Grignard Reactions in Cyclopentyl Methyl Ether - Wiley Online Library

DOI: 10.1002/ajoc.201600059

Full Paper

Grignard Reactions

Grignard Reactions in Cyclopentyl Methyl Ether Shoji Kobayashi,* Keisuke Shibukawa, Yuta Miyaguchi, and Araki Masuyama[a] of the recovered solvent revealed that CPME was stable under the Grignard conditions. Some of the Grignard reagents in CPME were stable for at least several months and therefore these solutions could replace existing commercial reagents. Tramadol hydrochloride, an analgesic that is used worldwide, and tamoxifen, an anti-estrogenic drug for estrogen-receptor-positive breast cancer were synthesized using Grignard reactions in CPME.

Abstract: We undertook a systematic study of Grignard reactions using cyclopentyl methyl ether (CPME) as a reaction solvent. Diisobutylaluminum hydride was found to be the most appropriate activator of magnesium and a number of Grignard reagents could be prepared as homogeneous or heterogeneous solutions in CPME. The CPME used in the Grignard reaction was efficiently recycled without affecting the reaction yield of the second experiment. GC–MS analysis

Introduction

About ten years ago, the Zeon Corporation introduced the other green hydrophobic ether, cyclopentyl methyl ether (CPME), as an alternative to conventional ether solvents.[3] Its application in organic synthesis has increasingly broadened,[4] and in some cases better selectivity was obtained with this solvent system.[5] We independently reported that CPME served as a radical reaction solvent and demonstrated a number of radical reactions including radical additions, radical deoxygenations, radical cyclizations, and radical-containing one-pot reactions.[6] Notably, CPME could be recycled by the normal workup procedures (extraction and distillation) without affecting the reaction yield of the subsequent experiment.[6] These promising results, along with objective assessments[7] regarding safety, health, and environmental aspects prompted us to examine more applications of CPME in organic reactions. Although several reports on Grignard reactions with CPME have been published,[8] most are focused on the specific substrates. To the best of our knowledge, there are no commercially available Grignard reagents in CPME solution. Additionally, it is unknown whether CPME has been deemed unsuitable for Grignard reactions or if its adaptability had not yet been explored. Herein, we report a comprehensive study of Grignard reactions in CPME. Our results show that many Grignard reagents could be prepared as CPME solutions with acceptable yield and quality, and some reagents could be stored for several months without decomposition. Recycling of the solvent after the Grignard reaction is also demonstrated.

The Grignard reaction is one of the most important reactions in organic chemistry because it constructs a carbon–carbon bond in a simple operation with broad substrate applicability. The most commonly used organic solvents for Grignard reactions are diethyl ether (Et2O) and THF; in fact, a number of Grignard reagents in both of these solvents are commercially available. However, these typical solvent systems often pose issues when applied to large-scale synthesis in the manufacturing process of chemical products. The major drawbacks of Et2O are its extreme flammability and anesthetic properties, whereas THF gradually forms explosive peroxide and is difficult to recover due to its miscibility with water. The biomass-derived 2methyltetrahydrofuran (2-MeTHF) has emerged as an alternative “green” solvent, and its broad application in organic reactions including Grignard and other organometallic reactions has been reported.[1] Importantly, switching the solvent to such an environmentally benign alternative has made existing industrial chemical processes more green and “sustainable”.[2] In fact, a significant number of companies are using 2-MeTHF in the manufacture of pharmaceutically active ingredients.[1]

[a] Dr. S. Kobayashi, K. Shibukawa, Y. Miyaguchi, Prof. Dr. A. Masuyama Department of Applied Chemistry Faculty of Engineering Osaka Institute of Technology 5-16-1 Ohmiya, Asahi-ku Osaka 535-8585 (Japan) E-mail: [email protected]

Results and Discussion

Supporting information for this article can be found under http:// dx.doi.org/10.1002/ajoc.201600059.

Optimization of Reaction Conditions

Ó 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial, and no modifications or adaptations are made.

Our study commenced with preparation of a Grignard reagent in CPME with 3-bromoanisole (1 b) as a representative substrate. It was our intention to eventually adopt the established protocol for the synthesis of tramadol (vide infra), an analgesic

Asian J. Org. Chem. 2016, 5, 636 – 645

636

Ó 2016 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper agent now being used in over one hundred countries. Initial attempts revealed that the reaction did not proceed at room temperature even in the presence of activators of magnesium and therefore gentle heating was necessary during the addition of aryl bromide 1 b. The optimal temperature was found to be 60 8C, and subsequent work revealed that degradation of the Grignard reagent occasionally occurred at higher temperatures. By adjusting the stoichiometry it was clear that a small excess of magnesium (1.5 equiv) relative to aryl bromide 1 b (1 equiv) was appropriate. The concentration was more critical, with initiation failing to take place under dilute conditions such as at 0.1 m. In contrast, at higher concentrations, a muddy supernatant was obtained, which indicated that the forming Grignard reagent precipitated out due to its low solubility in CPME. The optimal concentration for the preparation of 2 b was around 1 m with a deep-colored precipitate of fine particles settling at the bottom of the flask (Figure 1 a). The solution

Table 2. Screening of activators.[a]

Entry

Activator

t [min][b]

Conc. [m][c]

Yield [%][d] 2b 4b

1 2 3 4 5 6 7 8

– I2 BrCH2CH2Br DIBALH Red-Al LiAlH4 BH3·SMe2 NaBH4

60 30 30 20 14 40 40 60

0.73 0.76 0.74 0.81 0.77 0.79 0.79 0.80

58 61 59 65 62 63 63 64

92 89 83 87 90 88 88 88

[a] Typical procedure: the activator (0.15 mmol) was added to a suspension of well-ground Mg turnings (30 mmol) in CPME (3 mL). After 30 min at room temperature, the suspension was warmed to 60 8C followed by the addition of 1 b (20 mmol) in CPME (13 mL) through a dropping funnel at a rate of one drop every 5 s. Initiation of the reaction was visually confirmed by the change of color and the solution state. The reaction mixture was stirred at 60 8C for 3 h in total and left at room temperature until two phases were clearly separated. The supernatant liquid was titrated to determine the concentration of the Grignard reagent. Then the solution of 2 b (1.2 equiv) was added to 3 (1 equiv) in CPME (2 mL) and the mixture was stirred for 30 min at room temperature. The usual workup and purification afforded 4 b. [b] The amount of time required for initiation (visual confirmation). [c] Concentration of 2 b determined by titration. [d] The yield of 2 b was estimated from the concentration. The isolated yield of 4 b is based on 3.

Figure 1. Photographs of 2 b in a) CPME, b) THF, and c) 2-MeTHF solution.

Tilstam and Weinmann previously reported that diisobutylaluminum hydride (DIBALH) could activate the surface of magnesium and the initiation and the formation steps could be performed at lower temperatures.[11] More recently, Zhang et al. examined Grignard reactions in a variety of solvents and, in some cases, obtained better results by switching the activator from I2 to DIBALH.[12] Although DIBALH or other reducing agents are not commonly used for the activation of magnesium in laboratory experiments, we explored the potential of hydride reductants to facilitate Grignard formation in CPME (Table 2). As reported previously,[11, 12] smoother formation of the Grignard reagent 2 b occurred when DIBALH was used as the activator with a marginal improvement in the yield (Table 2, entry 4). Sodium bis(2-methoxyethoxy)aluminum hydride (Red-Al) was also effective for the activation of magnesium[11] (Table 2, entry 5). The activation potency of lithium aluminum hydride (LiAlH4) and boron–dimethyl sulfide complex (BH3·SMe2) was weak, whereas sodium borohydride (NaBH4), a milder reducing agent, gave no acceleration of the formation of 2 b (Table 2, entries 6–8). Although the data shown in Table 2 are specific to 3-bromoanisole (1 b), an improvement of approximately 10 % yield by DIBALH activation relative to I2 activation was similarly obtained in the formation of tert-butylmagnesium bromide (2 t, Table 3). It is worth noting that only 0.5 mol % of the activator relative to magnesium was required for smooth initiation of the reactions. By taking these results into account, together with the costs and the problem of dust arising from the use of powdered reagents, DIBALH was chosen as the most appropriate activator of the Grignard reactions in CPME and was used in subsequent experiments.

state was obviously different from that in THF or 2-MeTHF, as use of the latter solvents resulted in colored homogeneous mixtures (cf. Figure 1 a with 1 b and 1 c). Nevertheless, the Grignard reaction between 2 b and benzaldehyde (3) proceeded in high yields irrespective of the solvent used[9] (Table 1). During our investigations, we found that activators of magnesium moderately affected the efficiency of the Grignard formation (Table 2). In general, molecular iodine or 1,2-dibromoethane have been applied to activate the surface of magnesium turnings.[10] Indeed, the amount of time required for initiation was shortened with these activators (Table 2, entries 2 and 3). However, we later encountered difficulty in the preparation of some Grignard reagents despite the use of these activators.

Table 1. The Grignard reaction of 1 b in different ethers.

Entry

Solvent

Conc. [m][a]

Yield of 2 b [%][b]

Yield of 4 b [%][c]

1 2 3

CPME THF 2-MeTHF

0.81 0.97 1.02

65 78 82

87 92 91

[a] Concentration of 2 b determined by titration. [b] Yield of 2 b was estimated from the concentration. [c] Isolated yield of 4 b based on 3 (0.83 equiv relative to 2 b). Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

637


Full Paper data in Table 3, fluorobenzene derivatives (1 l–n) and alkyl bromides except for tertiary and vinyl bromides tended to produce homogeneous solutions. Fortunately, the solubility was not the major factor in determining the efficiency of the Grignard reactions. Indeed, inverse additions of 3 into the heterogeneous solutions did not affect the product yields significantly (data not shown). These facts indicate that fine precipitates appearing during the formation stage did not inhibit the addition reactions, although their composition was not identified. We therefore conclude that it is advisable to change the addition methods to suit the situation. We were surprised to find that 2-chlorophenyl bromide (1 i) and 2-fluorophenyl bromide (1 l) were not suitable substrates for the Grignard reactions, although titration indicated moderate formation of the Grignard reagents (Table 3, entries 9 and 12). For these substrates, the inverse addition method was not fruitful and gave complex mixtures of unidentifiable products. It can be assumed that extreme lability of the intermediates (2 i and 2 l) led to rapid decomposition of the reagents. In fact, 2 i and 2 l are not commercially available. Another remarkable point is the formation of tert-butyl-, allyl-, and benzylmagnesium bromides (Table 3, entries 20–22). Experimentally, these Grignard reagents were formed more readily than others at 60 8C but were prone to dimerizing into the Wurtz coupling products. After many trials, this side reaction could be minimized by a careful dropping of alkyl bromides at room temperature or into a water-cooled suspension of magnesium turnings, giving rise to the addition products 4 t–v in appreciable yields. (E)-(2-Bromovinyl)benzene (1 y) also tended to produce the Wurtz coupling product (45 % yield based on 1 y) under the normal conditions at 60 8C (Table 3, entry 25). In this case, however, the side reaction was not suppressed by lowering the temperature because the substrate 1 y was inert toward activation below 40 8C. Among the substrates examined, only the Grignard reagents derived from 3-bromothiophene (1 p)[13] and, surprisingly, propargyl bromide (1 w) could not be prepared. In the latter case, an exothermic initiation occurred upon addition of 1 w at room temperature; however, evolution of heat ceased after a short while and two immiscible phases appeared. Addition of catalytic amounts of HgCl2 or ZnBr2[14] to facilitate formation of the propargyl Grignard reagent was ineffective. After evaluating the performance of CPME for the reactions of aryl and alkyl bromides, we undertook the preparation of the Grignard reagents from chloride substrates. From the industrial aspect, chlorides are more desirable than bromides in terms of cost and stability. However, it is obvious that the reactivity of chlorides is much lower than that of bromides, which complicates the Grignard formation. Indeed, a previous report described only the formation of benzylmagnesium chloride in the Grignard reaction with CPME,[12] and the adaptability to other chloride substrates was not disclosed. Table 4 shows the results of the aryl and alkyl chloride Grignard reactions. In contrast to the bromides, alkyl chlorides tended to produce heterogeneous dispersions of the Grignard reagents. The supernatant liquids were colorless, which were clearly different from those obtained from alkyl bromides,

Table 3. Grignard reactions in CPME with a variety of aryl and alkyl bromides (1 a–y).[a]

Entry

RBr (1)

State[b]

Conc. [m][c]

Yield [%][d] 2 4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20[e] 21[h] 22[h] 23 24 25

2-MeOC6H4Br (1 a) 3-MeOC6H4Br (1 b) 4-MeOC6H4Br (1 c) 2-MeC6H4Br (1 d) 3-MeC6H4Br (1 e) 4-MeC6H4Br (1 f) C6H5Br (1 g) 1-bromonaphthalene (1 h) 2-ClC6H4Br (1 i) 3-ClC6H4Br (1 j) 4-ClC6H4Br (1 k) 2-FC6H4Br (1 l) 3-FC6H4Br (1 m) 4-FC6H4Br (1 n) 2-bromothiophene (1 o) 3-bromothiophene (1 p) CH3CH2CH2CH2Br (1 q) (CH3)2CHBr (1 r) bromocyclohexane (1 s) (CH3)3CBr (1 t) H2C=CHCH2Br (1 u) C6H5CH2Br (1 v) HCCCH2Br (1 w) H2C=C(CH3)Br (1 x) (E)-C6H5CH=CHBr (1 y)

homo hetero hetero hetero hetero hetero hetero homo hetero homo hetero homo homo homo homo – homo homo homo hetero homo homo – hetero hetero

1.02 0.81 0.88 0.97 0.92 0.82 1.06 1.09 0.64 1.17 1.04 0.45 1.12 1.03 1.06 – 1.08 1.09 1.03 (0.48)[f] 0.69 (0.89)[f] – 0.75 (0.51)[f]

82 65 70 78 74 66 85 87 51 94 83 36 90 82 85 – 86 87 82 (38)[f] 55 (71)[f] – 60 (41)[f]

87 87 86 89 82 88 84 97 ND 94 94 ND 96 87 93 ND 97 95 88 74[g] 92 83[g] ND 89 76[g]

[a] The experimental procedure was same as that described in the footnotes of Table 2. [b] “homo” = homogeneous, “hetero” = heterogeneous. [c] Concentration of 2 determined by titration. [d] The yield of 2 was estimated from the concentration. The isolated yield of 4 is based on 3. “ND” = not detected. [e] The Grignard formation was performed at room temperature. [f] Determined by an independent experiment. [g] The inverse addition method was applied. [h] The Grignard formation was performed under the water-cooled conditions.

Substrate Scope for the Grignard Reactions in CPME With the optimal conditions in hand, we explored the substrate scope to evaluate the performance of CPME in the formation and reaction of Grignard reagents. Table 3 shows comprehensive results of the Grignard reactions of aryl and alkyl bromides (1 a–y) with benzaldehyde (3). In all cases, the concentration of the Grignard reagent was determined by titration, after which a small excess of the Grignard reagent (1.2 equiv) was added to the benzaldehyde solution (1 equiv). If the Grignard reagents were unstable for titration, benzaldehyde (3) was inversely added into the solutions of the Grignard reagent (Table 3, entries 20, 22 and 25). The results clearly show that a number of Grignard reagents could be prepared in CPME in acceptable yields. Notably, some of the Grignard reagents were prepared as homogeneous solutions; however, the relationship between the substrate structures and the solution states was ambiguous. Nevertheless, as judged from the Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

638


Full Paper an appropriate co-solvent should be used if the intended Grignard reaction does not occur in CPME.

Table 4. Grignard reactions in CPME with a variety of aryl and alkyl chlorides (5 a–h)[a]

Stability of the Grignard Reagents in CPME

Entry

RCl (5)

Conc. [m][b]

Yield [%][c] 6 4

1[d] 2[d,g] 3 4 5[h] 6 7 8[i,j] 9[i]

C6H5Cl (5 a) C6H5Cl (5 a) CH3CH2CH2CH2Cl (5 b) Me3SiCH2Cl (5 c) (CH3)2CHCl (5 d) chlorocyclohexane (5 e) (CH3)3CCl (5 f) H2C=CHCH2Cl (5 g) C6H5CH2Cl (5 h)

– – 1.22 0.85 1.08 1.08 0.88 – (0.83)[k]

– – 98 68 86 86 70 – (67)[k]

Having explored the substrate scope of Grignard reactions in CPME, we then studied the stability of selected Grignard reagents, using n-butylmagnesium bromide (2 q) and n-butylmagnesium chloride (6 b) as examples. The data clearly showed that both Grignard reagents were stable in storage in a refrigerator at 0 8C for at least three months (Figure 2). Neither precipitation nor colorization was observed, and the stock reagents could be used for the addition reactions. These positive data suggest that some Grignard reagents prepared as CPME solutions could replace the existing commercial reagents, although a more detailed quality analysis is necessary.

26[e,f] 60[e,f] 85 81 87 88 83 83[e] 67[f]

[a] The experimental procedure was same as that described in the footnotes of Table 2. [b] Concentration of 6 determined by titration. [c] The yield of 6 was estimated from the concentration. The isolated yield of 4 is based on 3. [d] The Grignard formation was performed at 90 8C for 20 h. 1.5 mmol of DIBALH was used. [e] Isolated yield is based on 5. An equimolar amount of 3 relative to 5 was added. [f] The inverse addition method was applied. [g] A mixed solvent consisting of CPME and THF (1:1 v/v) was used. [h] The Grignard formation was performed at room temperature. [i] The Grignard formation was performed under watercooled conditions. [j] A 1:1 mixture of 5g and 3 was added simultaneously to the activated magnesium. [k] Determined by an independent experiment.

Figure 2. Stability assessment of the selected Grignard reagents 2 q and 6 b. Concentration was determined by titration.

which were yellow or dark in color (except for the colorless suspension of 2 t). The primary (5 b and 5 c), secondary (5 d and 5 e), and tertiary alkyl chlorides (5 f) afforded the corresponding Grignard reagents (6 b–f) in good yields (Table 4, entries 3–7). The benzyl Grignard reagent 6 h was prepared under water-cooled conditions to suppress the homo-coupling side reaction (Table 4, entry 9), although allylmagnesium chloride (6 g) could not be prepared even by using this method. Fortunately, the simultaneous addition of allyl chloride (5 g) and benzaldehyde (3) to the activated magnesium (Barbier reaction) gave rise to the expected addition product 4 u in 83 % yield. In contrast with alkyl chlorides, however, formation of arylmagnesium chlorides was challenging. After much experimentation, only a small amount of phenylmagnesium chloride (6 a) was detected after stirring of the reaction mixture for 20 h at 90 8C using a tenfold amount of DIBALH (5 mol % relative to magnesium). Upon treatment with 3, the desired addition product 4 g was obtained in 26 % yield (based on 5 a; Table 4, entry 1). Replacement of the solvating media of DIBALH from n-hexane to THF did not affect the reaction efficiency, although a previous report indicated a co-solvent effect in the Grignard reaction with 3-bromoanisole.[12] If the amount of THF was increased to the same volume as CPME, smoother Grignard formation occurred (Table 4, entry 2). It appears that CPME has a negative effect on Grignard formation from aromatic chlorides, whereas this incompatibility did not emerge from the reactions with aromatic bromides. We therefore propose that Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

Recycling of CPME after the Grignard Reactions One of the advantages of CPME over THF is its recyclability arising from immiscibility with water (solubility in water: 1.1 g per 100 g water at 23 8C). Such hydrophobic character enables recovery of the solvent by typical workup techniques. Indeed, CPME could be recovered without notable deterioration of the purity after iterative treatment of the tin hydride-mediated radical addition reactions.[6] Solvent recycling was also demonstrated by one-pot reactions consisting of radical addition, cross-coupling and organometallic reactions.[6] As recent innovations in green chemistry recommend minimum use of solvents with reduction of waste, recycling of CPME after the Grignard reaction was conducted. In practice, CPME used in the Grignard reaction between 1-bromo-3-chlorobenzene (1 j) and benzaldehyde (3) was recovered by extraction and distillation, and reused for the second experiment. The yield of product 4 j was constant (first reaction: 94 %, second: 94 %), and the purities of the recovered CPME after the first and second experiments were 99.48 % and 98.72 %, respectively.[15] It is notable that minor contaminants in the recovered solvent after the second experiment were only chlorobenzene (1.21 %, derived from protonation of 2 j) and n-hexane (0.03 %, originating from the DIBALH solution), and no degradation products derived from CPME were detected by GC–MS.[16] This result is in contrast with our previous data that indicated several degradation products of CPME in trace amounts after repeated radical 639


Full Paper addition reactions.[6] It is therefore concluded that CPME is stable under the Grignard conditions and can be recycled as part of the usual workup procedure. Syntheses of ( )-Tramadol Hydrochloride and Tamoxifen with the Grignard Reactions in CPME Finally, in an effort to demonstrate value to the pharmaceutical industry, we undertook the syntheses of two representative pharmaceutical agents—tramadol and tamoxifen—by applying Grignard reactions in CPME. Although several efficient approaches of tramadol have been published,[17] a green method amenable to the large-scale synthesis is still anticipated. Thus, a CPME solution of (3-methoxyphenyl)magnesium bromide (2 b) was treated with aminoketone 7[18] at ¢78 8C to room temperature to afford amino alcohol ()-8, which was, without purification, treated with 4 m hydrochloric acid in CPME to provide tramadol hydrochloride [()-9] as a colorless solid after recrystallization from acetone, with an overall yield of 61 % from 7 (Scheme 1). Although an exact diastereomeric ratio of ()-8 was not determined due to the complexity of the NMR spectrum of the crude product, smooth recrystallization of the desired isomer ()-9 indicated that the Grignard addition had occurred in a stereoselective manner.[19] The second example, the synthesis of tamoxifen (13), is summarized in Scheme 2.[20, 21] Tamoxifen is in worldwide demand as a pioneering medicine for the treatment of estrogen-receptor-positive breast cancer.[22] In practice, the freshly prepared phenylmagnesium bromide (2 g) in CPME was reacted with Weinreb amide 10 at 60 8C to give the desired ketone 11 in 92 % yield. The second Grignard reaction of 11 was, in contrast, difficult because the requisite nitrogen-containing Grignard reagent 12 a could not be prepared in only CPME. When a mixed solvent system consisting of a 1:1 mixture of CPME and THF was applied, a moderate amount of the Grignard reagent formed, which gave, after acid-catalyzed dehydration and basic workup, (Z/E)-tamoxifen (13) in 48 % yield. Unfortunately, isolation of 13 from the product mixture was difficult because a considerable amount (49 %) of the unreacted bromide 12 re-

Scheme 2. Synthesis of tamoxifen.

mained and could not be separated by chromatography. Therefore, we performed aryllithium addition via halogen– metal exchange between 12 and nBuLi, using a slight excess of 11. Although the bromine–lithium exchange reaction was incomplete in CPME at either ¢78 or 0 8C,[23] the use of THF as a co-solvent (CPME/THF, 3.3:1 v/v) facilitated the exchange reaction to give a clear pale yellow solution of 12 b, to which 11 was added to provide the desired tertiary alcohol. Without purification, the resulting alcohol was subjected to dehydration with hydrochloric acid in methanol to give tamoxifen (13), after basic workup, as a 2:1 mixture of Z and E isomers in 85 % overall yield based on 12. Importantly, the pharmaceutically active Z isomer could be separated as a colorless solid by recrystallization from n-hexane and EtOAc.[24]

Conclusions We undertook a systematic study of Grignard reactions using CPME as a reaction solvent. This study revealed that a number of Grignard reagents could be accessible as CPME solutions, despite some of them producing heterogeneous mixtures. DIBALH was found to be the most appropriate activator of magnesium, although it remains to be determined whether this activation system is effective only with magnesium in CPME or has generality for activating other metals in other solvents.[25] The CPME used in the Grignard reaction was efficiently recycled without affecting the reaction yields, and GC–MS analysis of the recovered solvent revealed no degradation of CPME under the Grignard conditions. Some of the Grignard reagents in CPME were stable for at least several months and therefore could replace existing commercial reagents.[26] Finally, we applied the Grignard reactions in CPME to the syntheses of tramadol hydrochloride and tamoxifen. This study gives general guidance to process chemists for the selection of the reaction solvent for Grignard and related organometallic reactions for large-scale synthesis.

Scheme 1. Synthesis of tramadol hydrochloride. Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

640


Full Paper Experimental Section

two-necked flask. The solution was cooled in an ice bath, followed by addition of the Grignard reagent 2 b (0.81 m, 1.5 mL, 1.2 mmol). The mixture was stirred for 30 min at room temperature and quenched by the addition of saturated aqueous NH4Cl solution at 0 8C. The resulting mixture was extracted with Et2O (3 Õ 6 mL), and the combined organic layer was washed with brine, dried over anhydrous MgSO4, filtered and concentrated. The residue was purified by flash chromatography on silica gel (n-hexane/EtOAc, 15:1! 10:1!8:1 v/v) to give alcohol 4 b as a pale yellow oil (186 mg, 0.870 mmol, 87 % based on 3).

General Techniques All reactions were performed under an atmosphere of argon, unless otherwise stated. 2-MeTHF and CPME were reagent grade and stored over 4 æ molecular sieves. MeOH was reagent grade and stored over 3 æ molecular sieves. Benzaldehyde was distilled under reduced pressure. Other commercially available reagents were used without further purification, unless otherwise stated. Reactions were monitored by thin-layer chromatography performed on 0.25 mm silica gel plates (60-F254) that were visualized by fluorescence upon 254 nm irradiation or by staining with p-anisaldeyde/AcOH/H2SO4/EtOH, 12 MoO3·H3PO4/EtOH, or ninhydrin/ AcOH/nBuOH. The products were purified by flash chromatography on silica gel (spherical, neutral, 40–50 mm) and, if necessary, were additionally purified by normal-phase HPLC equipped with a prepacked column (Mightysil, Si 60 250-20 (5 mm)) using n-hexane/ EtOAc as the eluent. NMR spectra were recorded on either a 300 MHz (1H: 300 MHz, 13C: 75 MHz) or a 400 MHz (1H: 400 MHz, 13 C: 100 MHz) spectrometer and referenced to the solvent peak at 7.26 ppm (1H) and 77.16 ppm (13C) for CDCl3. Splitting patterns are indicated as follows: br, broad; s, singlet; d, doublet; t, triplet; q, quartet; quint, quintet; m, multiplet. Infrared spectra were recorded on an FTIR spectrometer and are reported as wavenumber [cm¢1]. High-resolution ESI and atmospheric-pressure chemical ionization (APCI) mass spectra were recorded with an Orbitrap analyzer in the positive- or negative-ion mode.

(2-Methoxyphenyl)(phenyl)methanol (4 a).[28] This compound was obtained from 1-bromo-2-methoxybenzene (1 a) in 87 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.42–7.22 (m, 7 H), 6.96 (dt, J = 7.5, 1.2 Hz, 1 H), 6.90 (br d, J = 8.1 Hz, 1 H), 6.07 (s, 1 H), 3.82 (s, 3 H), 3.07 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 156.9, 143.4, 132.1, 128.9, 128.3, 128.0, 127.3, 126.7, 120.9, 110.9, 72.4, 55.5 ppm. (3-Methoxyphenyl)(phenyl)methanol (4 b).[29] This compound was obtained from 1-bromo-3-methoxybenzene (1 b) in 87 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.41–7.23 (m, 6 H), 6.97– 6.94 (m, 2 H), 6.83–6.79 (m, 1 H), 5.82 (s, 1 H), 3.79 (s, 3 H), 1.92 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 159.9, 145.6, 143.8, 129.7, 128.6, 127.7, 126.7, 119.0, 113.1, 112.2, 76.3, 55.3 ppm. (4-Methoxyphenyl)(phenyl)methanol (4 c).[29, 30] This compound was obtained from 1-bromo-4-methoxybenzene (1 c) in 86 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.40–7.26 (m, 7 H), 6.87 (d, J = 8.7 Hz, 2 H), 5.81 (s, 1 H), 3.79 (s, 3 H), 2.19 ppm (br s, 1 H; OH); 13 C NMR (75 MHz, CDCl3): d = 159.2, 144.1, 136.3, 128.6, 128.0, 127.6, 126.5, 114.0, 76.0, 55.4 ppm.

Typical Procedure for the Grignard Reaction in CPME with Bromides (Table 3, entry 2)

Phenyl(o-tolyl)methanol (4 d).[28] This compound was obtained from 1-bromo-2-methylbenzene (1 d) in 89 % yield. Pale yellow solid; 1H NMR (300 MHz, CDCl3): d = 7.53 (dd, J = 6.9, 1.5 Hz, 1 H), 7.35–7.13 (m, 8 H), 6.01 (s, 1 H), 2.26 (s, 3 H), 2.04 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 143.0, 141.6, 135.5, 130.7, 128.6, 127.72, 127.67, 127.2, 126.4, 126.3, 73.5, 19.5 ppm.

Formation of the Grignard reagent: Well-ground magnesium turnings (729 mg, 30.0 mmol) were placed in a well-dried 50 mL threenecked round-bottom flask equipped with a reflux condenser and a dropping funnel. The flask was further dried under reduced pressure with a heat gun and then flushed with argon, after which CPME (3 mL) and DIBALH (1.0 m solution in n-hexane, 150 mL, 150 mmol) were added. The suspension was stirred for 30 min at room temperature and then warmed to 60 8C. A solution of 3-bromoanisole (1 b; 3.74 g, 20.0 mmol) in CPME (11 mL + 2 mL for rinsing) was added through the dropping funnel at a rate of one drop every 5 s (it took about 40–50 min to complete the addition). The reaction mixture was stirred at 60 8C for 3 h in total and left at room temperature until two phases were clearly separated. The supernatant liquid was titrated to determine the concentration of 2 b. Titration of the Grignard reagent:[27] 1,10-Phenanthroline (3.6 mg, 0.020 mmol) was placed in a well-dried 30 mL two-necked flask. The flask was flushed with argon, after which anhydrous MeOH (40 mL, 0.99 mmol) and anhydrous THF (5 mL) were added. The supernatant liquid of the Grignard reagent was then slowly added with a syringe until the color of the solution tuned to red-purple. The end point was determined by coloration persisting for more than 1 min. The concentration of the Grignard reagent was determined using Equation (1). For 2 b, the concentration was determined to be 0.81 m (65 % yield based on 1 b).

½RMgX¤ ½m¤ ¼ amount of MeOH ½mmol¤=volume of Grignard reagent ½mL¤

Phenyl(m-tolyl)methanol (4 e).[31] This compound was obtained from 1-bromo-3-methylbenzene (1 e) in 82 % yield. Yellow oil; 1 H NMR (300 MHz, CDCl3): d = 7.41–7.08 (m, 9 H), 5.81 (s, 1 H), 2.35 (s, 3 H), 2.22 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 144.0, 143.9, 138.3, 128.6, 128.54, 128.50, 127.6, 127.3, 126.6, 123.8, 76.4, 21.6 ppm. Phenyl(p-tolyl)methanol (4 f).[30] This compound was obtained from 1-bromo-4-methylbenzene (1 f) in 88 % yield. Colorless solid; 1 H NMR (300 MHz, CDCl3): d = 7.41–7.25 (m, 7 H), 7.16 (br d, J = 8.1 Hz, 2 H), 5.82 (s, 1 H), 2.35 (s, 3 H), 2.28 ppm (br s, 1 H; OH); 13 C NMR (75 MHz, CDCl3): d = 144.1, 141.1, 137.4, 129.3, 128.6, 127.6, 126.7, 126.6, 76.2, 21.2 ppm. Diphenylmethanol (4 g).[30] This compound was obtained from bromobenzene (1 g) in 84 % yield. Colorless solid; 1H NMR (300 MHz, CDCl3): d = 7.41–7.24 (m, 10 H), 5.85 (s, 1 H), 1.89 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 143.9, 128.7, 127.7, 126.7, 76.4 ppm. Naphthalen-1-yl(phenyl)methanol (4 h).[30] This compound was obtained from 1-bromonaphthalene (1 h) in 97 % yield. Colorless solid; 1H NMR (300 MHz, CDCl3): d = 7.89 (br d, J = 8.1 Hz, 1 H), 7.77 (br d, J = 7.2 Hz, 1 H), 7.71 (br d, J = 8.1 Hz, 1 H), 7.49 (br d, J = 7.2 Hz, 1 H), 7.38–7.12 (m, 8 H), 6.34 (s, 1 H), 2.52 ppm (br s, 1 H; OH); 13 C NMR (75 MHz, CDCl3): d = 143.2, 138.9, 134.0, 130.8, 128.8, 128.6, 128.5, 127.7, 127.1, 126.2, 125.7, 125.4, 124.7, 124.1, 73.6 ppm.

ð1Þ

(3-Chlorophenyl)(phenyl)methanol (4 j).[32] This compound was obtained from 1-bromo-3-chlorobenzene (1 j) in 94 % yield. Color-

Addition of the Grignard reagent: Benzaldehyde (3; 106 mg, 0.999 mmol) and CPME (2 mL) were placed in a well-dried 20 mL Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

641


Full Paper less solid; 1H NMR (300 MHz, CDCl3): d = 7.41–7.24 (m, 9 H), 5.79 (s, 1 H), 2.25 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 145.8, 143.3, 134.5, 129.9, 128.8, 128.1, 127.8, 126.7, 124.8, 75.8 ppm.

1 H), 3.07 (dd, J = 14, 5.4 Hz, 1 H), 3.00 (dd, J = 14, 8.1 Hz, 1 H), 1.97 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 143.9, 138.2, 129.6, 128.6, 128.5, 127.7, 126.7, 126.0, 75.4, 46.2 ppm.

(4-Chlorophenyl)(phenyl)methanol (4 k).[30] This compound was obtained from 1-bromo-4-chlorobenzene (1 k) in 94 % yield. Colorless solid; 1H NMR (300 MHz, CDCl3): d = 7.36–7.27 (m, 9 H), 5.77 (s, 1 H), 2.49 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 143.5, 142.3, 133.4, 128.8, 128.7, 127.98, 127.96, 126.6, 75.7 ppm.

2-Methyl-1-phenylprop-2-en-1-ol (4 x).[38] This compound was obtained from 2-bromoprop-1-ene (1 x) in 89 % yield. Yellow oil; 1 H NMR (300 MHz, CDCl3): d = 7.40–7.26 (m, 5 H), 5.21 (m, 1 H), 5.14 (br s, 1 H), 4.96 (m, 1 H), 2.00 (br s, 1 H; OH), 1.62 ppm (s, 3 H); 13 C NMR (75 MHz, CDCl3): d = 146.9, 142.1, 128.5, 127.8, 126.6, 111.3, 78.0, 18.4 ppm.

(3-Fluorophenyl)(phenyl)methanol (4 m).[30] This compound was obtained from 1-bromo-3-fluorobenzene (1 m) in 96 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.37–7.27 (m, 6 H), 7.16–7.09 (m, 2 H), 6.99–6.92 (m, 1 H), 5.80 (s, 1 H), 2.34 ppm (br s, 1 H; OH); 13 C NMR (75 MHz, CDCl3): d = 163.1 (d, JCF = 244.9 Hz), 146.4 (d, JCF = 6.2 Hz), 143.4, 130.1 (d, JCF = 7.6 Hz), 128.8, 128.0, 126.7, 122.2 (d, JCF = 2.8 Hz), 114.5 (d, JCF = 21.3 Hz), 113.5 (d, JCF = 22.0 Hz), 75.8 ppm (d, JCF = 2.1 Hz).

(E)-1,3-Diphenylprop-2-en-1-ol [(E)-4 y][39] and (Z)-1,3-Diphenylprop-2-en-1-ol [(Z)-4 y].[40] These compounds were obtained from (E)-(2-bromovinyl)benzene (1 y) in 76 % yield with an E/Z ratio of 56:44. The two isomers were separated by HPLC for spectral analysis. The Z isomer was unstable in both CDCl3 and C6D6, and gradually decomposed during the 13C NMR measurement. (E)-4 y: yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.51–7.22 (m, 10 H), 6.70 (br d, J = 16 Hz, 1 H), 6.40 (dd, J = 16, 6.6 Hz, 1 H), 5.40 (br d, J = 6.6 Hz, 1 H), 2.14 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 142.9, 136.6, 131.6, 130.7, 128.8, 128.7, 127.94, 127.92, 126.7, 126.5, 75.3 ppm. (Z)-4 y: yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.51–7.28 (m, 10 H), 6.71 (br d, J = 12 Hz, 1 H), 5.95 (dd, J = 12, 9.3 Hz, 1 H), 5.65 (br d, J = 9.3 Hz, 1 H), 2.05 ppm (br s, 1 H; OH).

(4-Fluorophenyl)(phenyl)methanol (4 n).[30] This compound was obtained from 1-bromo-4-fluorobenzene (1 n) in 87 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.36–7.27 (m, 7 H), 7.02 (t, J = 8.4 Hz, 2 H), 5.80 (s, 1 H), 2.38 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 162.3 (d, JCF = 244.2 Hz), 143.7, 139.7 (d, JCF = 3.5 Hz), 128.7, 128.3 (d, JCF = 8.3 Hz), 127.9, 126.6, 115.4 (d, JCF = 21.3 Hz), 75.7 ppm. Phenyl(thiophen-2-yl)methanol (4 o).[30] This compound was obtained from 2-bromothiophene (1 o) in 93 % yield. Colorless solid; 1 H NMR (300 MHz, CDCl3): d = 7.47–7.29 (m, 5 H), 7.27 (dd, J = 5.1, 1.2 Hz, 1 H), 6.95 (dd, J = 5.1, 3.6 Hz, 1 H), 6.89 (ddd, J = 3.6, 1.2, 0.9 Hz, 1 H), 6.05 (s, 1 H), 2.54 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 148.2, 143.2, 128.6, 128.1, 126.8, 126.4, 125.5, 125.0, 72.5 ppm.

Typical Procedure for the Grignard Reaction in CPME with Chlorides (Table 4, entry 3) Formation of the Grignard reagent: Well-ground magnesium turnings (729 mg, 30.0 mmol) were placed in a well-dried 50 mL threenecked round-bottom flask equipped with a reflux condenser and a dropping funnel. The flask was flushed with argon, after which CPME (3 mL) and DIBALH (1.0 m solution in n-hexane, 150 mL, 150 mmol) were added. The suspension was stirred for 30 min at room temperature and then warmed to 60 8C. A solution of 1chlorobutane (5 b; 1.85 g, 20.0 mmol) in CPME (11 mL + 2 mL for rinsing) was added through the dropping funnel at a rate of one drop every 5 s (it took about 40–50 min to complete the addition). The reaction mixture was stirred at 60 8C for 3 h in total and then left at room temperature until two phases were clearly separated. The supernatant liquid was titrated to determine the concentration of 6 b (1.22 m, 98 % yield).

1-Phenylpentan-1-ol (4 q).[33] This compound was obtained from 1-bromobutane (1 q) in 97 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.36–7.25 (m, 5 H), 4.66 (dd, J = 7.5, 5.7 Hz, 1 H), 1.85 (br s, 1 H; OH), 1.85–1.65 (m, 2 H), 1.45–1.20 (m, 4 H), 0.89 ppm (t, J = 7.1 Hz, 3 H); 13C NMR (75 MHz, CDCl3): d = 145.1, 128.6, 127.6, 126.0, 74.8, 38.9, 28.1, 22.7, 14.1 ppm. 2-Methyl-1-phenylpropan-1-ol (4 r).[34] This compound was obtained from 2-bromopropane (1 r) in 95 % yield. Colorless oil; 1 H NMR (300 MHz, CDCl3): d = 7.38–7.24 (m, 5 H), 4.36 (d, J = 6.8 Hz, 1 H), 1.96 (octet, J = 6.8 Hz, 1 H), 1.80 (br s, 1 H; OH), 1.01 (d, J = 6.8 Hz, 1 H), 0.80 ppm (d, J = 6.8 Hz, 1 H); 13C NMR (75 MHz, CDCl3): d = 143.8, 128.3, 127.6, 126.7, 80.2, 35.4, 19.1, 18.4 ppm. Cyclohexyl(phenyl)methanol (4 s).[28, 30] This compound was obtained from bromocyclohexane (1 s) in 88 % yield. Pale yellow oil; 1 H NMR (300 MHz, CDCl3): d = 7.37–7.24 (m, 5 H), 4.35 (d, J = 7.2 Hz, 1 H), 2.04–1.95 (m, 1 H), 1.82–1.55 (m, 4 H), 1.42–0.86 ppm (m, 6 H); 13 C NMR (75 MHz, CDCl3): d = 143.7, 128.3, 127.5, 126.7, 79.4, 45.0, 29.4, 28.9, 26.5, 26.2, 26.1 ppm.

Addition of the Grignard reagent: Benzaldehyde (3; 106 mg, 0.999 mmol) and CPME (2 mL) were placed in a well-dried 20 mL two-necked flask. The solution was cooled in an ice bath, followed by the addition of the Grignard reagent (6 b; 1.22 m, 1.0 mL, 1.2 mmol). The mixture was stirred for 1 h at room temperature and quenched by the addition of saturated aqueous NH4Cl solution at 0 8C. The resulting mixture was extracted with Et2O (3 Õ 6 mL), and the combined organic layers were washed with brine, dried over anhydrous MgSO4, filtered and concentrated. The residue was purified by flash chromatography on silica gel (n-hexane/EtOAc, 10:1!7:1 v/v) to give alcohol 4 q as a pale yellow oil (140 mg, 0.852 mmol, 85 % based on 3).

2,2-Dimethyl-1-phenylpropan-1-ol (4 t).[35] This compound was obtained from 2-bromo-2-methylpropane (1 t) in 74 % yield. Pale yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.37–7.23 (m, 5 H), 4.39 (s, 1 H), 1.89 (br s, 1 H; OH), 0.93 ppm (s, 9 H); 13C NMR (75 MHz, CDCl3): d = 142.3, 127.74, 127.68, 127.4, 82.5, 35.7, 26.1 ppm. 1-Phenylbut-3-en-1-ol (4 u).[36] This compound was obtained from 3-bromoprop-1-ene (1 u) in 92 % yield. Yellow oil; 1H NMR (300 MHz, CDCl3): d = 7.44–7.25 (m, 5 H), 5.89–5.75 (m, 1 H), 5.21– 5.13 (m, 2 H), 4.74 (dd, J = 7.5, 5.4 Hz, 1 H), 2.60–2.44 (m, 2 H), 2.01 ppm (br s, 1 H; OH); 13C NMR (75 MHz, CDCl3): d = 144.0, 134.6, 128.5, 127.7, 125.9, 118.6, 73.4, 44.0 ppm.

1-Phenyl-2-(trimethylsilyl)ethan-1-ol (4 z).[41] This compound was obtained from (2-chloroethyl)trimethylsilane (5 c) in 81 % yield. Colorless oil; 1H NMR (300 MHz, CDCl3): d = 7.38–7.24 (m, 5 H), 4.85 (t, J = 7.5 Hz, 1 H), 1.75 (br s, 1 H; OH), 1.28 (dd, J = 14, 7.4 Hz, 1 H), 1.18 (dd, J = 14, 7.7 Hz, 1 H), ¢0.08 ppm (s, 9 H); 13C NMR (75 MHz, CDCl3): d = 146.5, 128.6, 127.7, 126.0, 73.0, 28.5, ¢1.00 ppm; HRMS (ESI +): m/z: calcd for C11H18ONaSi: 217.1019 [M+ +Na] + ; found: 217.1020.

1,2-Diphenylethan-1-ol (4 v).[37] This compound was obtained from benzyl bromide (1 v) in 83 % yield. Colorless solid; 1H NMR (300 MHz, CDCl3): d = 7.38–7.20 (m, 10 H), 4.90 (dd, J = 8.1, 5.4 Hz, Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

642


Full Paper Recycling of CPME in the Grignard Reaction between 1 j and 3

stirred for 1.5 h. The solution was quenched by the addition of saturated aqueous NH4Cl solution and then basified by the addition of saturated aqueous NaHCO3 solution. The resulting mixture was extracted with CPME (3 Õ 10 mL), and the combined organic layers were back-extracted with 1 m aqueous HCl. The aqueous layer was basified with 1 m aqueous NH3 and extracted with CPME (3 Õ 10 mL). The combined organic layer was washed with brine, dried over anhydrous MgSO4, filtered and concentrated to give tramadol [()-8] as a pale yellow oil (2.23 g). The crude ()-8 was dissolved in CPME (4 mL) and treated with 4 m HCl in CPME (5.0 mL, 20 mmol) for 5 min at room temperature. The solution was concentrated, and the residue was recrystallized from acetone at ¢20 8C in a freezer to give tramadol hydrochloride [()-9] as a colorless solid (1.71 g, 5.70 mmol, 61 % based on 7); m.p. 176–179 8C; 1 H NMR (300 MHz, CDCl3): d = 11.4 (br s, 1 H; OH), 7.28 (t, J = 8.1 Hz, 1 H), 7.07 (br s, 1 H), 6.99 (br d, J = 7.8 Hz, 1 H), 6.79 (br dd, J = 8.4, 2.7 Hz, 1 H), 3.82 (s, 3 H), 2.98 (ddd, J = 13, 9.3, 1.8 Hz, 1 H), 2.65 (d, J = 4.8 Hz, 3 H), 2.58 (ddd, J = 13, 8.7, 2.1 Hz, 1 H), 2.51–2.41 (m, 1 H), 2.44 (d, J = 4.8 Hz, 3 H), 2.38–2.26 (m, 1 H), 2.15–2.05 (m, 1 H), 1.95– 1.86 (m, 2 H), 1.78–1.61 (m, 3 H), 1.51–1.38 ppm (m, 1 H); 13C NMR (75 MHz, CDCl3): d = 160.0, 148.4, 129.9, 117.0, 112.1, 111.4, 75.1, 60.7, 55.5, 46.2, 42.2, 42.0, 41.0, 27.3, 25.1, 21.5 ppm; elemental analysis calcd (%) for C16H26ClNO2 : C 64.09, H 8.74, N 4.67; found: C 63.75, H 8.78, N 4.63.

Well-ground magnesium turnings (729 mg, 30.0 mmol) were placed in a well-dried 50 mL three-necked round-bottom flask equipped with a reflux condenser and a dropping funnel. The flask was further dried under reduced pressure with a heat gun and then flushed with argon, after which CPME (3 mL) and DIBALH (1.0 m solution in n-hexane, 150 mL, 150 mmol) were added. The suspension was stirred for 30 min at room temperature and then warmed to 60 8C. A solution of 1-bromo-3-chlorobenzene (1 j; 3.83 g, 20.0 mmol) in CPME (11 mL + 2 mL for rinsing) was added through the dropping funnel at a rate of one drop every 5 s. The reaction mixture was stirred at 60 8C for 3 h in total and then cooled in an ice bath. To this solution was added benzaldehyde (3; 1.06 g, 9.99 mmol) in CPME (10 mL) through a cannula and the mixture was stirred for 1 h at room temperature. The reaction mixture was quenched by the sequential addition of saturated aqueous NH4Cl solution and saturated Rochelle salt solution at 0 8C. The resulting mixture was extracted with CPME (3 Õ 30 mL), and the combined organic layers were washed with brine, dried over anhydrous MgSO4 and filtered (rinsed with CPME). CPME was recovered by distillation under reduced pressure (b.p. 40–50 8C, 13–17 kPa), transferred to a brown-colored bottle and kept at room temperature after addition of molecular sieves (4 æ). The residue was purified by flash chromatography on silica gel (n-hexane/EtOAc, 5:1 v/ v) to give alcohol 4 j (2.05 g, 9.37 mmol, 94 % based on 3) as a pale yellow oil. The purity of the recovered CPME was determined to be 99.48 % by GC analysis. The experiment was repeated using the recovered CPME, affording 4 j (2.05 g, 9.37 mmol, 94 %) with recovery of CPME (98.72 % purity).

Synthesis of Tamoxifen N-Methoxy-N-methyl-2-phenylbutanamide (10).[21a] (COCl)2 (6.25 mL, 73.1 mmol) and DMF (93.5 mL, 1.22 mmol) were added to a solution of 2-phenylacetic acid (10.0 g, 60.9 mmol) in CH2Cl2 (122 mL) at 0 8C. The mixture was warmed to room temperature and stirred for 2 h. Then the mixture was cooled to 0 8C, followed by the addition of N,O-dimethylhydroxylamine hydrochloride (5.95 g, 61.0 mmol) and Et3N (42.2 mL, 305 mmol). After 1.5 h at 0 8C, the reaction mixture was quenched by the addition of 4 m aqueous HCl (60.0 mL) and extracted with CH2Cl2 (5 Õ 40 mL). The combined organic layers were washed with saturated aqueous NaHCO3 solution, dried over anhydrous MgSO4, filtered and concentrated. The residue was purified by open chromatography on silica gel (n-hexane/EtOAc, 20:1!5:1 v/v) to give amide 10 as a reddish oil (10.1 g, 48.7 mmol, 80 %). 1H NMR (400 MHz, CDCl3): d = 7.34–7.20 (m, 5 H), 3.89 (m, 1 H), 3.47 (s, 3 H), 3.16 (s, 3 H), 2.09 (m, 1 H), 1.75 (m, 1 H), 0.88 ppm (t, J = 7.2 Hz, 3 H); 13C NMR (100 MHz, CDCl3): d = 174.9, 140.3, 128.6, 128.3, 126.9, 61.4, 49.5, 32.4, 27.3, 12.5 ppm. 1,2-Diphenylbutan-1-one (11).[42] Well-ground magnesium turnings (569 mg, 23.4 mmol) were placed in a well-dried 300 mL three-necked round-bottom flask equipped with a reflux condenser and a dropping funnel. The flask was further dried under reduced pressure with a heat gun and then flushed with argon, after which CPME (5.5 mL) and DIBALH (1.0 m solution in n-hexane, 117 mL, 117 mmol) were added. The suspension was stirred for 30 min at room temperature and then warmed to 60 8C. A solution of bromobenzene (1 g; 1.63 mL, 15.6 mmol) in CPME (5 mL) was added from the dropping funnel at a rate of one drop every 5 s. The reaction mixture was stirred at 60 8C for 3 h in total, after which the Weinreb amide 10 (2.08 g, 10.0 mmol) in CPME (20 mL) was added from the dropping funnel at a rate of one drop per second. The reaction mixture was stirred at 60 8C for 1 h in total and then cooled to 0 8C. 1 m aqueous HCl (40 mL) was slowly added from the dropping funnel and the resulting mixture was stirred for 12.5 h before extraction with CPME (2 Õ 10 mL). The combined organic layers were washed with brine, dried over anhydrous MgSO4, filtered and concentrated. The residue was purified by flash chromatography on

GC–MS Analysis of the Recovered CPME A model HP7890 gas chromatograph (Agilent Technologies), equipped with an InertCap-1 capillary column with dimensions 60 m Õ 0.25 mm Õ 1.50 mm, was used. The conditions were a temperature program from 40 to 290 8C, at increments of 10 8C min¢1. The injection volume of the sample was 0.2 mL with a split ratio of 100:1, using nitrogen as the carrier gas at a flow rate of 1.2 mL min¢1. Both injector and detector temperature were maintained at 290 8C. The percentage composition was calculated using a peak normalization method assuming equal detector response. The samples were then analyzed using a quadrupole mass spectrometer (model HP5975 C) with electron impact ionization at 70 eV. The separated compounds were characterized from their mass spectral data using the Wiley2009 and NIST2011 libraries.

Synthesis of Tramadol Hydrochloride [( )-9] Well-ground magnesium turnings (729 mg, 30.0 mmol) were placed in a well-dried 50 mL three-necked round-bottom flask equipped with a reflux condenser and a dropping funnel. The flask was further dried under reduced pressure with a heat gun and then flushed with argon, after which CPME (3 mL) and DIBALH (1.0 m solution in n-hexane, 150 mL, 150 mmol) were added. The suspension was stirred for 30 min at room temperature and then warmed to 60 8C. A solution of 1-bromo-3-methoxybenzene (1 b; 3.74 g, 20.0 mmol) in CPME (11 mL + 2 mL for rinsing) was added through the dropping funnel at a rate of one drop every 5 s. The reaction mixture was stirred at 60 8C for 3 h in total and then cooled to ¢78 8C. To this mixture was added a solution of aminoketone 7 in CPME (2 mL + 3 mL for rinsing) through a cannula, and the reaction mixture was warmed to room temperature and Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

643


Full Paper silica gel (n-hexane/EtOAc, 50:1 v/v) to give ketone 11 (2.06 g, 9.18 mmol, 92 % based on 10). Pale yellow oil; 1H NMR (400 MHz, CDCl3): d = 7.94–7.91 (m, 2 H), 7.44 (tt, J = 7.6, 1.2 Hz, 1 H), 7.37–7.32 (m, 2 H), 7.28–7.22 (m, 4 H), 7.19–7.13 (m, 1 H), 4.40 (t, J = 7.6 Hz, 2 H), 2.16 (dquint, J = 14, 7.6 Hz, 1 H), 1.82 (dquint, J = 14, 7.6 Hz, 1 H), 0.87 ppm (t, J = 7.6 Hz, 3 H); 13C NMR (100 MHz, CDCl3): d = 200.3, 139.8, 137.2, 132.9, 129.0, 128.8, 128.6, 128.4, 127.1, 55.6, 27.3, 12.5 ppm.

brine, then dried over anhydrous MgSO4, filtered and concentrated to give a tertiary alcohol (732 mg) as a colorless solid. The crude alcohol was dissolved in MeOH (12 mL) and treated with concentrated HCl (1.23 mL, 14.7 mmol). The mixture was stirred for 16 h at 60 8C. The solution was concentrated, and the residual solid was thoroughly washed with n-hexane/EtOAc (5:1 v/v). The solid was again extracted with CPME (3 Õ 10 mL) after basification with aqueous saturated NaHCO3 solution. The combined organic layers were washed with brine, dried over anhydrous MgSO4, filtered and concentrated to give 13 (Z/E = 2:1, 534 mg, 1.44 mmol, 85 % based on 12) as a brownish solid. The analytically pure Z isomer was obtained as follows: the Z/E mixture of 13 was dissolved in hot nhexane and insoluble solids were removed by filtration. The filtrate was concentrated and recrystallized from n-hexane and a few drops of EtOAc to give (Z)-13 as a colorless solid. M.p. 96–98 8C; 1 H NMR (300 MHz, CDCl3): d = 7.37–7.09 (m, 10 H), 6.77 (d, J = 8.8 Hz, 2 H), 6.55 (d, J = 8.8 Hz, 1.5 H), 4.07 (m, 2 H), 2.89 (m, 2 H), 2.48 (br s, 6 H), 2.45 (q, J = 7.4 Hz, 2 H), 0.92 ppm (t, J = 7.4 Hz, 3 H); 13 C NMR (100 MHz, CDCl3): d = 156.6, 143.9, 142.5, 141.5, 138.3, 135.9, 132.0, 129.8, 129.6, 128.2, 128.0, 126.7, 126.2, 113.5, 65.3, 58.2, 45.7, 29.1, 13.8 ppm; IR (film on ZnSe): n˜ = 3077, 3053, 3020, 2971, 2870, 2819, 2771, 1606, 1573, 1508, 1461, 1442, 1286, 1243 cm¢1; HRMS (APCI +): m/z: calcd for C26H30ON: 372.2322 [M+ +H] + ; found: 372.2311; elemental analysis calcd (%) for C26H29NO: C 84.06, H 7.87, N 3.77; found: C 83.84, H 7.90, N 3.51.

2-(4-Bromophenoxy)-N,N-dimethylethan-1-amine (12).[21a] 4-Bromophenol (6.40 g, 37.0 mmol) and 2-chloro-N,N-dimethylethylamine hydrochloride (5.82 g, 40.4 mmol), CPME (100 mL) and EtOH (100 mL) were placed in a well-dried 500 mL two-necked roundbottom flask. K2CO3 (20.5 g, 148 mmol) was added and the resulting mixture was stirred for 15 h at 80 8C. The reaction mixture was filtered through a Buchner funnel and the filtrate was concentrated. The residue was purified by open chromatography on silica gel (EtOAc/MeOH, 5:1 v/v) to give a title compound as a reddish oil (7.98 g, 32.7 mmol, 88 %). 1H NMR (400 MHz, CDCl3): d = 7.36 (d, 2 H, J = 8.8 Hz), 6.80 (d, 2 H, J = 8.8 Hz), 4.05 (t, 2 H, J = 6.0 Hz), 2.76 (t, 2 H, J = 6.0 Hz), 2.36 ppm (s, 6 H); 13C NMR (100 MHz, CDCl3): d = 157.8, 132.4, 116.5, 113.3, 65.8, 58.1, 45.7 ppm. (Z)-2-[4-(1,2-Diphenylbut-1-en-1-yl)phenoxy]-N,N-dimethylethan1-amine [tamoxifen, (Z)-13].[21] Synthesis by Grignard reaction: Well-ground magnesium turnings (109 mg, 4.48 mmol) were placed in a well-dried 50 mL two-necked round-bottom flask equipped with a reflux condenser and a dropping funnel. The flask was further dried with a heat gun under reduced pressure and the flask was flushed with argon, after which CPME/THF (1:1 v/v, 3 mL) and DIBALH (1.0 m solution in n-hexane, 23 mL, 23 mmol) were added. The suspension was stirred for 30 min at room temperature and then warmed to 60 8C. A solution of aryl bromide 12 (739 mg, 3.03 mmol) in CPME/THF (1:1 v/v, 2 mL + 2 mL for rinsing) was added through the dropping funnel over 45 min. The reaction mixture was stirred at 60 8C for 3 h in total and then cooled to 21 8C (water bath). A solution of ketone 11 (449 mg, 2.00 mmol) in CPME/THF (1:1 v/v, 2 mL + 2 mL for rinsing) was added to the mixture through the dropping funnel over 5 min, and the reaction mixture was stirred for 45 min. The solution was quenched by the addition of saturated aqueous NH4Cl solution (13 mL). The resulting mixture was extracted with CPME (3 Õ 10 mL), and the combined organic layers were washed with brine, dried over anhydrous MgSO4, filtered and concentrated to give a tertiary alcohol (947 mg) as a yellow solid. The crude alcohol was dissolved in MeOH (16 mL) and treated with concentrated HCl (1.62 mL, 19.4 mmol). The mixture was stirred at 60 8C for 16 h. The solution was concentrated, and the residual solid was thoroughly washed with n-hexane/EtOAc (5:1 v/v). The solid was again extracted with CPME (3 Õ 10 mL) after basification with aqueous saturated NaHCO3 solution. The combined organic layers were washed with brine, dried over anhydrous MgSO4, filtered and concentrated to give a 1:1.5 mixture of 13 (Z/E = 2:1, 360 mg, 0.969 mmol, 48 % based on 11) and 12 (359 mg, 1.47 mmol, 49 %) as a brownish oil.

Acknowledgements This work was supported in part by a Grant-in-Aid for Scientific Research (C) (No. 24550124) from the JSPS and the Science Research Promotion Fund from the Promotion and Mutual Aid Corporation for Private Schools of Japan. We are grateful to Mr. Shugo Matsuno and Mr. Takashi Sasanuma for the GC analysis. We also thank Ms. Ayako Sato, A Rabbit Science Japan Co., Ltd., for elemental analysis measurements, and Ms. Sayaka Kado, Center for Analytical Instrumentation, Chiba University, Japan, for measurement of mass spectra. Keywords: cyclopentyl methyl ether · diisobutylaluminum hydride · Grignard reaction · tamoxifen · tramadol [1] For excellent reviews on the application of 2-MeTHF as a reaction solvent for organic syntheses, see: a) D. F. Aycock, Org. Process Res. Dev. 2007, 11, 156 – 159; b) V. Pace, P. Hoyos, L. Castoldi, P. Domnguez de Mara, A. R. Alcntara, ChemSusChem 2012, 5, 1369 – 1379; c) V. Pace, Aust. J. Chem. 2012, 65, 301 – 302, and references therein. [2] a) P. Anastas, N. Eghbali, Chem. Soc. Rev. 2010, 39, 301 – 312; b) J. M. DeSimone, Science 2002, 297, 799 – 803; c) R. A. Sheldon, Green Chem. 2005, 7, 267 – 278; d) C. Capello, U. Fischer, K. Hungerbuhler, Green Chem. 2007, 9, 927 – 934; e) R. Gani, P. A. Gûmez, M. Folic´, C. Jim¦nezGonzlez, D. J. C. Constable, Comput. Chem. Eng. 2008, 32, 2420 – 2444; f) P. G. Jessop, Green Chem. 2011, 13, 1391 – 1398; g) P. J. Dunn, Chem. Soc. Rev. 2012, 41, 1452 – 1461. [3] K. Watanabe, N. Yamagiwa, Y. Torisawa, Org. Process Res. Dev. 2007, 11, 251 – 258. [4] For recent examples, see: a) K. L. Kimmel, J. D. Weaver, J. A. Ellman, Chem. Sci. 2012, 3, 121 – 125; b) T. Otani, M. Hachiya, D. Hashizume, T. Matsuo, K. Tamao, Chem. Asian J. 2011, 6, 350 – 354; c) Y.-Y. Huang, A. Chakrabarti, N. Morita, U. Schneider, S. Kobayashi, Angew. Chem. Int. Ed. 2011, 50, 11121 – 11124; Angew. Chem. 2011, 123, 11317 – 11320; d) T. Hashimoto, H. Kimura, K. Maruoka, Nat. Chem. 2011, 3, 642 – 646; e) K. Hayashida, H. Fujii, S. Hirayama, T. Nemoto, H. Nagase, Tetrahedron 2011, 67, 6682 – 6688; f) K. Asano, S. Matsubara, J. Am. Chem. Soc. 2011,

Synthesis by aryllithium addition: Aryl bromide 12 (415 mg, 1.70 mmol) was place in a well-dried 50 mL two-necked round bottom flask. After the addition of CPME/THF (3.3:1 v/v, 5 mL), the solution was cooled to ¢78 8C. nBuLi (1.6 m solution in n-hexane, 1.15 mL, 1.84 mmol) was added and the mixture was stirred for 40 min at ¢78 8C. A solution of ketone 11 (449 mg, 2.00 mmol) in CPME/THF (3.3:1 v/v, 3 mL and 3 mL for rinsing) was added and the solution was gradually warmed to 0 8C. The reaction mixture was stirred for 5 h in total and then quenched by the addition of water. The resulting mixture was extracted with EtOAc (3 Õ 20 mL), and the combined organic layers were washed with water and Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

644


Full Paper

[5]

[6] [7]

[8]

[9] [10] [11] [12] [13] [14] [15] [16] [17]

[18] [19] [20] [21]

133, 16711 – 16713; g) S. Gabrielli, A. Palmieri, A. Perosa, M. Selva, R. Ballini, Green Chem. 2011, 13, 2026 – 2028; h) K. Beydoun, H. Doucet, ChemSusChem 2011, 4, 526 – 534; i) K. Watanabe, N. Kogoshi, H. Miki, Y. Torisawa, Synth. Commun. 2009, 39, 2008 – 2013. For examples, see: a) A. Younai, B.-S. Zeng, H. Y. Meltzer, K. A. Scheidt, Angew. Chem. Int. Ed. 2015, 54, 6900 – 6904; Angew. Chem. 2015, 127, 7004 – 7008; b) K. L. Kimmel, J. D. Weaver, M. Lee, J. A. Ellman, J. Am. Chem. Soc. 2012, 134, 9058 – 9061; c) H. Cho, Y. Iwama, N. Mitsuhashi, K. Sugimoto, K. Okano, H. Tokuyama, Molecules 2012, 17, 7348 – 7355; d) T. Okabayashi, K. Iida, K. Takai, Y. Nawate, T. Misaki, Y. Tanabe, J. Org. Chem. 2007, 72, 8142 – 8145; e) A. Shimoyama, A. Saeki, N. Tanimura, H. Tsutsui, K. Miyake, Y. Suda, Y. Fujimoto, K. Fukase, Chem. Eur. J. 2011, 17, 14464 – 14474. S. Kobayashi, H. Kuroda, Y. Ohtsuka, T. Kashihara, A. Masuyama, K. Watanabe, Tetrahedron 2013, 69, 2251 – 2259. a) V. Antonucci, J. Coleman, J. B. Ferry, N. Johnson, M. Mathe, J. P. Scott, J. Xu, Org. Process Res. Dev. 2011, 15, 939 – 941; b) R. K. Henderson, C. Jim¦nez-Gonzlez, D. J. C. Constable, S. R. Alston, G. G. A. Inglis, G. Fisher, J. Sherwood, S. P. Binks, A. D. Curzons, Green Chem. 2011, 13, 854 – 862; c) D. Prat, O. Pardigon, H.-W. Flemming, S. Letestu, V. Ducandas, P. Isnard, E. Guntrum, T. Senac, S. Ruisseau, P. Cruciani, P. Hosek, Org. Process Res. Dev. 2013, 17, 1517 – 1525. For examples, see: a) S. Kanemura, A. Kondoh, H. Yorimitsu, K. Oshima, Synthesis 2008, 2659 – 2664; b) Z. Fei, Q. Wu, F. Zhang, Y. Cao, C. Liu, W.C. Shieh, S. Xue, J. McKenna, K. Prasad, M. Prashad, D. Baeschlin, K. Namoto, J. Org. Chem. 2008, 73, 9016 – 9021. For the Grignard reaction with CPME, a supernatant liquid was used. Y.-H. Lai, Synthesis 1981, 585. U. Tilstam, H. Weinmann, Org. Process Res. Dev. 2002, 6, 906 – 910. A. Kadam, M. Nguyen, M. Kopach, P. Richardson, F. Gallou, Z.-K. Wan, W. Zhang, Green Chem. 2013, 15, 1880 – 1888. To the best of our knowledge, 3-thienylmagnesium bromide (2 p) is not commercially available. H. P. Acharya, K. Miyoshi, Y. Kobayashi, Org. Lett. 2007, 9, 3535 – 3538. The purity of commercial CPME used in the first experiment was 99.97 % (determined by GC). Detailed GC–MS data are presented in the Supporting Information. a) C. Alvarado, Ý. Guzmn, E. Daz, R. PatiÇo, J. Mex. Chem. Soc. 2005, 49, 324 – 327; b) E. Riva, S. Gagliardi, M. Martinelli, D. Passarella, D. Vigo, A. Rencurosi, Tetrahedron 2010, 66, 3242 – 3247. X.-M. Gan, S. Parveen, W. L. Smith, E. N. Duesler, R. T. Paine, Inorg. Chem. 2000, 39, 4591 – 4598. If the Grignard addition was performed at 0 8C, an approximate 10 % lower yield was obtained. For a review of the synthesis, see: K. M. Kasiotis, S. A. Haroutounian, Curr. Org. Chem. 2012, 16, 335 – 352. Our synthetic approach referred to previous reports, see: a) P. R. D. Murray, D. L. Browne, J. C. Pastre, C. Butters, D. Guthrie, S. V. Ley, Org. Process Res. Dev. 2013, 17, 1192 – 1208; b) M. P. Drapeau, I. Fabre, L. Gri-

Asian J. Org. Chem. 2016, 5, 636 – 645

www.AsianJOC.org

[22]

[23]

[24]

[25]

[26]

[27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42]

maud, I. Ciofini, T. Ollevier, M. Taillefer, Angew. Chem. Int. Ed. 2015, 54, 10587 – 10591; Angew. Chem. 2015, 127, 10733 – 10737. a) C. V. Jordan in Comprehensive Medicinal Chemistry II, Vol. 8 (Ed.: J. B. Taylor, D. J. Triggle), Elsevier, 2007, pp. 83 – 102; b) P. Y. Maximov, R. E. McDaniel, V. C. Jordan, Tamoxifen: Pioneering Medicine in Breast Cancer, Springer, Basel, 2013. Pace et al. succeeded in halogen – metal exchange in CPME by using a combination of aryl bromide and tBuLi. However, we avoided the use of tBuLi in consideration of the safety, see: V. Pace, L. Castoldi, S. Monticelli, S. Safranek, A. Roller, T. Langer, W. Holzer, Chem. Eur. J. 2015, 21, 18966 – 18970. Separation of Z and E isomers was first reported by Bedford and Richardson, see: G. D. Bedford, D. N. Richardson, Nature 1966, 212, 733 – 734. The use of DIBALH for the activation of the surface of zinc in the Reformatsky reaction has been reported, see: a) M. J. Girgis, J. K. Liang, Z. Du, J. Slade, K. Prasad, Org. Process Res. Dev. 2009, 13, 1094 – 1099; b) G. Loh, R. Tanigawa, S. M. Shaik, K. Sa-ei, L. Wong, P. N. Sharratt, Org. Process Res. Dev. 2012, 16, 958 – 966. We searched for Grignard reagents commercially available from selected suppliers. The search result is summarized in the Supporting Information. H.-S. Lin, L. A. Paquette, Synth. Commun. 1994, 24, 2503 – 2506. M. Kuriyama, R. Shimazawa, T. Enomoto, R. Shirai, J. Org. Chem. 2008, 73, 6939 – 6942. C. Qin, H. Wu, J. Cheng, X. Chen, M. Liu, W. Zhang, W. Su, J. Ding, J. Org. Chem. 2007, 72, 4102 – 4107. J. Karthikeyan, M. Jeganmohan, C.-H. Cheng, Chem. Eur. J. 2010, 16, 8989 – 8992. T. Brodmann, P. Koos, A. Metzger, P. Knochel, S. V. Ley, Org. Process Res. Dev. 2012, 16, 1102 – 1113. C.-H. Xing, Q.-S. Hu, Tetrahedron Lett. 2010, 51, 924 – 927. M. B. Runge, M. T. Mwangi, L. Miller II, M. Perring, N. B. Bowden, Angew. Chem. Int. Ed. 2008, 47, 935 – 939; Angew. Chem. 2008, 120, 949 – 953. M. Hatano, O. Ito, S. Suzuki, K. Ishihara, J. Org. Chem. 2010, 75, 5008 – 5016. R. J. Rahaim, Jr., R. E. Maleczka, Jr., Org. Lett. 2011, 13, 584 – 587. C.-H. A. Lee, T.-P. Loh, Tetrahedron Lett. 2004, 45, 5819 – 5822. P. R. Blakemore, S. P. Marsden, H. D. Vater, Org. Lett. 2006, 8, 773 – 776. K. Miura, D. Wang, A. Hosomi, J. Am. Chem. Soc. 2005, 127, 9366 – 9367. A. Ogata, M. Nemoto, K. Kobayashi, A. Tsubouchi, T. Takeda, J. Org. Chem. 2007, 72, 3816 – 3822. L. Salvi, S.-J. Jeon, E. L. Fisher, P. J. Carroll, P. J. Walsh, J. Am. Chem. Soc. 2007, 129, 16119 – 16125. J. Barluenga, F. J. Fananas, M. Yus, J. Org. Chem. 1979, 44, 4798 – 4801. T. Matsuda, M. Makino, M. Murakami, Org. Lett. 2004, 6, 1257 – 1259.

Manuscript received: February 1, 2016 Final Article published: March 7, 2016

645
