Palladium Nanoparticles Tethered in Amine-Functionalized Hypercrosslinked Organic Tubes as an Efficient Catalyst for Suzuki Coupling in Water

Abstract

It is highly desirable to design functionalized supports in heterogeneous catalysis regarding the stabilization of active sites. Pd immobilization in porous polymers and henceforth its application is a rapidly growing field. In virtue of its’ scalable synthesis and high stability in reaction conditions, amorphous polymers are considered an excellent scaffold for metal mediated catalysis, but the majority of them are found as either agglomerated particles or composed of rough spheres. Owing to several important applications of hollow organic tubes in diverse research areas, we aimed to utilize them as support for the immobilization of Pd nanoparticles. Pd immobilization in nanoporous polymer tubes shows high activity in Suzuki cross coupling reactions between aryl halides and sodium phenyl trihydroxyborate in water, which deserves environmental merit.

1. Introduction

Porous organic polymers (POPs) are emerging as next-generation support materials for heterogeneous catalysis [1,2,3,4,5,6]. Cheap and readily available organic precursors, tailorable functionality arising from diverse building blocks, are generally advantageous for making high surface area POPs, which are not only used as “support” for metal-mediated catalysis, but also possess significant applications in diverse research, owing to the advantages of densely packed organic groups [7,8,9]. Therefore, it is customary to mention that the inherent advantages of having organic units in POPs is tremendous, as a support for metal complexes/nanoparticles or as a catalyst because of the virtue of having an electronic interaction between the organic units in POPs and metal nanoparticles. In fact, the main advantage of POPs with its competitive porous support viz. metal organic frameworks (MOFs) [10] and periodic mesoporous organosilica (PMOs) [11,12] is its stability in drastic reaction conditions, which has been immensely highlighted as POPs have shown usability in water medium for catalysis.

Considering adverse environmental impact, using volatile organic solvent for catalysis is a serious concern, which should be replaced by water as an environmentally more demanding. However, reactions using water as the only solvent is frequently encountered as fatal because of the moisture sensitive organic precursors, catalysts, and intermediates. Therefore, it is highly challenging to develop water-compatible catalysts that could be stable, active, and reusable for a number of times without being deteriorated [13,14]. In this context, a Suzuki–Miyaura cross coupling reaction in water as an eco-friendly solvent is meritorious because of its high stability and good solubility of phenyl boronic acid/salt [15]. Pd-catalyzed Suzuki–Miyaura cross coupling reactions have significant importance in organic chemistry, pharmaceuticals research, drug discovery, and development as an elegant tool for C–C bond formation reactions, mainly because of the wide availability of starting materials and relatively mild reaction conditions [16,17]. Although a majority of research has been done in homogeneous conditions using either organic/water as solvent, it suffers from the formation of Pd black as an inactive catalytic species. Again, the use of triphenyl phosphine for stabilization of Pd intermediates under homogeneous conditions often encounters toxicity/poison [18]. In this regard, high surface area heterogeneous catalysts possess tremendous applications in fine chemical industries, owing to its repetitive usability, tailorability of surface modification for stabilization of active sites, and so on. Pd-grafted heterogeneous catalysts comprising porous silica, zeolite, MOFs, and polymers are hereby reported as solid-supported catalysts for Suzuki reaction, but few show considerable recyclability and stability in a water medium [19,20,21]. On the other hand, microporous POPs are generally formed through precipitation polymerization as agglomerated solid particles/irregular flakes, instead of any well defined nanostructure [22]. Nevertheless, POPs having uniform morphology of hollow tubes/fibers are still scarce; moreover, their application is merely limited to device manufacturing [23]. Recently, Modak and Bhaumik reported interesting microporous polymer tubes (PP-1, PP-2, PP-3; PPs) that are uniform and show high heterogeneity for one-pot tandem catalysis [24]. However, PPs could also be interesting as support for metal nanoparticles’ owing to their amine functionality; therefore, we investigated its activity for Suzuki–Miyaura cross coupling reactions. The catalysis in water might be advantageous, since the hydrophobic 4-tritylaniline-based tubes are stable and prevent the active sites from agglomeration and inactivation. Hopefully, this research can provide tremendous scientific interest in the utilization of highly functionalized porous organic tubes/fibers as support of nanoparticles.

2. Results and Discussion

2.1. Synthesis and Characterization of Pd/PP-3 Tubes

The amine-functionalized hypercrosslinked polymer (PP-3) is formed through a one-pot polymerization condensation using 4-Tritylaniline as starting precursor, dimethoxymethane (DMM) as linker and FeCl₃ as catalyst/mediator (Scheme 1).

The resulting light brown precipitate shows a unique hollow tube shaped morphology as shown in both scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images (Figure 1). Unlike other tube-shaped porous polymers, synthesis of PP-3 is performed in relatively mild conditions utilizing FeCl₃ as a cheap and non-harmful chemical [23,25]. The size of the organic tubes are found to be ~5–7 μm in length, and the inner hollow diameter is ~80–100 nm, together with ~300–350 nm is the wall thickness. This thickness accounts for excessive polymerization/non-covalent interaction, which is due to the addition of a large quantity of DMM linker during the synthesis [24]. Pd immobilization to PP-3 was performed by loading with Pd(OAc)₂, followed by a reduction with aqueous NaBH₄ under mild conditions. It was observed in TEM images (Figure 2) that Pd nanoparticles with dimensions of 5–9 nm were distributed throughout PP-3, which is possibly because of the stabilization by built-in amine sites (Figure 2) [26,27,28,29].

Size distribution of Pd nanoparticles is shown in the inset of Figure 2, suggesting a broad distribution pattern (3–9 nm). Further characterization of Pd/PP-3 was achieved through powder X-ray diffraction (XRD), N₂-sorption, X-ray photoelectron spectroscopy (XPS), and inductively coupled plasma atomic emission spectroscopy (ICP) analysis and provides good justification for the presence of Pd nanoparticles in our hypercrosslinked PP-3 tubes.

PXRD of Pd/PP-3 is given in Figure 3a, which shows sharp diffraction at the 40.2°, 43.9°, and 47.3° regions, corresponding to different facets of Pd crystal particles. In comparison with Pd/PP-3, only PP-3 shows a broad peak because of the presence of an amorphous pore wall [30]. Porous properties of Pd/PP-3 were measured from the N₂ sorption isotherm, as shown in Figure 3b, which preferentially suggests Type I characteristics of the isotherm. Like other microporous materials, a high uptake at a low P/P₀ is observed, followed by a flat extrapolation at 0.2–0.8 P/P₀, along with a step uptake at 0.9 P/P₀. The increase of the isotherm at 0.9 P/P₀ depicts the inter-particle mesoporosity. The Brunauer–Emmett–Teller (BET) surface area of Pd/PP-3 was calculated to be 420 m²·g⁻¹, lower than 530 m²·g⁻¹ for PP-3 tubes, which is basically due to pore blocking by Pd nanoparticles. In Figure 4a, we provide a survey XPS spectrum of Pd/PP-3, which shows the presence of N, O, and Pd. Figure 4b shows the XPS of Pd 3d electrons, which indicates that Pd sites were reduced to a metallic state in PP-3 with the presence of NaBH₄. The confirmation of Pd(0) has been characterized at 334.4–334.8 eV and 339.8–340.2 eV, which were assigned to the Pd 3d_5/2 and Pd 3d_3/2 electrons, respectively [31,32]. Traces of a PdO peak at 341 eV have been observed [33]. All these results, however, demonstrate almost a complete conversion of Pd(OAc)₂ into Pd nanoparticles with the utilization of hollow PP-3 tubes.

2.2. Heterogeneous Catalysis for Suzuki–Miyaura Cross Coupling Reaction in Water

Based on this perspective, the benchmark Suzuki reaction between bromobenzene and sodium phenyltrihydroxyborate was tested in water as a solvent at 100 °C. Because of the better solubility of sodium phenyltrihydroxyborate (PHB) in water compared with phenylboronic acid, the use of PHB for Suzuki coupling without the aid of an additional base was considered advantageous. Owing to the easy preparation and highly stable PHB as organoboron salt, Pd-mediated C–C bond formation reactions are quite meritorious. [34]. Initially, we investigated the scope of Suzuki catalysis in several solvents such as toluene, dicholomethane, dioxane, water, and dimethylformamide (DMF) (

Table 1. Optimization in reaction conditions for Suzuki coupling reactions between 1 equiv bromobenzene and 1.2 equiv sodium phenyltrihydorxyborate, catalyzed by Pd/PP-3 support.

Entry	Solvent	Temperature/°C	Time/h	Yield a/%
1	Toluene	100	24	Traces
2	1,4-dioxane	20	24	35
3	DMF	100	24	>90
4	Dichloromethane	20	24	<10
5	Acetone	20	24	50
6	Water/DMF	25	24	>90
7	Water b	100	24	60
8	Water c	100	24	>90
9	Water d	25	24	50

^a Yield refers to isolated products after purification; ^b Without using TBAB; ^{c, d} 1 equiv of TBAB.

). We observed that the reaction was very sluggish in non-polar, aprotic solvents such as toluene and dicholomethane, partly successful in 1,4-dioxane, and takes place quite efficiently in DMF and DMF/water mixtures (Table 1). Reaction in pure water shows only a 60% yield of biphenyl at 24 h, which considerably improves to >90% at 24 h upon the addition of tetrabutylammonium bromide (TBAB) as phase transfer catalyst.

Using TBAB during the reaction is essential in order to increase the solubility of organic precursors in water. However, upon the addition of TBAB, the reaction can also take place at room temperature conditions, as shown in Table 1. In this regard, we show a temperature-dependent conversion of bromobenzene to biphenyl in Figure 5a, with or without TBAB. In the case of adding TBAB, our model reaction shows much faster kinetics than the reaction devoid of TBAB, demonstrating that phase transfer catalyst is indeed essential for reactions in water. Furthermore, the optimum loading of Pd(OAc)₂ and consequently the generation of Pd nanoparticles in PP-3 influences catalytic activity, and we observe the highest catalytic activity/reaction rate (12.5 h⁻¹) with 2 wt % Pd/PP-3, which shows the highest yield of biphenyl from bromobenzene (Figure 5b).

It is worthy to mention that the reaction was carried out in aerial conditions without using degassed water/co-solvent, which could partially lead to the formation of agglomerated Pd/Pd black, as this phenomenon is quite common with several known palladacycles [35]. Therefore, the problem of using Pd/PP-3 in recycling studies was outperformed when choosing degassed water under N₂ prior to adding substrates and catalysts, which worked out well in our study.

Next, the efficiency of Pd/PP-3 was tested by encountering a broad substrate scope ranging from electron rich to electron poor aromatic halides, as given in

Table 2. Substrate scope of Suzuki coupling reaction by Pd/PP-3.^a

Entry	Ar-X	Time/h	% Conv. of Ar-X b	% Selectivity	Rate c/h−1
1	Iodo benzene	20	90	>99	12.5
2	Bromobenzene	24	86	>99	11
3	Chlorobenzene	24	48	>80	6
4	4-bromo nitrobenzene	18	88	>95	13.8
5	4-chloro nitrobenzene	24	40	>78	8
6	4-bromo benzaldehyde	24	75	-	5.3
7	4-bromo acetophenone	20	76	>80	12.5
8	4-iodo acetophenone	15	88	>98	16.6
9	4-iodo anisole	19	85	>98	13
10	4-bromo anisole	24	79	>90	10
11	4-bromo benzonitrile	24	70	>90	9.2
12	4-chloro benzonitrile	24	30	>70	4.7
13	4-bromo toluene	20	70	>85	11
14	4-bromobenzoic acid	18	60	>80	12.8
15	Brombenzene d, e	24	70 d, 55 e	>90 d, >86 e	-

^a 1 mmol aryl halide, 1.2 mmol sodium phenyltrihydroxyborate, 5 mL water, 100 °C, 0.02 g Pd/PP-3 (2 wt % Pd); ^b Yield refer to the isolated product; ^c Reaction rate (mol product per mol of total Pd per time) at 10 min; ^d Reaction was carried out with Pd/MCM-41; ^e Pd/C catalyzed reaction.

. Aromatic iodo and/or bromo compounds worked efficiently for the formation of substituted biphenyl; however, for chloro derivates, a partially sluggish reaction was observed. Nonetheless, there are many reports for heterogeneous catalysts such as Suzuki coupling reactions, which can work efficiently with expensive aromatic iodo/bromo derivatives, [36], but very few catalysts can activate aromatic chloro compounds, which are cheap and have more economically feasible applications [31].

It is worthy to mention that the Suzuki coupling reaction of aromatic chloride by Pd/PP-3 showed almost a 40% yield of biphenyl in water, which indicates that Pd/PP-3 also has a potential for utilizing much cheaper chloroaromatics for making biphenyl in a cost-effective way. All products of the Suzuki reaction, given in Table 2, were characterized by ¹H NMR and ¹³C NMR (see Supplementary Materials). It is pertinent to mention that the model Suzuki reaction, when compared with Pd/MCM-41 (2 wt %·Pd; 5–8 nm) and Pd/C (2 wt % Pd; 10–15 nm) in water, shows much lower activity than Pd/PP-3, signifying the importance of amine functionality decorated in hydrophobic PP-3 for the stabilization of active sites, as well as organometallic intermediates. Similarly, Pd nanoparticles stabilized by phenol resign also show comparable catalytic activity as that of Pd/PP-3, which again proves that the surface functional sites could endow a significant binding interaction with guest metal particles for stabilization and improvement in its catalytic property [37,38].

In Figure 5c,d, we present the kinetic aspect of the Pd/PP-3 catalyzed Suzuki reaction, which shows that the reaction is devoid of any long induction time. The reaction rate shows first order dependency with respect to bromobenzene; the rate constant is calculated to be k_obs = 0.081 ± 0.006 h⁻¹, and the half life period, t_1/2, is 6.93 h. The first order reaction rate can be explained on the basis of an initial oxidative addition by bromobenzene with surface-exposed Pd nanoparticles in a slow step, accompanied by rapid elimination of the cross coupled product.

Finally, the heterogeneity of Pd/PP-3 was proved through a hot filtration test. In this regard, we initially separate the catalyst after the 10 h reaction is over. The hot filtrate without any Pd/PP-3 was then investigated for the Suzuki reaction, which did not produce any cross coupling products, signifying that the filtrate is devoid of any Pd-based impurity (Figure 6a). Again, ICP measurements of the filtrate solution did not detect any Pd content, i.e., the Pd amount in the filtrate is beyond the scope of its detection limit. All these results clearly suggest that Pd sites in Pd/PP-3 are stable for liquid-phase catalytic reactions.

Furthermore, we studied poisoning experiments in order to check if the reaction is still catalyzed by leached Pd nanoparticles or not. In this regard, we added 2–3 drops of metallic Hg at the middle of the reaction, and the addition of Hg hardly affects the rate as well as the overall yield of final product, demonstrating that Pd/PP-3 has good heterogeneous characteristics for successive reactions.

Later, while investigating the recycling experiments, we found that Pd/PP-3 retains its catalytic activity for five cycles without much deterioration in its activity (Figure 6b). All these results, however, demonstrate that Pd/PP-3 is a stable heterogeneous catalyst for long-term applications. Since Pd/PP-3 was used to repeat Suzuki coupling experiments in a water medium, there might be another possibility of damage in either morphology or active Pd sites. In this regard, we investigated the stability of reused Pd/PP-3 through TEM and XPS analysis. XPS investigation of the 5th reused Pd/PP-3 shows marginal change in Pd 3d electrons, possibly because of the partial oxidation or continuous exposure in water (Figure 7). On the other hand, TEM analysis of the 3rd reused Pd/PP-3 suggests no collapse of hollow tube morphology and almost no change in the Pd size distribution (3–6 nm particles), as shown in Figure 8.

3. Materials and Methods

3.1. Instrumentation

Porosity was measured at 77 K using a Quantachrome Instrument (Quantachrome Instrument; Boynton Beach, FL, USA), Autosorb-1, where all samples was degassed at 100 °C for 4 h before the measurement. The Brauner–Emmett–Teller (BET) surface area was calculated over the entire pressure region from ~0.05 to ~0.18 P/P₀. Transmission electron microscopy (TEM) was obtained from Hitachi HT-7700 (Nishi-shimbashi, Minato-Ku, Tokyo, Japan) with an acceleration voltage at 100 kV after the samples were dispersed in ethanol via sonication and placed onto an ultrathin carbon film supported on a copper grid. Powder X-ray diffraction (PXRD) was performed with a Rigaku D/Max2500 PC diffractometer with CuKα radiation (λ = 1.5418 Å) over the 2θ range of 5°–70° at a scan speed of 5° per min at room temperature. A Bruker DPX-300 NMR spectrometer was used to measure the ¹H and ¹³C NMR of catalytic products in a liquid state. X-ray photoelectron spectroscopy (XPS) was recorded on a VG ESCALAB MK2 apparatus using AlKα (hν = 1486.6 eV) as the excitation light source. Pd content was determined via PLASAM-SPEC-II inductively coupled plasma atomic emission spectrometry (ICP). Particle sizes were determined from Nano Measurer 1.2 software, 2008 by Jie Xu, Fudan University, China.

3.2. Methods

Synthesis of PP-3 was followed in accordance with a previously reported procedure [24]. For the preparation of Pd/PP-3, we initially dissolved 0.010 g of palladium acetate in a 20 mL glass vial containing 10 mL of distilled water and stirred until the solution became yellow. Next, 0.05 g of PP-3 was mixed, and the solution was stirred for 7 h. The mixture was centrifuged several times and washed with H₂O and ethanol followed by drying at 60 °C for 12 h, which was denoted as Pd/PP-3. Pd loading was found to be ~2 wt %, as confirmed by ICP analysis.

3.3. General Procedure for Suzuki–Miyaura Coupling Reaction

In the typical synthesis condition, a mixture of Arylhalide (1 mmol), sodium phenyltrihydroxyborate (1.2 mmol), 0.02 g Pd/PP-3 (2 wt % Pd), and TBAB (1 mmol) was stirred in water (5 mL). The mixture was refluxed with magnetic stirring for several hours, as shown in Table 2. After the reaction was complete (monitored by thin layer chromatography technique), the mixture was filtered to separate the catalyst, and the filtrate was subjected to extract (10–20 min) with diethyl ether (20 mL). The combined organic layers were then washed with brine (10 mL), dried by anhydrous Na₂SO₄, and evaporated. The residue was purified on a short column of silica using petroleum ether as the eluent to afford the desired substituted biphenyl as pure product.

4. Conclusions

Herein, we report on a Suzuki–Miyaura cross coupling reaction in water with Pd-grafted PP-3 as a porous polymer support, which is thought to be promising and environmentally appealing. High catalytic activity, good stability, and reusability of Pd/PP-3 essentially signify its advantages as a heterogeneous catalyst for the liquid phase synthesis of fine chemicals. Moreover, the hollow tube geometry of PP-3 is suitable for anchoring Pd sites by exploiting both the inner and outer hollow spaces, providing an enormous stabilization of Pd and preventing the formation of inactive Pd clusters. Owing to the importance of particle morphologies in solid catalytic research, we believe our efforts could motivate others’ for developing organic nanotubes/carbon tubes and utilization of its hollow space for developing nanoreactors, which could ultimately lead to a sustainable and environmentally benign solid catalysis research.