催化学报  2014, Vol. 35 Issue (11): 1864-1873   PDF (753 KB)    
扩展功能
加入收藏夹
复制引文信息
加入引用管理器
Email Alert
RSS
本文作者相关文章
张淑芳
苗成霞
徐大乾
孙伟
夏春谷
CuI/N4 ligand/TEMPO derivatives: A mild and highly efficient system for aerobic oxidation of primary alcohols
Shufang Zhanga,b, Chengxia Miaoa, Daqian Xua, Wei Suna , Chungu Xiaa     
a State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, Gansu, China;
b University of Chinese Academy of Sciences, Beijing 100049, China
Received: 2014-5-1. Accepted: 2014-6-3. Published: 2014-11-20.
* Corresponding author. Fax: +86-931-8277088; E-mail: wsun@licp.cas.cn
E-mail: chxmiao@licp.cas.cn
Foundation Item: This work was supported by the National Natural Science Foundation of China (21103207, 21133011).
Abstract: A new system consisting of a copper(I) complex generated in situ from a tetradentate nitrogen ligand and CuI in combination with 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) derivatives was successfully developed. The system was suitable for efficient and selective aerobic oxidation of primary benzyl and allyl alcohols with a wide range of functional groups to the corresponding aldehydes at room temperature. The best result was obtained with N,N'-dimethyl-N,N'-bis(2- pyridylmethyl)ethane-1,2-diamine as the ligand and 4-OH-TEMPO as a cocatalyst in CH3CN. In addition, high-resolution mass spectrometry, ultraviolet-visible spectroscopy, and electrochemical experiments were used to provide evidence of intermediates.
© 2014, Dalian Institute of Chemical Physics, Chinese Academy of Sciences.
Published by Elsevier B.V. All rights reserved.
Key words: Alcohol     Aerobic oxidation     Tetradentate nitrogen ligand     Copper     2,2,6,6-tetramethylpiperidine-1-oxyl    

1. Introduction

Metal complexes with nitrogen ligands have been extensively studied as models of monooxygenase active sites and as oxidation catalysts in organic synthesis[1, 2, 3]. Metal porphyrins such as cytochrome P-450 [1], which play a central role in many biological processes, have been mimicked using various synthetic ligands. Recently, biologically inspired and more flexible tetradentate nitrogen (N4) ligands and their corresponding metal complexes have been attracting increasing attention [4, 5, 6, 7]. Our group has also developed a series of N4 ligands and corresponding metal complexes for asymmetric epoxidation of olefins [8, 9, 10]. Until now, there have been few reports on the oxidation of alcohols catalyzed by metal complexes with N4 ligands.

The selective oxidation of alcohols to the corresponding aldehydes or ketones is a common transformation in organic synthesis because aldehydes and ketones are important as intermediates in pharmaceuticals, plastic additives, perfumes, flavoring compounds, and certain dyes in the textile industry. Recently, the use of the stable nitroxyl radical 2,2,6,6- tetramethylpiperidine-1-oxyl (TEMPO) in combination with various oxidants in the oxidation of alcohols has attracted interest [11, 12, 13]. However, many TEMPO-based systems use stoichiometric amounts of terminal oxidants such as sodium chlorite and hypervalent iodine(III) compounds, so large amounts of waste are produced [14, 15], and longer reaction time or high temperature is also often needed [16].

Air, oxygen, and hydrogen peroxide are clean oxidants, and only water is formed as the by-product, so they have been attracting much interest for use in green chemistry. However, TEMPO, which is expensive, cannot be regenerated directly by molecular oxygen alone, so a cocatalyst is required for activation of molecular oxygen [17].

Copper complexes are more attractive as they are biomimetic functional models of the mononuclear copper enzyme galactose [18]. In 1984, Semmelhack et al. [19] reported the aerobic oxidation of allyl and benzyl alcohols to the corresponding aldehydes and ketones catalyzed by 10 mol% CuI/ TEMPO in N,N-dimethylformamide (DMF) at room temperature. Subsequent research mainly focused on the choice and design of nitrogen-containing ligands such as 1,10- phenanthroline [20], 2,2'-bipyridine (bpy) [21, 22, 23, 24, 25], 1,4- diazabicyclo[2.2.2] octane [26], 4-pyrrolidinopyridine [27], pyrazole- pyridine [28], salen-H4 [29], and 1,4,7-triazacyclononane [30] to improve the catalytic activity and extend the substrate scope. Recently, a breakthrough was achieved by Stahl and coworkers [21, 22, 23, 24, 25]. They used N-methylimidazole as an additive in a CuI-bpy/ TEMPO catalyst system for the highly selective and efficient transformation of a broad range of alcohols, including allylic, benzyl, and aliphatic derivatives, with heterocycles and other heteroatom- containing groups. It should be noted that this catalyst system used non-commercial CuI(OTf) as the copper source.

Although significant progress has been made in developing TEMPO/Cu catalyst systems, readily available systems for the efficient oxidation of alcohols under mild conditions are still needed. In our continuing efforts to develop new systems for oxidation of alcohols catalyzed by TEMPO derivatives [31, 32, 33, 34] and new applications of N4 ligands in different reactions [8, 9, 10], we considered that a tetradentate nitrogen ligand (Scheme 1) combined with copper(I) ions and TEMPO derivatives might be active in oxidation of alcohols. We found that an easily obtainable catalyst system comprising CuI, N,N'- dimethyl- N,N'-bis(2-pyridylmethyl)ethane-1,2-diamine (BPMEN), and 4-OH-TEMPO exhibited high efficiency and wide substrate scope, including allylic, benzyl, and aliphatic primary alcohols, at room temperature using air as the oxidant without any additives such as a base.

Schemes 1. Oxidation of alcohols catalyzed by copper salt/TEMPO derivative/N4 ligand. BPMEN = N,N'- dimethyl- N,N'- bis(2- pyridylmethyl) ethane-1,2-diamine, BPMPN = N,N'-dimethyl- N,N'- bis(2- pyridinylmethyl) propane-1,3-diamine, BPMCN = trans-N,N'- dimethyl- N,N'- bis(2- pyridylmethyl)cyclohexane-1,2-diamine.
2. Experimental
2.1. General information

Gas chromatography-mass spectrometry (GC-MS) was performed using an Agilent Technologies 7890A/5975C system. 1H and 13C nuclear magnetic resonance (NMR) spectra were recorded using a Bruker Avance III 400 MHz spectrometer. The chemical shifts (δ) are reported in parts per million (ppm) and coupling constants (J) in Hz. GC analysis was performed using an Agilent Technologies 6820 instrument with a flame- ionization detector. High-resolution (HR) MS was performed using a Bruker Daltonics micrOTOF-QII mass spectrometer. Ultraviolet (UV)-visible spectra were recorded using an Agilent Cary 60 spectrometer. Electrochemical measurements were conducted with a CHI 660E potentiostat using a platinum-button working electrode, non-aqueous Ag/Ag+ reference, and a platinum wire counter electrode at a scan rate of 100 mV/s. Elemental analysis was performed using a Vario EL cube elemental analyzer.

2.2. Preparation and characterization of ligands and Cu(BPMEN)I complex

The ligands, i.e., BPMEN, BPMPN, and BPMCN, were synthesized as previously reported [35, 36].

BPMEN: 1H NMR (CDCl3) δ = 8.54 (ddd, J = 4.9, 1.7, 0.8 Hz, 2H), 7.63 (td, J = 7.7, 1.8 Hz, 2H), 7.42 (d, J = 7.8 Hz, 2H), 7.14 (ddd, J = 7.4, 4.9, 1.0 Hz, 2H), 3.69 (s, 4H), 2.65 (s, 4H), 2.28 (s, 6H); 13C NMR (CDCl3) δ = 159.4, 149.1, 136.3, 123.0, 121.9, 64.2, 55.5, 42.9.

BPMPN: 1H NMR (CDCl3) δ = 8.53 (d, J = 4.5 Hz, 2H), 7.63 (t, J = 8.4 Hz, 2H), 7.39 (d,J = 7.8 Hz, 2H), 7.14 (dd, J = 6.8, 5.5 Hz, 2H), 3.65 (s, 4H), 2.54-2.44 (m, 4H), 2.25 (s, 6H), 1.83-1.73 (m, 2H); 13C NMR (CDCl3) δ = 159.5, 149.0, 136.3, 123.0, 121.9, 63.9, 55.8, 42.5, 25.2.

BPMCN: 1H NMR (CDCl3) δ = 8.42 (dt, J = 4.9, 1.3 Hz, 2H), 7.53-7.49 (m, 4H), 7.08-7.01 (m, 2H), 3.79 (dd, J = 48.3, 14.6 Hz, 4H), 2.65-2.55 (m, 2H), 2.22 (s, 6H), 1.92 (dd, J = 10.4, 2.4 Hz, 2H), 1.74-1.65 (m, 2H), 1.22 (td, J = 12.3, 6.4 Hz, 2H), 1.13-1.05 (m, 2H); 13C NMR (CDCl3) δ = 161.2, 148.6, 136.3, 123.0, 121.6, 64.5, 60.4, 36.7, 25.8, 25.8.

CuI (0.25 mmol) and BPMEN (0.25 mmol) were added to CH3CN (1 mL) in an Ar atmosphere, and the mixture was stirred at room temperature for 2 h. After the reaction, the solvent was removed under vacuum to yield a Cu(BPMEN)I complex, which was then washed with CH3CN and diethyl ether and dried under vacuum. HRMS (ESI-MS) calcd for C16H22CuN4 [M − I]+: 333.1121; found: 333.1135. Anal. calcd for C16H22CuN4I· 0.7MeCN: C 42.84%, H 5.12%, N 13.38%; found C 42.69%, H 4.96%, N 13.45%.

2.3. Typical procedure for oxidation of alcohols

The copper salt and the ligand (each 0.025 mmol) were added to CH3CN (1 mL) in an Ar atmosphere and stirred for 30 min. Then 4-OH-TEMPO (0.025 mmol) and substrate (0.5 mmol) were added successively, and the mixture was stirred at room temperature. The reaction progress was checked using thin-layer chromatography. The reaction conversion and yield were obtained from GC measurements using nitrobenzene or nonane as an internal standard, or by column chromatography.

2.4. NMR data of some products

Phenylpropargyl aldehyde: 1H NMR (CDCl3) δ = 9.34 (s, 1H), 7.51 (dd, J = 5.2, 3.2 Hz, 2H), 7.43-7.37 (m, 1H), 7.32 (t, J = 7.5 Hz, 2H); 13C NMR (CDCl3) δ = 175.7, 132.3, 130.3, 127.7, 118.4, 94.1, 87.4.

α-Methylcinnamaldehyde: 1H NMR (CDCl3) δ = 9.47 (s, 1H), 7.45-7.39 (m, 2H), 7.38-7.25 (m, 3H), 7.16 (d, J = 6.7 Hz, 1H), 1.97 (d, J = 1.3 Hz, 3H); 13C NMR (CDCl3) δ = 195.5, 149.8, 138.3, 135.2, 130.1, 129.6, 128.7, 10.9.

Neral: 1H NMR (CDCl3) δ = 9.82 (d, J = 8.2 Hz, 1H), 5.80 (d, J = 8.2 Hz, 1H), 5.07-4.99 (m, 1H), 2.51 (t, J = 7.5 Hz, 2H), 2.16 (q, J = 7.4 Hz, 2H), 1.91 (d, J = 1.2 Hz, 3H), 1.61 (s, 3H), 1.52 (s, 3H); 13C NMR (CDCl3) δ = 190.7, 163.7, 133.6, 128.6, 122.3, 32.5, 27.0, 25.6, 25.0, 17.7.

3. Results and discussion
3.1. Optimization of reaction conditions

Exploratory experiments were performed to test this protocol and screen the reaction conditions using benzyl alcohol as the model substrate (Table 1). The copper complex was generated from a copper salt and N4 ligand, which were stirred for 30 min before every reaction. The copper salt, TEMPO, and N4 ligand were essential in the aerobic oxidation of benzyl alcohol (Table 1, entries 1-3 vs 15-17). Initially, three different ligands, BPMEN, BPMPN, and BPMCN (Scheme 1), in combination with CuBr were used in the oxidation of benzyl alcohol under the same conditions; BPMEN proved to be the best ligand choice (Table 1, entries 1-3), giving 99% yield after prolonging the reaction time to 50 min (Table 1, entry 3). A series of organic solvents were then systematically investigated (Table 1, entries 3-11). The results indicated that the solvent had a significant effect on the oxidation reaction, and 69% yield was obtained in CH3CN within 30 min. The oxidation did not proceed in hexane; this may be attributable to the insolubility of the in situ-generated copper complex or failure to form the corresponding copper complex. A series of copper salts, i.e., CuCl, CuI, and Cu(MeCN)4BF4, were also screened; the best catalytic performance was obtained using CuI as the copper source (Tabel 1, entries 3 and 12-14). CuCl2 and CuBr2 were inert in the reaction, unlike the case for the bpy-Cu system reported by Stahl et al. [21].

Table 1
Oxidation of benzyl alcohol catalyzed by TEMPO/copper salt/N4 ligand, using air as oxidant.

The effects of TEMPO derivatives on the reaction were also examined. Most tested TEMPO derivatives gave above 90% conversions and yields, except 4-CH3SO2O-TEMPO (Table 2, entries 1-7). 4-OH-TEMPO gave nearly quantitative conversion and yield and proved to be the most efficient TEMPO derivative when the oxidation was carried out in the presence of 10% CuI and BPMEN in air for 30 min. The 4-OH-TEMPO, CuI, and BPMEN loadings were then screened (Table 2, entries 8-10). A 91% yield was still achieved with 5% CuI, 5% BPMEN, and 10% 4-OH-TEMPO under the same conditions (Table 2, entry 8). When the loadings of CuI and BPMEN were further reduced to 2%, a comparable yield was obtained by prolonging the reaction time to 1.5 h (Table 2, entry 9). Overall, the reaction proceeded smoothly with 5% each of 4-OH-TEMPO, CuI, and BPMEN, giving 99% yield of aldehyde in 1 h (Table 2, entry 10).

Table 2
Oxidation of benzyl alcohol catalyzed by different TEMPO derivatives.
3.2. Substrate scope and limitations

To examine the utility and generality of the catalyst system in the oxidation of alcohols, we used it for the oxidation of various alcohols; the results are summarized in Table 3. All the primary aromatic alcohols with either electron-donating or electron-withdrawing groups on the benzene ring were smoothly converted to aldehydes, giving more than 85% yields of the desired products; no over-oxidization products were detected. It was found that the electronic properties and steric hindrance of the substituents on the benzene ring influence the reactivity of primary benzyl alcohols; for example, the reactivity of p-OMe-benzyl alcohol was better than those of o- and m-OMe-benzyl alcohol (Table 3, entries 2-4). Benzyl alcohols bearing more electron-donating groups on the benzene ring exhibited higher reactivitiy (Tabel 3, entries 2-4 vs 5 and 6). Moreover, stronger electron-donating groups promoted oxidation (Table 3, entries 2-4 vs 7 and 8).

Table 3
Catalytic aerobic oxidation of various alcohols.

Primary aromatic alcohols with electron-withdrawing groups on the benzene ring had poorer activity than the opposite ones, but comparable yields were obtained by prolonging the reaction time to 24 h in an oxygen atmosphere (Table 3, entries 9-12). A series of heterocyclic benzyl alcohols were also tested. Furfuryl alcohol and 2-thiophenemethanol gave yields of more than 90% (Table 3, entries 13 and 14), but only 8% yield of the desired product was obtained using pyridine-2-methanol as the substrate (Table 3, entry 15). The present protocol was also applicable to the oxidation of allylic alcohols such as cinnamyl alcohol, nerol, and 2-methyl-3- phenyl-2-propen-1-ol and gave good conversions (Table 3, entries 16-18). Propargyl alcohols such as 3-phenylprop-2-yn-1-ol also tolerated the optimized reaction conditions, affording the corresponding aldehyde in 51% yield (Table 3, entry 19). Unfortunately, the catalyst system was unsuitable for the oxidation of aliphatic alcohols and secondary alcohols, giving only 15% phenylacetaldehyde and 7% acetophenone, respectively (Table 3, entries 20 and 21).

The different reactivity of benzyl, allylic, and secondary alcohols suggest that the designed oxidation system should be chemoselective. We chose an equimolar mixture of benzyl alcohol and 1-phenylethanol (0.5 mmol each) as intermolecular oxidation substrates for the reaction under the optimized reaction conditions. The benzyl alcohol was completely transformed to benzaldehyde and 1-phenylethanol was recovered completely after 0.5 h. These results clearly demonstrate the excellent chemoselectivity of the Cu(BPMEN)I/TEMPO system in the aerobic oxidation of alcohols, providing considerable potential advantages in synthetic organic chemistry.

Semmelhack et al. [19] proposed that an oxoammonium species was involved in CuCl/TEMPO-catalyzed aerobic alcohol oxidation with DMF as the solvent, and oxoammonium involvement in alcohol oxidation is well known in many catalyst systems [37, 38, 39]. Based on kinetic experiments, Stahl et al. [21, 22, 23, 24, 25] showed that TEMPO+ is not the active oxidant in the CuI(bpy)/TEMPO catalyst system. Their electrochemical studies also indicated that copper(II) cannot oxidize TEMPO to TEMPO+ under their reaction conditions. Based on the similarity to Stahl’s system [21, 22, 23, 24, 25], electrochemical experiments were performed on a CH3CN solution of CuI, BPMEN, and 4-OH-TEMPO (each 0.06 mmol) after stirring at room temperature for 5 min in Ar atmosphere. The cyclic voltammogram of our developed system shows reversible redox potentials at 0.35 and −0.09 V (vs Fc+/Fc), corresponding to 4-OH-TEMPO+/ 4-OH-TEMPO and CuII/CuI, respectively (Fig. 1). In the case of 4-OH-TEMPO+/4-OH-TEMPO, the one-electron reduction potential is approximately equal to that observed with a solution containing only 4-OH-TEMPO (Fig. 2). However, the one- electron reduction potential shifts from 0.3 to −0.09 V (vs Fc+/Fc) in the presence of BPMEN (Figs. 3 and 4). This indicates that the BPMEN ligand coordinated with CuI, leading to an increase in the overall electron density of the metal ion, resulting in a large negative shift in the reduction potential. This finding is in good accord with Stahl’s system [21]. Copper(II) cannot oxidize 4-OH-TEMPO to 4-OH-TEMPO+ in our catalyst system.

Fig. 1. Cyclic voltammogram of Cu(BPMEN)I/4-OH-TEMPO in CH3CN under Ar. Conditions: 0.6 mmol CuI, 0.6 mmol BPMEN, 0.6 mmol TEMPO, 300 mg LiClO4, scan rate 100 mV/s.

Fig. 2. Cyclic voltammogram of 4-OH-TEMPO under Ar.

Fig. 3. Cyclic voltammogram of CuI under Ar.

Fig. 4. Cyclic voltammogram of CuI + BPMEN (1:1) under Ar.

Electron-spray ionization (ESI)-MS is an effective method for the characterization of reaction mechanisms [40]. An equimolar mixture of CuI and BPMEN was stirred for 30 min, and a major peak appeared at Mr/z = 333.1121; it was assigned to [Cu(BPMEN)]+ (calculated: Mr/z = 333.1135). The formation of the Cu(BPMEN)I complex was also confirmed by elemental analysis. Then 4-OH-TEMPO was added to a copper complex generated from CuI and BPMEN and exposed to air or Ar. The peak corresponding to 4-OH-TEMPO ([M + Na]+ Mr/z = 195.1227) was not detected when the system was exposed to air, but the 4-OH-TEMPO peak was clearly observed under Ar. These results indicate that the transformation of 4-OH-TEMPO must be assisted by both the Cu(BPMEN)I complex and oxygen.

UV-visible spectroscopy was also performed to obtain an insight into the mechanism. An equimolar mixture of CuI and BPMEN (5 mmol/L in CH3CN) was stirred in Ar atmosphere for 30 min; the maximum absorption was located at 705 nm (Cu(BPMEN)I complex); the maximum absorption of 4-OH-TEMPO (5 mmol/L in CH3CN) added to the above mixture in air was 810 nm (Fig. 5). When oxidation took place after the addition of benzyl alcohol, the intensity of the absorption peak at 810 nm gradually decreased, and a new shoulder appeared at 675 nm as the reaction proceeded. This indicates that the species causing the absorption at 810 nm may be the real intermediate responsible for the oxidation of alcohols.

Fig. 5. UV-vis spectra in CH3CN. Conditions: 5 mmol/L BPMEN, 5 mmol/L CuI, 5 mmol/L 4-OH-TEMPO in CH3CN.
4. Conclusions

We developed an easily obtained and efficient catalyst system consisting of CuI, BPMEN, and 4-OH-TEMPO for aerobic oxidation of primary alcohols under mild conditions. In all cases, primary benzyl and allylic alcohols were selectively oxidized to the corresponding aldehydes, and no over-oxidized products such as carboxylic acids were observed. The catalyst system exhibited excellent chemoselectivity for the oxidation of primary alcohols. Electrochemical experiments further revealed that copper(II) was not able to oxidize 4-OH-TEMPO to 4-OH- TEMPO+. HRMS confirmed that the transformation of 4-OH- TEMPO must be assisted by both the Cu(BPMEN)I complex and air. The UV-visible spectra clearly indicated that a new species was formed in the presence of the Cu(BPMEN)I complex and 4-OH-TEMPO in air, which showed good reactivity in alcohol oxidation. More investigations of the proposed mechanism and application of the Cu(BPMEN)I/4-OH-TEMPO) catalyst system to oxidation reactions are underway in our laboratory.

References
[1] de Montellano P R O. Cytochrome P450: Structure, Mechanism, and Biochemistry. 2nd Ed. New York: Plenum Press, 1995
[2] Turlington C R, White P S, Brookhat M, Templeton J L. J Am Chem Soc, 2014, 136: 3981
[3] Jacobsen E N. In: Abel E W, Stone F G A, Wilkinson G Eds. Comprehensive Organometallic Chemistry Ⅱ. Oxford (UK): Pergamon, 1995
[4] McDonald A R, Que L Jr. Coord Chem Rev, 2013, 257: 414
[5] White M C, Doyle A G, Jacobsen E N. J Am Chem Soc, 2001, 123: 7194
[6] Prat I, Font D, Company A, Junge K, Ribas X, Beller M, Costas M. Adv Syn Catal, 2013, 355: 947
[7] Chen M S, White M C. Science, 2007, 318: 783
[8] Wu M, Wang B, Wang S F, Xia C G, Sun W. Org Lett, 2009, 11: 3622
[9] Wang B, Miao C X, Wang S F, Xia C G, Sun W. Chem Eur J, 2012, 18: 6750
[10] Wang B, Wang S F, Xia C G, Sun W. Chem Eur J, 2012, 18: 7332
[11] Sheldon R A, Arends I W C E, ten Brink G J, Dijksman A. Acc Chem Res, 2002, 35: 774
[12] Zhang H, Fu L L, Zhong H M. Chin J Catal (张华, 付罗岭, 钟红敏. 催化学报), 2013, 34: 1848
[13] Wang L Y, Li J, Lv Y, Zhao G D, Gao S. Appl Organomet Chem, 2012, 26: 37
[14] Fritz-Langhals E. Org Process Res Dev, 2005, 9: 577
[15] Mu R Z, Liu Z Q, Yang Z J, Liu Z G, Wu L M, Liu Z L. Adv Synth Catal, 2005, 347: 1333
[16] Yu Y L, Gao B J, Li Y F. Chin J Catal (余依玲, 高保娇, 李艳飞. 催化学报), 2013, 34: 1776
[17] Vogler T, Studer A. Synthesis, 2008: 1979
[18] Sasano Y, Nagasawa S, Yamazaki M, Shibuya M, Park J, Iwabuchi Y. Angew Chem Int Ed, 2014, 53: 3236
[19] Semmelhack M F, Schmid C R, Cortes D A, Chou C S. J Am Chem Soc, 1984, 106: 3374
[20] Markό I E, Giles P R, Tsukazaki M, Chellé-Regnaut I, Gautier A, Brown S M, Urch C J. J Org Chem, 1999, 64: 2433
[21] Hoover J M, Stahl S S. J Am Chem Soc, 2011, 133: 16901
[22] Greene J F, Hoover J M, Mannel D S, Root T W, Stahl S S. Org Process Res Dev, 2013, 17: 1247
[23] Steves J E, Stahl S S. J Am Chem Soc, 2013, 135: 15742
[24] Hoover J M, Ryland B L, Stahl S S. ACS Catal, 2013, 3: 2599
[25] Hoover J M, Ryland B L, Stahl S S. J Am Chem Soc, 2013, 135: 2357
[26] Mannam S, Alamsetti S K, Sekar G. Adv Synth Catal, 2007, 349: 2253
[27] Jiang N, Ragauskas A J. ChemSusChem, 2008, 1: 823
[28] Lu Z L, Costa J S, Roubeau O, Mutikainen I, Turpeinen U, Teat S J, Gamez P, Reedijk J. Dalton Trans, 2008: 3567
[29] Velusamy S, Punniyamurthy T. Eur J Org Chem, 2003: 3913
[30] Contel M, Izuel C, Laguna M, Villuendas P R, Alonso P J, Fish R H. Chem Eur J, 2003, 9: 4168
[31] Miao C X, Wang J Q, Yu B, Cheng W G, Sun J, Chanfreau S, He L N, Zhang S J. Chem Commun, 2011, 47: 2697
[32] Miao C X, He L N, Wang J L, Wu F. J Org Chem, 2010, 75: 257
[33] Miao C X, He L N, Wang J Q, Gao J. Synlett, 2009: 3291
[34] Miao C X, He L N, Wang J Q, Wang J L. Adv Synth Catal, 2009, 351: 2209
[35] Karmakar T K, Aromi G, Ghosh B K, Usman A, Fun H K, Mallah T, Behrens U, Solans X, Chandra S K. J Mater Chem, 2006, 16: 278
[36] Yu S J, Miao C X, Wang D Q, Wang S F, Xia C G, Sun W. J Mol Catal A, 2012, 353-354: 185
[37] Semmelhack M F, Chou C S, Cortés D A. J Am Chem Soc, 1983, 105: 4492
[38] Wang X L, Liu R H, Jin Y, Liang X M. Chem Eur J, 2008, 14: 2679
[39] Liu R H, Liang X M, Dong C Y, Hu X Q. J Am Chem Soc, 2004, 126: 4112
[40] Cole R B. Electrospray Ionization Mass Spectrometry: Fundamentals, Instrumentation, and Applications. New York: Wiley, 1997
基于四氮配体的铜配合物/TEMPO催化的苄醇与烯丙基醇的氧化反应
张淑芳a,b, 苗成霞a, 徐大乾a, 孙伟a , 夏春谷a     
a 中国科学院兰州化学物理研究所羰基合成与选择氧化国家重点实验室, 甘肃兰州730000;
b 中国科学院大学, 北京100049
收稿日期: 2014-5-1. 接受日期: 2014-6-3. 出版日期: ##FaBuRiQi##.
基金来源：国家自然科学基金(21103207,21133011).
*通讯联系人.苗成霞, 电子信箱: chxmiao@licp.cas.cn
孙伟, 传真: (0931)8277088; 电子信箱: wsun@licp.cas.cn
摘要：发展了非血红素类四氮配体的铜配合物和2,2,6,6-四甲基哌啶-氮-氧自由基(TEMPO)相结合的催化体系, 应用于分子氧参与的伯醇氧化反应. 该体系具有条件温和、高效、高选择性、无需任何助剂和底物(包括苄醇、烯丙基醇和含杂原子伯醇)使用性强等优点. 此外, 利用高分辨质谱和紫外-光谱等对反应活性中间体进行了初步探讨.
关键词： 醇     需氧氧化     四氮配体     铜     2,2,6,6-四甲基哌啶-氮-氧自由基    

1. 前言

含氮配体的金属配合物作为单氧酶活性位点的模型在有机反应尤其是氧化反应中备受关注[1, 2, 3]. 近年来, 随着仿生催化技术的发展, 为模拟在生物代谢过程中发挥重要作用的金属卟啉, 如细胞色素P450[1], 科学家们合成了一系列四氮配体并将其配合物应用到许多反应体系中[4, 5, 6, 7]. 我们课题组也在四氮配体及其金属配合物催化烯烃不对称环氧化方面取得了一定进展[8, 9, 10]. 然而, 有关四氮配体配合物用于催化醇的氧化反应却未见报道.

醛酮类化合物是药物、塑料、香料及染料生产中重要的中间体, 因此醇被选择性地氧化成相应的醛、酮化合物在有机合成中发挥着重要作用. 近年来, 基于2,2,6,6-四甲基哌啶-氮-氧自由基(TEMPO)的催化体系成为实现醇氧化反应的重要方法[11, 12, 13]. 但这类体系催化的氧化反应往往需要化学计量的氧化剂, 如次氯酸钠和高价碘(III)等, 因此在反应过程中会产生大量有毒副产物[14, 15], 且所需反应时间较长, 反应温度较高[16].

从绿色化学角度考虑, 选择高效、环保的醇氧化方法, 以空气、氧气和过氧化氢为清洁氧源尤为重要. 然而, 成本较高的TEMPO并不能直接活化氧气, 必须有共催化剂参与[17].

铜配合物作为半乳糖单核铜酶的功能化模型成为人们的研究热点[18]. 早在1984年, Semmelhack课题组[19]首次报道了CuCl/TEMPO/DMF催化体系, 当向体系中加入10% CuCl/TEMPO时, 室温下就可高效、高选择性地实现苄醇和烯丙基醇的氧化. 之后, 有关醇氧化的研究主要集中在含氮配体的设计上, 如1,10-邻二氮菲[20]、2,2’-联吡啶[21, 22, 23, 24, 25]、1,4-二氮杂二环[2.2.2]辛烷[26]、4-吡咯烷基吡啶[27]、咪唑-吡啶类[28]、salen-H4[29]和1,4,7-氮杂壬烷[30]等. Stahl课题组[21, 22, 23, 24, 25]发现, 在Cu(OTf)-联吡啶/TEMPO催化体系中加入一定量的氮甲基咪唑, 即可在空气或氧气存在下高效、高选择性地实现苄醇、烯丙基醇、脂肪族伯醇及含杂原子的伯醇的氧化. 该体系的不足之处在于所用铜盐Cu(OTf)为非商品化药品, 且制备过程较为复杂.

尽管TEMPO/Cu体系在催化分子氧参与的醇氧化反应方面取得了较大进展, 但在温和条件下利用简单的催化体系高效、高选择性地实现醇氧化反应仍是难点.

鉴于我们课题组在TEMPO催化醇氧化体系[31, 32, 33, 34]及四氮配体合成和应用方面的经验[8, 9, 10], 我们推断四氮配体/亚铜离子和TEMPO相结合的体系可能较好地催化醇氧化反应(图式1), 实验结果发现Cu(BPMEN)I/TEMPO催化体系在室温下, 以空气为氧化剂, 无需添加任何助剂就可以高效、高选择性地实现苄醇和烯丙基醇的氧化.

2. 实验部分
2.1. 仪器信息

采用Agilent公司7890A/5975C型气相色谱-质谱联用仪记录气相色谱-质谱图. 用Bruker Avance III型400 MHz核磁共振仪测定1H NMR和13C NMR谱图(化学位移和耦合常数分别用百万分之一(ppm)和Hz表示). 气相色谱图用Agilent公司带有离子火焰化检测器的6820型仪器记录. 采用Bruker Daltonics micro TOF-QII型质谱仪记录高分辩质谱, 利用Agilent Cary 60型光谱仪检测紫外-可见光谱. 以铂电极作工作电极, 非水Ag+/Ag作参比电极, 铂丝作对电极, 采用CHI 660E型恒电势器测定电化学电极电势, 扫描速率为100 mV/s. 化合物的元素分析在Vario EL型元素分析仪上测定.

2.2. 配体和配合物的制备与表征

配体N,N'-二甲基-N,N'-二(2-吡啶甲基)-1,2-乙二胺(BPMEN), N,N'-二甲基-N,N'-二(2-吡啶甲基)-1,3-丙二胺(BPMPN)和反式-N,N'-二甲基-N,N'-二(2-吡啶甲基)-1,2-环己二胺(BPMCN)根据文献[35, 36]合成.

BPMEN: 1H NMR (CDCl3) δ = 8.54 (ddd, J = 4.9, 1.7, 0.8 Hz, 2H), 7.63 (td, J = 7.7, 1.8 Hz, 2H), 7.42 (d, J = 7.8 Hz, 2H), 7.14 (ddd, J = 7.4, 4.9, 1.0 Hz, 2H), 3.69 (s, 4H), 2.65 (s, 4H), 2.28 (s, 6H); 13C NMR (CDCl3) δ = 159.4, 149.1, 136.3, 123.0, 121.9, 64.2, 55.5, 42.9.

BPMPN: 1H NMR (CDCl3) δ = 8.53 (d, J = 4.5 Hz, 2H), 7.63 (t, J = 8.4 Hz, 2H), 7.39 (d, J = 7.8 Hz, 2H), 7.14 (dd, J = 6.8, 5.5 Hz, 2H), 3.65 (s, 4H), 2.54-2.44 (m, 4H), 2.25 (s, 6H), 1.83-1.73 (m, 2H); 13C NMR (CDCl3) δ = 159.5, 149.0, 136.3, 123.0, 121.9, 63.9, 55.8, 42.5, 25.2.

BPMCN: 1H NMR (CDCl3) δ = 8.42 (dt, J = 4.9, 1.3 Hz, 2H), 7.53-7.49 (m, 4H), 7.08-7.01 (m, 2H), 3.79 (dd, J = 48.3, 14.6 Hz, 4H), 2.65-2.55 (m, 2H), 2.22 (s, 6H), 1.92 (dd, J = 10.4, 2.4 Hz, 2H), 1.74-1.65 (m, 2H), 1.22 (td, J = 12.3, 6.4 Hz, 2H), 1.13-1.05 (m, 2H); 13C NMR (CDCl3) δ = 161.2, 148.6, 136.3, 123.0, 121.6, 64.5, 60.4, 36.7, 25.8, 25.8.

在氩气保护下, 向装有1.0 mL CH3CN的反应管中加入CuI (47.6 mg, 0.25 mmol)和BPMEN配体(67.6 mg, 0.25 mmol), 室温下搅拌2.0 h. 反应结束后, 蒸干溶剂, 分别用乙腈和乙醚洗涤配合物, 除去溶剂, 得到配合物Cu(BPMEN)I. 其高分辨质谱表征C16H22CuN4 [M-I]+: 333.1135, 理论计算值为333.1121; 其元素分析表征 C16H22CuN4I·0.7MeCN: C 42.69%, H 4.96%, N 13.45%; 理论计算值: C 42.84%, H 5.12%, N 13.38%.

2.3. 氧化反应的一般步骤

在Ar气保护下, 向装有1.0 mL CH3CN的反应管中加入铜盐(0.025 mmol)和配体(0.025 mmol), 反应0.5 h. 然后依次加入TEMPO衍生物(0.025 mmol)和醇(0.5 mmol), 在室温下空气中反应, 通过薄层色谱观察反应进度. 反应结束后, 以硝基苯或正壬烷为内标, 用气相色谱测得反应转化率和收率, 或者柱分离得到分离收率.

2.4. 部分产物的核磁数据

苯丙炔醛: 1H NMR (CDCl3) δ = 9.34 (s, 1H), 7.51 (dd, J = 5.2, 3.2 Hz, 2H), 7.43-7.37 (m, 1H), 7.32 (t, J = 7.5 Hz, 2H); 13C NMR (CDCl3) δ= 175.7, 132.3, 130.3, 127.7, 118.4, 94.1, 87.4.

α-甲基肉桂醛: 1H NMR (CDCl3) δ = 9.47 (s, 1H), 7.45-7.39 (m, 2H), 7.38-7.25 (m, 3H), 7.16 (d, J = 6.7 Hz, 1H), 1.97 (d, J = 1.3 Hz, 3H); 13C NMR (CDCl3) δ = 195.5, 149.8, 138.3, 135.2, 130.1, 129.6, 128.7, 10.9.

橙花醛: 1H NMR (CDCl3) δ = 9.82 (d, J = 8.2 Hz, 1H), 5.80 (d, J = 8.2 Hz, 1H), 5.07-4.99 (m, 1H), 2.51 (t, J = 7.5 Hz, 2H), 2.16 (q, J = 7.4 Hz, 2H), 1.91 (d, J = 1.2 Hz, 3H), 1.61 (s, 3H), 1.52 (s, 3H); 13C NMR (CDCl3) δ = 190.7, 163.7, 133.6, 128.6, 122.3, 32.5, 27.0, 25.6, 25.0, 17.7.

3. 结果与讨论
3.1. 优化反应条件

以苯甲醇氧化为模板反应, 对反应条件进行考察(表1). 在反应之前先将等当量的铜盐与配体原位反应0.5 h, 生成相应的配合物. 空白实验表明, 体系中不加铜盐、TEMPO和四氮配体中任一组分均不能实现苯甲醇的氧化反应(表1, 实验1-3对15-17). 在相同条件下, 以CuBr为铜盐, 对不同配体进行筛选, 结果发现BPMEN为最优配体(表1, 实验1-3), 当反应时间为50 min, GC测得收率可达99%. 之后, 利用CuBr-BPMEN对溶剂进行筛选(表1, 实验3-11), 发现溶剂对氧化反应也有较大影响. 当以CH3CN为介质时, 反应0.5 h收率为69%, 体系表现出最好的活性, 而以正己烷为溶剂时, 由于配合物的溶解性较差, 该催化体系并未表现出活性. 此外, 还考察了不同铜盐对氧化反应的影响, 发现当以CuI为铜源, 反应0.5 h, 转化率和收率分别可达94%和92%, 表现出最好的催化活性(表1, 实验3和12-14), 而二价铜盐CuCl2和CuBr2对苯甲醇的氧化几乎没有催化活性. 该结果与Stahl等[21]报道的联吡啶/Cu(OTf)体系有所不同.

此外, 我们还比较了不同TEMPO衍生物的催化活性. 在氧化反应中, 大部分TEMPO的衍生物表现出相近的催化活性, 反应0.5 h, 除4-CH3SO2O-TEMPO催化的氧化反应转化率和收率低于80%外, 几乎所有TEMPO衍生物参与的氧化反应转化率和收率都达到90%以上(表2, 实验1-7). 当向体系中加入10% CuI和10% BPMEN, 以空气为氧化剂, 反应0.5 h, 4-OH-TEMPO表现出最好的催化活性, 实现了苯甲醇的定量转化. 此外, 还对催化剂和TEMPO用量进行了筛选(表2, 实验8-10). 当将CuI和BPMEN用量降低到5%, TEMPO仍为10%, 反应0.5 h时, 苯甲醇氧化收率也可达到91% (表2, 实验8). 进一步降低催化剂量至2%, 反应1.5 h, 也可达到相当高的收率(表2, 实验9). 当CuI, BPMEN和TEMPO用量都为5%时, 反应1 h后, 也可实现苯甲醇定量转化(表2, 实验10).

3.2. 底物拓展

在最优条件下对该催化体系进行了底物拓展(表3). 结果发现, 几乎所有的苄醇都可以被高效氧化, 收率可达85%以上, 且无酸生成. 但是, 该体系易受电子效应和空间位阻影响. 对甲氧基苯甲醇相比邻、间位表现出更好的反应活性(表3, 实验2-4), 且苯环上带有的供电子基越多, 反应活性越高(表3, 实验2-4对5-6), 供电子能力越强, 越有利于氧化反应发生(表3, 实验2-4对7-8).

对于苄醇氧化, 当苯环带有供电子基时, 反应要比带吸电子基的苄醇的氧化容易得多. 如苯环上带有甲基、甲氧基和异丙基时, 反应可以在空气中2.0-4.0 h完成; 而当苯环上带有氯或溴等吸电子基时, 反应则必须在氧气中进行, 且需要24 h才能达到相当的转化率和收率(表3, 实验9-12). 除此之外, 还考察了带有杂原子伯醇的氧化, 如呋喃甲醇和噻吩甲醇都可以在常温下空气中氧化, 可以得到90%的醛(表3, 实验13-14), 但对于吡啶甲醇的氧化, 收率只有8% (表3, 实验15). 该催化体系除了适用于苄醇氧化之外, 还适用于烯丙基醇的氧化, 如肉桂醇和橙花醇等(表3, 实验16-18). 另外, 炔丙基醇如3-苯基-2-丙炔-1-醇在该体系中氧化也能得到51%的收率(表3, 实验19). 该体系对脂肪族伯、仲醇氧化的催化效率较低, 生成的苯乙醛和苯乙酮收率仅为15%和7% (表3, 实验20和21).

研究结果表明, Cu(BPMEN)I/4-OH-TEMPO体系适用于催化苄醇和烯丙基伯醇的氧化, 而对仲醇的氧化效果较差. 因此, 利用苯甲醇和α-苯乙醇在同一体系中竞争氧化, 考察了该催化体系的化学选择性. 将苯甲醇(54.1 mg, 0.5 mmol)和α-苯乙醇(61.1 mg, 0.5 mmol)同时加入到催化体系中, 反应0.5 h, 薄层色谱跟踪检测, 苯甲醇全部转化为苯甲醛, 而α-苯乙醇不发生反应. 此结果表明, Cu(BPMEN)I/4-OH-TEMPO体系对催化苄醇和烯丙基醇氧化具有很好的选择性.

Semmelhack等[19]认为在CuCl/TEMPO/DMF催化醇氧化反应中有TEMPO+的参与; 且TEMPO+在许多催化体系中都得到认可[37, 38, 39]. Stahl课题组[21, 22, 23, 24, 25]基于动力学研究, 发现TEMPO+并不是Cu(OTf)-联吡啶/TEMPO催化体系的活性物种. 电化学表征也表明在该反应条件下CuII不能氧化TEMPO生成TEMPO+. 由于Cu(BPMEN)I/ 4-OH-TEMPO催化体系与Cu(OTf)-联吡啶/TEMPO体系存在一定的相似性[21, 22, 23, 24, 25], 我们也利用电化学方法(以非水银电极作参比)测定了Cu(BPMEN)I/ TEMPO催化体系中的氧化还原电势. 将CuI (11.4 mg, 0.06 mmol), BPMEN (16.2 mg, 0.06 mmol)和4-OH- TEMPO (10.3 mg, 0.06 mmol)加入到3.0 mL CH3CN中, 并在Ar气氛围下搅拌5 min, 实验测得该体系中两个电极电势0.35 V和-0.09 V (vs Fc+/Fc)分别对应于4-OH-TEMPO+/4-OH-TEMPO和Cu(BPMEN)II/Cu(BPMEN)I (图1), 反应体系中4-OH- TEMPO+/4-OH-TEMPO的电极电势接近于4-OH- TEMPO+/ 4-OH-TEMPO本身的电极电势(图2). 此外, 向CuI的CH3CN溶液中加入等当量配体BPMEN, 由于发生配位作用, 增加了中心金属离子的电子云密度, 因此电极电势从0.3 V降低到-0.09 V (图3和图4), 该实验结果与Stahl体系中所测结果一致[21], 当有配体存在时CuII不能氧化TEMPO生成TEMPO+.

我们还利用高分辨质谱(ESI-MS)[40]对反应机理进行了研究. 在惰气气氛下, 等当量的CuI和BPMEN在CH3CN中搅拌0.5 h, 利用高分辨质谱发现, CuI和BPMEN可发生配位作用生成[Cu(BPMEN)]+(Mr/z = 333.1121). 元素分析也证实了该体系中配合物Cu(BPMEN)I的生成. 然后分别向空气存在和惰气保护的体系中加入等当量的4-OH-TEMPO, 在惰气保护的体系中, 高分辨质谱很容易捕捉到4-OH-TEMPO([M+Na]+ Mr/z = 195.1227), 而在空气存在的体系中没有捕捉到4-OH-TEMPO, 说明4-OH-TEMPO必须在氧气参与下才能与CuI和BPMEN作用生成可能的活性中间体.

紫外-可见光谱(UV-vis)也为醇氧化机理解释提供了一定依据. 将等当量的CuI和BPMEN溶于CH3CN (5.0 mmol/L), 在Ar气氛围下搅拌0.5 h, 利用UV-vis检测发现, 在705 nm处有一个很强的吸收峰对应于CuI(BPMEN)的配合物. 当向体系中通入空气并加入4-OH-TEMPO (5 mmol/L)时, 705 nm处的吸收峰逐渐消失, 而在810 nm出现了最大吸收峰, 加入底物之后, 随着反应的进行, 810 nm处的吸收峰逐渐降低直至消失, 最后在675 nm出现了最大吸收. 因此我们推测对应于810 nm的吸收峰可能为该体系活性中间体的峰(图5).

4. 结论

发展了一个简单、有效且高选择性的CuI/BPMEN/ 4-OH- TEMPO催化体系, 该体系能够在温和条件下催化苄醇和烯丙基醇的有氧氧化生成相应的醛, 且无酸生成. 电化学实验表明CuII不能氧化TEMPO生成TEMPO+. 高分辨质谱证实4-OH-TEMPO必须在配合物Cu(BPMEN)I和空气同时参与的情况下才能发生转化. 紫外-可见光谱表明, Cu(BPMEN)I和4-OH-TEMPO能够在空气氛围下生成可能的活性中间体. 有关该体系的催化醇氧化机理及其在其他氧化反应中的应用有待于进一步研究.