 1a (0.25 mmol), copper catalyst (20 mol %), and additives (indicated amounts) in DMSO (0.36 M) was allowed to react at 120 °C under 1 atm of oxygen. The yields were determined by 1H NMR using maleic acid as an internal standard.

O2 was bubbled through the solution. Reaction condition: the mixture of 1a (0.25 mmol), n class="Chemical">copper catalyst (20 mol %), and additives (indicated amounts) in DMSO (0.36 M) was allowed to react at 120 °C under 1 atm of oxygen. The yields were determined by 1H NMR using maleic acid as an internal standard.

After optimizing the conditions to cleave the C–C bond in compound 1a, compounds possessing different substituents were examined (Table ). 2-Ethoxy-1,2-diphenylethan-1-ol and n class="Chemical">2-isopropoxy-1,2-diphenylethan-1-ol were converted to benzaldehyde and ethyl- and isopropyl benzoate with good yields, respectively (entries 2 and 3). Whereas the reactions of 1a showed lower yield of ester 1c than aldehyde 1b, compounds 2a and 3a possessing ethoxy and propyl alkoxide groups showed similar yields of aldehydes and esters. The electron-deficient aromatic group at the α-carbon (R1) of 4a decreased the yield compared to electron-rich aromatic substituted compounds (1a, 5a, and 6a) (entries 1 and 4–6). The electron-rich substitution at the β-carbon (R3) of 7a also reduced the yields of aldehydes and esters (entry 7).

Table 2

Substrate scope of Copper-catalyzed C–C Cleavage I

entry	reactants	R1	R2	R3	yield
1	1a	C6H5	CH3	C6H5	94% (1b), 88% (1c)
2	2a	C6H5	C2H5	C6H5	95% (1b), 93% (2c)
3	3a	C6H5	C3H7	C6H5	96% (1b), 91% (3c)
4	4a	4-Cl-C6H4	CH3	C6H5	80% (2b), 71% (1c)
5	5a	4-OMe-C6H4	CH3	C6H5	97% (3b), 71% (1c)
6	6a	3,4-diOMe-C6H3	CH3	C6H5	98% (4b), 78% (1c)
7	7a	C6H5	CH3	4-Me-C6H4	71% (1b), 76% (4c)

After the oxidative cleavage of β-alkoxy alcohols including 1a, the n class="Chemical">copper-catalyzed oxidative cleavage of 1-phenyl-2-(p-tolyl)propane-1,3-diol (1d), a β-1 model compound, was attempted, as listed in Table . The reactions involving CuBr2 with/without Me-TBD afforded products 1b and 5b with low yields (entries 1 and 2). Other than indicated products, the starting materials remained the same (see the Supporting Information). Compared to the result for 1a involving C–C bond cleavage, 1d undergoes two successive C–C bond cleavages to afford 1b and 5b. Thus, the copper catalyst, bases, and temperature were varied to improve the yield. Changing copper catalysts to copper(I) increased the yield of each product (entries 3 and 4). The increasing temperature during the shorter reaction time promoted the C–C bond cleavage of 1d with yields of 46% (1b) and 65% (5b) (entries 5 and 6). Further increasing the catalyst load and addition of Me-TBD did not improve the yield of the product (entries 7 and 8). Products were observed with the remaining starting materials (see Supporting Information, Table S2).

Table 3

Optimization of the Oxidative Cleavage of 1d

entry	Cu catalyst (mol %)	additives (mol %)	temp (°C)	yield (1b) (%)	yield (1c) (%)
1	CuBr2 (10)	Me-TBD (30)	120	9	14
2	CuBr2 (20)		120	2
3	CuBr (20)		120	20	28
4	CuCl (20)		120	25	36
5	CuCl (20)		150	38	55
6a	CuCl (20)		150	46	65
7a	CuCl (30)		150	44	59
8a	CuCl (20)	Me-TBD (30)	150	17	24

Reaction time: 12 h. Reaction condition: the mixture of 1d (0.25 mmol), copper catalyst (indicated amounts), and additives (indicated amounts) in DMSO (0.36 M) was allowed to react at the indicated temperature under 1 atm of oxygen. The yields were determined by 1H NMR using 1,3-trimethoxybenzene as the internal standard.

Reaction time: 12 h. Reaction condition: the mixture of 1d (0.25 mmol), n class="Chemical">copper catalyst (indicated amounts), and additives (indicated amounts) in DMSO (0.36 M) was allowed to react at the indicated temperature under 1 atm of oxygen. The yields were determined by 1H NMR using 1,3-trimethoxybenzene as the internal standard.

The scope of cleaving the β-1 model compounds was examined, as listed in Table . As the number of electron-donating substituents was increased, the overall yields of bond-cleavage products increased. Compound 2d possessing two methoxy groups underwent cleavage to afford 4b and 1b with 49 and 41% yields, respectively, which was similar to the result for n class="Chemical">1d having a methyl group (entries 1 and 2). Compound 3d having a methoxy and a methyl group was converted to 3b and 5b with 50 and 52% yields, respectively (entry 3). Based on the oxidative cleavage yields of 4d, 5d, 6d, and 7d, as the number of electron-donors increases, better yields were observed (entries 4–7). In addition to lignin model compounds, various 1,2- and 1,3-diols were examined, thereby demonstrating that our copper-catalyzed aerobic C–C cleavage protocol could be expanded to the reactions of these compounds. The transition-metal-catalyzed oxidative cleavage of 1,2-diols was reported using ruthenium, vanadium, and silver catalysts.[42−45] Homogeneous copper catalysts for this transformation has not been reported. In addition, transition-metal catalyzed aerobic cleavage of simple 1,3-diol 11d has not been reported yet. The reactions of 1,2-diols were successful with both CuBr2 and CuCl (entries 8–10), but the cleavage yields of 1,3-diols were variable depending on the copper catalyst (entry 11). The CuCl-catalyzed reaction afforded benzaldehyde 1b with a 73% yield, whereas the reaction with CuBr2 and Me-TBD provided 1b with a 10% yield. The oxidative cleavage of 1,2 and 1,3-alcohols also afforded only aldehydes without forming carboxylic acids. Finally, the reaction of 1a was conducted under the conditions of Table to compare the catalytic performance between CuBr2/Me-TBD and CuCl. The C–C cleavage of 1a afforded 1b with 98% yield and 1c with 55% yield in the presence of CuCl, showing that CuCl was inferior to CuBr2/Me-TBD for the formation of 1c (entry 11 of Table ).

Table 4

Substrate Scope of Copper-Catalyzed C–C Cleavage II

The reaction mechanisms of the C–C bond cleavage of 1a and 1d were probed (Scheme ). Since the reaction was carried out under n class="Chemical">copper-catalyzed aerobic oxidation conditions, it could be initiated by the oxidation of alcohols,[46−49] followed by C–C bond cleavage. To assess our hypothesis, independently prepared α-oxyketone 5a′ was subjected to the reaction conditions (Scheme , eq 1). 5a′ underwent C–C bond cleavage to produce acid 3b′ instead of aldehyde 3b under the optimized conditions. Even in the absence of CuBr2, the same combination of products was observed with similar yields. In the previous copper-catalyzed C–C cleavage of lignin models, aldehydes were formed rather than acids.[30,32] The starting materials were retained without both CuBr2 and Me-TBD. In the case of 1d′, aldehydes 1b and 5b were obtained with low yields, and no cleavage was observed without CuCl (Scheme , eq 2). Besides, ketones formed by the oxidation of 1a and 1d were not found during the copper-catalyzed cleavage reactions. Accordingly, it was concluded that the C–C cleavage reactions of 1a and 1d might not begin with the oxidation of sec-alcohols.[35]

Scheme 1

Control Experiments

Plausible reaction mechanisms to account for C–C cleavage of 1a and 1d are illustrated in cycle A of Scheme .[50,51] The reaction of 1a was promoted by n class="Chemical">CuBr2 and Me-TBD. Based on XPS data of the mixtures of CuBr2/Me-TBD and CuBr2/Me-TBD/1a, Me-TBD is presumed to reduce Cu(II) to Cu(I) (Figure ). The Cu 2p3/2 core levels taken from CuBr2/Me-TBD (Figure a) and CuBr2/Me-TBD/1a (Figure b) show the presence of Cu(I) and Cu(II) chemical states at 932.7 and 934 eV, respectively. Along with the observation of Br 3d at 69 eV as well as the survey spectra (Figure S1), the results indicate that two Cu chemical states can be associated with CuBr and CuBr2 species, respectively.[52,53] The presence of Cu(I) [not Cu(0)] as well as Cu(II) is further corroborated by the Cu L3M4,5M4,5 Auger features (Figure S2), which overlap well with the reference Auger spectra of CuBr and CuBr2; the Auger features of CuBr and Cu are well resolved despite the similar Cu 2p3/2 binding energies of the Cu(I) and Cu(0) oxidation states (Figure S2). The relative intensity of Cu(I) to that of Cu(II) differs between (a) and (b) in Figure , but higher Cu(I)-to-Cu(II) ratio is maintained even after the reaction of 1a (b), suggesting the dominant role of Cu(I) in the catalytic cycle of the oxidative C–C cleavage of 1a. Thus, both reactions (1a and 1d) begin with the cleavage by Cu(I). A free hydroxy group coordinates with the copper catalyst to form intermediate I, which undergoes oxygen-mediated cleavage to afford benzaldehyde and copper-coordinated peroxide intermediate II.[54] Next, the fragmentation of II provides methyl benzoate along with Cu(I) regeneration (CuOH). In the case of 1d, α-hydroxy ketones released from the cycle undergo decarboxylation to afford aldehydes.[50] Compound 11d which does not include a secondary hydroxy group undergoes decarboxylation to form intermediate III, which is followed by copper-peroxide IV formation and fragmentation (cycle B).

Scheme 2

Plausible Catalytic Cycles for the Cleavage of 1a, 1d, and 11d

Figure 2

XPS spectra of (a) CuBr2 and Me-TBD and (b) CuBr2, Me-TBD, and 1a.

XPS spectra of (a) CuBr2 and n class="Chemical">Me-TBD and (b) CuBr2, Me-TBD, and 1a.

Conclusions

We report the copper-catalyzed oxidative C–C cleavage of various n class="Chemical">alcohols, including lignin model compounds. The optimal copper catalyst was varied depending on the structure of the alcohols. CuBr2/Me-TBD was efficient for vicinal dioxy compounds (β-alkoxy alcohols and 1,2-diols), whereas CuCl was valid with β-1 compounds and 1,2- and 1,3-diols. The yields of CuCl-catalyzed oxidative cleavage of β-1 compounds are somewhat lower than those of CuBr2/Me-TBD-catalyzed cleavage of β-alkoxy alcohols. However, electron-rich β-1 compounds abundant in lignin undergo oxidative cleavage with improved yields. The reaction mechanisms involving copper(I)-catalyzed aerobic oxidative cleavage were presented based on XPS data. Compared to previously reported catalytic systems using large amounts of bases and showing good catalytic activities on a specific linkage, our catalysts can cleave the C–C bonds of β-alkoxy alcohols, 1,2-diols, 1,3-diols, and β-1 substrates without extra bases and elaborated ligands. This finding could be a general and useful tool for depolymerizing lignin, which possesses various linkages and bonds, to afford different aromatic feedstocks.

Experimental Section

Synthetic Procedure for the Synthesis of 1a–3a

Compounds 1a–3a were known and prepared according to literature procedure.[48] 2-Methoxy-1, 2-diphenylethan-1-one (3.185 mmol), n class="Chemical">Al(OiPr)3 (0.956 mmol), and iPrOH (35.04 mmol) were dissolved in toluene (4.0 mL, 1.3 M) and stirred overnight at 50 °C. After reaction, the reaction mixture was cooled down, quenched with 1 N HCl, and extracted with ethyl acetate. The organic layer was washed with H2O and evaporated in vacuo. The residue was purified by column chromatography (ethyl acetate: hexane) to afford 1a as a white solid with 91% yield. Compounds 2a and 3a were obtained with 70 and 76% yields, respectively.

Synthetic Procedure for the Synthesis of 4a–7a

Copper oxide (5.0 mmol) and n class="Chemical">iodine (5.0 mmol) were added to a well-stirred solution of 1-(4-chlorophenyl)-2-phenylethan-1-one (5.0 mmol) in MeOH (0.25 M). The mixture was refluxed overnight. Then, the mixture was filtered. Iodine was removed by Na2S2O3 solution. The mixture was extracted with ethyl acetate, and the organic layer was dried with Na2SO4. The residue was further purified by column chromatography (ethyl acetate: hexane) using silica gel to form 1-(4-chlorophenyl)-2-methoxy-2-phenylethan-1-one as a white solid with 60% yield. 1-(4-Chlorophenyl)-2-methoxy-2-phenylethan-1-one (3.4 mmol) and NaBH4 (5.1 mmol) were dissolved in MeOH (0.13 M) and stirred for 4 h at room temperature. The solution was washed with aqueous NH4Cl and extracted with ethyl acetate. The organic phase was extracted and dried with MgSO4. The product was purified by column chromatography (ethyl acetate: hexane) to afford 4a as a colorless oil with 69% yield. Compounds 5a, 6a, and 7a were obtained via reduction of ketone compounds with yields of 89, 81, and 99%, respectively.

1-(4-Chlorophenyl)-2-methoxy-2-phenylethan-1-ol (4a)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz, diastereomeric mixture): δ 7.32–7.18 (m, 8H), 7.04 (m, 1H), 5.52 (d, J = 4.1 Hz, 0.23H), 5.42 (d, J = 5.4 Hz, 0.77H), 4.69 (dd, J = 6.6, 4.2 Hz, 0.23H), 4.65 (t, J = 5.5 Hz, 0.77H), 4.25 (d, J = 6.6 Hz, 0.23H), 4.20 (d, J = 6.0 Hz, 0.77H), 3.15 (s, 0.69H), 3.03 (s, 2.3H). 13C NMR (DMSO-d6, 151 MHz, diastereomeric mixture): δ 142.01, 140.82, 139.08, 138.37, 131.28, 131.22, 129.04, 128.98, 127.97, 127.73, 127.65, 127.38, 127.23, 87.62, 87.02, 75.45, 74.94, 56.57, 56.36. HRMS (FAB, [M – H2O + H]+) calcd for C15H14ClO, 245.0733; found, 245.0736. FTIR (neat) 3436, 2932, 1901, 1089, 700 cm–1.

2-Methoxy-1-(4-methoxyphenyl)-2-phenylethan-1-ol (5a)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz, diastereomeric mixture): δ 7.30–7.15 (m, 6.64H), δ 7.01 (d, J = 7.2 Hz, 0.18H), 6.95 (d, J = 8.4 Hz, 0.18H), 6.82 (m, 1.82H), 6.69 (d, J = 9.0 Hz, 0.18H), 5.26 (d, J = 4.2 Hz, 0.09H), 5.17 (d, J = 4.8 Hz, 0.91H), 4.60 (dd, J = 4.2 Hz, 7.2 Hz, 0.09H), 4.57 (dd, J = 4.8 Hz, 6.0 Hz, 0.91H), 4.20 (d, J = 6.6 Hz, 0.09H), 4.16 (d, J = 6.0 Hz, 0.91H), 3.72 (s, 2.73H), 3.66 (s, 0.27H), 3.14 (s, 0.27H), 3.01 (s, 2.73H). 13C NMR (DMSO-d6, 151 MHz): δ 158.2, 139.5, 135.1, 128.3, 128.0, 127.6, 127.2, 112.8, 87.4, 75.2, 56.3, 54.9. HRMS (FAB, [M – H2O + H]+) calcd for C16H17O2, 241.1229; found, 241.1228. FTIR (neat) 3454, 2934, 2057, 1099 cm–1.

1-(3,4-Dimethoxyphenyl)-2-methoxy-2-phenylethan-1-ol (6a)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz, diastereomeric mixture): δ 7.27–7.20 (m, 4.78H), δ 7.03–7.02 (m, 0.22H), 6.83 (d, J = 8.4 Hz, 0.89H), 6.78–6.74 (m, 1.78H), 6.71 (d, J = 8.4 Hz, 0.11H), 6.58 (dd, J = 1.2 Hz, 8.4 Hz, 0.11H), 6.51 (d, J = 2.4 Hz, 0.11H), 5.30 (d, J = 4.2 Hz, 0.11H), 5.19 (d, J = 4.8 Hz, 0.89H), 4.60 (t, J = 5.4 Hz, 0.89H), 4.58 (t, J = 3.6 Hz, 0.11H), 4.20 (d, J = 7.8 Hz, 0.11H), 4.18 (d, J = 6.0 Hz, 0.89H), 3.72 (s, 2.67H), 3.66 (s, 0.33H), 3.64 (s, 2.67H), 3.53 (s, 0.33H), 3.16 (s, 0.33H), 3.04 (s, 2.67H). 13C NMR (DMSO-d6, 151 MHz): δ 147.9, 147.7, 139.4, 135.5, 128.1, 127.5, 127.2, 119.3, 111.0, 110.9, 87.3, 75.3, 56.3, 55.4, 55.3. HRMS (FAB, [M – H2O + H]+) calcd for C17H19O3, 271.1334; found, 271.1336. FTIR (neat) 3440, 2908, 1944, 1136 cm–1. mp 74 °C.

2-Methoxy-1-phenyl-2-(p-tolyl)ethan-1-ol (7a)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz, diastereomeric mixture): δ 7.25 (m, 4.15H), 7.14–7.11 (m, 0.51H), 7.09 (s, 3.32H), 7.05 (m, 0.34H), 6.99 (d, J = 7.8 Hz, 0.34H), 6.90 (d, J = 7.2 Hz, 0.34H), 4.62 (d, J = 6.6 Hz, 0.17H), 4.60 (d, J = 7.2 Hz, 0.83H), 4.19 (d, J = 7.2 Hz, 0.17H), 4.16 (d, J = 6.6 Hz, 0.83H), 3.12 (s, 0.51H), 2.98 (s, 2.49H), 2.28 (s, 2.49H), 2.21 (s, 0.51H). 13C NMR (DMSO-d6, 151 MHz, diastereomeric mixture): δ 143.18, 141.81, 136.32, 136.28, 135.61, 128.21, 128.00, 127.73, 127.44, 127.32, 127.25, 127.23, 126.77, 87.84, 87.11, 76.43, 75.66, 56.38, 56.17, 20.77, 20.72. HRMS (FAB, [M – H2O + H]+) calcd for C16H17O, 225.1279; found, 225.1281. FTIR (neat) 3431, 2929, 1899, 1103 cm–1. mp 90 °C.

Synthetic Procedure for the Synthesis of 1d–7d

Compounds 2d, 4d, and 6d were known and prepared according to literature procedure.[49,50] 1,2-Diaryl-1-ethanone (2 mmol), n class="Chemical">formaldehyde (2.15 mmol), and K2CO3 (3.65 mmol) were dissolved in acetone (0.185 M) and stirred for 3 h at room temperature. The solid was filtered and extracted with ethyl acetate. The product was purified by column chromatography (ethyl acetate/hexane) with 98% yield. 3-Hydroxy-1,2-diphenyl-1-propanone (1.97 mmol) and NaBH4 (2.167 mmol) were dissolved in MeOH (0.13 M) and stirred for 3 h at room temperature. The solution was washed with aqueous NH4Cl and extracted with ethyl acetate. The organic phase was extracted and dried with MgSO4. The product was purified by column chromatography (ethyl acetate/hexane) to afford 1d as a white solid with 80% yield. Compounds 2d, 3d, 4d, 5d, 6d, and 7d were obtained via the reduction of ketone compounds with yields of 63, 93, 67, 93, 92, and 77%, respectively.

1-Phenyl-2-(p-tolyl)propane-1,3-diol (1d)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz): δ 7.20 (t, J = 7.5 Hz, 2H), 7.13 (m, 3H), 6.96 (m, 4H), 5.13 (d, J = 4.8 Hz, 1H), 4.97 (t, J = 4.8 Hz, 1H), 4.49 (t, J = 5.1 Hz, 1H), 3.71–3.68 (m, 1H), 3.49–3.46 (m, 1H), 2.85 (dd, J = 6.6 Hz, 12 Hz, 1H), 2.22 (s, 3H). 13C NMR (DMSO-d6): δ 145.1, 137.2, 134.5, 129.3, 127.9, 127.5, 126.3, 126.2, 72.0, 62.7, 55.2, 20.6. HRMS (FAB, [M – H2O + H]+) calcd for C16H17O, 225.1279; found, 225.1282. FTIR (neat) 3295, 2932, 1903 cm–1. mp 132 °C.

1-(4-Methoxyphenyl)-2-(p-tolyl)propane-1,3-diol (3d)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz) 7.05 (m, 2H), 6.97 (m, 4H), 6.77 (m, 2H), 5.01 (d, J = 4.8 Hz, 1H), 4.89 (t, J = 4.8 Hz, 1H), 4.43 (t, J = 5.1 Hz, 1H), 3.69 (s, 3H), 3.67–3.63 (m, 1H), 3.46–3.43 (m, 1H), 2.81 (q, J = 6.2 Hz, 1H), 2.22 (s, 3H). 13C NMR (DMSO-d6, 151 MHz): δ 157.8, 137.5, 137.1, 134.5, 129.3, 127.9, 127.3, 112.9, 71.8, 62.8, 55.3, 54.9, 20.7. HRMS (FAB, [M – H2O + H]+): calcd for C17H19O2, 255.1385; found, 255.1387. FTIR (neat) 3398, 2889, 1981 cm–1. mp 112 °C.

1-(3,4-Dimethoxyphenyl)-2-(p-tolyl)propane-1,3-diol (5d)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz): δ 6.97–6.91 (m, 4H), 6.76–6.58 (m, 3H), 5.02 (d, J = 4.2 Hz, 1H), 4.87 (t, J = 5.1 Hz, 1H), 4.42 (t, J = 5.4 Hz, 1H), 3.69–3.67 (m, 1H), 3.66 (s, 3H), 3.54 (s, 3H), 3.48–3.45 (m, 1H), 2.79 (dd, J = 6.9 Hz, 11.7 Hz, 1H), 2.20 (s, 3H). 13C NMR (DMSO-d6, 151 MHz): δ 147.99, 147.32, 137.60, 137.54, 134.58, 129.46, 127.93, 118.25, 110.93, 110.32, 72.10, 62.77, 55.42, 55.39, 55.14, 20.67. HRMS (FAB, [M – H2O + H]+): calcd for C18H21O3, 285.1491; found, 285.1489. FTIR (neat) 3298, 2961, 1903 cm–1. mp 115 °C.

2-(3,4-Dimethoxyphenyl)-1-(2,3,4-trimethoxyphenyl)propane-1,3-diol (7d)

1H n class="Chemical">NMR (DMSO-d6, 600 MHz): δ 6.73 (d, J = 7.2 Hz, 1H), 6.65 (d, J = 8.4 Hz, 1H), 6.59–6.57 (m, 3H), 5.18 (t, J = 4.8 Hz, 1H), 4.85 (d, J = 4.2 Hz, 1H), 4.43 (t, J = 5.1 Hz, 1H), 3.81 (s, 3H), 3.76–3.73 (m, 1H), 3.72 (s, 3H), 3.71 (s, 3H), 3.67 (s, 3H), 3.60 (s, 3H), 3.56 (m, 1H), 2.86 (dd, J = 6.9 Hz, 11.7 Hz, 1H). 13C NMR (DMSO-d6, 151 MHz): δ 151.7, 149.8, 147.6, 147.0, 140.9, 133.0, 130.5, 121.9, 121.4, 113.7, 111.1, 107.1, 66.4, 63.0, 60.7, 60.2, 55.7, 55.4, 55.2, 53.8. HRMS (FAB, [M – H2O + H]+): calcd for C20H25O6, 361.1651; found, 361.1648. FTIR (neat) 3466, 2937, 1994 cm–1.

Synthetic Procedure for the Synthesis of 9d–10d

Compounds 9d and 10d were known and prepared according to literature procedure.[51,55] Benzoin (6 mmol) and n class="Chemical">NaBH4 (9.0 mmol) were dissolved in MeOH (0.13 M) and stirred for 1 h at 0 °C. The solution was washed with aqueous NH4Cl and extracted with ethyl acetate. The organic phase was extracted and dried with anhydrous Na2SO4. The product was purified by column chromatography (ethyl acetate: hexane) to afford 9d as a white solid with 99% yield. The yield of 10d was 99%.

Catalytic Reaction Procedure of Tables and 2

The mixture of 1a (0.25 mmol), copper catalyst (indicated amounts), and additives (indicated amounts) in n class="Chemical">DMSO (0.36 M) was allowed to react at 120 °C for 16 h under 1 atm of oxygen.

Catalytic Reaction Procedure of Tables and 4

The mixture of 1d (0.25 mmol), n class="Chemical">copper catalyst (indicated amounts), and additives (indicated amounts) in DMSO (0.36 M) was allowed to react at the indicated temperature for the given time under 1 atm of oxygen.