Multifunctional hollow polydopamine-based composites (Fe₃O₄/PDA@Ag) for efficient degradation of organic dyes

Abstract

As a newly appeared biomimetic platform and universal surface modification agent, polydopamine (PDA) has been paid great attention recently. However, the fabrication of multifunctional hollow PDA-based nanostructures remains a challenge and has been rarely reported. Herein, a facile method to prepare multifunctional hollow Fe₃O₄/PDA@Ag spherical composites is reported. The in situ embedding of Fe₃O₄ nanoparticles and the formation of a PDA shell have been achieved in a single system employing carboxylic-capped polystyrene (PS-COOH) spheres as a hard template, which can be selectively dissolved by tetrahydrofuran. Subsequently, silver nanoparticles were anchored on the PDA surface via in situ reduction by the PDA layers, leading to a multifunctional hollow PDA-based composite, i.e., Fe₃O₄/PDA@Ag. This newly prepared composite could efficiently catalyze the degradation of 4-nitrophenol and rhodamine B with NaBH₄ as a reducing agent, and the reaction rate constants were calculated using the pseudo-first-order reaction equation. Compared to non-hollow PS@Fe₃O₄/PDA@Ag, the hollow composite exhibited an enhanced performance toward both reactions without a significant loss in activity even after ten cycles. This multifunctional hollow PDA-based composite may find potential applications in other domains like heavy metal removal or antibacterial applications.

---

1. Introduction

Water pollution has always been one of the major environmental issues as a consequence of continued and reckless human activity such as the emission of large amounts of dye-bearing wastewater from tanneries and textile, paper and pulp industries.¹ Most water soluble dyes are non-biodegradable and pollute both surface water as well as groundwater, causing a serious threat to aquatic and human life.² Thus, the removal of dyes plays a significant role in wastewater treatment. The most commonly used dyes, such as 4-nitrophenol (4-NP) and rhodamine B (RhB), are highly carcinogenic and mutagenic.^3,4 To treat these dyes, various physicochemical methods, such as Fenton oxidation, adsorptive removal, photocatalytic decolorization and chemical reduction, have been employed.^5–9 Comparatively speaking, chemical reduction is a fast, effective and commonly used method for the decolorization of 4-NP and RhB from aqueous solutions. In most cases, metal and metal-oxide nanoparticles, such as silver, gold, palladium, platinum or bimetallic composites, have been used as efficient nanocatalysts toward the chemical reduction of both dyes.^10–13 By contrast, a great deal of research has been conducted employing the catalytic function of silver nanoparticles (Ag NPs) toward the degradation and reduction of 4-NP and RhB^14–18 owing to multiple aspects of the availability, cost and facile preparation processes. Nevertheless, the catalytic efficiency and stability of Ag NPs are often hampered by the aggregation or congregation of particles due to interparticle interactions through van der Waals forces and high surface energies.¹⁹

To overcome these shortcomings, Ag NPs are generally immobilized onto a variety of matrices including polymer, metal oxides, graphene oxide etc.^20,21 For instance, Xie and co-workers reported the fabrication of magnetic recyclable Fe₃O₄@polydopamine (PDA)-Ag core–shell microspheres, in which the catalytic Ag NPs are deposited onto the surface via an in situ reduction by mussel-inspired PDA layers. Meanwhile, the PDA layers could also stabilize the incorporated NPs and promote the adsorption of reactant molecules, leading to enhanced catalytic results, and the catalytic activity could be retained even after 27 cycles, regardless of the loss of catalyst caused by magnetic recovery.² However, hollow structured nanocatalysts possess a broader spectrum of attractive characteristics especially for diffusion/adsorption-involved catalytic applications where the accessibility of active sites is required.²² Indeed, the two major advantages of such structures, i.e., a hollow central cavity and a porous shell, can offer a versatile set of functional options. On the one hand, the central cavity can serve as a store for reactants and products and increase the flow rate.¹³ On the other hand, not only can the porous shell play an important role by acting as a physical separation between the active sites and thus improving the catalyst stability, it can also support or encapsulate other functional components such as magnetic particles.²³ Recently, Zhang et al. reported the fabrication of hollow C@PDA@Ag using a template-assisted strategy to prepare hollow carbon spheres, which exhibited an expected catalytic performance toward CO conversion and the reduction of 4-NP by NaBH₄ in aqueous solutions.²⁴ However, trying to remove the hard-template by heat treatment not only wasted lots of energy, but also led to the significant reduction of potentially useful nitrogen and phenolic hydroxyl groups derived from PDA layers. Accordingly, the search for an alternative “green” template and a mild, controllable solution-processed strategy to remove the template is of significance,²⁵ that is, it would fully retain the PDA initial performance and reduce energy consumption. Moreover, we tried to use magnetic separation to promote the efficiency of separating the catalysts from the reaction solution, leading to the avoidance of complicated filtration and centrifugation procedures.

Dopamine is a mimic of 3,4-dihydroxy-L-phenylalanine (L-DOPA) found in the adhesive protein (i.e., Mytilus edulis foot protein 5) secreted from mussels. PDA has the ability of adhering to almost all material surfaces to form a thin and surface-adherent polydopamine (PDA) coating under mild conditions.²⁶ It is well known that PDA layers have emerged as a versatile platform for secondary reactions for application in cell adhesion, protein immobilization, nanoparticle stabilization and membrane preparation.^27–30 Furthermore, PDA coatings have plenty of functional groups, such as amino and hydroxyl groups as well as π–π bonds, thus establishing further multiple modification. The good electrochemical behavior of a PDA coating with π–π stacking interactions can accelerate electron transfer, which can also serve as a reducing and capping agent to reduce noble metallic salts into metallic NPs.³¹

Herein, we present a facile, effective and controllable synthesis strategy to prepare multifunctional hollow PDA-based composites (Fe₃O₄/PDA@Ag); specifically, solution-dissolvable carboxylic-capped polystyrene (PS-COOH) microspheres were employed as a hard template for Fe₃O₄, and the PDA shell and subsequently preformed PDA coating acted as both a reducing agent and the substrate for the embedding of the Ag NPs. To illustrate this concept, Scheme 1 shows the synthetic procedure for the multifunctional hollow Fe₃O₄/PDA@Ag composite. This novel Fe₃O₄/PDA@Ag was tested for the degradation of 4-NP and RhB with NaBH₄ under aqueous conditions, and the reaction rate constants were calculated through the pseudo-first-order reaction equation. To demonstrate the superiority of the synthesized samples toward both reactions, the non-hollow PS@Fe₃O₄/PDA@Ag composite and different components of the hollow Fe₃O₄/PDA@Ag catalyst were also comparatively studied. Additionally, by virtue of the magnetic properties of the incorporated Fe₃O₄ NPs, the catalysts could be easily separated from the reaction solution using a magnet and then reused.

Scheme 1 Schematic demonstration of the preparation process of the multifunctional hollow Fe₃O₄/PDA@Ag composites.

2. Experimental section

2.1. Chemicals

Acrylic acid, THF, FeCl₃·6H₂O, FeCl₂·4H₂O, NH₃·H₂O (28 wt%) were obtained from Tianjin Kermel Chemical Reagent Factory, China. Styrene, methylacrylic acid, AgNO₃ and RhB were purchased from Sinopharm Chemical Reagent Co., Ltd., China. Dopamine hydrochloride was obtained from Sigma Chemistry Co., Ltd., China. NaBH₄ and 4-NP were provided by Aladdin Chemistry Co., Ltd., China. Absolute ethanol was bought from Tianjin Fuye Fine Chemical Reagent Co., Ltd., China. All chemicals were of analytical grade and used as received. De-ionized water was used throughout the experiments.

2.2. Preparation of hollow Fe₃O₄/PDA@Ag composite

The carboxylic-capped PS microspheres were synthesized according to our previous work.³² Firstly, 5 mL of a PS-COOH microspheres (1 g L⁻¹) ethanol solution and 70 mg FeCl₃·6H₂O were injected into 45 mL de-ionized water under mechanical stirring at 70 °C. The reaction system was maintained for 1 h before the solution was purged with nitrogen to remove oxygen. Then, 25 mg FeCl₂·4H₂O was added into the system. After 2 h, NH₃·H₂O was added into the system to adjust the pH to ∼9. The reaction was allowed to proceed for another 40 min under nitrogen protection, and the PS@Fe₃O₄ composites were prepared. Subsequently, when the system was ultrasonificated for 10 min, 0.15 g dopamine hydrochloride was added into the system and stirred for 24 h. Lastly, the PS@Fe₃O₄/PDA composites were separated using a magnet and washed with de-ionized water.

The as-prepared PS@Fe₃O₄/PDA composites were suspended into 30 mL THF solution. After stirring for 12 h, the PS-COOH microspheres were removed, and the resulting Fe₃O₄/PDA hollow spheres were separated with a magnet and washed with ethanol and de-ionized water. Then the products were dried at 80 °C in air overnight.

The preformed Fe₃O₄/PDA hollow spheres were treated with 50 mL AgNO₃ solution (2 g L⁻¹, 4 g L⁻¹ and 6 g L⁻¹) for 24 h at 25 °C. After being magnetically separated and washed with de-ionized water several times, Fe₃O₄/PDA@Ag hollow spheres were obtained and dried at 80 °C. The as-obtained Fe₃O₄/PDA@Ag hollow spheres with an initial dosage of Ag precursor varying from 0.1 to 0.3 g were marked as Fe₃O₄/PDA@Ag I, Fe₃O₄/PDA@Ag II and Fe₃O₄/PDA@Ag III, respectively.

For comparison purpose, Ag NPs were also supported on the surface of the PS@Fe₃O₄/PDA composites prepared with the same method and dosage under identical synthesis conditions. The fabrication methodology of the Fe₃O₄/PDA(200) hollow spheres was similar to that of the Fe₃O₄/PDA hollow spheres, but the dosage of dopamine was doubled. The pure Fe₃O₄ NPs were also synthesized under identical conditions.

2.3. Catalytic testing of the Fe₃O₄/PDA@Ag hollow spheres

The catalytic performance and versatility of the as-prepared Fe₃O₄/PDA@Ag hollow spheres were carefully investigated through the degradation of different organic dyes, i.e., the catalytic decolorization of 4-NP and RhB in the presence of excessive NaBH₄ at room temperature. The typical catalytic reaction proceeded as follows: 0.7 mL aqueous dye solution (5 mM L⁻¹ for 4-NP and RhB), 17 mL de-ionized water and 1.3 mL of a freshly prepared NaBH₄ aqueous solution (0.2 M L⁻¹) were mixed in a three-necked flask under nitrogen atmosphere followed by the addition of 1 mg mL⁻¹ of the Fe₃O₄/PDA@Ag hollow spheres aqueous solution (1.0 mL for 4-NP and RhB). As the catalytic reaction proceeded, the color of the corresponding solution faded gradually with time and the catalytic activity was monitored by an ultraviolet-visible (UV-Vis) spectrophotometer at the maximum adsorption peaks of the dyes (400 nm for 4-NP and 555 nm for RhB).

2.4. Recycling test of the Fe₃O₄/PDA@Ag hollow spheres

The recyclability of the catalyst was also investigated by consecutively reusing the catalysts. After the reduction of the dyes for 5.0 min was achieved, the catalysts were collected from the mixture with a magnet, washed with de-ionized water and then dried for the next cycle. This procedure was repeated 10 times.

2.5. Characterization

Field emission scanning electron microscopy (FESEM, JSM-7800F electron microscope, JEOL, Japan) and transmission electron microscopy (TEM, JEM-2000EX electron microscope, JEOL, Japan) were applied to characterize the morphology and size of the prepared samples. The surface elemental composition was obtained using an X-Max50 energy dispersive X-ray analyzer (EDS, Oxford, UK) in the form of pressed pellets and selecting all the area in the field of the electron microscope. X-ray diffraction (XRD) patterns were recorded by a Shimadzu XRD-6100 diffractometer with CuKa radiation (λ = 1.54060 Å) from 10° to 80° at a scanning speed of 5° min⁻¹. Fourier transform infrared (FT-IR) spectra were recorded on a one-B FT-IR spectrometer over KBr pressed pellets. X-ray photoelectron spectroscopy (XPS) measurements were carried out using a Thermo Scientific ESCALAB250 spectrometer (Thermo VG, USA) with monochromatic AlKα radiation (1486.6 eV). UV-Vis absorption spectra were recorded on an Agilent Cary 60 spectrophotometer. The specific surface areas of the materials were calculated from the adsorption data in a pressure range of P/P₀ = 0.05–0.3 using the Brunauer–Emmett–Teller (BET) model. Magnetism was analyzed using a Lake Shore 7410 vibrating sample magnetometer (VSM) at room temperature.

3. Results and discussion

3.1. Structural confirmation of the multifunctional hollow composites

The FT-IR spectrum of the PS-COOH microspheres is shown in Fig. 1a, and the main peaks are located at 3026, 2923, 1493, 1453 and 697 cm⁻¹. In addition, a peak appears at 1735 cm⁻¹, suggesting that the carboxylic groups were successfully modified on the surface, which is beneficial for the deposition of Fe₃O₄ NPs in the next step.³³ As a matter of fact, once NH₃·H₂O was introduced into the reaction system, the alkalinity of the mixture increased, resulting in the hydrolysis of both the Fe³⁺ and Fe²⁺ ions. Therefore, Fe(OH)₃ and Fe(OH)₂ were produced under alkaline conditions (the pH was adjusted to 9 by adding NH₃·H₂O gradually), and finally Fe₃O₄ NPs formed through a dehydration process.³⁴ By this controllable means, the newly formed Fe₃O₄ NPs could be deposited on the surface of the PS-COOH microspheres.

Fig. 1 FT-IR spectra of the (a) PS microspheres; (b) PS@Fe₃O₄/PDA composites; (c) Fe₃O₄/PDA hollow spheres; and (d) the Fe₃O₄/PDA@Ag hollow spheres.

Subsequently, through directly utilizing this basic reaction system, the self-polymerization of dopamine could also be conducted, leading to the formation of a stable coating on the surface of the PS@Fe₃O₄ spheres.³⁵ The FT-IR spectrum of the PS@Fe₃O₄/PDA composites is shown in Fig. 1b. It is clear that, except for the feature peaks at 3026, 2923, 1493, 1453 and 697 cm⁻¹ for the PS-COOH template, the distinct absorption peaks at 1645 cm⁻¹ and 1290 cm⁻¹ can be attributed to the N–H stretching and the C–O stretching of the phenolic hydroxyl group, respectively.^29,36 By contrast, the peak at 3420 cm⁻¹ for the PS@Fe₃O₄/PDA composite was broader than that of the PS-COOH spheres, possibly as a result of the overlapping of the hydroxyl groups, the water adsorbed in the PDA polymer and the amine groups of the PDA layers.³⁷ Moreover, a peak at 585 cm⁻¹ is due to the stretching mode of the Fe–O bond.³⁸ Additionally, Fig. 2b shows the XRD pattern of the resulting Fe₃O₄/PDA hollow spheres; six diffraction peaks appear at 2θ = 30.2°, 35.5°, 43.2°, 53.6°, 57.1° and 62.6°, which can be assigned to the (220), (311), (400), (422), (511) and (440) Bragg diffractions of the cubic lattice of the Fe₃O₄ NPs (JCPDS no. 19-0629), respectively. This indicates that Fe₃O₄ NPs were synthesized inside the PDA shell. Noticeably, combining the FT-IR and XRD analysis results, it can be considered that a non-hollow PS@Fe₃O₄/PDA composite was successfully prepared from a single preparation system.

Fig. 2 XRD patterns of the (a) PS@Fe₃O₄/PDA composites, (b) Fe₃O₄/PDA hollow spheres and (c) the Fe₃O₄/PDA@Ag hollow spheres.

After the PS@Fe₃O₄/PDA composites were added into the THF solution and stirred for 12 h, not only the PS-COOH template spheres were removed, but the PDA coating also retained the initial performance. As shown in Fig. 2, compared to a distinct peak at 2θ = 22.2° which could be due to the polymeric PS template for the PS@Fe₃O₄/PDA composites (Fig. 2a), this peak in the Fe₃O₄/PDA hollow spheres was considerably reduced (Fig. 2b); this indicates that hollow Fe₃O₄/PDA spheres were successfully obtained at this stage. Meanwhile, several weak peaks can be observed at 697, 1453, 1493, 2923 and 3026 cm⁻¹ in Fig. 1c, this corresponds to tiny residual PS chains. Possibly, a number of pores in the PDA shell were blocked by the Fe₃O₄ NPs, therefore rendering it difficult to dissolve the PS chains completely to allow them to diffuse out of the sphere shell. It should be noted, however, that in Fig. 1c, the peak located at 1645 for the N–H stretching was a little decreased. This indicates that the nitrogen functional group could have been fully preserved.

Subsequently, utilizing a hollow Fe₃O₄/PDA as a matrix with the dual roles of reducing and stabilizing functions, the Ag NPs could be anchored onto the PDA surface easily, leading to the controllable deposition and dispersion of Ag NPs within the PDA layers.³⁹ That is, the use of PDA permitted the organization of size-controllable Ag NPs assemblies through the balance of attractive and repulsive forces between building blocks. As shown in Fig. 2c, a typical XRD pattern of the Fe₃O₄/PDA@Ag hollow spheres exhibits peaks at 2θ angles of 37.9°, 44.1°, 64.3° and 77.2°, corresponding to the (111), (200), (220) and (311) crystalline planes of the fcc structure of Ag (JCPDS no. 04-0783), respectively. There was no other peak in the XRD pattern except for those that related to magnetite and the Ag NPs, revealing the high purity of the Ag loaded magnetic particles.

In addition, Fig. 3a shows the SEM image of the PS@Fe₃O₄/PDA@Ag composites. They exhibit a uniform spherical shape. The magnified image reveals that the surface of the PS@Fe₃O₄/PDA@Ag composites is slightly rough and embossed (inset of Fig. 3a); this is a feature of the Ag NPs modifying the surface of the PDA. The diameter of the PS@Fe₃O₄/PDA@Ag composites was measured to be 910 nm, suggesting the original PS nanosphere was successfully coated by the PDA shell. Fig. 3b shows the corresponding TEM image of the PS@Fe₃O₄/PDA@Ag composites. It is difficult to resolve the interface between the PDA shell and PS core due to their similar electron contrast, and only solid composites with a spherical shape can be seen, whereas many inorganic nanoparticles randomly decorate the PDA shell. Fig. 3c shows the morphology of the resulting hollow spheres. Some of the spheres collapsed with many random folds due to the high-vacuum conditions under the SEM measurements (inset of Fig. 3c), suggesting the removal of the PS core and survival of the PDA shell during the etching process. In the corresponding TEM image (Fig. 3d), the appearance of a large void further confirms that the Fe₃O₄/PDA@Ag spheres possess a well-defined hollow structure, and the thickness of Fe₃O₄/PDA@Ag shell was estimated to be 90 nm.

Fig. 3 (a) SEM image of the PS@Fe₃O₄/PDA@Ag composites; (b) TEM image of the PS@Fe₃O₄/PDA@Ag composites; (c) SEM image of the Fe₃O₄/PDA@Ag hollow spheres; (d) TEM image of the Fe₃O₄/PDA@Ag hollow spheres; (e) magnified TEM image of the Fe₃O₄/PDA@Ag hollow spheres; (f)–(h) SEM image and corresponding element mappings of the Fe₃O₄/PDA@Ag hollow spheres. The insets show the corresponding magnified image.

A magnified image shows that Ag and Fe₃O₄ NPs randomly decorate the hollow spheres (Fig. 3e). Also, as shown by Fig. 3f–h, for the element mappings of Ag and Fe in the sample, a homogenous distribution of the Ag and Fe elements in the Fe₃O₄/PDA@Ag composite can be visualized; this confirms that the Ag and Fe₃O₄ NPs were distributed homogenously rather than forming respective aggregates. Because of their similar electron contrast and spherical shape, it is difficult to identify them via different shades. To solve the problem, EDS analysis (Fig. S2a†) indicates that the apparent Ag concentration was 20.98%, and its weight percentage is about 11.04%. Hence, Fe₃O₄/PDA hollow spheres with Ag particles assembled on the outer surface were synthesized successfully. In addition, when the Fe₃O₄ NPs started to form under alkaline conditions, the polymerization of the dopamine monomer proceeded; therefore, the Fe₃O₄ NPs were coated by a thin PDA shell, and the size was measured to be 23.7 nm. The Ag NPs are indicated with red boxes and the Fe₃O₄ NPs were indicated by blue boxes in Fig. 3e. This shows that the small Ag NPs (∼35 nm) were densely and uniformly distributed on the PDA surface. No obvious aggregation of the Ag NPs can be observed in the TEM view.

Simultaneously, XPS measurements were employed to characterize the surface chemical compositions and valence states of the Fe₃O₄/PDA hollow spheres and the Fe₃O₄/PDA@Ag hollow spheres (Fig. 4A). The survey spectrum of the Fe₃O₄/PDA@Ag hollow spheres in Fig. 4A(a) reveals the presence of C, N, O and Ag. Clearly, as compared to the XPS spectrum of the Fe₃O₄/PDA hollow spheres (Fig. 4A(b)), the Ag element was present. Accordingly, in the high-resolution Ag 3d XPS spectrum (inset Fig. 4A), two bands at ca. 368.19 and 374.19 eV were observed; these two bands could be ascribed to the Ag 3d_5/2 and Ag 3d_3/2 binding energies of the metallic Ag (Ag°), confirming the existence of metallic Ag in these newly designed multifunctional hollow Fe₃O₄/PDA@Ag spheres. The N 1s core-level spectrum of the XPS measurements (Fig. 4B) can be curve-fitted into two peak components with binding energy at 398.6 and 399.9 eV, attributable to the pyridine-N and pyrrolic-N, respectively,^40,41 which is in good agreement with the analysis results of the N–H stretching peak in the FT-IR spectra, indicating that the PDA coating retained the initial performance.

Fig. 4 (A) XPS survey spectra of (a) the Fe₃O₄/PDA hollow spheres and (b) the Fe₃O₄/PDA@Ag hollow spheres. The insets show the corresponding narrow scan for the Ag 3d peaks. (B) Core-level spectrum of N 1s in the Fe₃O₄/PDA@Ag hollow spheres.

3.2. Catalytic test for dye degradation over the synthesized samples

The reduction of the heterocyclic aromatic compounds is commonly applied to study the catalytic performance of metal-based nanocatalysts. On the other hand, considering the environmental and health concerns caused by the dyes and pigments mentioned above, we thus chose the reduction and decoloration of different dyes, such as 4-NP and RhB, to evaluate the catalytic activity and versatility of the as-prepared Fe₃O₄/PDA@Ag hollow spheres.

Firstly, we took the reduction of 4-NP as an example to introduce the catalytic reaction system. The aqueous solution of 4-NP exhibited an absorbance peak at 317 nm.^16,42 As demonstrated in Fig. 5a, after the addition of NaBH₄, the peak immediately red-shifts to 400 nm due to the formation of the 4-nitrophenolate ion under alkaline conditions, which corresponds to a color change from light yellow to intense yellow. When the catalysts were added, the intensity of the absorption peak at 400 nm gradually decreased with time; meanwhile, a new absorption peak relating to 4-aminophenol (4-AP) appeared at 300 nm (Fig. 5a). The solution color turned from intense yellow to colorless. Additionally, the UV-Vis spectra show an isosbestic point for the two absorption bands, indicating that the nitro compound was gradually converted to aminophenol without any side reactions. After completion of the reaction, the peak at 400 nm completely disappeared. Therefore, the reaction progress can be monitored by recording the UV-Vis absorption spectra of the reaction solution with respect to reaction time.

Fig. 5 UV-Vis absorption spectra of (a) 4-NP under different conditions; time-dependent UV-Vis absorption spectra (b) without catalyst and in the presence of (c) the Fe₃O₄/PDA hollow spheres; (d) the PS@Fe₃O₄/PDA@Ag composites; (e) the Fe₃O₄/PDA@Ag I hollow spheres; (f) the Fe₃O₄/PDA@Ag II hollow spheres; and (g) Fe₃O₄/PDA@Ag III hollow spheres. (h) The relationship between ln(A_t/A₀) and the reaction time at different conditions.

To study the reduction rate of 4-NP, the evolution of the catalytic reduction reaction was monitored by recording the absorbance of 4-NP. Fig. 5c–g illustrates the reduction reaction of 4-NP observed at different time intervals using 1 mg of catalyst. The reactions were performed under ambient temperature in all cases at pH = 7. The peak of 4-NP gradually decreased when the peak of 4-AP locating at 300 nm appears and increases in intensity.

To prove the role the Fe₃O₄/PDA@Ag hollow spheres played in the reaction system and the synergistic effect between the Ag NPs, PDA species and hollow morphology of the composite, contrast experiments were also carried out. Firstly, as shown in Fig. 5b and S6a,† the reactions were completed within 150 min in the absence of catalyst and in the presence of pure Fe₃O₄ NPs, respectively. Intriguingly, the result of the catalytic effect using the Fe₃O₄ NPs was the same as the reaction without catalyst. Therefore, the amount of Fe₃O₄ NPs could not influence the catalytic reaction. The study then focused on magnetic separation and improving the recovery efficiency.

Secondly, the preparation procedure of the Fe₃O₄/PDA(200) hollow spheres was the same as the Fe₃O₄/PDA hollow spheres but the dosage of dopamine was doubled. The TEM images of the Fe₃O₄/PDA hollow spheres and the Fe₃O₄/PDA(200) hollow spheres are shown in Fig. S7.† The thickness of the Fe₃O₄/PDA hollow spheres and the Fe₃O₄/PDA(200) hollow spheres was about 100 and 200 nm, respectively. As shown in Fig. S8,† 87 and 55 min were required to completely reduce 4-NP, respectively. Compared to the Fe₃O₄/PDA hollow spheres, the Fe₃O₄/PDA(200) hollow spheres exhibited an enhanced performance toward both reactions. This indicated that the amount of PDA could influence the catalytic reaction.

More importantly, it can be seen from Fig. 5c–e that 87, 5 and 7 min were needed for the reaction systems of the Fe₃O₄/PDA hollow spheres, Fe₃O₄/PDA@Ag hollow spheres and the PS@Fe₃O₄/PDA@Ag composites, respectively. It was illustrated that the Fe₃O₄/PDA@Ag hollow spheres had the highest catalytic activity in those types of catalysis for the reduction of 4-NP. Undoubtedly, this confirms that the catalytic activity was mainly due to the Ag species. This phenomenon can be explained by the following aspects: firstly, the existence of the PDA support improved the dispersion of Ag owing to the fact that the amino and hydroxyl functional groups could reduce metallic species in situ without introducing additional reducing agents or surfactants, rendering a bigger active contact surface; secondly, synergistic effects may exist between Ag, PDA species and the hollow morphology of the composite. A detailed investigation of this phenomenon will be conducted in our further work.

Nitrogen sorption isotherms were used to investigate the surface areas and pore structures of the as-prepared hybrid materials. The specific surface area, pore volumes and pore sizes are summarized in Table 1. As shown in Fig. 6 and Table 1, both the Brunauer–Emmett–Teller (BET) surface areas and the total pore volumes of the Fe₃O₄/PDA@Ag hollow spheres were higher than that of the non-hollow PS@Fe₃O₄/PDA@Ag composites. Possibly, as the aqueous reaction system required favorable nanostructure and interfacial characteristics, it benefited from the hollow structure of the resultant nanocomposite. Accordingly, taking advantage of the hollow nanostructures, relative to PS@Fe₃O₄/PDA@Ag, the designed hollow Fe₃O₄/PDA@Ag exhibited an enhanced catalytic performance in the reduction of 4-NP (Fig. 5d–e).

Table 1 Nitrogen adsorption analysis results of the Fe₃O₄/PDA@Ag hollow spheres and the PS@Fe₃O₄/PDA@Ag composites

Samples	BET surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore size (nm)
Fe3O4/PDA@Ag	26.76	0.05385	8.04851
PS@Fe3O4/PDA@Ag	12.00	0.01943	6.47762

---

Fig. 6 Nitrogen adsorption–desorption isotherms and the corresponding pore size distribution curve (inset) of the Fe₃O₄/PDA@Ag and PS@Fe₃O₄/PDA@Ag composites.

Meanwhile, as shown in Fig. 5e–g, with an increase in the Ag precursor dosage, the time needed for the completion of the decoloration was decreased. The content of Ag in the nanocomposite and reaction rate constant k both increased (Table S1† and 5h). Additionally, the EDS analysis results also verified this conclusion (Fig. S2†). In Fig. S2,† the initial dosage of the Ag precursor increased from 0.1 to 0.3 g and the Ag NPs were 11 wt%, 18 wt% and 25 wt% in Fe₃O₄/PDA@Ag I, Fe₃O₄/PDA@Ag II and Fe₃O₄/PDA@Ag III, respectively. As such, it can be concluded that as the Ag dosage was increased, the exposed active site or surface increased and therefore the catalytic reaction was facilitated. In this context, the optimal dose of the Ag precursor in our system was 0.3 g.

On the other hand, as the initial concentration of NaBH₄ was far higher than the reactants, it is considered that the reaction rate remained constant throughout the whole reaction process. Thus, the reaction is assumed to be independent of the concentration of NaBH₄,^43,44 and a pseudo first-order kinetic equation could be applied to evaluate the catalytic rate, and the L-H model can also be simplified to a pseudo-first-order expression.⁴⁵ The absorption intensity of 4-NP is proportional to the concentration in the medium. Hence, the kinetic equation for the reduction could be written as follows:

A_t and A₀ are the absorbance values of 4-NP initially at time t and 0, respectively. As shown in Fig. 5e, a good linear relationship between ln(A_t/A₀) and the reaction time was displayed, which is consistent with pseudo-first order kinetics. Therefore, the apparent reaction rate constant k can be directly obtained from the slope of the linear plots. The rate constant k (Fig. 5e(▲)) of the Fe₃O₄/PDA@Ag catalysts was calculated to be 0.600 min⁻¹ at 25 °C, which was clearly higher than that of the Fe₃O₄/PDA catalysts with a k value of 0.032 min⁻¹ (Fig. 5e(■)) and PS@Fe₃O₄/PDA@Ag with a k value of 0.410 min⁻¹ (Fig. 5e(●)).

To further demonstrate the potential of the synthesized samples, the catalytic performance of the resultant catalysts for the reduction of RhB was also investigated. For obtaining accurate data by means of UV-Vis spectroscopy based on Lambert–Beer Law, the concentrations of RhB and the amount of catalyst were all same to the reaction system of 4-NP, and the time-dependent UV-Vis absorption spectra and reaction kinetic data are presented in Fig. S3 and Table S1.† Noticeably, in the presence of the Fe₃O₄/PDA@Ag hollow spheres, just 6 min was required to completely reduce RhB, whereas it took 70 and 12 min to reduce RhB for Fe₃O₄/PDA and PS@Fe₃O₄/PDA@Ag, respectively. All the results indicated that Fe₃O₄/PDA@Ag possesses excellent catalytic properties for the degradation of organic dyes under aqueous conditions.

According to the above discussion, one of the obstacles hindering the application of metal NPs is that they are difficult to separate from the reaction solution. Therefore, the reusability of the Fe₃O₄/PDA@Ag hollow spheres needs to be investigated. In our experiments, Fe₃O₄ NPs were synthesized during the preparation procedure, which endowed the catalysts with magnetic properties. Fig. 7 shows the magnetization curve of the Fe₃O₄/PDA@Ag hollow spheres. The nearly zero coercivity and the reversible hysteresis behavior indicate their paramagnetic nature at room temperature. At 295 K, the saturation magnetization value of the Fe₃O₄/PDA@Ag hollow spheres is 10 emu g⁻¹ (the content of Fe₃O₄ is about 6.64 wt%), which permits the catalysts to be easily recycled from the reaction solution by an external magnetic field, and enables the catalytic decolorization of 4-NP and RhB in the presence of excess NaBH₄ at room temperature (inset of Fig. 7).

Fig. 7 Room temperature magnetization curve of Fe₃O₄/PDA@Ag.

During the reduction of 4-NP and RhB, when the reduction reaction finished, the catalysts could be magnetically separated from the reaction solution and rinsed with water twice; then they were reused for the next run under identical conditions. After reaction with 1.0 mg catalysts for 5.0 min each cycle, the absorbance intensity of the reaction solution was immediately measured. The results demonstrated that there was no significant loss of activity for the reduction of 4-NP and RhB over the Fe₃O₄/PDA@Ag hollow spheres even after ten successive cycles and the conversion of each cycle was above 95%, clearly indicating that the Fe₃O₄/PDA@Ag catalysts have good reusability (Fig. 8). This was mainly due to the fact that the Ag NPs in the hollow spheres were stabilized by the PDA nanonetwork with nearly no loss of complexing groups, leading to a high dispersion and stabilization during the reused cycles by virtue of a mild solution strategy to remove the hard template; that is, distinct aggregation would have not occurred and accordingly their catalytic activity was well maintained during successive runs.

Fig. 8 Recycling catalytic test of Fe₃O₄/PDA@Ag for probing dyes.

4. Conclusion

A facile, controllable synthesis strategy to fabricate multifunctional hollow PDA-based composites of Fe₃O₄/PDA@Ag has been successfully developed. Active Ag NPs were anchored onto PDA shells via in situ reduction and stabilization by PDA layers, leading to the formation of Ag NPs assemblies with promising applications for the catalytic degradation of organic dyes. In contrast to the non-hollow counterpart of PS@Fe₃O₄/PDA@Ag, the designed hollow Fe₃O₄/PDA@Ag composite exhibited a much enhanced catalytic performance; and the rate constant was calculated through the traditional pseudo-first-order reaction equation. More interestingly, after the reaction, Fe₃O₄/PDA@Ag could be quickly recovered from the reaction solution owing to the room temperature paramagnetic properties endowed by the Fe₃O₄ NPs that were pre-loaded onto the interior surface of the hollow PDA spheres. During the recycling test, no significant decrease in catalytic activity was detected, even after recycling 10 times. Due to the combination of excellent catalytic activity, good reusability and high stability, together with a mild preparation procedure, this type of hollow composites is of potential interest for specific fields.

Acknowledgements

Financial support from the National Natural Science Foundation of China (21446001), the Program for Liaoning Innovative Research Team in University (LT2013012) and the Program for Liaoning Excellent Talents in University (LJQ2014056) is highly appreciated.

References

1. B. Ramalingam, M. M. Khan, B. Mondal, A. B. Mandal and S. K. Das, ACS Sustainable Chem. Eng., 2015, 3, 2291–2302.
2. Y. J. Xie, B. Yan, H. L. Xu, J. Chen, Q. X. Liu, Y. H. Deng and H. B. Zeng, ACS Appl. Mater. Interfaces, 2014, 6, 8845–8852.
3. P. Mahamallik and A. Pal, RSC Adv., 2015, 5, 78006–78016.
4. S. W. Chook, C. H. Chia, C. H. Chan, S. X. Chin, S. Zakaria and M. S. Sajab, RSC Adv., 2015, 5, 88915–88920.
5. Z. M. Cui, Z. Chen, C. Y. Cao, L. Jiang and W. G. Song, Chem. Commun., 2013, 49, 2332–2334.
6. S. H. Yuan, Y. Fan, Y. C. Zhang, M. Tong and P. Liao, Environ. Sci. Technol., 2011, 45, 8514–8520.
7. A. Benhouria, M. A. Islam, H. Zaghouane-Boudiaf, M. Boutahala and B. H. Hameed, Chem. Eng. J., 2015, 270, 621–630.
8. L. Obeid, N. E. Kolli, D. Talbot, M. Welschbillig and A. Bée, J. Colloid Interface Sci., 2015, 457, 218–224.
9. M. Zhang, C. C. Chen, W. H. Ma and J. C. Zhao, Angew. Chem., Int. Ed., 2008, 47, 9730–9733.
10. W. Zhou, Y. Zhou, Y. Liang, X. H. Feng and H. Zhou, RSC Adv., 2015, 5, 50505–50511.
11. Y. Chen, Q. H. Wang and T. M. Wang, Dalton Trans., 2015, 44, 8867–8875.
12. C. Zhang, Y. M. Zhou, Y. W. Zhang, Z. W. Zhang, Y. M. Xu and Q. L. Wang, Powder Technol., 2015, 284, 387–395.
13. S. L. Li, H. Li, J. Liu, H. L. Zhang, Y. M. Yang, Z. Y. Yang, L. Y. Wang and B. D. Wang, Dalton Trans., 2015, 44, 9193–9199.
14. Z. Z. Wang, S. R. Zhai, B. Zhai, Z. Y. Xiao, F. Zhang and Q. D. An, New J. Chem., 2014, 38, 3999–4006.
15. J. Cao, S. L. Mei, H. Jia, A. Ott, M. Ballauff and Y. Lu, Langmuir, 2015, 31, 9483–9491.
16. N. N. Meng, S. J. Zhang, Y. F. Zhou, W. Y. Nie and P. P. Chen, RSC Adv., 2015, 5, 70968–70971.
17. M. Russo, F. Armetta, S. Riela, D. C. Martino, P. L. Meo and R. Noto, J. Mol. Catal. A: Chem., 2015, 408, 250–261.
18. Z. Z. Wang, S. R. Zhai, B. Zhai, Q. D. An and S. W. Li, J. Sol-Gel Sci. Technol., 2015, 3, 680–692.
19. K. Zhao, C. J. Wu, Z. W. Deng, Y. C. Guo and B. Peng, RSC Adv., 2015, 5, 52726–52736.
20. Y. Zhou, Y. H. Zhu, X. L. Yang, J. F. Huang, W. Chen, X. M. Lv, C. Y. Li and C. Z. Li, RSC Adv., 2015, 5, 50454–50461.
21. T. T. S. Bui, Y. Kim, S. Kim and H. Lee, RSC Adv., 2015, 5, 75272–75280.
22. J. C. Song, F. F. Xue, Z. Y. Luc and Z. Y. Sun, Chem. Commun., 2015, 51, 10517–10520.
23. S. Mezzavilla, C. Baldizzone, K. J. J. Mayrhofer and F. Schuth, ACS Appl. Mater. Interfaces, 2015, 7, 12914–12922.
24. Q. Zhang, C. Zhou and A. S. Huang, RSC Adv., 2015, 5, 91056–91061.
25. T. J. Yao, Q. Zuo, H. Wang, J. Wu, B. F. Xin, F. Cui and T. Y. Cui, J. Colloid Interface Sci., 2015, 450, 366–373.
26. S. D. Ma, J. Feng, W. J. Qin, Y. Y. Ju and X. G. Chen, RSC Adv., 2015, 5, 53514–53523.
27. Q. Liu, N. Y. Wang, J. Caro and A. S. Huang, J. Am. Chem. Soc., 2013, 135, 17679–17682.
28. A. S. Huang, Q. Liu, N. Y. Wang and J. Caro, J. Mater. Chem. A, 2014, 2, 8246–8251.
29. N. Y. Wang, Y. Liu, Z. W. Qiao, L. Diestel, J. Zhou, A. S. Huang and J. Caro, J. Mater. Chem. A, 2015, 3, 4722–4728.
30. C. F. Yuan, Q. Liu, H. F. Chen and A. S. Huang, RSC Adv., 2014, 4, 41982–41988.
31. J. J. Zhou, B. Duan, Z. Fang, J. B. Song, C. X. Wang, P. B. Messersmith and H. W. Duan, Adv. Mater., 2014, 26, 701705.
32. J. M. Li, W. F. Ma, C. Wei, J. Guo, J. Hu and C. C. Wang, J. Mater. Chem., 2011, 21, 5992–5998.
33. Y. Yang, Y. Chu, F. Y. Yang and Y. P. Zhang, Mater. Chem. Phys., 2005, 92, 164–171.
34. T. J. Yao, T. Y. Cui, H. Wang, L. X. Xu, F. Cui and J. Wu, Nanoscale, 2014, 6, 7666–7674.
35. B. Y. Wang, M. Zhang, W. Z. Li, L. L. Wang, J. Zheng, W. J. Gan and J. L. Xu, Dalton Trans., 2015, 44, 17020–17025.
36. Z. N. Li, C. J. Wu, K. Zhao, B. Peng and Z. W. Deng, Colloids Surf., A, 2015, 470, 80–91.
37. S. D. Ma, J. Feng, W. J. Qin, Y. Y. Ju and X. G. Chen, RSC Adv., 2015, 5, 53514–53523.
38. B. Yu, F. Zhai, H. L. Cong, D. Wang, Q. H. Peng, S. J. Yang and R. X. Yang, RSC Adv., 2015, 5, 77860–77865.
39. S. Q. Xiong, Y. Wang, J. Zhu, J. R. Yu and Z. M. Hu, Langmuir, 2015, 31, 5504–5512.
40. K. G. Qu, Y. Zheng, S. Dai and S. Z. Qiao, Nanoscale, 2015, 7, 12598–12605.
41. S. N. Du, Z. J. Liao, Z. L. Qin, F. Zuo and X. H. Li, Catal. Commun., 2015, 72, 86–90.
42. Z. Z. Wang, S. R. Zhai, J. L. Lv, H. X. Qi, W. Zheng, B. Zhai and Q. D. An, RSC Adv., 2015, 5, 74575–74584.
43. J. Ge, Q. Zhang, T. Zhang and Y. Yin, Angew. Chem., Int. Ed., 2008, 120, 9056–9060.
44. K. S. Shin, Y. K. Cho, J.-Y. Choi and K. Kim, Appl. Catal., A, 2012, 413, 170–175.
45. N. Muthuchamy, A. Gopalan and K. P. Lee, RSC Adv., 2015, 5, 76170–76181.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra05954f

---