Catalysis by multifunctional polyelectrolyte capsules

Abstract

Gold nanoparticles and nanocomposites have high catalytic performance for several chemical reactions. Here we present gold and iron oxide nanoparticle modified polymer capsules as porous and multifunctional platforms for catalysis. Layer-by-layer polyelectrolyte microcapsules were formed on calcium carbonate template cores loaded with gold nanoparticles, allowing for high gold loading of the capsules. Magnetic nanoparticles were incorporated in the polymeric shells of the capsules, allowing for magnetic separation. The influence on the catalytic behaviour of gold was studied in terms of the nanoparticle size, the presence of a polymeric shell, and the presence of the magnetic nanoparticles in the shell, by using the model electron transfer reaction between hexacyanoferrate(III) and borohydride.

---

Introduction

Polyelectrolyte multilayer (PEM) capsules have a broad range of applications. This is mainly due to the possibility of finely tailoring their composition, size, shape, and structure at the nano and micro scale. One of the most popular applications is drug delivery.^1–6 This is because the interior of the hollow shell can be loaded with a large amount of drugs that are protected from the environment and the mechanism of drug release can be chosen on demand (e.g. degradation or stimuli responsive opening of the capsule shell).^7–10 However, PEM capsules can be used as well as microreactors, since they are porous and it is possible to load them with catalysts such as enzymes,^11–15 transition metal complexes¹⁶ or catalytic nanoparticles.^17,18 Microreactors based on enzymes are sensitive to denaturation and therefore more prone to decrease the catalytic activity with the time and under rush conditions than microreactors based on inorganic catalysts. However, it has been demonstrated that the polyelectrolyte shell can make the enzyme more stable than the free enzyme in solution.¹⁹ Most of the industrial catalysts are composed of metallic clusters or nanoparticles supported in inorganic matrices. This can be explained if we consider several facts: (i) metallic clusters and nanoparticles are good catalyst due to their high surface energy and the metal–metal bond deficiencies of surface atoms, which favors the exchange of electrons between molecules and metallic atoms, and (ii) the supported material is necessary to stabilize the nanoparticles against agglomeration. Although it is possible to use free metallic clusters and nanoparticles stabilized with capping molecules as catalysts,^20,21 the use of hybrid materials or bulk materials supporting them is more extended.^22–24 This is because it is not straightforward to keep well dispersed metallic clusters or very small nanoparticles in reaction conditions (often including high temperatures, high ionic strength, etc.) and this fact comprises the catalytic activity of the metals. Moreover, the nanoparticle support can add new features to the catalyst, such as reusability or regulation of the catalytic activity.^25–27

In this regard, oxides of silicon and titanium, and carbon and carbon nitrides have been largely used to add stability and increase the catalytic activity of certain metal nanoparticles.^28–32 Reusability is often performed by separation of the catalysts from the reaction solution, which can be achieved by precipitation or adding of magnetic nanoparticles in the formulation of the catalyst for magnetic separation.^33–35 However, the formation of a hybrid material might hinder the catalytic activity of the metallic catalyst due to steric hindrance, unwanted interactions etc.

In this context, we have analysed the effect of adding magnetic nanoparticles (MNPs) in the composition of hybrid catalysts based on polymers and gold nanoparticles (AuNPs) by studying the catalytic behaviour of the AuNPs integrated into PEM capsules of different compositions and compared it with the catalytic activity of free AuNPs in solution. We produced PEM capsules loaded with spherical AuNPs of 4.3 or 14.9 nm in diameter, with and without MNPs made of iron oxide (27.9 nm) in the shell of the capsules, and used them as catalyst in the reaction of reduction of hexacyanoferrate(III) by borohydride. The two different sizes of AuNPs were selected to study the effect of the nanoparticle size on the catalytic activity of the PEM capsules. We selected PEM capsules because they are highly porous and easy to load with nanoparticles.⁹ AuNPs were used as catalyst because it is known that the size of the nanoparticles affects their catalytic activity.^36,37 It is worth mentioning that other metals such as silver or platinum could be used as well as efficient catalyst which own size and shape dependent catalytic activity.^32,38,39 Finally, MNPs were used to perform magnetic separation which could be used for recycling or to locally concentrate the catalytic AuNPs entrapped in such PEM capsules.

Experimental

Materials

Gold(III) chloride trihydrate (HAuCl₄·3H₂O), bis(p-sulfonatophenyl)phenylphosphine dihydrate dipotassium salt (#698539), trisodium citrate dehydrate (#S1804), calcium chloride dehydrate (CaCl₂, #223506), sodium carbonate (Na₂CO₃, #S7795), poly(allylamine hydrochloride) (PAH, M_w ≈ 56 kDa, #283223), poly(sodium 4-styrenesulfonate) (PSS, M_w ≈ 70 kDa, #243051), ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA disodium salt, #E5134), sodium chloride (NaCl, #S7653), dextran from Leuconostoc spp. (2000 kDa, #95771), sodium borohydride were purchased from Sigma-Aldrich, Germany. Milli-Q water having resistivity greater than 18.2 MΩ cm⁻¹ was used. Citrate capped Au NPs (14.9 nm in diameter, #EM.GC15) were purchased from BBI Solutions, UK.

Synthesis and surface functionalization of gold nanoparticles

AuNPs of 4.3 nm in diameter were prepared following the protocol reported by Jana et al.⁴⁰ We added 375 μL of 100 mM solution of HAuCl₄·3H₂O in 100 mL of ice-cooled water (4 °C) in a round bottom flask under vigorous stirring and after 5 min, 500 μL of a 200 mM sodium citrate solution were added into the reaction flask and left stirring for 2 hours. Then, 5 mL of a 2.64 mM sodium borohydride solution were added and the reaction was left stirring for further 5 hours. This resulted in the formation of a colloidal solution of citrate capped AuNPs having 4.3 nm of diameter as determined by TEM. The surface plasmon resonance (SPR) band of this colloidal solution showed its maximum at 516 nm. Because these AuNPs were not colloidally stable at high ionic strength, the citrate molecules on their surface were exchange for bis(p-sulfonatophenyl)phenylphosphine. For that, 25 mL of a 15 mM bis(p-sulfonatophenyl)phenylphosphine dehydrate solution were added under magnetical stirring to the citrate capped gold nanoparticle solution and left stirring 24 hours before washing. The excess of ligands were removed using filtration with a membrane filter (10 000 molecular weight cut-off (MWCO)). The AuNPs were washed several times with Milli-Q water. The obtained AuNPs were characterized with UV-Vis absorption spectroscopy before and after ligand exchange and transmission electron microscopy (TEM) (this information is shown in the ESI†). Their average diameter was 4.3 ± 0.9 nm. Similarly, commercial 14.9 nm Au NPs (obtained from BBI International) were functionalized with bis(p-sulfonatophenyl)phenylphosphine to have similar surface chemistry than the smaller AuNPs.

Synthesis of magnetic PEM capsules with encapsulated gold nanoparticles

The bis(p-sulfonatophenyl)phenylphosphine functionalized AuNPs with an average diameter of 4.3 nm (AuNP1) and the commercially available AuNPs with an average diameter of 14.9 nm (AuNP2) were used to be integrated into the cores of CaCO₃ particles during their formation. Briefly, 300 μL of a dextran solution (6.5 mg mL⁻¹) were mixed with 615 μL of a 0.33 M CaCl₂ solution under constant magnetical stirring at 1000 rpm. Then, 100 μL of AuNPs (both types i.e. the 4.3 nm and the 14.9 nm AuNPs having ≈750 μg of elemental gold, as determined by ICP-MS) were added to the mixture solution. Finally, 615 μL of a 0.33 M Na₂CO₃ solution was added fast and the formation CaCO₃ started immediately. The resulting CaCO₃ cores loaded with AuNPs were washed twice by centrifugation to precipitate the particles and redissolved them in fresh Milli-Q water. After that, the PEM shell was prepared on the CaCO₃ template via self-assembly layer-by-layer (LbL) of two kinds of polyelectrolytes with opposite charge.⁴¹ Seven monolayers of polyelectrolytes were deposited on the top of the CaCO₃ cores, starting from poly(allylamine hydrochloride) (PAH) and followed by poly(sodium 4-styrenesulfonate) (PSS). This process was repeated three times. Finally, a last layer of PAH was added. The final composition of the polyelectrolyte shell was (PAH/PSS)₃PAH. The CaCO₃ cores were dissolved using a 0.2 M solution of ethylenediaminetetraacetic acid disodium salt (EDTA) at pH = 7. This resulted in the formation of the hollow PEM shells having AuNPs in their cavity CAP1 and CAP2. The formation of the magnetic PEM capsules MCAP1 and MCAP2 was similar. The MNPs made of iron oxide were self-assembled between the polyelectrolyte monolayers due to their negative charge.⁹ The MNPs having a diameter of 27.9 ± 4.1 nm were synthesized using the protocol reported by Yu et al.⁴² and coated with the amphiphilic polymer poly(maleic anhydride-alt-1-tetradecene) as reported by Pellegrino et al.⁴³ The obtained MNPs were reported to be Fe₃O₄ (determined by XRD). However, in our study since MNPs did not show any catalytic effect in our reaction conditions, the relative amount of maghemite respect to magnetite was not determined (both are superparamagnetic). The architecture of the final shell was in this case (PAH/PSS)(PAH/MNPs)(PAH/PSS)PAH. All capsule concentrations used in this study were measured using a hemocytometer due to the large size of the capsules. The final solutions of capsules were adjusted in volume to contain the same amount of capsules. The concentration of elemental gold was determined by inductively coupled plasma mass spectrometry (ICP-MS).

Kinetic measurements

The catalytic reactions were carried out in a UV-Vis diode-array spectrophotometer Agilent 8453. The reactant solutions were prepared just before measurements. The BH₄⁻ solution was prepared with Milli-Q water at high pH adjusted with NaOH/HCl to obtain pH 12 in the final reaction volume to avoid the hydrolysis of this reactant. The solutions of BH₄⁻ and catalysts were firstly mixed and finally the solution of Fe(CN)₆³⁻ was added right before starting collecting spectra. Kinetic data were satisfactorily fitted to first order integrated rate equations.

Results and discussion

To compare the catalytic behaviour of AuNPs within different compositions of PEM capsules four kinds of hybrid capsules were produced and their catalytic activities were compared with the catalytic activity of free AuNPs in solution. Moreover, two sizes of AuNPs were synthesized to study their catalytic activity after encapsulation in PEM capsule: (i) AuNPs of 4.3 nm core diameter, in the following termed AuNP1, and (ii) AuNPs of 14.9 nm core diameter, in the following termed AuNP2. Based on these AuNPs the four kinds of PEM microcapsules were prepared: (i) PEM capsules with 4.3 nm AuNPs in their interior named CAP1, (ii) PEM capsules with 4.3 nm AuNPs in their interior and MNPs within their shell named MCAP1, (iii) PEM capsules with 14.9 nm AuNPs in their interior named CAP2, and (iv) PEM capsules with 14.9 nm AuNPs in their interior and MNPs within their shell named MCAP2.

The protocol to fabricate the different PEM microcapsules followed standard procedures, except for the loading of AuNPs into the CaCO₃ cores. CaCO₃ particles were loaded with AuNPs by coprecipitation and used as sacrificial templates. In this way, we obtained a high loading of AuNPs. Optionally, also negatively charged MNPs prepared following the synthetic protocol of Yu et al.⁴² and coated with amphiphilic polymer poly(maleic anhydride-alt-1-tetradecene) as reported by Pellegrino et al.⁴³ could be integrated to the shell via LbL assembly¹⁰ obtaining the samples MCAP1 and MCAP2. Fig. 1 shows the schematic representation of the synthetic protocol of the magnetic and catalytic capsules and transmission electron microscopy (TEM) images of the CaCO₃ template and the final capsule. The synthetic protocol of capsules without MNPs was similar.

Fig. 1 (A) Schematic representation of the synthesis of polyelectrolyte capsules loaded with AuNPs and MNPs. (B.1, B.2) TEM images of template cores comprising CaCO₃ coprecipitated with 14.9 nm AuNPs at two different magnifications. The scale bars correspond to 1 μm and 200 nm respectively. (C.1, C.2) TEM images of polyelectrolyte capsules MCAP2 loaded with 14.9 nm AuNPs and MNPs at two different magnifications after dissolution of the CaCO₃ template with EDTA. In C.2 the two kinds of nanoparticles are visible, the small and black are the AuNPs and the grey big NPs are the MNPs. The scale bars correspond to 1 μm and 100 nm respectively.

Basic characterization information about the six samples, including the AuNPs that were used to compare their catalytic behavior, is enlisted in Table 1.

Table 1 Size, short name, shell composition, diameter, concentration and nanoparticle loading of the four different PEM capsules

Sample composition	AuNP diameter (nm)	Short name	Capsule diameter (μm)	Amount of Au/capsule (pg per capsule)
Spherical AuNPs	4.3 ± 0.9	AuNP1	—	—
Spherical AuNPs	14.9 ± 1.1	AuNP2	—	—
AuNP1 encapsulated by (PAH/PSS)3PAH	4.3 ± 0.9	CAP1	4.2 ± 0.4	3.8
AuNP1 encapsulated by (PAH/PSS)(PAH/MNPs/PSS)(PAH/PSS)PAH	4.3 ± 0.9	MCAP1	5.5 ± 0.4	7.5
AuNP2 encapsulated by (PAH/PSS)3PAH	14.9 ± 1.1	CAP2	4.6 ± 0.5	4.7
AuNP2 encapsulated by (PAH/PSS)(PAH/MNPs/PSS)(PAH/PSS)PAH	14.9 ± 1.1	MCAP2	3.7 ± 0.6	3.5

---

Fig. 2 shows TEM images of the four different PEM microcapsules (CAP1, MCAP1, CAP2 and MCAP2) and the distribution of NPs in the capsules.

Fig. 2 TEM images of all polyelectrolyte capsules. (A.1, A.2) correspond to CAP1, (B.1, B.2) correspond to MCAP1, (C.1, C.2) correspond to CAP2 and (D.1, D.2) correspond to MCAP2. The two kinds of nanoparticles can be recognized in the TEM images: the small and black are the AuNPs of two sizes (4.3 nm in A and B and 14.9 nm in C and D) and the grey big NPs are the MNPs. The scale bars in the images from A.1 to D.1 correspond to 2 μm and in the images from A.2 to D.2 correspond to 50 nm.

The magnetic separation was verified for MCAP1 and MCAP2. Fig. 3 shows the separation of the capsules (MCAP1) from their solution (initially brown) when a magnet of 0.4 T was located on the right hand side of the sample. It is visible in the images that the solution became clear while the capsules deposited on the wall of the tube close to the magnet. The shape of the capsules was not modified due to the application of the magnetic field.

Fig. 3 Images of the magnetic separation with PEM capsules loaded with 4.3 nm AuNPs within the hollow shell and MNPs between the polyelectrolyte layers. The image on the left shows the initial solution at t = 0 min, in the centre t = 2 min and on the right t = 4 min. The MCAP1 were concentrated on the right surface of the tube after 4 min close to the magnet. This was visible because the initial grey solution in the tube became a clear solution with a dark spot on the right which were the concentrated capsules.

Regarding catalysis with PEM capsules, several metallic nanoparticles such as AuNPs or PtNPs are able to catalyse the reduction of hexacyanoferrate(III) by borohydride ions.⁴⁴ The advantage of using this model reaction is that the whole reaction can be monitored spectroscopically due to the absorption of hexacyanoferrate(III) at 420 nm. Even more, the reaction does not show any induction time that could spoil the kinetic analysis. Therefore, the effect of using different nanocomposites or nanoparticles as catalysts can be readily compared and conclusions about the optimal design of such nanocatalysts can be obtained.⁴⁴ The reduction of hexacyanoferrate(III) by borohydride ions in aqueous solution can be written as:

BH₄⁻ + 8Fe(CN)₆³⁻ + 3H₂O → H₂BO₃⁻ + 8Fe(CN)₆⁴⁻ + 8H⁺ (1)

The coordination complex of hexacyanoferrate is extremely stable with respect to dissociation in both oxidation states (+2 and +3), and the reactant and product have the same geometry and chemical composition. As a result, the mechanism of this reaction catalyzed by AuNPs could be reported and compared with the uncatalyzed reaction.²¹ In the presence of AuNPs nanoparticles, a 5000 fold decrease of the half-life of the reaction compared with the uncatalyzed reaction was reported.^45,46 Interestingly, the presence of gold nanoparticles also changed the mechanism of the reaction compared with the direct reduction of [Fe(CN)₆³⁻] by [BH₄⁻].²¹

The experiments were carried out under pseudo-nth-order conditions such that [Fe(CN)₆³⁻] ≪ [BH₄⁻]. It must be taken into account that this reaction should proceed at high pH to inhibit the competitive reaction of borohydride hydrolysis.⁴⁷ Under these conditions, the reaction of the ferrocyanide ion obeys

where k_obs is the pseudo-nth-order rate constant for the reaction. For the following reaction conditions n was reported to be 1: pH > 11.5 and T = 15 °C.²¹

Fig. 4 shows that similarly to the already reported results obtained with citrate stabilized AuNPs, the catalysis in the presence of PEM capsules with encapsulated AuNPs could also be followed at 420 nm and first-order kinetics were determined. Due to the obtained high-quality first order kinetics respect the [Fe(CN)₆³⁻], the data of absorbance versus time could be linearized following eqn (3):

where A is the absorbance at 420 nm. The absorbance band around 550 nm due to the surface plasmon resonance of the AuNPs remained totally unchanged, demonstrating that the entrapped gold nanoparticles was not oxidized or further agglomerated in the reaction media. The results shown in Fig. 4 were obtained with the sample MCAP2 that contained 14.9 nm AuNPs and MNPs in the shell. For all six samples first-order kinetics were obtained.

Fig. 4 (A) Schematic representation of the reduction of hexacyanoferrate(III) with borohydride catalyzed by the encapsulated AuNPs within a magnetic and porous microcapsule. Due to the porosity of the polymeric shell the two reactants can diffuse through it and react on the surface of AuNPs. (B) Spectral evolution of a mixture of hexacyanoferrate(III) and MCAP2 loaded with 15 nm upon borohydride addition. The absorbance maximum at 420 nm corresponds to hexacyanoferrate(III) and the maximum around 550 nm corresponds to the presence of AuNPs. It is worth nothing that this latter absorbance peak remained unchanged during the reaction which means that AuNPs did not oxide or further agglomerate. (C) Kinetic trace of the absorbance at 420 nm corresponding to (B) and on the right Y axis the linearized data for first order analysis according to eqn (3). [Fe(CN)₆³⁻] = 4.16 × 10⁻⁴ M, [BH₄⁻] = 0.05 M, V_t = 3 mL, pH = 12, m_Au = 20.8 μg, k_obs = 0.011 s⁻¹, RT.

The influence of the mass of catalysts in the k_obs was studied for the different kinds of capsules and for the free nanoparticles in solution. The catalytic effect of the AuNPs depended linearly with the concentration and mass of gold for MCAP2 (Fig. 5A) and the same trend was observed for all samples. Fig. 5B shows the results obtained with 14.9 nm AuNPs. These results were in agreement with previously reported data obtained with citrate stabilized gold nanoparticles.²¹

Fig. 5 (A) Influence of the amount of Au m_Au (as determined with ICP-MS) on the observed rate constant k_obs for sample MCAP2. A linear trend was obtained meaning the catalytic effect is directly proportional to the mass of Au in the reaction solution. (B) Influence of mass of catalyst. The k_obs values were obtained for: () AuNP2, () CAP2, and () MCAP2. The different slopes meant different catalytic activities. Higher slope for the linear trend meant higher catalytic activity for the chosen amount of Au. [Fe(CN)₆³⁻] = 4.16 × 10⁻⁴ M, [BH₄⁻] = 0.05 M, pH = 12, RT.

Fig. 5B shows the linear trends obtained with free AuNPs (AuNP2) and AuNPs encapsulated in PEM capsules (CAP2) contrasted with AuNPs1 encapsulated in magnetic PEM capsules (MCAP2). The catalytic activity of such samples can be evaluated by comparing the obtained observed rate constant k_obs for samples with the same mass of AuNPs.²⁶ As it can be seen all samples were highly active for catalysis and the encapsulation of AuNP2 within PEM shells only produced 1.3 and 2.3 fold decrease of the k_obs of the reaction for CAP2 and MCAP2 respectively compared with AuNP2 when using the same mass of catalyst. This decrease is not high as compared with the 5000 fold increase of the half-life of the reaction for the uncatalyzed reaction.^45,46 Interestingly, the effect of the shell and the presence of magnetic nanoparticles were clearly observed. There was a slight decrease of the catalytic activity when AuNPs were entrapped inside a polyelectrolyte shell and a further slight decrease of the catalytic activity when the polyelectrolyte shell was filled with MNPs. This fact can be due to several reasons. The first one is a possible change in the diffusion of reactants and products through the polyelectrolyte shell as compared with this diffusion when the reaction is catalyzed by free AuNPs. Secondly, the catalytic activity could be decreased because of the interaction between the AuNPs and the polyelectrolytes and the interaction between AuNPs which are now confined in a reduced space. These two kinds of interactions will reduce the amount of gold surface available for catalysis. The change in diffusion and the conditions of the Au surface could also affect the catalysis in parallel. What is clear from the results is that the presence of MNPs further reduced the catalytic activity. This effect of the presence of nanoparticles within the PEM shells, is in agreement with already reported data which showed that the permeability of macromolecules through polyelectrolytes shells can be decreased by the presence of nanoparticles.⁴⁸ Moreover, we recently observed different conducting pathways and different conductivity behaviors for ions crossing planar polyelectrolyte films with and without nanoparticles between the polyelectrolyte layers which indicates that the transport of molecules even small enough to permeate through the polyelectrolyte shell is affected by the presence of nanoparticles.⁴⁹

Despite the fact that the catalytic activity was slightly reduced, the polymeric shells around the AuNPs provide a confined space for the reaction to take place, a permeable membrane that can filter the reactants and products of the reaction, stability for the AuNPs against further agglomeration and the possibility of adding multiple functionalities on demand such as labeling with dyes or magnetism with magnetic nanoparticles for magnetic separation as it was shown in Fig. 3.

In general, there is a lot of room for modifications in the formulation of catalytic capsules based on polyelectrolytes that can make them more or less active in catalysis such as the composition of the shell or the catalyst. If the catalysts are nanoparticles, then the morphology of the nanoparticle can be tailored to modify their catalytic activity and selectivity.^50–53 We studied the effect of the nanoparticle size on the catalytic activity. It is known that for a certain nanoparticle shape, the size also influences the exposed crystallographic planes and their associated corners and defects that benefit or hinder the adsorption of reactants and their reactivity on the nanoparticles surface.⁵⁴ Fig. 6A includes in red, the k_obs values plotted versus the mass of Au for the samples containing 4.3 nm AuNPs compared with the data already shown for 14.9 nm AuNPs. As it can be seen, again linear trends were obtained. The effect of changing the AuNP size produced 6.0, 1.7 and negligible fold increase of the k_obs of the reaction for AuNP1, CAP1 and MCAP1 as compared with AuNP2, CAP2 and MCAP2 respectively when using the same reaction conditions and mass of AuNPs. This means that the decrease of nanoparticle size led to an increase of the catalytic activity for the same mass of catalysts but this effect was voided when the polyelectrolyte shell was loaded with nanoparticles. Smaller nanoparticles display larger catalytic surface for the same mass of Au due to their higher surface-to-volume ratio which supports the observed increase of catalytic activity for capsules containing smaller nanoparticles. However, the effect of the nanoparticle size is also related with the different surface reactivity for the different nanoparticles. The k_obs can be compared per unit of surface to determine if the nanoparticles have different reactivity. For the calculation we took into consideration the average diameter of the NPs obtained from TEM images (information included in the ESI†). AuNPs were considered perfectly spherical and monodispersed. We compared the k_obs per unit of surface for AuNP1 and AuNP2 plotting k_obs versus the total surface area of AuNPs (S_total). Fig. 6B shows a 1.4 increase of the catalytic activity for AuNP1 as compared to AuNP2 which agrees with the work of Zhou et al. with spherical Au NPs synthesized by the Turkevivh method.⁵⁵ Using single-molecule microscopy of a fluorogenic reaction, they were able to study at a single particle level the catalysis with different sizes of AuNPs. They reported a higher catalytic reactivity per surface area for 6 nm as compared to 14 nm spherical AuNPs.

Fig. 6 Influence of AuNP size. The k_obs values were obtained for two sizes of AuNPs (4.3 and 14.9 nm) free in solution or encapsulated in the polyelectrolyte capsules. (A) The k_obs values were plotted versus the total mass of AuNPs. Red and black dots and lines correspond to 4.3 and 14.9 nm AuNP respectively. Results correspond to the different samples: () AuNP1, () CAP1 and () MCAP1, () AuNP2, () CAP2 and () MCAP2. (B) The k_obs values were plotted versus the total surface area of AuNPs for () AuNP1 and () AuNP2. [Fe(CN)₆³⁻] = 4.16 × 10⁻⁴ M, [BH₄⁻] = 0.05 M, pH = 12, RT.

A similar analysis was done for the AuNPs encapsulated in PEM shells. In this case, we had to consider that the whole surface of AuNPs was available for catalysis. In the same plot of Fig. 6B it can be seen than AuNP2 were more reactive per unit of surface than AuNP1 when encapsulated in PEM shells. The k_obs were 2.0 and 3.5 higher for CAP2 and MCAP2 than the slopes for CAP1 and MCAP1 respectively.

The two facts: (1) the presence of the polyelectrolyte shell containing magnetic nanoparticles equalized the catalytic activity in terms of mass of catalyst of MCAP1 and MCAP2 and (2) the higher catalytic activity in terms of unit of surface was found for the capsules containing the bigger AuNPs, showed how tangled are the parameters that play a role in catalysis with hybrid materials. On one hand, the catalyzed reaction could be a diffusion limited reaction. This would mean that the slowest step of the reaction would be the diffusion of the reaction species through the PEM shell. In this case, an increase of the catalyst concentration and a change of the nanoparticle size will affect less the k_obs than expected. For diffusion controlled reaction, the mass transfer through the pores of the catalyst play an important role on the rate of the reaction. Diffusion limited reactions are typical for heterogeneous catalysis with solid catalysts where the reaction species are in a different phase than the catalyst.⁵⁶ In our case, the catalytic capsules were in solution but the catalytic surface of gold was in the inner cavity of the porous capsules. Therefore reactants and products had to cross such membrane to reach the gold surface. For citrate coated AuNPs with 15 nm diameter free in solution a change of regime from surface-controlled reaction to diffusion-controlled reaction was reported to happen at concentrations above [BH₄⁻] ∼ 0.1 M.²¹ In our current study, the reaction was carried out at [BH₄⁻] = 0.05 M. In this context, it is reasonable expecting a change between a surface-controlled to a diffusion-controlled reaction between AuNPs free in solution and encapsulated in a PEM shell respectively. On the other hand, the fact that the catalytic activity per unit of surface was higher for 4.3 nm than 14.9 nm AuNPs when the AuNPs were free in solution but when encapsulated the smaller AuNPs were less catalytic could mean that the real available surface for catalysis in the case of the smaller AuNPs was less than for the bigger AuNPs. The smaller AuNPs could interact more with the polyelectrolyte shell and between each other than the bigger AuNPs leading to a decrease of the effective catalytic surface. In the two situations the polymeric shell would affect the transport of reactants or the surface of the catalyst.

The design of our PEM catalytic capsules allowed encapsulating prefabricated AuNPs in the interior of the hollow shell and made possible the comparison of different catalyst formulations. With our work, we have shown that decreasing the size of the catalyst increased the catalytic activity per unit of mass (k_obs (CAP1) > k_obs (CAP2)), nevertheless the reactivity of the AuNP surface for the smaller AuNPs decreased respect to the bigger AuNPs when the PEM shells contained MNPs. Therefore, when designing the formulation of catalysts based on polyelectrolyte capsules and other polymers it should be taken into consideration that the modification of the porous shell could be more efficient for the modulation of the catalytic activity than the tailoring of some physicochemical properties of the nanoparticles used as catalyst.^26,57

Conclusions

In conclusion, we have shown that PEM capsules loaded with AuNPs and multifunctional PEM capsules loaded with MNPs and AuNPs can be used as efficient catalysts in the reduction of hexacyanoferrate(III) by borohydride ions. The conversion of the redox reaction was 100%. The MNPs made the capsules easily recoverable from solution. First order kinetics were obtained for the reactions catalyzed by all kinds of PEM capsules shown in this work, similarly to the reported kinetics obtained with citrate stabilized AuNPs. However, a clear hindering of the catalytic activity was observed for the AuNPs within the PEM capsules, as compared with similar AuNPs free in solution. When the PEM shells were further loaded with MNPs a higher decrease of the catalytic activity was observed. Moreover, when the formulation of the catalysts included AuNPs of different sizes an increase of the catalytic activity was observed for 4.3 nm AuNPs compared with 14.9 nm AuNPs. However, when MNPs were entrapped within the PEM shell, the effect of the nanoparticle size was screened. These facts point out the effect of nanoparticle size and shell composition in the catalytic activity of PEM capsules.

Despite the findings regarding the slight decrease of catalytic activity for encapsulated AuNPs, there are potential advantages of these hybrid materials respect to the AuNPs free in solution. First, the evident modification of the catalytic activity by the presence of the polyelectrolyte shell and MNPS around the gold nanoparticles can be exploited to design catalysts with switchable catalytic activity by using polyelectrolyte shells with a controlled permeability.⁵⁸ In case MNPs are loaded in their shell, PEM capsules can be accumulated with a magnet, which can be used to control or enhance the catalytic activity in a specific location within a fluidic channel or a reactor. In other words, the catalyst could be transported to a designed target region, allowing for reaching local instead of global catalytic activity. In addition, as mentioned above, MCAPs also could be extracted from solution for recycling after usage by accumulation with a magnet.

Acknowledgements

This work was supported by LOEWE (grant Synchembio to WJP) and by the DAAD (grant to IH and WJP). Authors thank Dr Pablo del Pino and Dr Qian Zhang for the preparation of MNPs.

References

1. B. G. De Geest, N. N. Sanders, G. B. Sukhorukov, J. Demeester and S. C. De Smedt, Chem. Soc. Rev., 2007, 36, 636–649.
2. A. L. Becker, A. P. R. Johnston and F. Caruso, Small, 2010, 6, 1836–1852.
3. S. De Koker, L. J. De Cock, P. Rivera_Gil, W. J. Parak, R. A. Velty, C. Vervaet, J. P. Remon, J. Grooten and B. G. De Geest, Adv. Drug Delivery Rev., 2011, 63, 748–761.
4. M. Delcea, H. Möhwald and A. G. Skirtach, Adv. Drug Delivery Rev., 2011, 63, 730–747.
5. S. Carregal-Romero, M. Ochs, P. Rivera-Gil, C. Ganas, A. M. Pavlov, G. B. Sukhorukov and W. J. Parak, J. Controlled Release, 2012, 159, 120–127.
6. L. L. del Mercato, M. M. Ferraro, F. Baldassarre, S. Mancarella, V. Greco, R. Rinaldi and S. Leporatti, Adv. Colloid Interface Sci., 2014, 207, 139–154.
7. P. Rivera_Gil, S. D. Koker, B. G. De_Geest and W. J. Parak, Nano Lett., 2009, 9, 4398–4402.
8. W.-C. Liao, C.-H. Lu, R. Hartmann, F. Wang, Y. S. Sohn, W. J. Parak and I. Willner, ACS Nano, 2015, 9, 9078–9086.
9. P. Rivera-Gil, L. L. del Mercato, P. del Pino, A. Muñoz-Javier and W. J. Parak, Nano Today, 2008, 3, 12–21.
10. M. Ochs, S. Carregal-Romero, J. Rejman, K. Braeckmans, S. C. De Smedt and W. J. Parak, Angew. Chem., Int. Ed., 2013, 52, 695–699.
11. F. Caruso, D. Trau, H. Möhwald and R. Renneberg, Langmuir, 2000, 16, 1485–1488.
12. C. Y. Gao, X. Y. Liu, J. C. Shen and H. Möhwald, Chem. Commun., 2002, 1928–1929.
13. J. F. Shi, S. H. Zhang, X. L. Wang and Z. Y. Jiang, Chem. Commun., 2014, 50, 12500–12503.
14. A. M. Yu, I. Gentle, G. Q. Lu and F. Caruso, Chem. Commun., 2006, 2150–2152.
15. M. J. McShane, Methods Mol. Biol., 2011, 679, 147–154.
16. L. L. del Mercato, M. Carraro, A. Zizzari, M. Bianco, R. Miglietta, V. Arima, I. Viola, C. Nobile, A. Soraru, D. Vilona, G. Gigli, M. Bonchio and R. Rinaldi, Chem.–Eur. J., 2014, 20, 10910–10914.
17. D. G. Shchukin, E. Ustinovich, D. V. Sviridov, Y. M. Lvov and G. B. Sukhorukov, Photochem. Photobiol. Sci., 2003, 2, 975–977.
18. A. G. Skirtach, B. G. De_Geest, A. Mamedov, A. A. Antipov, N. A. Kotov and G. B. Sukhorukov, J. Mater. Chem., 2007, 17, 1050–1054.
19. C. S. Karamitros, A. M. Yashchenok, H. Moehwald, A. G. Skirtach and M. Konrad, Biomacromolecules, 2013, 14, 4398–4406.
20. T. S. Ahmadi, Z. L. Wang, A. Henglein and M. A. El-Sayed, Chem. Mater., 1996, 8, 1161–1163.
21. S. Carregal-Romero, J. Perez-Juste, P. Herves, L. Liz-Marzan and P. Mulvaney, Langmuir, 2009, 26, 1271–1277.
22. C. Vazquez-Vazquez, B. Vaz, V. Giannini, M. Perez-Lorenzo, R. A. Alvarez-Puebla and M. A. Correa-Duarte, J. Am. Chem. Soc., 2013, 135, 13616–13619.
23. Y. Mei, G. Sharma, Y. Lu, M. Ballauff, M. Drechsler, T. Irrgang and R. Kempe, Langmuir, 2005, 21, 12229–12234.
24. J. Oliver-Messeguer, L. C. Liu, S. Garcia-Garcia, C. Canos-Gimenez, I. Dominguez, R. Gavara, A. Domenech-Carbo, P. Concepcion, A. Leyva-Perez and A. Corma, J. Am. Chem. Soc., 2015, 137, 3894–3900.
25. M. Zahmakiran, Y. Tonbul and S. Ozkar, J. Am. Chem. Soc., 2010, 132, 6541–6549.
26. S. Carregal-Romero, N. Buurma, J. Perez-Juste, L. Liz-Marzan and P. Herves, Chem. Mater., 2010, 22, 3051–3059.
27. G. Y. Tonga, Y. D. Jeong, B. Duncan, T. Mizuhara, R. Mout, R. Das, S. T. Kim, Y. C. Yeh, B. Yan, S. Hou and V. M. Rotello, Nat. Chem., 2015, 7, 597–603.
28. S. H. Joo, J. Y. Park, C.-K. Tsung, Y. Yamada, P. Yang and G. A. Somorjai, Nat. Mater., 2009, 8, 126–131.
29. Z. W. Seh, S. H. Liu, S. Y. Zhang, K. W. Shah and M. Y. Han, Chem. Commun., 2011, 47, 6689–6691.
30. I. Pastoriza-Santos, J. Perez-Juste, S. Carregal-Romero, P. Herves and L. Liz-Marzan, Chem.–Asian J., 2006, 1, 730–736.
31. X. H. Li and M. Antonietti, Chem. Soc. Rev., 2013, 42, 6593–6604.
32. M. Sanles-Sobrido, M. A. Correa-Duarte, S. Carregal-Romero, B. Rodriguez-Gonzalez, R. A. Alvarez-Puebla, P. Herves and L. M. Liz-Marzan, Chem. Mater., 2009, 21, 1531–1535.
33. L. Zhou, C. Gao and W. J. Xu, Langmuir, 2010, 26, 11217–11225.
34. R. L. Oliveira, P. K. Kiyohara and L. M. Rossi, Green Chem., 2010, 12, 144–149.
35. W. T. Hu, B. C. Liu, Q. Wang, Y. Liu, Y. X. Liu, P. Jing, S. L. Yu, L. X. Liu and J. Zhang, Chem. Commun., 2013, 49, 7596–7598.
36. S. K. Shaikhutdinov, R. Meyer, M. Naschitzki, M. Baumer and H. J. Freund, Catal. Lett., 2003, 86, 211–219.
37. K. Q. Sun, S. W. Luo, N. Xu and B. Q. Xu, Catal. Lett., 2008, 124, 238–242.
38. B.-E. Pinchasik, H. Moehwald and A. G. Skirtach, Small, 2014, 10, 2670–2677.
39. Z. Lin, Z. Wu, X. Lin and Q. He, Chem.–Eur. J., 2016, 22, 1587–1591.
40. N. R. Jana, L. Gearheart and C. J. Murphy, Langmuir, 2001, 17, 6782–6786.
41. G. B. Sukhorukov, D. V. Volodkin, A. M. Günther, A. I. Petrov, D. B. Shenoy and H. Möhwald, J. Mater. Chem., 2004, 14, 2073–2081.
42. W. W. Yu, J. C. Falkner, C. T. Yavuz and V. L. Colvin, Chem. Commun., 2004, 2306–2307.
43. T. Pellegrino, L. Manna, S. Kudera, T. Liedl, D. Koktysh, A. L. Rogach, S. Keller, J. Rädler, G. Natile and W. J. Parak, Nano Lett., 2004, 4, 703–707.
44. P. Herves, M. Perez-Lorenzo, L. M. Liz-Marzan, J. Dzubiella, Y. Lu and M. Ballauff, Chem. Soc. Rev., 2012, 41, 5577–5587.
45. T. Freund, J. Inorg. Nucl. Chem., 1959, 9, 246–251.
46. M. Bhattacharjee, A. K. Bhattacharjee and M. K. Mahanti, Bull. Chem. Soc. Jpn., 1981, 54, 3566–3569.
47. M. Chatenet, F. Micoud, I. Roche and E. Chainet, Electrochim. Acta, 2006, 51, 5459–5467.
48. M. F. Bedard, D. Braun, G. B. Sukhorukov and A. G. Skirtach, ACS Nano, 2008, 2, 1807–1816.
49. S. Carregal-Romero, P. Rinklin, S. Schulze, M. Schäfer, A. Ott, D. Hühn, X. Yu, B. Wolfrum, K.-M. Weitzel and W. J. Parak, Macromol. Rapid Commun., 2013, 34, 1820–1826.
50. C. K. Tsung, J. N. Kuhn, W. Y. Huang, C. Aliaga, L. I. Hung, G. A. Somorjai and P. D. Yang, J. Am. Chem. Soc., 2009, 131, 5816–5822.
51. R. Reske, H. Mistry, F. Behafarid, B. R. Cuenya and P. Strasser, J. Am. Chem. Soc., 2014, 136, 6978–6986.
52. H. Tsunoyama, H. Sakurai, Y. Negishi and T. Tsukuda, J. Am. Chem. Soc., 2005, 127, 9374–9375.
53. M. Haruta and M. Date, Appl. Catal., A, 2001, 222, 427–437.
54. S. Back, M. S. Yeom and Y. Jung, ACS Catal., 2015, 5, 5089–5096.
55. X. Zhou, W. Xu, G. Liu, D. Panda and P. Chen, J. Am. Chem. Soc., 2010, 132, 138–146.
56. X. B. Wang, Y. Y. Wang, F. Tian, H. J. Liang, K. Wang, X. H. Zhao, Z. S. Lu, K. Jiang, L. Yang and X. D. Lou, J. Phys. Chem. C, 2015, 119, 15963–15976.
57. Y. Lu, Y. Mei, M. Drechsler and M. Ballauff, Angew. Chem., Int. Ed., 2006, 45, 813–816.
58. A. A. Antipov, G. B. Sukhorukov, S. Leporatti, I. L. Radtchenko, E. Donath and H. Möhwald, Colloids Surf., A, 2002, 198, 535–541.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra08171a

---