Polyamino amphiphile mediated support of platinum nanoparticles on polyHIPE as an over 1500-time recyclable catalyst

Abstract

For supported metal nanoparticles, the ligand/support is crucial to their catalytic activity, stability and recyclability. Here we show that a polyamino dendritic amphiphile is a versatile mediator for supported platinum nanoparticles (PtNPs). PEI@PS-EHA, a branched polyethylenimine (PEI) with a shell of poly(styrene-co-2-ethylhexyl acrylate) (PS-EHA), will self-assemble along the interface of a water-in-oil (W/O) emulsion, and stabilize a corresponding high internal phase emulsion (HIPE); after transformation of the HIPE into an open-cellular material termed polyHIPE, PEI@PS-EHA will act as a surface modifier to immobilize PtNPs. The supported catalyst is very fit for reduction reactions because the Pt species hardly leaches in a reductive environment (judged just from the leaching data, the catalyst can be recycled over 10 million times). No ripening of the supported PtNPs is found for over 1500 equivalent cycles, as judged from the apparent rate constant. The ready separation, high activity, extremely low metal leakage and high durability make the catalyst promising.

---

1. Introduction

Noble metal nanoparticles (MNPs) are representative catalysts widely used in chemical synthesis but their recycling, leakage and degradation remain a central concern.^1–4 The leakage leads to increased production costs, deteriorates the product quality, and exerts a negative impact on the eco-environment. Supporting MNPs can facilitate separation and recycling. An ideal support needs to feature several aspects such as large surface area, mechanical robustness, good durability and strong ligation of MNPs. Elastic foam support,⁵ N-doped carbon,^6–8 organo-silica,⁹ cellulose,¹⁰ graphene oxide and its derivatives^11–13 are popular supports, where some also serve as ligands. High internal phase emulsion (HIPE) can be transformed into a material termed polyHIPE^14,15 and the latter represents a kind of soft material with large interconnecting pores. PolyHIPE is typically characterized with several features: (1) dense and interconnected pores throughout the 3D structure, with diameters ranging from scores of nanometers to hundreds microns, which are beneficial to mass transfer; (2) a moderately high surface area, typically at 10¹ to 10² m² g⁻¹; (3) large-scale production in a one-pot manner, typically employing water droplets as a soft template; (4) polyHIPE technology can lead to macroscopic assembly of amphiphiles. However, surface engineering of a polyHIPE is usually a tedious process because most stabilizers cannot simultaneously act as surface modifiers as well as stabilizers. Recently, amphiphilic block copolymers¹⁶ and reverse-micelle-like dendritic amphiphiles¹⁷ have been reported to play dual roles as a stabilizer and a surface modifier, leading to one-pot preparation of surface-functionalized polyHIPE. Moreover, dendritic amphiphiles with impregnated metal nanoparticles can also serve as a HIPE stabilizer, directly leading to metal nanoparticles-decorated polyHIPE.^18,19 The unique feature of a polyHIPE support is its availability at any size, which favours ready separation.

The ligand which couples a support and MNPs is crucial to the stability and catalytic activity of the latter. A strong ligand can stabilize a MNP but may suppress its catalytic activity; while a weak ligand favours an active MNP but the latter may suffer from rapid ripening and leaching. In this aspect, a multiligand usually can efficiently stabilize a MNP while allow the latter to be catalytically active. Representative multiligands are dendrimers such as poly(amidoamine),^20–24 and derivatives of hyperbranched polymers (also termed dendritic polymers) such as polyethylenimine (PEI)-based derivatives.^25–28 However, dendritic molecules along stabilized MNP nanocomposites are too small to readily separate. The leaching and ripening of MNP are also influenced by the ligand. The leaching mechanism and catalytic mechanism of metal species are different one from another. For example, palladium nanoparticles mainly leach via atomic Pd and the catalytic centre is usually the atomic Pd;^20–22 while PtNPs mainly leach via ionic Pt and the catalytic centre is usually PtNP.^29,30 The commercially available PEI and supported PEI are becoming popular multiligands for PdNP³¹ and PtNP.^{18,19,25–28} To now, active and highly stable MNPs, supported on macroscopic, robust and porous matrix, are less reported, especially those with extremely low leakage and recyclable for hundreds times. Recently, we showed that a PEI-based dendritic amphiphile can simultaneously act as a HIPE stabilizer, a PtNP multiligand and a coupling agent to link a PtNP on a support, rendering one-pot preparation of polyHIPE-supported PtNPs available.¹⁹ The catalyst was over 20 times recyclable.¹⁹ Here, we use a different method to prepare smaller and more active PtNPs on a durable polyHIPE support, the resulting catalyst is with extremely low leaching degree (below parts per billion (ppb) scale), and for a reduction reaction, it is recyclable for over 1500 times without any observable decrease in terms of catalytic activity.

2. Experimental section

2.1. Materials

K₂PtCl₄, p-nitrophenol (p-NP) and 2-ethylhexyl acrylate (EHA, 99%) were purchased from Aladdin (China). Azodiisobutyronitrile (AIBN) and sodium borohydride (96%) were purchased from SCRC (China). EHA was distilled under vacuum to remove the inhibitor and stored at 4 °C until use. AIBN was recrystallized in ethanol and stored at 4 °C prior to use. Poly(ethylene glycol) diacrylate (PEGDA, CH₂=CHCOO(CH₂CH₂O)_nH, n ≈ 4) was purchased from Aladdin (China) and used as received.

PEI@P(S-EHA) (1) was synthesized on literature,¹⁹ with a structure of PEI₂₃₂-P(S₆₃-co-EHA₂₂)₃₀, which means 30 chains of P(S-EHA) is attached to one PEI, and the PEI has on average 232 repeat units, while each chain of P(S-EHA) has 63 styrene units and 22 EHA units in the form of random sequence in polymer.

2.2. Measurements

UV/vis spectra were recorded on a Mapada UV-6300 spectrophotometer (Shanghai Mapada Instruments Co., Ltd.). The size of metal NPs and high resolution lattice structure were recorded using aJEM-2011F transmission electronic microscopy (TEM and HRTEM) operating at an acceleration voltage of 200 kV on a carbon-coated copper grid. Crystal lattice structure of Pt was analysed with software of Digital Micrograph based on Fast Fourier Transformation (FFT). Field emission scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX) was recorded on Magellan 400 (FEI Co. Ltd). The porosity and surface area of polyHIPE were measured with a mercury intrusion porosimetry (MIP) of Pore Master 60-GT (Quantachrome Corp). Inductively coupled plasma mass spectra (ICP-MS) were recorded on Nexlon 300X (PE, USA). X-ray photoelectron spectroscopy (XPS) was recorded on ESCALAB 250Xi (USA) with an Al Ka X-ray source.

2.3. Synthesis of Pt–polyHIPE (4)

Typically, to a homogeneous oil phase consisting of stabilizer 1 PEI@P(S-EHA, 0.10 g), AIBN (5.8 mg), EHA (0.43 g) and PEGDA (0.13 g) under rigorous stirring, water (4 mL) containing K₂PtCl₄ (1.0 mg) was dropped. The resulting concentric emulsion was kept under stirring for another 15 minute after the addition. The emulsion was transferred into a beaker-like vessel and heated at 70 °C in an oil bath for 24 h. The resulting porous elastomer was dropped into fresh aqueous sodium borohydride (1 g NaBH₄ in 10 mL of water), and subjected to mild stirring at room temperature for 4 h. The elastomer was sufficiently washed with fresh water and ethanol in sequence to yield 0.58 g product (4). The Pt content in 4 was 810 parts per million (ppm) or 4.1 μmol g⁻¹.

2.4. Introduction thiol groups to 4 (S-1 to S-5)

Typically for the case of S-1(thiol/Pt = 1 at molar ratio), a solution of ethylene sulfide (3.8 μL) in ethanol (1 mL) was prepared and an aliquot of 10 μL was injected into a degased ethanol (5 mL) containing a slice of 4 (0.107 g). The reaction system was sealed and mildly stirred for 6 h. The elastic solid was fully washed with fresh ethanol before dried in a vacuum oven. Other samples (thiol/Pt = 2, 3, 4, 5 at molar ratio) were prepared similarly (S-2 to S-5).

2.5. Catalytic reduction of p-NP by 4

Typically, for Pt at 0.34 molar equivalent of p-NP, a slice of monolith 4 (0.015 g, equivalent to 12 μg Pt, stored in N₂ atmosphere prior to use) was suspended in a N₂-bubbled aqueous solution containing p-NP (3 mL, 0.06 mM) and NaBH₄ (0.077 g), and the mixture was charged in a cuvette and monitored with a UV/vis spectrometer. The reaction typically finished within 0.5 h, and 4 was taken out with tweezers, quickly sucked with a filter paper and used in another cycle. S-1 to S-5 samples were similarly tested.

2.6. Detection of Pt leakage

Typically, 4 (0.10 g, 81 μg Pt) was soaked in aqueous NaBH₄ (0.68 M, 6 mL) for 10 d. 4 was fished out and the water sample was detected with ICP-MS to detect Pt species released into water. Similar operation was applied to S-1 to S-5.

3. Results and discussion

3.1. Preparation and characterization of polyHIPE-supported PtNP (4)

The preparation of supported PtNP (4) is outlined in Scheme 1. 1 is a dendritic amphiphile synthesized by alkylation of hydrophilic PEI core with hydrophobic glycidyl-capped P(S-EHA), as previously reported.¹⁹ 1 is oil-soluble but upon exposure to a water/oil mixture under stirring, can adopt a conformation of Janus-like particles and self-assemble along the W/O interface to stabilize the resulting emulsion. Since the dispersed water droplets amounts to around 80% of the total volume of the emulsion, it is termed high internal phase emulsion (HIPE, 2). The continuous oil phase consists of EHA monomer, a crosslinker of PEGDA (EHA/PEGDA = 3.3, weight ratio), and an initiator of AIBN at 1 wt% with respect to the polymerizable components. Heating of 2 triggers a radical polymerization and transforms 2 into a porous elastomer termed polyHIPE (3). The elastic property of 3 arises from the low glass transition temperature of EHA units (with a glass transition temperature of −58 °C (ref. 32)) and the flexible PEGDA units. When K₂PtCl₄ is present in the water, similar polyHIPE is available. The polyHIPE absorbs Pt salt by electrostatic complement between the PtCl₄²⁻ and protonated 1. The material is further treated by soaking into aqueous sodium borohydride so that the ionic Pt species are quickly transformed into metallic PtNPs. Removal of the water droplets and washed in sequence with fresh water and ethanol leads to 4. To learn the size of the PtNPs, a control sample is similarly synthesized but the PEGDA is replaced with the same volume of EHA. In such a case, the final Pt–polyHIPE collapses upon vacuum drying at 40 °C and is no longer porous, and is well soluble in chloroform. Several drops of the chloroform solution are dropped on a carbon-coated grid for TEM measurement. TEM shows particles with an average size of 32 ± 6 nm (micrograph not shown), however, with the addition of several drops of trifluoroacetic acid into the chloroform solution, the corresponding PtNPs change into smaller size of 6.6 ± 1.4 nm (Fig. 1). This fact indicates that Pt–polyHIPE tends to aggregate unless sufficient protonation, where cationic PEI promotes disaggregation. HRTEM shows the single crystal nature of the PtNPs, and FFT analysis shows the existence of 111 and 200 single-crystal planes (Fig. 1).

Scheme 1 Outline of the synthesis of PtNP-decorated polyHIPE (4). Conditions: (a) self-assembly of 1 (PEI@P(S-EHA)) along water/oil interface to form a HIPE (2); (b) polymerization of 2 to form polyHIPE (3); (c) reduction of Pt²⁺ to PtNPs and remove the water droplets to produce Pt–polyHIPE (4).

Fig. 1 TEM micrograph of PtNPs (top), HRTEM of several PtNPs and FFT analysis of HRTEM of Pt crystal (bottom).

The chloroform solution is reddish brown and UV/vis measurement shows a featureless wide band, where a red shift with respect to 1 is observed (Fig. S1, ESI†), also supports the formation of PtNPs. Unlike a conventional polyHIPE which is usually mechanically weak and chalky, 4 is elastic and remain integrity with repeated pinching (inset in Fig. 2), which is very favorable for separation and recycling. The similar C=C structure of EHA and PEGDA means close reactivity ratios, so a rather homogeneous crosslinking can be expected.

Fig. 2 Pore size distribution of 4 as measured by MIP, the inset shows flexibility of 4 upon repeated pressing and releasing.

The porous structure of 4 is supported by several facts: (1) 4 shows a typical apparent density of 0.18 ± 0.02 g mL⁻¹, as determined by a method based on Archimedes principle; (2) 4 shows a water uptake ability of 2.55 g g⁻¹, supporting the open-cellular structure and a hydrophilic surface; (3) MIP measurement supports an open-cellular structure (Fig. 2), with a specific surface area typically of 36.0 m² g⁻¹, a porosity volume of 76%, and an average pore diameter of 38 μm; (4) SEM measurement provides an intuitive evidence of the porous structure, as shown in Fig. 3.

Fig. 3 SEM and EDX elemental mapping of 4. The scale bar is 200 μm.

The EDX mapping of carbon or oxygen element outlines the porous framework of 4; Pt signal can be well observed, which appears to rather homogeneously distribute on the surface (Fig. 3). The large interconnected pores are very favorable for mass transfer, and the good water compatibility of the surface renders the catalyst well applicable in water, a green medium.

3.2. Leaching mechanism of Pt species from 4

To learn the leaching mechanism of Pt species, a slice of 4 is immersed into pure water in the absence and presence of a reducing agent of NaBH₄, respectively, and Pt species released into water is detected by ICP-MS. Typically, 4 (0.1 g) is dropped into aqueous NaBH₄ (6 mL, 0.68 M), similar to the condition for reduction of p-NP to p-aminophenol (p-AP). When 4 is soaked for 5 d or 10 d, Pt species released into water is detected to be at a scale of several part per billion (ppb) (Table 1). This result suggests for each reaction cycle (around 0.5 h), the leakage is at a scale of 10⁻² ppb, which is too low to detect. However, in the absence of NaBH₄, the leakage is 122 ppb for 5 d. A further test shows that even within 1 d, the concentration of soluble Pt species in water is 139 ppb. This means Pt releases quickly and reaches saturated solubility in water within 1 d (saturated content in water was reported to be 14–362 ppb, slightly increased with pH (ref. 33 and 34)). Leaching from bulk metallic Pt into water was in form of Pt²⁺, and would took hundreds days to reach a saturated solubility.^33,34 Our data suggests that in oxidative water (most probably due to O₂), PtNPs leaches rather quickly, probably due to the high specific surface area. And further erosion of PtNP is inhibited by the saturated solubility, as compared with those for 5 d and 10 d. While in reductive water, leakage of Pt is almost blocked, implying that the leakage is through ionic Pt species rather than atomic Pt.

Table 1 Pt species leached from Pt–polyHIPE into water and calculated recycles based on the leaking data^a

Sample	SH/Pt (molar ratio)	Leached Pt in water (ppb)		Recyclable times in aq. NaBH4/10^7^d
		5 d/10 d (aq. NaBH4)^b	1 d/5 d (pure water)^c
a Conditions: Pt–polyHIPE (0.1 g) soaked in water (6 mL) for specified time and the leached Pt species is detected with ICP-MS; the data is the average of triple measurement.b In aq. NaBH4 (0.68 M) for 10 d.c In neutral water.d Calculated based on 10% erosion of Pt species from the support in the case of aq. NaBH4 (10 d data). The measuring errors of ICP-MS are within 68%.
4	0	1.1/4.3	139/122	6.8
S-1	1	1.0/5.1	176/136	5.1
S-2	2	1.6/1.5	163/120	17
S-3	3	1.1/4.3	82/77	6.8
S-4	4	1.3/1.6	66/38	17
S-5	5	1.0/2.6	65/75	10

---

Based on the leaching rate of Pt species in reductive water (Table 1), the catalyst can be recycled for over 10⁷ times at 10% erosion of the PtNPs (based on the experimental data that each cycle takes about half an hour). This is an unprecedentedly high value.

To learn if a ligand exerts any influence on the leaching of Pt species, certain amount of thiol groups are introduced onto 1. A thiol group is known as a very strong ligand of noble metal species. Thiol groups can be introduced onto an amino group by ring-opening of ethylene sulfide with a primary or secondary amino group.³⁵ Here samples (S-1 to S-5) are prepared by treating 4 with ethylene sulfide at a feed molar ratio of sulfur/Pt = 1, 2, 3, 4 and 5, respectively (Scheme S1, ESI†). Since the corresponding molar ratio of sulfur/nitrogen is very low (0.027–0.14), discrete thiol groups rather than poly(ethylene sulfide)s will be formed. The introduced thiol species accounts for a very small weight fraction [(1–5) × 10⁻⁵] of the total weight of 4 and is hardly detectable by most techniques, but can be well detected by EDX (Fig. S2, ESI†). This is understandable because, under promotion of supramolecular interfacial enrichment, PEI and hence the thiol groups should mainly distribute on the surface of the polyHIPE matrix. The thiol groups, as a soft base, tend to ligate the soft acid of PtNP, if allowed by conformational motion. It is noticed that in reductive water, the presence of thiol groups exerts little role on the leaching of Pt species (Table 1), but in oxidative water, the sulfur species (probably exist in form of –SH, –S–Pt or –S–S–) appear to somewhat suppress the leakage (Table 1). The mechanism is not understood yet. It is also found the thiol groups sharply suppress the catalytic activity of PtNP. For example, in the case of 4, it takes less than 0.5 h to lead to about 99% conversion of p-NP to p-AP; while in the case of S-5, it takes 10 h to only 91% conversion.

3.3. Repeated catalyzing the reduction of p-NP by 4

Besides leakage, degradation such as Ostwald ripening of PtNP can lead to decreased catalytic property. 4 is used to catalyze the reduction of p-NP to p-AP in the presence of sodium borohydride. This reduction reaction has become a model reaction^36–38 to evaluate a catalyst for the following features: a single reactant to a single product; hardly any reduction occurs in the absence of a catalyst; reaction proceeds at mild temperatures. When a slice of 4 (0.015 g, equivalent to 12 μg Pt, stored in N₂ atmosphere prior to use) is dropped in a N₂-bubbled aqueous solution (3 mL) containing p-NP (0.06 mM) and NaBH₄ (0.68 M), the reduction proceeds rather quick, as judged from UV/vis spectra where the absorbance at 400 nm (due to anionic p-NP) decreases and a small absorbance grows up at 305 nm (due to p-AP) (Fig. 4a). The appearance of an isobestic point at 322 nm indicates the reaction is via a route of a single reactant to a single product.

Fig. 4 A typical UV/vis spectral monitoring the reduction of p-NP in presence of 4 (a) and the corresponding apparent rate constant (b); 4 is be repeatedly used for the reduction in aqueous NaBH₄ (c). The apparent rate constants are also evaluated after the catalyst 4 is soaked in aqueous NaBH₄ for up to 30 d (d). Conditions: [NaBH₄] = 0.68 M; room temperature; the aqueous solution refreshed every 5 d.

Regarding the catalytic mechanism, it is widely believed that p-NP and BH₄⁻ are both adsorbed by a PtNP where the latter conducts electron transfer, followed by dissociation of the product (Langmuir–Hinshelwood model).³⁸ Accordingly, in case the reductant is in much excess, the kinetics complies with a pseudo-first-order form (eqn (1)):

−ln(A_t/A₀) = k₁S_Ptt = k_appt (1)

where A_t and A₀ represent the absorbance at 400 nm at time t and 0, respectively; k₁ represents the rate constant normalized to the surface area (S_Pt) of PtNP and k_app represents the apparent rate constant. Treatment of the kinetic data does lead to a linear relationship, with a k_app derives to be 2.58 × 10⁻³ s⁻¹ (Fig. 4b), supporting the Langmuir–Hinshelwood model. The catalyst is repeated 10 recycles without any essential activity decrease (Fig. 4c). The Pt content in the reaction system is very low (4 ppm), and the turnover frequency (TOF = [p-NP]/([Pt][conversion])/t) is derived to be 9.4 h⁻¹, which is rather high among supported catalysts. To further evaluate whether the catalyst can be recycled for thousands time, a slice of 4 is soaked in aqueous sodium borohydride (0.68 M), with the solution refreshed every 5 d, and at each refreshing, the catalyst is evaluated in terms of the k_app. It is found that up to 30 d, the k_app still remains essentially the same (Fig. 4d). This result suggests that the surface area (S_Pt) of PtNPs remains intact, i.e., Ostwald ripening is negligible. The 30 d soaking test suggests over 1500 cycles is possible. This equivalence is reasonable because the metal leakage is at ppb scale (far below the saturated solubility of 130 ppb) and PtNP ripening is negligible. The ultrahigh stability should be attributed to the integrated coordination effect³⁹ of multiligand 1 upon interacting with a PtNP. The reducing agent of sodium borohydride can be replaced by hydrogen gas for similar reduction but cannot be replaced by a weaker reducing agent such as hydrazine, thiosulfate and sodium cyanoborohydride.

4. Conclusions

PEI-based dendritic amphiphiles can play triple role as stabilizer, PtNP multiligand, and anchoring agent to immobilize PtNP on polyHIPE, thus can lead to one-pot preparation of polyHIPE-supported PtNPs. It is found that leakage of Pt species from PtNPs is almost exclusively through ionic Pt, and can be blocked in a reductive environment. It is also found that the PEI multiligand can suppress Ostwald ripening of PtNPs while allow a high catalytic activity of the PtNPs. As a result, the surface area of the polyHIPE-supported PtNPs remains intact within an equivalent of 1500 cycles for a reduction reaction, as exemplified by the reduction of 4-nitrophenol. The facile separation, high activity and robust durability will favour cost-effective and environment-friendly production, and ensure quality products, especially where pharmacy involves.

Acknowledgements

The Natural Science Foundation of China (No. 51573138 and 51273149), and the Open Measuring Fund for Large Instrument and Equipment, Tongji University (No. 0002015031 and 0002015032) are gratefully acknowledged.

Notes and references

1. D. Astruc, F. Lu and J. R. Aranzaes, Angew. Chem., Int. Ed., 2005, 44, 7852–7872.
2. A. Roucoux, J. Schulz and H. Patin, Chem. Rev., 2002, 102, 3757–3778.
3. L. L. Chng, N. Erathodiyil and J. Y. Ying, Acc. Chem. Res., 2013, 46, 1825–1837.
4. P. Anastas and M. M. Kirchhoff, Acc. Chem. Res., 2002, 35, 686–694.
5. K. Y. Wu, J. Guo and C. C. Wang, Angew. Chem., Int. Ed., 2016, 55, 6013–6017.
6. Y. K. Zhou, K. Neyerlin, T. S. Olson, S. Pylypenko, J. Bult, H. N. Dinh, T. Gennett, Z. P. Shao and R. O'Hayre, Energy Environ. Sci., 2010, 3, 1437–1446.
7. H. K. Sadhanala, R. Nandan and K. K. Nanda, Green Chem., 2016, 18, 2115–2121.
8. Y. Z. Chen, G. R. Cai, Y. M. Wang, Q. Xu, S. H. Yu and H. L. Jiang, Green Chem., 2016, 18, 1212–1217.
9. O. D. Lyons, N. E. Musselwhite, L. M. Carl, K. A. Manbeck and A. L. Marsh, Langmuir, 2010, 26, 16481–16485.
10. G. C. Zheng, K. Kaefer, S. Mourdikoudis, L. Polavarapu, B. Vaz, S. E. Cartmell, A. Bouleghlimat, N. J. Buurma, L. Yate and A. R. de Lera, J. Phys. Chem. Lett., 2015, 6, 230–238.
11. E. Antolini, Appl. Catal., B, 2012, 123, 52–68.
12. S. K. Movahed, M. Dabiri and A. Bazgir, Appl. Catal., A, 2014, 488, 265–274.
13. J. L. Zhu, G. Q. He, Z. Q. Tian, L. Z. Liang and P. K. Shen, Electrochim. Acta, 2016, 194, 276–282.
14. H. Bartl and W. V. Bonin, Makromol. Chem., 1962, 57, 74–95.
15. M. S. Silverstein, Prog. Polym. Sci., 2014, 39, 199–234.
16. C. Gao and D. Y. Yan, Prog. Polym. Sci., 2004, 29, 183–275.
17. Y. L. Ye, D. C. Wan, J. Du, M. Jin and H. T. Pu, J. Mater. Chem. A, 2015, 3, 6297–6300.
18. Y. L. Ye, M. Jin and D. C. Wan, J. Mater. Chem. A, 2015, 3, 13519–13525.
19. H. H. Liu, D. C. Wan, J. Du and M. Jin, ACS Appl. Mater. Interfaces, 2015, 7, 20885–20892.
20. A. K. Diallo, C. Ornelas, L. Salmon, J. R. Aranzaes and D. Astruc, Angew. Chem., Int. Ed., 2007, 46, 8644–8648.
21. B. D. Briggs, N. M. Bedford, S. Seifert, H. Koerner, H. Ramezani-Dakhel, H. Heinz, R. R. Naik, A. I. Frenkel and M. R. Knecht, Chem. Sci., 2015, 6, 6413–6419.
22. C. Deraedt and D. Astruc, Acc. Chem. Res., 2014, 47, 494–503.
23. D. Astruc and F. Chardac, Dendritic catalysts and dendrimers in catalysis, Chem. Rev., 2001, 101, 2991–3023.
24. D. Astruc, E. Boisselier and C. Ornelas, Chem. Rev., 2010, 110, 1857–1959.
25. Y. Liu, Y. Fan, Y. Yuan, Y. Chen, F. Cheng and S. C. Jiang, J. Mater. Chem., 2012, 22, 21173–21182.
26. N. Hu, J. Y. Yin, Q. Tang and Y. Chen, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 3826–3834.
27. M. L. Wang, T. T. Jiang, Y. Lu, H. J. Liu and Y. Chen, J. Mater. Chem. A, 2013, 1, 5923–5933.
28. M. Dhiman, B. Chalke and V. Polshettiwar, ACS Sustainable Chem. Eng., 2015, 3, 3224–3230.
29. S. Shrestha, Y. Liu and W. E. Mustain, Catal. Rev.: Sci. Eng., 2011, 53, 256–336.
30. Y. Shao-Horn, W. C. Sheng, S. Chen, P. J. Ferreira, E. F. Holby and D. Morgan, Top. Catal., 2007, 46, 285–305.
31. S. Ogasawara and S. Kato, J. Am. Chem. Soc., 2010, 132, 4608–4613.
32. Polymer Handbook, ed. J. Brandrup, E. H. Immergut and E. A. Grulke, Wiley, New York, 4th edn, 1999.
33. S. A. Wood, Geochim. Cosmochim. Acta, 1991, 55, 1759–1767.
34. M. Azaroual, B. Romand, P. Freyssinet and J.-R. Disnar, Geochim. Cosmochim. Acta, 2001, 65, 4453–4466.
35. J. X. Huang, F. Wang, D. M. Du and J. X. Xu, Synthesis, 2005, 2122–2128.
36. S. Wunder, F. Polzer, Y. Lu, Y. Mei and M. Ballauff, J. Phys. Chem. C, 2010, 114, 8814–8820.
37. P. X. Zhao, X. W. Feng, D. S. Huang, G. Y. Yang and D. Astruc, Coord. Chem. Rev., 2015, 287, 114–136.
38. Y. Mei, G. Sharma, Y. Lu and M. Ballauff, Langmuir, 2005, 21, 12229–12234.
39. M. Mammen, S.-K. Choi and G. M. Whitesides, Angew. Chem., Int. Ed., 1998, 37, 2754–2794.

---

Footnote

† Electronic supplementary information (ESI) available: UV/vis of PtNP, EDX of S-5 and schematic synthesis of S-1 to S-5. See DOI: 10.1039/c6ra19013h

---