Synthetic study of vic-bromination of diarylacetylenes, easy purification and separation

Introduction

The addition reaction of bromine to diphenylacetylene is one of basic reactions and is recognized as important, because resulting 1,2-dibromo-1,2-diphenylethylene can serve as a precursor for various molecular transformations in organic synthesis [1,2]. This reaction seems to be simple, but some reported studies suggest that it is surprisingly difficult to obtain the E isomer selectively (Scheme 1). For selected examples, Espenson and co-workers reported in 1999 that E-selective bromination of diphenylacetylene took place by using H₂O₂, NaBr, and AcOH in the presence of MTO (methyltrioxorhenium(VII)) to give (E)-1,2-dibromo-1,2-diphenylethylene, selectively (Scheme 1a) [3]. However, MTO is an expensive reagent and a more versatile synthetic method is highly needed. As a simple method, the reaction of diphenylacetylene and Br₂ in CH₂Cl₂ was reported by Glorius in 2007 [4], but the selectivity was reported for the Z isomer (Scheme 1b). As another example in 2008, Adimurthy and co-workers proposed the combination of NaBr-NaBrO₃ in AcOH in order to prepare in-situ generated Br₂, which was reacted with diphenylacetylene to give no selectivity between E and Z isomers of 1,2-dibromo-1,2-diphenylethylenes (Scheme 1c) [5]. Previous studies reported the selective synthesis of (E)-1,2-dibromo-1,2-diphenylethylene and related compounds, but Br₂ or expensive reagent, which was difficult to handle, was often used [6-15].

Based on these research backgrounds, we have also been interested in the selective synthesis of the E isomer of 1,2-dibromo-1,2-diphenylethylene, together with simple purification and separation toward process chemistry, because the compound is an attractive synthetic intermediate for further transformations. During the course of our studies, we have found that the combination of NBS (N-bromosuccinimide) and FeBr₃ in CH₂Cl₂ was good for the reaction with diphenylacetylene [16-19] to give 1,2-dibromo-1,2-diphenylethylene with high E-selectivity. The present method has some advantages. First, is mild reaction conditions, and second is stereoselective synthesis. In addition, we have found an easy purification and separation method of E and Z isomers of the crude product by using heptane. The method utilizes the difference in solubility of E and Z isomers. Because of this technique, the purification, separation, and scale-up were easy to perform. In this paper, we wish to report the details of the results.

Results and Discussion

In order to obtain the E isomer of 1,2-dibromo-1,2-diphenylethylene, we have examined the reaction regarding optimized conditions (Table 1). The typical procedure is as follows: the reaction of diphenylacetylene (1a, 0.5 mmol) with NBS (N-bromosuccinimide, X equiv), FeBr₃ (1.0 equiv), n-Bu₄NBr (Y equiv), and TEMPO (2,2,6,6-tetramethylpiperidine 1-oxyl free radical, Z equiv) in solvent (2 mL) at the desired temperature was conducted for 23 h or 2 h. After the reaction, the typical work-up procedure was carried out and the crude product was analyzed for the ratio of E and Z isomers and starting material 1a, by using HPLC. Purification and separation from the crude product was conducted by the filtration using heptane (Table 1). At first, we have recognized that E-2a is solid and Z-2a is liquid in heptane. After various considerations, we found that E-2a could be isolated by only filtration, utilizing the difference in solubility of E-2a and Z-2a in heptane from the crude product.

In Table 1, entry 1, the combination of NBS (2.2 equiv) and FeBr₃ (1.0 equiv) for 2 h gave the ratio of E-2a/Z-2a/1a such as 86:14:0 by HPLC, in which E-2a was obtained in 78% yield. Similarly, to examine the reproducibility of the chemical yield and the ratio of E-2a/Z-2a/1a by GC analysis, Table 1, entry 2 was shown. The reaction time of 2 h gave the same ratio of E-2a/Z-2a/1a by GC analysis, and E-2a was obtained in 78% yield (Table 1, entry 2). The use of a smaller amount of NBS (1.2 equiv) did not result in a decrease of yield, indicating that FeBr₃ is thought to be one of the sources of bromine (Table 1, entry 3). To increase the chemical yield and ratio of E-2a/Z-2a/1a, various parameters such as NBS, n-Bu₄NBr, TEMPO, temperature, and solvent were examined (Table 1, entries 3–12). Because there was a possibility that a bromine radical was involved, TEMPO was used for the purpose of the ratio-control of E and Z isomers. TEMPO was found to have a tendency to suppress the generation of Z-2a in CH₂Cl₂ (Table 1, entries 6 and 8) [20]. Among Table 1, entries 5–11, the addition of n-Bu₄NBr did not seem to have any effect. Instead of CH₂Cl₂, C₂H₈Cl₂ (1,2-dichloroethane) was used as the solvent, but the yield of E-2a was decreased (Table 1, entries 11 and 12). Generally, decomposition of the product or over reaction was not observed by GC and/or GC–MS analysis, in the case of the longer reaction time of 23 h. We have examined other Lewis acids such as ZnBr₂, LiBr, and FeBr₂ instead of the use of FeBr₃ under similar reaction conditions. The use of ZnBr₂ for reaction time of 4 h showed a ratio of E-2a/Z-2a/1a = 86:14:0 by HPLC analysis and 68% yield of E-2a. In the case of LiBr, the desired reaction progressed only slightly, even after two days, and the ratio of E-2a/Z-2a/1a = 17:3:80 was indicated by GC analysis. In addition, the use of FeBr₂ for a reaction time of 3 h gave an E-2a/Z-2a/1a ratio of 67:13:20 by GC analysis and 57% yield of E-2a. Based on these investigations, we have adopted the conditions of Table 1, entry 2, because better chemical yield of E-2a as well as a higher ratio of E-2a/Z-2a/1a with short reaction time were attractive for chemical synthesis in the laboratory [21].

Next, we have investigated the scope and limitations by using various diarylacetylenes bearing the substituents on the benzene ring at para-position. The results were summarized in Table 2. Depending on the product's characteristics, either heptane or methanol was used for purification and separation. The reaction of 1,2-di-p-tolylethyne (1b) under the optimized conditions gave the corresponding product with a ratio of 68:32 of E and Z isomers. The isolated yield of E-2b was <54% yield (Table 2, entry 1). Likewise, 1,2-bis(4-(tert-butyl)phenyl)ethyne (1c) showed the similar tendency to give E-2c in 45% yield (Table 2, entry 2). Poor selectivity as shown in entries 1 and 2 might be due to the stability of the intermediate. In the cases of Table 2, entries 3–6 residues such as MeO, F, Br and CN as the substituent at para position of the benzene ring in 1d–g, the selectivity of E and Z and the isolated yield of the E isomer seemed to be good. In the case of unsymmetrical diarylacetylene, 1-(tert-butyl)-4-((4-fluorophenyl)ethynyl)benzene (1h) was tested and the corresponding E-2h was isolated and purified with MeOH. E-2h was obtained in 41% yield (Table 2, entry 7).

A gram-scale synthesis of E-2a and Z-2a from 1a was performed (Scheme 2a). 1a (20.0 mmol) was allowed to react under optimized conditions for 5 hours. The purification and separation by using heptane from the crude product afforded E-2a (4.1 g, 12.2 mmol, 61% yield) and Z-2a (2.5 g, 7.2 mmol, 36% yield), indicating that purification and separation by heptane make scale-up very easy. The E-selectivity was dropped in gram-scale synthesis, although the reason is unclear at present.

For the understanding of the reaction mechanism, some control experiments were carried out. The reaction of 1a with NBS and n-Bu₄NBr without FeBr₃ gave trace amount of E-2a (Scheme 2b). In addition, the reaction of 1a with FeBr₃ and n-Bu₄NBr without NBS showed no reaction (Scheme 2c). Instead of NBS, Br₂ (1.1 equiv) was reacted with 1a and FeBr₃ (1.0 equiv) to give E-2a in 56% yield, with a ratio of E-2a/Z-2a/1a = 79:21:0. These results indicate that the combination of NBS and FeBr₃ might be important and in situ Br₂ might be generated and formed. FeBr₃ might react with Br₂ to form [FeBr₄]⁻ and "Br⁺", which can react with 1a.

As for the reaction mechanism, the plausible pathway might be the following (Scheme 3). The reaction of NBS and FeBr₃ might generate Br₂ or "Br⁺", which can react with 1a to form intermediate A. A reacts with Br⁻ to give E-2a. Another possibility is the reaction of NBS and FeBr₃ gives coordinated intermediate B, which can react with 1a to give A, leading to the formation of E-2a [21,22]. Both pathways may be mixed together. One reason why E and Z isomers are formed together might be due to the equilibrium between A and C, although there are many unknown points.

Conclusion

In summary, we have developed a simple and selective procedure of the synthesis of (E)-1,2-dibromo-1,2-diphenylethylene from diphenylacetylene by using NBS and FeBr₃ in CH₂Cl₂. In addition, heptane played an important and critical role for purification and separation of the crude product. Scope and limitations for bromination of various diarylacetylenes together with the gram-scale synthesis and control experiments were also investigated. These results and findings can contribute to process chemistry for the synthesis of (E)-1,2-dibromo-1,2-diphenylethylene and their derivatives. Further synthetic investigations are currently underway in our laboratory.

Experimental

The experimental procedures and characterization data (Table 1, entry 1). The glass flask was dried and heated by a heating gun. After cooled to room temperature, the flask was filled with N₂. FeBr₃ (154.2 mg, 0.52 mmol) and NBS (N-bromosuccinimide, 196.0 mg, 1.10 mmol) were added to the glass flask. Then, CH₂Cl₂ (dry, 2.0 mL) and 1,2-diphenylacetylene (1a, 89.4 mg, 0.502 mmol) were added and the mixture was stirred at room temperature for 2 hours. A 10% aqueous solution of Na₂S₂O₃ (20 mL) was added to stop the reaction. The mixture was extracted with CH₂Cl₂ (20 mL × 1), and separated. The aqueous phase was extracted with CH₂Cl₂ (20 mL × 2). The combined organic phase was washed with H₂O (20 mL) and brine (20 mL), and dried over Na₂SO₄. After this, filtration and concentration were performed; the organic phase was passed through a short column of silica gel using CH₂Cl₂ (100 mL) to remove inorganic materials and others; then it was concentrated under reduced pressure to give the crude product. This crude product was purified three times with heptane to obtain (E)-1,2-dibromo-1,2-diphenylethylene (E-2a, 132.6 mg, 0.392 mmol, 78% yield) of high-purity.

(E)-1,2-Dibromo-1,2-diphenylethylene (E-2a) [4]: White solid. ¹H NMR (400 MHz, CDCl₃) δ 7.34–7.46 (m, 6H), 7.50–7.57 (m, 4H) ppm; ¹³C NMR (100 MHz CDCl₃) δ 118.0, 128.4, 128.9, 129.1, 140.7 ppm; HRMS (ESI) m/z: [M + Na]⁺ calcd for C₁₄H₁₀Br₂Na, 358.9041; found, 358.9028.

(Z)-1,2-Dibromo-1,2-diphenylethylene (Z-2a) [7]: Yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 7.10–7.22 (m, 10H) ppm; ¹³C NMR (75 MHz, CDCl₃) δ 125.7, 128.0, 128.3, 129.8, 139.4 ppm; HRMS (ESI) m/z: [M + Na]⁺ calcd for C₁₄H₁₁Br₂, 336.9222; found, 336.9215.