Size-Dependent Catalytic Activity and Fate of Palladium Nanoparticles in Suzuki-Miyaura Coupling Reactions.

Stable, catalytically active palladium nanoparticles of various average diameters (1.9-7.4 nm) have been synthesized and characterized by X-ray diffraction, spectroscopy, and microscopy techniques to demonstrate remarkable size-dependent and renewed catalytic activity toward the Suzuki-Miyaura coupling reaction in green protocol. The catalytic activity is found to depend on the amount of the reducing agent, stabilizer-precursor ratio, solvent composition, and aryl halides used. The product obtained by this reaction is characterized by 1H NMR, 13C NMR, and IR spectroscopy analyses. A newly developed kinetic equation illustrates that while the catalyst particles of the lowest dimension are gradually exposed to the reactants and hence activated due to partial removal of capping polymer from the catalyst surface, others are deactivated due to agglomeration during the progress of the reaction, as conformed by the microscopic profiles of the used and unused catalysts.

Introduction

Nanoparticles play tremendous roles in various fields because of their large surface-to-volume ratio, more energized surface molecules, electron confining ability, capability of tuning the properties of material, providing bridge between bulk material and single molecules, etc.[1−4] On the other hand, transition metals are used as catalysts in a large number of organic reactions, including the carbon–carbon (C–C) coupling reaction using metals like Cu, Cr, Co, Pd, Ni, etc.[5−7] Among these, Pd is special and found to be the metal of the most comprehensive use.

Among different C–C coupling reactions, Suzuki–Miyaura is an efficient one[8−18] because handling and removal of reagents and products are easy compared to those of the corresponding compounds in organometallic reactions.[19−22] Moreover, the reaction is not affected by the number of functional groups present in aryl halides or arylboronic acids. Pd-catalyzed C–C and carbon–heteroatom coupling reactions are performed mostly by expensive, toxic, and air-sensitive P- and/or N-donating ligands like tetrakis(triphenylphosphine)palladium(0), tris(dibenzylideneacetone)dipalladium(0) in organic solvents, and poly(2-oxazoline) palladium carbine complex in water or by in situ reduction of Pd(II) complexes.[23−29] The separation of products from these catalysts is very difficult, which seems to be the major drawback of the method. So, ligand-free and loosely bound nanocatalysts are the focus of this study because these can work in a very small amount at moderate temperature, are easily achievable, and powerful enough to show activity in “green” experimental conditions, as well as in water and aerated solutions.[30−35] In fact, despite routine use of nonaqueous solvents in many organic syntheses, the most abundant, versatile, nontoxic solvent water is now regaining importance for economic, environmental, and safety concerns.[36−42]

The last 2 decades have witnessed extensive use of homogeneous Pd catalysts because of their higher activity and selectivity.[43−46] But considering the contamination of ligand and Pd metal in the products, as well as tedious separation and recycling involved in homogeneous catalysis in large scale and industrial syntheses, microheterogeneous catalysis especially by nanoparticles is now favored.[47−50] In previous studies,[51,52] we observed that strong ligands provide small but catalytically less active nanoparticles, since these ligands cover the nanoparticle surface almost entirely preventing any reactant molecules to approach. Conversely, linear polymers like poly(vinyl alcohol) (PVA), poly(vinylpyrrolidone), sodium polyacrylate, etc. can control the shapes and sizes of the nanoparticles, without affecting their inherent catalytic activity.[53,54] Reportedly, PVA is mainly a steric stabilizer of nanoparticles, although it weakly interacts with metal nanoparticles by its functional group and covers the surface of nanoparticles like ligands. Moreover, polymer-stabilized lyophilic metal nanoparticles can be uniformly dispersed with reactants and products in solvents resembling a homogeneous catalytic system.[55]

Our aim here is to study the Suzuki–Miyaura coupling reaction by using a very low content of synthesized palladium nanoparticles of various diameters under green protocol and also to determine the size-dependent catalytic activity and change of the PVA-protected synthesized nanocatalysts during the progress of the reaction.[56]

Results and Discussion

X-ray Powder Diffraction (PXRD) Analysis

The X-ray powder diffraction (PXRD) (Bruker AXS D8-Advance instrument) pattern of Pd nanoparticles of catalyst-A is shown in Figure . Only two characteristic peaks of carbon-loaded palladium are marked at the 2θ values of 40.01 and 68.06° corresponding to (111) and (220) planes of face-centered cubic (FCC) crystal of Pd (JCPDS FN: PDF#870637). Moreover, two characteristics peaks of graphite carbon are marked at the 2θ values of 36.59 and 44.52° corresponding to (020) and (101) planes of orthorhombic primitive and hexagonal primitive lattices (JCPDS FN: PDF#898491, 898487).

Figure 1

X-ray powder diffraction pattern of Pd(0) nanoparticles of catalyst-A (in the presence of graphite powder).

UV–Visible Absorption Spectra Analysis

The UV–visible spectra of all five sets of nanoparticles show typical characteristic spectra of nanoparticles of Pd metals (Figure ), with obvious absence of surface plasmon peak.[51,57] Each profile shows continuous increase in absorbance at lower wavelength (higher frequency), indicating the formation of nanoparticle with higher band gap. The absorbance values particularly at the UV region, of different sets of global nanoparticles synthesized, vary as A > B > C > D > E, according to the reverse order of average diameter, as presented in Scheme .

Figure 2

UV–visible spectra of 10 times diluted solution of catalyst (A–E).

Scheme 1

Size Control Syntheses of Palladium Nanoparticles

Transmission Electron Microscopy (TEM) Characterization of Pd/PVA Catalysts

The transmission electron microscopy (TEM) images (a–e) and the corresponding histograms (i–v) of almost globular synthesized Pd nanoparticles marked as A–E, respectively, are presented in Figure . The histograms are drawn on the basis of single or multiple micrographs of the same catalyst particles and presented on the right-hand side of the corresponding micrographs. The corresponding average diameter of the particles decreases with increase of the amount of reducing agent and stabilizer used, as expected.[57−59] The high-resolution TEM images of catalysts A, C, and E (Figure ) reveal that d spacing is around 0.232 nm, which corresponds to (111) planes of FCC Pd nanoparticles (PDF#011312).

Figure 3

Transmission electron microscopy (TEM) images (a–e) and the corresponding histograms (i–v) of almost globular synthesized Pd nanoparticles marked as A–E.

Figure 4

High-resolution TEM images of palladium nanoparticles of (a) catalyst-A, (b) catalyst-C, and (c) catalyst-E.

Salient Features of the Suzuki–Miyaura Coupling Reaction

Palladium nanoparticles of smaller diameters are more effective catalysts compare to larger ones, as evidenced from Figure S1 and Tables S1–S5 presented in the Supporting Information. These also act excellently in green solvents compared to hazardous aprotic organic solvents like CH3CN, where the catalytic efficiency of the Pd nanoparticles is decreased plausibly due to agglomeration. Table shows the effect of medium and the leaving group of the aryl halides in the Suzuki–Miyaura reaction. Initially, three pure solvents C2H5OH, H2O, and CH3CN are studied, and it is found that protic organic solvent, ethanol, is the best, as revealed from the percent yield of biphenyl, turn over number (TON), and turn over frequency (TOF). The presence or generation of heterogeneity in the reaction mixture due to less solubility of any component of the reaction mixture or less dispersion of the catalyst seems to be the cause of decreased kinetics in highly polar protic solvent H2O and dipolar aprotic CH3CN. Notably, base K2CO3 and product biphenyl are less soluble in CH3CN and water, respectively. Since the relative stability of the transition state with respect to that of the reactants is required for better kinetics,[60] the reaction has also been studied in 1:1 and 1:3 v/v ratios of aqueous ethanol. It is found that the latter provides much better kinetics than that obtained from pure H2O and C2H5OH. It is found that the TON and TOF values can be improved significantly by reducing the content of the catalyst in the reaction mixture and increasing the time of reaction, as evident in Table . Reaction conducted at room temperature (30 °C) shows 63% yield of biphenyl in 4.5 h (Table ). The percentage of yield and apparent TON are gradually increased with time for each catalyst, indicating the ability of the catalyst to regenerate and continue the reaction (Tables S1–S5). The TOF value decreases with increase of time, indicating decrease in the concentration of the reactants with time (Tables S1–S5). The rate is found to be independent of the concentration of phenylboronic acid, as can be revealed in Table S6. The coupling product obtained from column chromatography of reaction is identified by 1H NMR, 13C NMR, and IR spectroscopy techniques (Figures S2–S4, SD1). Since the rate-determining step of the reaction is the oxidative addition of Pd to ArX with the formation of Ar–Pd–X as intermediate,[61,62] the reaction is of first order with respect to ArX, where X = Cl, Br, I, etc., and the C–X bond energy (presented in parenthesis in kJ mol–1) is in the order C–Cl (338) > C–Br (276) > C–I (238), the yield and apparent TON values are greater for catalyst-A in the reverse order of aryl halides.[63,64] The percentage of yield for biphenyl gradually decreases in the order of the catalysts: A > B > C > D > E. This variation shows exactly the size correspondence,[53] that is, smaller the average diameter of the set of particles, greater the yield of biphenyl compound in the same time interval. However, the apparent turn over frequency (TOF) or rate of formation of biphenyl for each catalyst decreases with time because of decreasing concentration of aryl halide (ArX) and, in most cases, the active sites of catalyst with time, the concentration of phenylboronic acid being taken in excess. Since the concentration of the catalyst remains same throughout the reaction, the apparent first-order rate constants (k1) are determined for iodobenzene for all of the catalysts, and the data are presented in Table , as obtained from the initial linear portion of the profiles presented in Figure S5. The % yield and the rate constants are found to decrease with increase in the diameter of the catalyst. To get an insight into the fate of the catalyst during reaction, the rate of change of the activation of the catalyst (b) or the deactivation of catalyst during reaction must be considered. Thus, the effective concentration of the catalyst at any time t after onset of the reaction is given bywhere C0 is the initial concentration of the catalyst in the solution at time t = 0. Notably, b might be positive or negative depending on the activation or deactivation of the catalyst.

Table 1

Results of the Suzuki–Miyaura Coupling Reaction of Aryl Halide with Phenylboronic Acid at 60 °C by Using Catalyst-A in Different Solventsa

entry	aryl halide	solvent	time (min)	Pd (mol %)	yieldb (%)	TON	TOF (min–1)
1	iodobenzene	C2H5OH	30	0.2	44	220	7.3
2	iodobenzene	H2O	30	0.2	28	140	4.7
3	iodobenzene	CH3CN	30	0.2	10	50	1.7
4	iodobenzene	C2H5OH/H2Oc	30	0.2	36	180	6.0
5	iodobenzene	C2H5OH/H2Od	30	0.2	52	260	8.7
6	iodobenzene	C2H5OH/H2Od	90	0.01	50	5000	55.6
7	iodobenzene	C2H5OH/H2Od	150	0.01	79	7900	52.7
8	iodobenzene	C2H5OH/H2Od	60	0.2	78	390	6.5
9	bromobenzene	C2H5OH/H2Od	60	0.2	47	235	3.9
10	chlorobenzene	C2H5OH/H2Od	60	0.2	32	160	2.7

Reaction conditions: 1.0 mmol aryl halide, 1.3 mmol phenylboronic acid, 2.0 mmol anhydrous K2CO3, 5 mL of solvent, and temperature 60 °C.

Isolated yield.

C2H5OH/H2O (1:1) (v/v).

C2H5OH/H2O (3:1) (v/v).

Table 2

Results of the Suzuki–Miyaura Coupling Reaction of Iodobenzene with Phenylboronic Acid at Room Temperature (30 °C) in 75% EtOH–H2O by (0.2 mol %) Pd(0) Nanoparticles (Catalyst-A)

entry	time (h)	yielda (%)	TON	TOF (min–1)
1	1.0	19	095	1.6
2	1.5	27	135	1.5
3	2.5	42	210	1.4
4	4.5	63	315	1.2

Isolated yields.

Table 3

Kinetic Parameters Exhibiting Different Catalytic Activities of the Sets of Pd Nanoparticles Synthesized, Designated by A–E in the Suzuki–Miyaura Coupling Reaction of Iodobenzene with Phenylboronic Acida

catalysts	diameter of Pd nanoparticles (nm)	yieldb (%)	rate constant (k1) (min–1)
A	1.9	95	0.0348
B	3.4	49	0.0059
C	4.2	38	0.0036
D	6.4	33	0.0034
E	7.4	26	0.0025

Reaction conditions: iodobenzene (1 mmol), phenylboronic acid (1.3 mmol), anhydrous K2CO3 (2 mmol), Pd catalyst (0.2 mol %), C2H5OH/H2O (5 mL) (3:1) (v/v), reaction time 1.5 h, and temperature 60 °C.

Isolated yield.

The differential form of the rate equation may be written aswhere k2 is the second-order rate constant, independent of catalyst concentration, “a” is the initial concentration of the reactant, “x” is the concentration of the product that increases with time t, and k1 is the first-order rate constant dependent on catalyst concentration in the reaction. Therefore, the integrated form of rate equation becomes (SD2)So, k2 and b are obtained from the intercepts and slopes of the plots of versus time (t) (Figure S6). The values of k2 and b for different catalysts are presented in Table .

Table 4

Intercept and Slope of the Plots of (1/t) ln(a/(a – x)) versus t Following Equation and the Corresponding Rate Constant and Rate of Activation/Deactivation of Catalysta

	catalysts
parameters	A	B	C	D	E
[intercept (C0k2)] × 103	19.43	8.75	5.99	5.32	3.70
	1.51	–0.11	–0.09	–0.11	–0.04
[second-order rate constant (k2)] × 103	65.7	29.6	20.2	18.0	12.5
[rate of activation of catalyst (b)] × 103	4.60	–0.72	–0.86	–1.19	–0.70
Adj. R2	0.97	0.98	0.98	0.96	0.96

Initial concentration of the catalyst in the solution (C0) is 0.296 mM.

Table shows that the rate constant (k2) of the Suzuki–Miyaura reaction increases 5 (65.7/12.5) times, whereas the average particle diameter (Table ) decreases and the increment in total surface area is ca. 4 times (7.4/1.9) for changing the catalyst from A to E. This indicates that greater catalytic activity of catalyst A is not only due to increased surface area but also due to the presence of more active surface molecules in A compared to other catalysts, B–E during the progress of the reaction. The positive sign of b in catalyst A indicates more activation of the catalyst, whereas the negative sign might be stated as effective deactivation for catalysts, B–E. The activation found for catalyst-A might be due to the removal of PVA from the part of surface of the nanoparticles and associated disintegration causing decrease in the diameter of the particles during reaction at the time of removal of PVA. From B to D, the deactivation increases plausibly because of immediate growth of particles. For catalyst E, the effective deactivation was not large seemingly because the average diameter of particles was initially very large. The values reveal that the second-order rate constants gradually increase with decrease in the diameter of the particles, indicating greater binding ability of the particles with ArX. The activation parameter b reveals that it decreases with increase in diameter. The set of nanoparticles with greater average diameter has relatively less number of active sites in the reaction. So, the progress of the reaction cannot make their fate much worse in terms of the number of active sites.

By comparing the value of rate constant of the Suzuki–Miyaura coupling reaction with iodobenzene and phenylboronic acid using 0.2 mol % Pd nanoparticles (catalyst-A) at two different temperatures (30 and 60 °C) (Tables and S1), the activation energy of the reaction was found to be 60.6 kJ mol–1.

Recyclability of the Catalyst

To study the level of reusability for practical applications, the best- and worst-performing catalysts A and E were selected for the recycling in the Suzuki–Miyaura reaction (Table ).[65−67] The result shows that the % yield in 1.5 h increases slightly in the third cycle in comparison to that in the first cycle for catalyst-A, whereas it decreases gradually for catalyst-E with increase of the cycle number. This indicates that catalyst-A can be used for a greater number of times in recycling of catalyst than catalyst-E.

Table 5

Recyclabilities of Catalysts A and E in the Suzuki–Miyaura Coupling Reaction of Iodobenzene and Phenylboronic Acid at 60 °C by Using by 0.2 mol % Pd(0) Nanoparticlesa

	no. of recycle with yieldb (%)
catalysts	1st	2nd	3rd
A	95	96	96
E	26	22	15

Reaction conditions: iodobenzene (1 mmol), phenylboronic acid (1.3 mmol), anhydrous K2CO3 (2 mmol), Pd catalyst (0.2 mol %), C2H5OH/H2O (5 mL) (3:1) (v/v), reaction time 1.5 h, and temperature 60 °C.

Isolated yield.

The scanning electron microscopy (SEM) images of catalyst-A before and after single run (use) are depicted in Figure a,b, respectively. In Figure a, it is evident that the globular nanoparticles with average diameter of 20.1 nm are mostly present in PVA matrix, whereas the particles are mostly distinct and uniformly distributed in the absence of such matrix, as evident in Figure b. The average diameter of the particles of used catalyst-A is 8.3 nm, less than that obtained before use of the catalyst. This type of removal of PVA from the surface of the catalyst with an effect of increased rate of reaction (current density) in cyclic voltammetric experiment was observed in our previous study.[60] For catalyst-E, the average diameter of the Pd nanoparticles is 27.4 nm before use of the catalyst (Figure c). It is decreased to 15.2 nm after the first run due to the removal of PVA during reaction (Figure d). Pd nanoparticles are initially stabilized (wrapped) by PVA for both catalysts A and E, but for catalyst A to a greater extent than catalyst E because of the higher surface energy of the former. Due to approach of reactant molecules toward Pd nanoparticle surface for undergoing reaction, PVA is removed from certain relevant portion of the surface of nanoparticle and covers the other nearby part of the nanoparticle to a greater extent, facilitating disintegration of nanoparticles. Thus, the particle size is reduced in both cases at the initial state of the reaction. However, the particles with greater size, containing less stabilizing layer initially, would agglomerate first with the progress of the reaction. Thus, the particles of catalyst-E quickly agglomerate to a sufficient extent to form several flakes reducing the number of active sites during reaction, as evident from Figure d.

Figure 5

SEM images of palladium nanoparticles of catalyst A and E before use (a, c) and after single run (1.5 h) (b, d). The histograms are presented on the right-hand side of the corresponding images.

Conclusions

Sets of PVA-encapsulated Pd nanoparticles of average diameter 1.9–7.4 nm have been synthesized and used in size-dependent catalysis of the Suzuki–Miyaura reaction. The catalyst shows a high efficiency in 75% (v/v) ethanol–water, which seems to be the best medium of optimum composition in ethanol–water binary solvent system. Besides evaluation of the effect of the leaving group and catalytic parameters, a new kinetic equation is proposed to study the catalyst activation and deactivation during the progress of reaction. It is understood from the new analysis strategy that the effective catalyst concentration is changed with the progress of the reaction. For catalyst with the least average diameter (1.9 nm), PVA encapsulated is partly removed during the reaction, causing disintegration, so effective catalyst concentration is increased, whereas for other catalysts synthesized and studied, it is decreased due to agglomeration in the absence of a sufficient stabilizing layer. Microscopic studies of the catalysts before and after use conform to the findings.

Experimental Section

General Information

The UV–visible spectra of the as-synthesized particles were recorded using a UV-1800 Shimadzu UV spectrophotometer. NMR spectra of the product of the Suzuki–Miyaura coupling reaction were obtained using a Bruker Advance 300 MHz spectrometer equipped with a high-resolution multinuclear probe in CDCl3. X-ray powder diffraction (PXRD) patterns of carbon powder-adsorbed Pd catalyst were recorded on a Bruker AXS D8-Advance instrument with Ni-filtered Cu Kα radiation (λ = 1.5418 Å) at 40 kV and 40 mA, employing a scanning rate of 0.2 s per step. SEM measurements were performed under vacuum by INSPECT F 50, using 20.00 kV. Transmission electron microscopy (TEM) images were recorded with an H7600 TEM, HITACHI instrument. IR spectra were recorded using NICOLET MAGNA IR 750.

Materials

PdCl2 with a purity of 99% was purchased from Arora-Matthey Ltd; water from Synergy of Millipore; NaBH4, phenylboronic acid, iodobenzene, bromobenzene, and chlorobenzene from Spectrochem Pvt. Ltd; and PVA (number-average molecular weight: 1 25 000) from Lab Rasayan Co. The solvent and other reagents were of AR/GR grade from Merck, India.

Size Control Syntheses of Palladium Nanoparticles

The typical preparation of catalyst-A is described as follows: 0.5 mL of 0.05636 (M) K2PdCl4 aqueous solution was taken in a 100 mL beaker along with 24.5 mL of 1 mass % (mass/volume) PVA solution. The whole solution was continuously stirred with addition of 3.19 mg (8.454 × 10–5 mol) of NaBH4 for 10 min. The yellow color of Pd(II) solution changed to dark brown at room temperature (30 °C), indicating the formation of palladium nanoparticles. For the size control syntheses of Pd/PVA catalysts designated as A, B, and C, NaBH4 was varied in the ratio of 6:3:2 keeping fixed concentration of K2PdCl4 and PVA, as presented in Scheme . In the preparation of catalyst samples A, D, and E, the PVA solution was varied in the ratio of 24.5:5:1 keeping a fixed concentration of NaBH4 and K2PdCl4 at the start of the reaction.[68] These particles were then characterized by using various techniques.

Palladium Nanoparticles-Catalyzed Suzuki–Miyaura Coupling Reaction

The Suzuki–Miyaura coupling reaction (Scheme S1 in Supporting Information) was carried out in a closed 25 mL round-bottom flask by taking 1 × 10–3 mol aryl halide, 1.3 × 10–3 mol phenylboronic acid, 2 × 10–3 mol anhydrous K2CO3, and nanocatalyst containing 2 × 10–6 mol of Pd atom at room temperature (30 °C) and 60 °C at different intervals for a definite period of time. Afterward, biphenyl was separated by extraction with diethyl ether (thrice with 15 mL of diethyl ether) dried by addition of anhydrous sodium sulfate and analyzed by column chromatography.[69] In column chromatography, a slurry of silica gel in petroleum ether (60–80 °C) was taken as stationary phase and 5% ethyl acetate–petroleum ether as mobile phase at a rate of flow of 0.1 mL min–1, during the separation and purification. The product (biphenyl) was identified by determination of melting point and 1H NMR spectroscopy analysis at every stage. During the course of reaction at 60 °C, it was observed that the whole reaction mixture becomes pale brownish throughout the solution. After a definite time, 15 mL of ice-cold brine solution was added and the whole solution becomes white, with some white shining crystal deposits at the bottom of the flask. The kinetic analyses of catalysts A–E were done by estimating the amounts of biphenyl product at different time intervals through column chromatography.

The recycling process was performed with catalysts A and E, at 60 °C by separation of biphenyl from the reaction mixture and by extraction with diethyl ether (three times) without addition of brine solution. Then, the catalyst was separated by centrifuge, dried, redispersed, and used in the next cycle.