Hideo Okaa,
Katsuya Kitaia,
Takeyuki Suzukib and
Yasushi Obora*a
aDepartment of Chemistry and Materials Engineering, Faculty of Chemistry, Materials and Bioengineering, Kansai University, Suita, Osaka 564-8680, Japan. E-mail: obora@kansai-u.ac.jp
bComprehensive Analysis Center, The Institute of Scientific and Industrial Research (ISIR), Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka 567-0057, Japan
First published on 26th April 2017
We report the catalysis of a Sonogashira–Hagihara cross-coupling reaction using a DMF-stabilized copper nanoparticle catalyst. The reaction proceeded with low catalyst loadings, and a turnover number of 4.0 × 103 was recorded for 0.01 mol% catalyst loading. DMF-stabilized copper nanoparticles thus showed high catalytic activity.
Transition metal nanoparticles (M NPs) have been extensively studied with regard to their application in catalytic reactions, with the aim of identifying the specific advantages of their high surface area and unique properties.24,25 The synthesis of metal NPs commonly requires a metal precursor, reductant, protectant, and solvent. M NPs stabilized by dendrimers and supported on metal oxides have also been reported.26–29 Copper NPs (Cu NPs) with various nanostructures have also been developed, and research into their applicability as catalysts has been carried out.26,27,30 There are many reports of Sonogashira coupling catalyzed by M NPs.31–37 However, in almost all cases, these reports describe Pd NP catalyst systems.31–33,38–41 Cu NPs are superior in terms of cost, but there are a limited number of examples in which Cu NPs are used as catalysts for the Sonogashira reaction.34–37 In addition, a greater amount of catalyst is required for Cu NP-catalyzed reactions, and multi-step procedures are required for the synthesis of these catalysts.28,29,34–37 Rothenberg and co-workers were the first to report a copper-catalyzed palladium-free and ligand-free Sonogashira cross coupling with various aryl iodides and bromides.35 For this they used Cu nanoclusters prepared from chloride salt precursors by using tetra-n-octylammonium formate (TOAF) as a stabilizer in DMF at 65 °C. However, a multi-step operation was required for the synthesis of the Cu nanoparticles and the Sonogashira reaction was carried out with 5 mol% of the Cu NP catalyst.35 Tang, Zhang, and co-workers reported that octahedral Cu2O NPs could be used in a reusable Cu2O/PPh3/n-Bu4NBr (TBAB) system to catalyze Sonogashira coupling.37 However, a multistep synthesis is required to access Cu NPs in this system; this includes the use of hazardous reductants, such as hydrazine Cu NPs. Furthermore, this Cu NP system required a relatively large amount of the Cu NP catalyst (10 mol%), and TBAB surfactant was also needed.37 Therefore, the development of more effective Cu NPs that require smaller catalytic amounts and less demanding synthetic procedures is desired.
We have reported the synthesis of Cu NPs using the DMF-reduction method without the need for specialized reductants or surfactants, as well as catalytic reactions using these Cu NPs.42 DMF-stabilized M NPs synthesized with this method have very small (<10 nm) particle sizes. The specific surface area of the M NPs increases as the particle diameter decreases, and this leads to an improvement in the catalytic activity. We have previously reported that DMF-stabilized Pd NCs catalyze Suzuki–Miyaura cross coupling,43 the Mizoroki–Heck reaction,43 and Migita–Kosugi–Stille coupling,44 and that DMF-stabilized Cu NPs catalyze the Ullmann coupling reaction.45 For these catalytic reactions, we recorded higher turnover numbers (TONs), and thus concluded that DMF-stabilized M NPs have high catalytic activity. Herein, we report that DMF-stabilized Cu NPs can catalyze the Sonogashira coupling reaction. DMF-stabilized Cu NPs prepared by a simple methodology from CuCl2 and DMF (as solvent, reductant, and stabilizer) promoted the reaction of terminal alkynes with aryl halides at low catalyst loadings (0.01 mol%) with a TON of 4.0 × 103. This is an improvement over previously reported TBAB- and TOAF-protected Cu NPs, which require 5–10 mol% catalyst loadings.35,37
The transmission electron microscopy (TEM) image and particle size distribution of the prepared Cu NPs revealed that their diameters mainly ranged from 2 to 7 nm (Fig. 1). In addition, X-ray diffraction (XRD) spectra showed that the NPs had an amorphous nature (Fig. S1†).
Fig. 1 (a) Transmission electron microscopy image of DMF-stabilized copper nanoparticles (scale bar = 10 nm); (b) particle size distribution of the nanoparticles. |
The FT-IR spectrum of the DMF-stabilized Cu NPs showed a strong absorption at around 1670 cm−1 (Fig. S2(a)†), which corresponds to the ν(CO) vibration of the DMF molecules. These results suggest that the DMF molecules interact with the Cu NPs.46,47 Thermogravimetric analysis-differential thermal analysis (TG-DTA) of the Cu NPs confirmed that the nanoparticles are thermally stable up to around 150 °C (Fig. S3†).
The synthesis of acetylene by the coupling of aryl halides with terminal alkynes has a wide scope, and is utilized in areas such as the pharmaceutical and electronic materials fields. In terms of cost advantage, it would be more useful to develop an M NP-catalyzed version of this reaction than of any other reaction. Furthermore, a wide variety of additives and solvents are tolerated in this reaction (Table 1). First, the Sonogashira reaction of the model compounds iodobenzene (1a, 0.5 mmol) and phenylacetylene (2a) was catalyzed by Cu NPs in DMF at 135 °C for 48 hours, using K2CO3 as the base. However, this reaction showed low conversion and we only obtained a trace of the desired product (entry 1). Addition of PPh3 promoted the reaction and improved conversion; we obtained desired product 3a in quantitative yield (entry 2). We then examined other phosphine, nitrogen, and diketone compounds because of the great improvement observed when using the PPh3 additive. Phosphine compounds other than PPh3, for example P(o-tolyl)3, PCy3, and dppe did not promote the reaction (entries 3–5). In contrast, nitrogen compounds such as bipyridine (bpy) and 4,4′-di-tert-butyl-2,2′-bipyridine (dbbpy) were very effective and the reaction proceeded quantitatively when these additives were used (entries 6 and 7). 1,10-Phenanthroline (1,10-phen) and N,N,N′,N′-tetramethylethylenediamine (TMEDA) improved the yield slightly (entries 8 and 9), and using 2,2,6,6-tetramethyl-3,5-heptadione (TMHD) gave the product in moderate yields (entry 10). However, when acetylacetone (acac) was used instead of PPh3 the product yield was 19% (entry 11).
Entry | Additive | Base | Solvent | Conv. (%) | Yieldb (%) |
---|---|---|---|---|---|
a Conditions: 1a (0.5 mmol), 2a (0.75 mmol), Cu NPs (2 × 10−1 mol%), additive (10 mol%), base (1.0 mmol), solvent (1 mL), 135 °C, 48 h, under Ar.b GC yield. The number in parentheses shows the isolated yield.c With 0.01 mol% Cu NPs.d At 80 °C.e At 120 °C.f Without catalyst. | |||||
1 | — | K2CO3 | DMF | 9 | 7 |
2 | PPh3 | K2CO3 | DMF | >99 | >99 (96) |
3 | P(o-tol)3 | K2CO3 | DMF | 73 | 14 |
4 | PCy3 | K2CO3 | DMF | 47 | Trace |
5 | dppe | K2CO3 | DMF | 65 | Trace |
6 | bpy | K2CO3 | DMF | >99 | >99 |
7 | dbbpy | K2CO3 | DMF | >99 | >99 |
8 | 1,10-Phen | K2CO3 | DMF | 82 | 40 |
9 | TMEDA | K2CO3 | DMF | 63 | 25 |
10 | TMHD | K2CO3 | DMF | 84 | 45 |
11 | acac | K2CO3 | DMF | 60 | 19 |
12 | PPh3 | Na2CO3 | DMF | 66 | 55 |
13 | PPh3 | Cs2CO3 | DMF | 65 | Trace |
14 | PPh3 | KOH | DMF | 27 | 18 |
15 | PPh3 | NEt3 | DMF | <1 | Trace |
16 | PPh3 | — | DMF | 59 | Trace |
17 | PPh3 | K2CO3 | NMP | 89 | 73 |
18 | PPh3 | K2CO3 | DMSO | >99 | >99 |
19 | PPh3 | K2CO3 | H2O | >99 | 88 |
20 | PPh3 | K2CO3 | Toluene | >99 | >99 |
21c | PPh3 | K2CO3 | DMF | 84 | 41 |
22d | PPh3 | K2CO3 | DMF | 3 | Trace |
23e | PPh3 | K2CO3 | DMF | 64 | 51 |
24f | PPh3 | K2CO3 | DMF | <1 | n.d. |
To investigate the effect of the DMF molecules on the surface of the Cu NPs during the catalytic reaction, FT-IR measurements featuring the DMF ν(CO) vibration peak at around 1670 cm−1 were conducted on Cu NP samples that had undergone complete removal of the DMF solvent using an NaCl plate method (Fig. S2†). These analyses indicated that DMF was liberated from the Cu NPs after catalytic reactions with PPh3 (entry 2 and Fig. S2(b) and (c)†) and bipyridine (entry 6 and Fig. S2(d)†). However, the ν(CO) vibration peak remained for a mixture of Cu NPs with PCy3 after the reaction had finished. This result indicates that addition of PCy3 did not induce the liberation of DMF, probably owing to steric constraints (entry 4 and Fig. S2(e)†).
We next measured the particle size distribution of the Cu NPs by TEM and dynamic light scattering under the conditions shown in entry 2 of Table 1 (Fig. 2 and S4†). A slight growth in the particle size was observed after the reaction. However, by inhibiting Ostwald ripening42,48,49 during the course of the catalytic reaction, the nanoparticles retained their original size of <10 nm.
Fig. 2 (a) Transmission electron microscopy image (scale bar = 10 nm) and (b) particle size distribution of the nanocatalyst after reaction under the conditions shown in entry 2, Table 1. |
The Cu NPs are themselves inactive as catalysts in this reaction owing to the strong coverage of DMF molecules on their surfaces (Fig. S5† and entry 1, Table 1). The liberation of the DMF molecules during the course of the reaction, which is assisted by additives such as PPh3, generates the active NP catalyst. PPh3 (or bipyridine) presumably serves to stabilize the uncoordinated sites on the Cu nanoparticles during the course of the reaction (Fig. 2 and S4†).
As shown in entry 2 of Table 1, K2CO3 was a good choice of base and gave excellent yields when used with the PPh3 additive. We also investigated other carbonate salts under the same conditions. Using a weak base, such as Na2CO3, gave the desired product in moderate yield (entry 12), but strong bases, such as Cs2CO3 or KOH, were not suitable for this reaction (entries 13 and 14). The substrates did not convert at all when using triethylamine or base-free conditions (entries 15 and 16). We next investigated the effect of changing the solvent and found that aprotic polar solvents, such as N-methyl-2-pyrrolidone (NMP) and dimethyl sulfoxide (DMSO), were most appropriate for this catalytic system (entries 17 and 18). Using H2O as the solvent gave good yields (entry 19), and quantitative yields of the product were also obtained with toluene (entry 20), which is less toxic than NMP and DMSO. Thus the reaction proceeded smoothly with various solvents under these conditions (entries 2, 18–20).
For this reaction, TONs of up to 4.0 × 103 were obtained by reducing the amount of the catalyst to 0.01 mol% (entry 21), although a small amount of a dehalogenated byproduct was observed. The time–yield curves for the formation of 3a with the Cu NP catalyst under the conditions in entry 2 of Table 1 (Fig. S6†) show that the reaction was mostly complete after about 24 h, and that the turnover frequency at the initial stage of the reaction was 29 h−1.
The reaction temperature is another important factor in this catalytic system. The reaction substrates did not convert at 80 °C (entry 22, Table 1) and only moderate yields were obtained at 120 °C (entry 23). The reaction did not proceed without the catalyst (entry 24).
We next attempted Cu NP-catalyzed Sonogashira cross-coupling reactions using various reaction substrates (Table 2), and found that DMF-stabilized Cu NPs had a broad scope in Sonogashira coupling reactions when the optimized reaction conditions were used. First, the reactivity of iodobenzene derivatives in reactions with phenylacetylene was studied (entries 1–9). Sterically hindered substrates and iodobenzenes bearing electron-donating or electron-withdrawing substituents gave the corresponding internal alkyne in moderate-to-good yields (3b–f, entries 1–5). Heteroaromatic iodides, such as 3-iodopyridine or 2-iodothiophene, provided the corresponding heteroaromatic internal alkynes (3g and 3h) in good yields (95% and 75%, respectively, entries 6 and 7).
Entry | Ar-X (1) | Alkyne (2) | Product (3) | Yielda (%) |
---|---|---|---|---|
a Conditions: same as entry 2, Table 1. Yields of the isolated product after purification.b Bromobenzene (2.0 mmol) and phenylacetylene (0.5 mmol) were used.c Not detected by GC. | ||||
1 | 57 | |||
2 | 2a | 90 | ||
3 | 2a | 97 | ||
4 | 2a | 84 | ||
5 | 2a | 93 | ||
6 | 2a | 95 | ||
7 | 2a | 75 | ||
8b | 2a | 90 | ||
9 | 2a | 3a | n.d.c | |
10 | 3e | 84 | ||
11 | 1a | 92 | ||
12 | 1a | 57 | ||
13 | 1a | 3g | 87 |
We also applied this catalytic system to the less reactive substrates bromobenzene and chlorobenzene in an attempt to extend the scope of the halide leaving group. Sonogashira coupling products were obtained from bromobenzene by using excess aryl halide reagent. However, no coupling products were obtained from chlorobenzene. Both electron-donating and electron-withdrawing groups were tolerated on the terminal alkyne and gave desired products 3e and 3i in good yields (entries 10 and 11). This is a result of the high acidity of the C–H bond attached to the electron-withdrawing alkyne group. In contrast, because aliphatic terminal alkynes lower the acidity of the C–H position, we obtained the internal alkyne with an aliphatic moiety (3j) in only moderate yield (entry 12). In addition, heteroaromatic substrate 3-ethynylpyridine gave corresponding heteroaromatic internal alkyne 3-(phenylethynyl) pyridine (3g) in good yield (87%, entry 13).
To gain further insight into the function of PPh3 on the catalytic activity of the Cu NPs, we carried out a poisoning test in the presence of mercury (5 equiv.) under the conditions shown in entry 2 of Table 1. This reaction gave coupling product 3a in only 3% yield and with low conversion (20%). This result indicates that PPh3-stabilized Cu NPs have more uncoordinated active sites available on their Cu surfaces, whereas the DMF seems to passivate the surface of the nanoparticles.
The reusability of the Cu NP catalyst was examined in toluene, and the results showed that the catalyst could be reused at least five times without loss of activity (Fig. 3).
Fig. 3 Repeated runs of the same catalyst under the reaction conditions shown in entry 20 of Table 1 (red: conversion of 1a; blue: yield of 3a). |
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra27910d |
This journal is © The Royal Society of Chemistry 2017 |