Poly(tetrafluoroethylene)-Stabilized Metal Nanoparticles: Preparation and Evaluation of Catalytic Activity for Suzuki, Heck, and Arene Hydrogenation in Water.

Poly(tetrafluoroethylene)-stabilized Pd nanoparticles (PTFE-PdNPs) were prepared in water with 4-methylphenylboronic acid as a reductant and characterized using powder X-ray diffraction, transmission electron microscopy (TEM), X-ray photoelectron spectroscopy, and inductively coupled plasma-atomic emission spectroscopy (ICP-AES). Small PdNPs with a fairly uniform size were obtained in the presence of PTFE, whereas aggregation of palladium was observed in the absence of PTFE. PTFE-PdNPs showed high catalytic activity for the Suzuki coupling reaction in water and were reused without any loss of activity. No palladium species were observed by ICP-AES analysis in the reaction solution after the reaction, nor was any change in particle size observed after the recycle experiment. PTFE-PdNPs also exhibited excellent catalytic activity and reusability for the Heck reaction in water. Although palladium species were not detected in the reaction solution after the reaction, aggregates and smaller sizes of PdNPs were observed in the TEM image of the recovered catalyst. PTFE was also useful as the stabilizer of rhodium nanoparticles (RhNPs) prepared by reduction with NaBH4. PTFE-RhNPs showed high catalytic activity and reusability toward arene hydrogenation under mild conditions.

Introduction

In the past several decades, chemical transformations in aqueous media have attracted tremendous attention from both academic researchers and industrial scientists because of advantages such as reaction rate acceleration and work-up procedure simplification.[1,2] In addition, water is safe, abundant, nontoxic, and inexpensive. The exploration of catalytic reactions with n class="Chemical">metal nanoparticle-based heterogeneous catalysts in neat water is increasing dramatically, because they exhibit unique surface properties, high specific surface areas, and high catalytic activities in water.[3−5] One major drawback, however, is that metal nanoparticles tend to lose their catalytic activity during use. This is a consequence of the formation of aggregates, owing to their high surface energy, via so-called Ostwald ripening. For this reason, a number of different solid support materials for metal nanoparticles have been investigated, including natural polymers,[6−12] synthetic polymers,[13−18] surfactants,[19,20] inorganic supports,[21,22] metal–organic frameworks,[23,24] covalent organic frameworks,[25,26] and mesoporous supports.[27−29]

Palladium-catalyzed cross-coupling reactions of organic n class="Chemical">halides with organo-boronic acids (Suzuki coupling reaction)[30] and olefins (Heck reaction)[31] for carbon–carbon bond formation are extremely useful to the chemical industry and in research. Both reactions are efficient methods to construct the sp2 C–sp2 C bond with high tolerance to the presence of functional groups as the substituents and have been frequently used to evaluate the catalytic activity of the developed palladium nanoparticles (PdNPs).[6−10,13,15,19,21] For example, Dewan et al. reported a green and economical synthesis of PdNPs, which showed high catalytic activity for the Suzuki coupling reaction at room temperature.[12] Arene hydrogenation has taken part in the facile production of substituted cyclohexane derivatives on both laboratory and industrial scales.[32] One of the most useful transition metal as a catalyst is rhodium, and the hydrogenation of aromatic rings in water with rhodium nanoparticles (RhNPs) was also reported recently.[33,34]

On the basis of the idea that metal nanoparticles with a fairly uniform size would be stabilized on n class="Chemical">polymer supports by hydrophobic interactions in water, our group successfully developed linear polystyrene (PS)-stabilized metal nanoparticles with applications to several reactions in water.[35−38] Linear PS-stabilized metal nanoparticles, however, could not be used for the reaction including C–H activation of aromatic ring and hydrogenation of aromatic ring because PS was used as a support. Whereas, poly(tetrafluoroethylene) (PTFE)–PS core–shell nanospheres were obtained from emulsifier-free styrene emulsion polymerization in water in the presence of PTFE latexes as seeds.[39] RhNPs can easily be deposited on the PTFE surface of the magnetic stir bar by stirring them in an RhNPs/BMImBF4 dispersion at room temperature.[40] According to these reports, we chose PTFE as a new polymer support and examined the catalytic activity of PTFE-supported Pd and RhNPs for the Suzuki coupling reaction, Heck reaction, and arene hydrogenation in water.

Results and Discussion

PTFE-stabilized n class="Chemical">PdNPs (PTFE-PdNPs) were prepared according to our previous method for PS-stabilized PdNPs.[35] A mixture of Pd(OAc)2, 4-methylphenylboronic acid, and PTFE was added to a 1.5 mol/L aqueous KOH solution. After the mixture was stirred vigorously (1350 rpm) at 90 °C for 5 h, the solution became colorless and a black precipitate formed. An X-ray diffraction (XRD) pattern for the black precipitate is presented in Figure . In addition to the diffraction peak at 2θ = 18° ascribed to PTFE, other diffraction peaks assigned to Pd are observed. The transmission electron microscopy (TEM) image indicated that particles with uniform sizes of 2.7 ± 0.4 nm were formed by this preparation method (Figure ). In contrast, although the nanoparticle size was similar, most of the Pd formed aggregates when PdNPs were prepared in the absence of PTFE, suggesting PTFE acted as the stabilizer for PdNPs (Figure ). Unfortunately, we could not confirm any obvious differences between PTFE and PTFE-PdNPs in the Fourier-transform infrared spectra. However, another broad peak was observed in addition to the peak ascribed to PTFE in X-ray photoelectron spectroscopy (XPS) spectra. When the XPS analysis for Pd 3d spectra of PTFE-PdNPs was conducted, the existence of several kinds of Pd species would be estimated and it was difficult to confirm the interaction between Pd and PTFE (Figure S1). In contrast, fluorine has more a partial positive charge because of the coordination to PdNPs, thereby justifying the shifting of the peak position to a higher energy range in PTFE-PdNPs. That is, it is considered that PdNPs would be immobilized on PTFE through the interaction with fluorine (Figure ). The loading of Pd determined by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was 2.59 mmol/g.

Figure 1

(a) XRD pattern of PTFE-PdNPs; (b) Powder Diffraction File #60-1504 (International Centre for Diffraction Data) for PTFE; (c) Powder Diffraction File #46-1043 (International Centre for Diffraction Data) for Pd.

Figure 2

(a) TEM image of PTFE-PdNPs (scale bar = 20 nm); (b) size distribution of PTFE-PdNPs.

Figure 3

(a) TEM image of PdNPs prepared in the absence of PTFE (scale bar = 50 nm); (b) magnification of the TEM image (a) (scale bar = 20 nm); (c) size distribution of PdNPs in the TEM image (b).

Figure 4

XPS analysis for F 1s spectra of (a) PTFE and (b) PTFE-PdNPs.

(a) XRD pattern of PTFE-PdNPs; (b) Powder Diffraction File #60-1504 (International Centre for Diffraction Data) for n class="Chemical">PTFE; (c) Powder Diffraction File #46-1043 (International Centre for Diffraction Data) for Pd.

(a) TEM image of PTFE-PdNPs (scale bar = 20 nm); (b) size distribution of n class="Chemical">PTFE-PdNPs.

(a) TEM image of Pdn class="Chemical">NPs prepared in the absence of PTFE (scale bar = 50 nm); (b) magnification of the TEM image (a) (scale bar = 20 nm); (c) size distribution of PdNPs in the TEM image (b).

XPS analysis for F 1s spectra of (a) n class="Chemical">PTFE and (b) PTFE-PdNPs.

To evaluate the catalytic activity of the PTFE-PdNPs, the Suzuki coupling reaction and the Heck reaction were carried out in n class="Chemical">water. When the reaction of bromobenzene with 4-methylphenylboronic acid was performed in a 1.5 mol/L KOH aqueous solution at 80 °C for 1 h in the presence of PTFE-PdNPs (1.0 mol % Pd based on bromobenzene), an 86% yield of 4-methylbiphenyl was obtained (Table , entry 1). After washing with water and diethyl ether and subsequent drying, the recovered catalyst was used for the next run under the same conditions. The catalyst was recycled at least nine times without any loss of activity (entries 2 and 3). No palladium species were observed in the reaction solution after the reaction, as confirmed by ICP-AES. Similarly sized PdNPs and no aggregates were observed by TEM after the recycling experiments (Figure ).

Table 1

Suzuki Coupling Reaction of Aryl Bromides with Arylboronic Acids in Water

entry	R1	R2	yield (%)a	TOF (h–1)
1	H	4-CH3	86	86
2b	H	4-CH3	81	81
3c	H	4-CH3	84	84
4	4-CH3O	4-CH3	56	56
5	4-CH3	4-CH3	79	79
6	4-CF3	4-CH3	62	62
7	4-NO2	4-CH3	24 (95)d	24 (95)d
8	H	4-CH3O	86	86
9	4-CH3	H	77	77
10	H	4-CF3	38	38

NMR yield.

10th run.

Average yield of 1st to 10th consecutive runs.

In the presence of TBAB.

Figure 5

(a) TEM image of the recovered catalyst after the 10th run of the Suzuki coupling reaction (scale bar = 20 nm); (b) size distribution of the recovered catalyst.

Both the electron-rich and electron-deficient aryl bromides were reactive, affording the desired coupling products in high to moderate yields (entries 4–6). The low yield in the case of n class="Chemical">4-bromonitrobenzene would be caused by the reaction of 4-bromonitrobenzene with the catalyst would be solid–solid reaction because of the high melting point of 4-bromonitrobenzene. Indeed, the coupling product was obtained in 95% yield when performing the reaction in the presence of tetrabutylammonium bromide (TBAB) (entry 7). When the Suzuki coupling reaction was performed using several arylboronic acids, the yield of the coupling product decreased with decreasing electron density on the aromatic ring (entries 8–10).

We next investigated the catalytic activity of PTFE-PdNPs for the Heck reaction in n class="Chemical">water. The coupling of iodobenzene with styrene took place smoothly in water at 90 °C in the presence of PTFE-PdNPs (1.0 mol % Pd based on iodobenzene) and TBAB to give (E)-stilbene in 91% yield (Table , entry 1). When the recyclability of the catalyst was examined under the same conditions, a slight catalytic deactivation was observed (entries 2–5). No detectable palladium species were observed by ICP-AES analysis in the reaction solution after the reaction. From the TEM image of the recovered catalyst, several aggregates (>5 nm) and a polymer portion in which PdNPs are not present were observed (circled areas in Figure a,b, respectively). The PdNPs in the recovered catalyst were slightly smaller than those in the fresh catalyst (Figure c). These data suggest that Ostwald ripening occurred during the reaction. A variety of aryl iodides were also successfully coupled with styrene to afford the desired product in good yields (entries 6–9). Although coupling products were obtained in excellent yields from the reaction with 4-chlorostyrene (entries 10–14), moderate yields were observed in the case of 4-methoxystyrene (entries 15–19).

Table 2

Heck Reaction of Aryl Iodides with Styrene Derivatives in Water

entry	R1	R2	yield (%)a	TOF (h–1)
1	H	H	91	6.1
2b	H	H	85	5.7
3c	H	H	85	5.7
4d	H	H	81	5.4
5e	H	H	71	4.7
6	4-CH3	H	86	5.7
7	4-CH3O	H	80	5.3
8	4-CF3	H	86	5.7
9	2-CH3	H	89	5.9
10	H	Cl	97	6.5
11	4-CH3	Cl	91	6.1
12	4-CH3O	Cl	91	6.1
13	4-CF3	Cl	94	6.3
14	2-CH3	Cl	99	6.6
15	H	CH3O	75	5.0
16	4-CH3	CH3O	80	5.3
17	4-CH3O	CH3O	45	3.0
18	4-CF3	CH3O	76	5.1
19	2-CH3	CH3O	61	4.1

NMR yield.

2nd run.

3rd run.

4th run.

5th run.

Figure 6

(a) TEM image of the recovered catalyst after the 5th run of the Heck reaction (scale bar = 50 nm); (b) magnification of the TEM image (a) (scale bar = 20 nm); (c) size distribution of PdNPs in the TEM image (b).

(a) TEM image of the recovered catalyst after the 5th run of the Heck reaction (scale bar = 50 nm); (b) magnification of the TEM image (a) (scale bar = 20 nm); (c) size distribution of Pdn class="Chemical">NPs in the TEM image (b).

PTFE-stabilized n class="Chemical">RhNPs were prepared using NaBH4 as a reductant in the presence of TBAB. The formation of PTFE-RhNPs was confirmed by TEM and XRD (Figures and 8). The loading value of Rh (2.52 mmol/g) was estimated by elemental analysis because RhNPs did not dissolve completely in acidic solution.

Figure 7

(a) TEM image of PTFE-RhNPs (scale bar = 20 nm); (b) size distribution of PTFE-RhNPs.

Figure 8

(a) XRD patterns of PTFE-RhNPs; (b) Powder Diffraction File #87-0714 (International Centre for Diffraction Data) for Rh.

(a) TEM image of PTFE-RhNPs (scale bar = 20 nm); (b) size distribution of n class="Chemical">PTFE-RhNPs.

(a) XRD patterns of PTFE-RhNPs; (b) Powder Diffraction File #87-0714 (International Centre for Diffraction Data) for n class="Chemical">Rh.

When the mixture of n-octylbenzene and n class="Chemical">PTFE-RhNPs (1.5 mol % Rh based on n-octylbenzene) was stirred at room temperature in water under H2 pressure (6 atm) for 15 h, a 97% yield of cyclohexyloctane was obtained. After the first reaction, PTFE-RhNPs were recovered and successively subjected to a second through fifth run under the same conditions to afford the desired products in 96, 96, 95, and 97% yields, respectively (Scheme ). ICP-AES analysis confirmed that the aqueous phase contained barely detectable levels of Rh. However, a larger RhNP size (3.1 ± 0.6 nm) and a broader size distribution were found by TEM observations of the recovered catalyst (Figure ).

Scheme 1

Recycling of the Catalyst for Hydrogenation of n-Octylbenzene

Figure 9

(a) TEM image of RhNPs in the recovered catalyst after the 5th run of the hydrogenation of n-octylbenzene (scale bar = 20 nm); (b) size distribution of RhNPs.

(a) TEM image of RhNPs in the recovered catalyst after the 5th run of the hydrogenation of n class="Chemical">n-octylbenzene (scale bar = 20 nm); (b) size distribution of RhNPs.

Conclusions

In summary, PTFE-stabilized n class="Chemical">Pd and RhNPs were prepared. Most Pd species aggregated when the reduction was performed in the absence of PTFE, indicating that PTFE was a useful stabilizer for metal nanoparticles. PTFE-PdNPs had high catalytic activity and reusability for the Suzuki coupling reaction and the Heck reaction in water. PTFE-RhNPs showed high catalytic activity for the reduction of aromatic rings in water, and the catalyst was reused without any loss of catalytic activity. PTFE was a stable support under the Suzuki coupling reaction conditions because no increase in the size of PdNPs was observed after the reaction. In contrast, the stability of PdNPs on PTFE was not as high under the conditions for the Heck reaction and in the hydrogenation of aromatic rings.

Experimental Section

General Comments

1H n class="Chemical">NMR spectra in CDCl3 were recorded with a 400 MHz NMR spectrometer (JNM-ECZ400, JEOL Ltd., Tokyo, Japan) using tetramethylsilane (δ = 0) as an internal standard. An ICPS-8100 (Shimadzu Co., Kyoto, Japan) was used to perform ICP-AES analysis. A JEM 2100F transmission electron microscope (JEOL Ltd., Tokyo, Japan) was used to confirm the size of metal nanoparticles. The samples were prepared by placing a drop of the solution on carbon-coated copper grids and allowed to dry in air. The reaction was performed using HE-20GB (Koike Precision Instruments, Hyogo, Japan). TSKgel standard PS was purchased from Tosoh Co., Ltd. (Tokyo, Japan). Pd(OAc)2, RhCl3, and PTFE (powder, 1 μm particle size) were obtained from Sigma-Aldrich Co. (Missouri, USA).

Preparation of PTFE-PdNPs

Pd(OAc)2 (8.4 mg, 37.5 μmol), 11.3 mg of n class="Chemical">PTFE (0.113 mmol of TFE unit), 15.3 mg of 4-methylphenylboronic acid (0.11 mmol), and an aqueous KOH solution (1.5 mol/L, 3 mL) were added to a screw-capped vial (no. 1, Maruemu Co., Osaka, Japan) with a stirring bar. After stirring vigorously (1350 rpm) at 90 °C for 5 h, the solid and the aqueous solution were separated by centrifugation. PTFE-stabilized PdNPs (13.3 mg) were obtained after washing with an aqueous KOH solution (1.5 mol/L, 1.0 mL × 5), methanol (1.0 mL × 3), and diethylether (1.0 mL × 5).

Determination of Loading of Pd

PTFE-PdNPs (1.2 mg) were placed in a screw-capped vial (no. 7, Maruemu Co., Osaka, Japan) and then added n class="Chemical">nitric acid (13 M, 5 mL). In case Pd species are difficult to dissolve, the mixture was heated at 80 °C. After dissolving Pd species, the solution was adjusted to 50 mL by distilled water and then measured the amount of Pd metal by ICP-AES analysis (6.40 ppm, 3.11 μmol Pd).

After the catalytic reaction, the aqueous phase was adjusted to 10 mL with hydrochloric acid (1 M), and then, ICP-AES analysis was conducted.

Preparation of PTFE-RhNPs

PTFE (11.3 mg, 0.113 mmol of n class="Chemical">TFE unit), 242 mg of TBAB (0.75 mmol), 4.0 M aqueous solution of RhCl3 (9.38 μL, 37.5 μmol), and water (2 mL) were added to a screw-capped vial (no. 1, Maruemu Co., Osaka, Japan) with a stirring bar. To the mixture which was being stirred vigorously (1350 rpm) at 25 °C was added methanol solution of NaBH4 (0.375 mol/L, 1 mL) dropwise, further stirred for 1 h, and then, the solid and the aqueous solution were separated by centrifugation. PTFE-stabilized RhNPs (14 mg) were obtained after washing with water (1.0 mL × 5), methanol (1.0 mL × 1), and diethylether (1.0 mL × 3). The results of elemental analysis of PTFE-RhNPs are as follows: C (%): 17.35; N (%): 1.39; F (%): 55.37.

Typical Procedures for the Suzuki Coupling Reaction

PTFE-PdNPs (2.0 mg, 1.0 mol % n class="Chemical">Pd based on bromobenzene), 4-methylphenylboronic acid (102 mg, 0.75 mmol), bromobenzene (78.5 mg, 0.5 mmol), and 1.5 mol/L aqueous KOH solution (1 mL) were added to a screw-capped vial (no. 02, Maruemu Co., Osaka, Japan) with a stirring bar. After stirring vigorously (1350 rpm) at 80 °C for 1 h under air, the vial was immersed in water (∼20 °C) for about 10 min to cool the reaction mixture to room temperature. After separating the catalyst and the aqueous solution by centrifugation, the recovered catalyst was washed with H2O (3.0 mL × 5) and diethyl ether (3.0 mL × 5), then dried in vacuo, and reused. The aqueous solution was extracted eight times with diethyl ether. The combined organic extracts were dried over anhydrous MgSO4 and evaporation of the solvent afforded the desired product. Characterization of the product was established by comparison of their 1H NMR spectra with those of authentic samples. Furthermore, the amount of Pd metal in the aqueous phase measured by ICP-AES was <0.1 ppm.

Typical Procedures for the Heck Reaction

PTFE-PdNPs (2.0 mg, 1.0 mol % n class="Chemical">Pd based on iodobenzene), TBAB (161 mg, 0.5 mmol), iodobenzene (102 mg, 0.5 mmol), styrene (78.1 mg, 0.75 mmol), and 1.5 mol/L aqueous KOH solution (1 mL) were added to a screw-capped vial (no. 02, Maruemu Co., Osaka, Japan) with a stirring bar. After stirring vigorously (1350 rpm) at 90 °C for 15 h under air, the vial was immersed in water (∼20 °C) for about 10 min to cool the reaction mixture to room temperature. After separating the catalyst and the aqueous solution by centrifugation, the recovered catalyst was washed with 1.5 mol/L aqueous KOH solution (3.0 mL × 3), diethyl ether (3.0 mL × 5), and methanol (3.0 mL × 3), then dried in vacuo, and reused. The aqueous solution was extracted eight times with diethyl ether. The combined organic extracts were dried over anhydrous MgSO4 and evaporation of the solvent afforded the desired product. Characterization of the product was established by comparison of their 1H NMR spectra with those of authentic samples. Furthermore, the amount of Pd metal in the aqueous phase measured by ICP-AES was <0.1 ppm.

Typical Procedures for the Hydrogenation of n-Octylbenzene

To a test tube with a stirring bar were added PTFE-RhNPs (6.0 mg, 1.5 mol % n class="Chemical">Rh based on n-octylbenzene), n-octylbenzene (190 mg, 1.0 mmol), and water (3 mL). The test tube was then placed into a steel autoclave (TVS-1, Taiatsu Techno Co., Tokyo, Japan). After stirring at room temperature for 15 h under H2 pressure (6 atm), the catalyst and the aqueous solution were separated by centrifugation. The recovered catalyst was washed with diethyl ether (3.0 mL × 3), then dried in vacuo, and reused. The aqueous solution was extracted eight times with diethyl ether. The combined organic extracts were dried over anhydrous MgSO4 and evaporation of the solvent afforded the desired product. Characterization of the product was established by comparison of their 1H NMR spectra with those of authentic samples. Furthermore, the amount of Rh metal in the aqueous phase measured by ICP-AES was <0.1 ppm.