Introduction

The engineered nanoparticles (NPs) are increasingly used in industrial and consumer products due to their unique properties. Among engineered NPs, silver NPs (AgNPs) are one of the commercially used nanomaterials in detergents, plastics, and textiles due to their excellent antimicrobial properties1,2,3. The production and the increasing use of AgNPs obviously results in their release into the environment, leading to a risk to the environment due to their toxic effects4,5. Thus, the removal of AgNPs from water is urgently needed.

To date, a variety of approaches such as filtration6, cloud point extraction7,8, and adsorption technology9,10,11,12,13,14,15,16,17,18, have been developed for the separation and extraction of metal NPs from water. Among these, adsorption is a promising one for the removal of AgNPs from aqueous solution due to its low cost and ease operation. For example, Valiyaveettil’s group used biomimetic metal oxides10, cellulose and polyvinyl alcohol/gluten-based nanofibers11,12,13, polyethyleneimine modified carbon spheres14 and amine-modified block copolymers15 for adsorption of AgNPs from water. Khan et al. illustrated the potential of the resistant bacterial species Aeromonas punctata for the effective removal of AgNPs16. More recently, the group of Černík reported the use of methane plasma treated electrospun nanofibres for the removal of various NPs in water17,18. Above studies demonstrate the high potential of adsorption technology for the effective removal of the engineered NPs.

On the other hand, the noble metallic NPs have been received great attention in the field of catalysis due to their superior chemical and physical properties. In particular, catalysis over AgNPs has received the most attention due to their highly catalytic performance19,20,21,22. However, their use in practical applications is highly limited due to high cost and limited resources. Therefore, from the view point of economic and environmental reasons, recovery and reuse of expensive noble metal catalysts play a key role in both heterogeneous and homogenous catalysis. However, no attempts have been made to explore the magnetic separation of AgNPs from aqueous media as a new strategy toward stable and recyclable catalyst for the reduction of toxic pollutants.

Here, for the first time, we reported a novel magnetic Fe3O4@polydopamine (Fe3O4@PDA) core-shell microsphere for the removal of AgNPs from the aqueous solution, and the obtained AgNPs-Fe3O4@PDA was used as catalyst for the reduction of methylene blue (MB) by NaBH4. As compared to other technologies, the magnetic separation removal of pollutants from water promises advantages of low-cost Fe3O4 as raw materials, easy to prepare and scale up, and easy separation from aqueous solution by a magnetic field23. PDA is utilized because it is a versatile and intriguing starting material for solid surface modification and autopolymerized to form PDA under mild conditions24,25, and has advantages such as: (1) a robust interfacial binding force between the coating and the substrate through covalent bonds and other strong intermolecular interactions26; (2) the PDA coatings are stable even in harsh environments such as a strong acid or alkaline solution; (3) PDA coatings have good hydrophilic and biocompatible properties. Our results reveal that Fe3O4@PDA composite is a promising adsorbent for the extraction and recovery of AgNPs with a maximum adsorptive capacity of 169.5 mg/g. Interestingly, the obtained AgNPs-Fe3O4@PDA was shown to exhibit highly catalytic ability for MB reduction by NaBH4. More importantly, the AgNPs-Fe3O4@PDA holds excellent cyclic performance via magnetic separation and can be reused for more than eight times, showing good potentials in practical applications. Further, the AgNPs-Fe3O4@PDA is acid resistant, showing good stability.

Results and Discussion

Synthesis and characterization of Fe3O4@PDA

The Fe3O4@PDA was facilely prepared by direct coating of PDA onto the surface of Fe3O4 via a simple one-step reaction. The Fe3O4 NPs were nearly spherical with a diameter of 10~20 nm (Fig. 1a) and the PDA shell layers formed around the Fe3O4 cores (Fig. 1b). A clear aggregation was formed on the surface of Fe3O4@PDA after adding of AgNPs suspension, indicating successful adsorption of AgNPs by Fe3O4@PDA (Fig. 1c), which was confirmed by the results of FT-IR spectra and XRD. Fe3O4@PDA exhibited bands at 3050 cm−1 and 2948 cm−1 (Fig. 2a), attributed to N-H, C-H vibration peak of benzene. The additional bands at 1000–1700 cm−1 may be ascribed to the aromatic rings of PDA and the amide I, amide II and C-N stretching bands27,28. These peaks weakened or disappeared after adsorption of gum arabic (GA) capped AgNPs (GA-AgNPs), indicating that AgNPs interacts with O and N atoms of PDA during adsorption process. A similar XRD pattern to Fe3O4 was observed for the Fe3O4@PDA (Fig. 2b), indicating that the crystalline structure of Fe3O4 was not affected by coating with PDA. The diffraction peaks of AgNPs-Fe3O4@PDA can be indexed to Ag (JCPDS 04-0783), confirming the presence of AgNPs in the Fe3O4@PDA29,30. TG data (Fig. 2c) indicate the weight loss of water from 0 to 100 °C. Above 100 °C, Fe3O4 NPs were very stable, while Fe3O4@PDA and AgNPs-Fe3O4@PDA had about 15% loss of weight. This may be due to the part weight loss of PDA coating. At 700 °C, AgNPs-Fe3O4@PDA had another 15% loss of weight due to the part weight loss of GA. The result confirms the successful synthesis and excellent thermal stability of Fe3O4@PDA as well as adsorption of GA-AgNPs by Fe3O4@PDA. The saturation magnetization values were 60.4, 27.5, and 25.9 emu/g for Fe3O4, Fe3O4@PDA and AgNPs-Fe3O4@PDA, respectively (Fig. 2d). Compared with Fe3O4, the decrease in saturation magnetization value of the Fe3O4@PDA was mainly due to the coating of Fe3O4 with PDA. However, no obvious loss of saturation magnetization was observed after adsorption of GA-AgNPs by Fe3O4@PDA. Hence, the AgNPs-Fe3O4@PDA can be easily separated from solution by an external magnetic field (inset in Fig. 2d).

Figure 1
figure 1

TEM images of (a) Fe3O4 NPs, (b) Fe3O4@PDA core-shell NPs and (c) AgNPs-Fe3O4@PDA.

Full size image
Figure 2
figure 2

(a) FT-IR spectra of Fe3O4, Fe3O4@PDA, and AgNPs-Fe3O4@PDA. (b) X-ray diffraction patterns of Fe3O4, Fe3O4@PDA, and AgNPs-Fe3O4@PDA. (c) TGA data of Fe3O4, Fe3O4@PDA, and AgNPs-Fe3O4@PDA. (d) Magnetic curves of Fe3O4, Fe3O4@PDA, and AgNPs-Fe3O4@PDA, inset photographs: separation of Fe3O4@PDA (upper left) and AgNPs-Fe3O4@PDA (lower right) from aqueous solution by using an external magnet.

Full size image

In this study, GA capped AgNPs was used as a target to investigate its magnetic separation by Fe3O4@PDA because GA-AgNPs was found in commercial products of AgNPs31. Supplementary Figure 1a and b show the TEM image and corresponding size distribution histogram of the synthesized GA-AgNPs, respectively. GA-AgNPs were sphere-like with an average diameter of 5.0 ± 1.7 nm based on the statistic result of 172 particles.

Adsorption of the prepared adsorbents for AgNPs

After PDA coating, the dynamic uptake capacity of the Fe3O4@PDA for GA-AgNPs increased by 7-fold compared with that of Fe3O4 (Supplementary Figure 2). The result indicates that the Fe3O4 core plays the role of magnetic separation and the PDA shell contributs mainly to the adsorption removal of GA-AgNPs. To understand the possible adsorption mechanism, the zeta potentials of the synthesized GA-AgNPs and Fe3O4@PDA were measured. In pH 10 solution, GA-AgNP is negatively charged and its zeta potential is −25.0 mV. While Fe3O4@PDA is also negatively charged and its zeta potential is −43.6 mV at pH 10. This indicates that the electrostatic force is not the main driving force for the adsorption of GA-AgNPs by Fe3O4@PDA. In contrast, the specific high affinity of AgNPs to PDA through its complexation with catechol group28 on Fe3O4@PDA surface may be responsible for adsorption separation of GA-AgNPs. In addition, the Fe3O4@PDA is suitable for removal of AgNPs capped with other capping agents, including polyvinyl alcohol (PVA), polyvinylpyrrolidone (PVP), humic acid (HA), citrate (Cit) and polyethyleneimine (PEI) (Supplementary Table 1). It is noted that the adsorption capacity of Fe3O4@PDA for AgNPs depends on the used capping agents. This may be due to the difference in the zeta potential and hydrodynamic diameter of the obtained AgNPs. However, the effects of the zeta potential and hydrodynamic diameter of the obtained AgNPs on adsorption capacity of Fe3O4@PDA for AgNPs did not show a clear trend. Also, the Fe3O4@PDA works for removal of AuNPs capped with GA, PVP and Cit (Supplementary Figure 3).

Effect of pH

The effect of pH on the adsorption of AgNPs to Fe3O4@PDA was explored. The adsorption of AgNPs increased with the increase of pH to 10, above which it decreased (Fig. 3a). This can be explained by the formation of Ag–catechol bonds through the specific high affinity between AgNPs with PDA on the surface of Fe3O432. It has been reported that the amino and phenolic hydroxyl groups in the PDA coating of PDA@Fe3O4 particles are expected to be deprotonated in the pH range of 8.0–10.033. At pH < 8, minor –OH groups are ionized to form –O− groups, and the adsorption capacity of AgNPs keeps minor change in pH range from 6 to 8. While at pH > 8, more –OH groups of PDA are ionized to form –O− groups with increase in solution pH due to deprotonation of amino and phenolic hydroxyl groups in the PDA coating, and the adsorption of AgNPs increased with increase in solution pH from 8 to 10. However, higher pH above 10 caused decrease in the adsorption of AgNPs. This is probably attributed to hinderation of approach of the negatively charged AgNPs to the PDA surface in high pH solution33.

Figure 3
figure 3

(a) Effect of pH. Reaction conditions: 5 mg adsorbent, 50 mL of 10.8 mg/L GA-AgNPs solution, adsorption time 26 h. (b) Adsorption kinetics. Reaction conditions: 5 mg adsorbent, 50 mL of 10.8 mg/L GA-AgNPs solution, pH 10.0. (c) Adsorption isotherm. Reaction conditions: 5 mg adsorbent, 50 mL of GA-AgNPs solution, pH 10.0, and adsorption time 26 h. (d) Effect of salt concentration. Reaction conditions: 5 mg adsorbent, 50 mL of 10.8 mg/L GA-AgNPs solution, pH 10.0, adsorption time 26 h. Error bars represent one standard deviation for three measurements.

Full size image

Adsorption kinetics and adsorption isotherms

The time-dependent adsorption capacity was obtained to study the kinetics of GA-AgNPs adsorption on Fe3O4@PDA. The sorption rate of GA-AgNPs onto Fe3O4@PDA increased with increase in contact time to 26 h, and then the adsorption capacity increased slightly with contact time up to 36 h (Fig. 3b). Thus, the sorption process for Fe3O4@PDA reaches equilibrium at about 26 h. A better fit of the pseudo-second-order model (Table 1) was obtained. This indicates that chemisorption is the dominant rate-limiting step. This result is reasonable because as indicated previously, the major driving force for the adsorption of AgNPs onto Fe3O4@PDA is the formation of Ag–catechol bonds through the specific high affinity between AgNPs with PDA on the surface of Fe3O4@PDA. It is a chelating process and is always controlled either by particle diffusion mechanism or by a second-order chemical reaction34.

Table 1 Kinetic parameters of GA-AgNPs adsorption by Fe3O4@PDA.
Full size table

The adsorption isotherm was obtained after the mixture of Fe3O4@PDA and AgNPs was shaken for 26 h (Fig. 3c). The correlation coefficient for the Langmuir model is quite high (>0.99) (Table 2), showing a better fit of the Langmuir model with the experimental data as compared to the Freundlich model. The maximum adsorption capacity for AgNPs was 169.5 mg/g, which was higher than those obtained with other adsorbents10,11,12,13,14,15,17,18 (Table 3). In addition, based on result of β value given in Table 2, the mean energy value of GA-AgNPs adsorption determined by the Dubinin-Radushkevich model was 23.57 kJ/mol. Hence, GA-AgNPs are chemisorbed on the Fe3O4@PDA because a chemical adsorption takes place if a value of adsorption energy is in the range of 8–16 kJ/mol, while a physical adsorption does if it is below 8 kJ/mol35.

Table 2 Isotherm constants for adsorption of GA-AgNPs onto Fe3O4@PDA.
Full size table
Table 3 Comparison of the maximum AgNPs adsorption capacity with different adsorbents.
Full size table

Effect of ionic strength

Effect of ion strength on the adsorption of AgNPs was examined by changing the concentration of NaNO3 in the 0–50 mM range. Minor effect was observed even when the concentration reached 50 mM (Fig. 3d), indicating that the electrostatic interaction played a minor role in AgNPs adsorption.

Effect of natural water matrices

The removal ability of Fe3O4@PDA for GA-AgNPs was investigated with different water matrices (including ultra pure water and river water) spiked with GA-AgNPs. The Jialingjiang River water in Beibei section was collected and used as a practical sample. Before use, the water sample was filtered through a 0.45 μm membrane. The adsorption capacity of Fe3O4@PDA for GA-AgNPs in river water was almost same as that in ultra pure water, showing the minor effect of river water matrices on removal of GA-AgNPs (Supplementary Table 2). This is probably due to the specific high affinity of AgNPs to PDA through its complexation with catechol group28 on Fe3O4@PDA surface.

Catalytic reduction of MB by AgNPs-Fe3O4@PDA

Due to good catalytic activity of AgNPs, the magnetically separated AgNPs-Fe3O4@PDA was used for the catalytic reduction of MB by NaBH4 as reducing agent. MB was selected as a target in present work because of its wide use in coloring paper, temporary hair colorant, dyeing cottons, and so on. Also, it inhibits caspases by oxidation of the catalytic cysteine36. The adsorption peak at 663 nm decreased with increase in reaction time in the presence of AgNPs-Fe3O4@PDA (Fig. 4a). MB was almost totally removed within 30 min, indicating the successful reduction of MB by AgNPs-Fe3O4@PDA. Also, direct adsorption of MB by Fe3O4@PDA was observed due to the electrostatic interaction between negatively charged Fe3O4@ PDA and positively charged MB. The absorption capacity of AgNPs-Fe3O4@PDA for MB is about 50% of the removed MB amount by AgNPs-Fe3O4@PDA/NaBH4 system (Fig. 4b), indicating that AgNPs-Fe3O4@PDA exhibits good catalytic performance. pH variance from 4 to 10 has no obvious effect on the catalytic activity of the AgNPs-Fe3O4@PDA toward MB reduction (Supplementary Figure 4a). Hence, no pH adjustment was needed for MB solution. 5.6 was selected as optimum pH.

Figure 4
figure 4

(a) Effect of contact time on MB removal by AgNPs-Fe3O4@PDA/NaBH4; (b) MB removal by AgNPs-Fe3O4@PDA/NaBH4 and Fe3O4@PDA/NaBH4 systems.

Full size image

The catalytic reduction of MB by AgNPs-Fe3O4@PDA/NaBH4 system can be considered to follow pseudo-first order kinetics because the high initial concentration of NaBH434 used in the experiment. Hence, Eq. (1) was used to fit the experimental data:

where C0 and Ct are the concentration of MB at the initial stage and at time t, respectively. k represents the reaction rate constant.

The kinetic data obtained with the AgNPs-Fe3O4@PDA as catalyst (C0 = 7.5 mg/L, 20 mL of MB solution, 5 mg of catalyst dosage) was fitted to the pseudo-first order kinetics model using a linear fitting and the obtained k was 1.44 × 10−3/s (Supplementary Figure 4b), which was much higher than those obtained with many other catalysts (Table 4), including AgNPs/P(NIPAM-co-DMA) microgels37, Ag nanowire38, Sacha inchi (SI) oil templated AgNPs39, GO/AgNPs40, Ag colloid41, Pd/polypyrrole-cellulose42, biogenic AuNPs43, Sterculia acuminata fruit extract templated AuNPs44, Au-PBCGO5545, dendrimer encapsulated AgNPs (AgDENs)46, dendrimer encapsulated AuNPs (AuDENs)46, and Fe3O4@Tween20@Ag47. High catalytic activity of the AgNPs-Fe3O4@PDA for MB removal is probably attributed to the presence of monodisperse AgNPs on the surface of Fe3O4@PDA (Fig. 1c), leading to a bigger active contact surface. In addition, the electrostatic interactions between PDA and MB are also in favor of this effect. MB was more easily absorbed on the surface of AgNPs-Fe3O4@PDA through π–π interaction and hydrogen bonding because of the presence of large amount of functional groups (amino and catechol groups) on PDA layer. Finally, in order to demonstrate whether the AgNPs-Fe3O4@PDA catalyst obtained from real water matrices could be effective for MB degradation, 5 mg of such AgNPs-Fe3O4@PDA was added to 20 mL of 7.5 mg/L MB solution in the presence of 0.5 mL of 0.1 M NaBH4. The obtained k values for two Jianlingjiang river water samples are 1.35 × 10−3/s, and 1.38 × 10−3/s (Supplementary Figure 4c and 4d), respectively. These results suggest the minor effect of river water matrices on catalytic performance of the obtained AgNPs-Fe3O4@PDA for the degradation of MB.

Table 4 Comparison of kinetic constant (k) of different noble metal catalysts in the degradation of MB reported in previous literatures.
Full size table

Recyclability and stability of AgNPs-Fe3O4@PDA

The cycling tests were carried out to study the reusability of AgNPs-Fe3O4@PDA catalyst. After catalytic reaction, AgNPs-Fe3O4@PDA was regenerated by treatment using 0.1 M HNO3, ethanol and ultra-pure water, then reused in the next catalytic reduction of MB for eight times under the same conditions. Above 85% MB elimination was retained after eight cycles (Supplementary Figure 5), indicating no apparent loss in catalytic activity of the AgNPs-Fe3O4@PDA for MB removal. Thus, the AgNPs-Fe3O4@PDA promises good recyclability and great potential in practical applications. Also, the stability of the AgNPs-Fe3O4@PDA was examined. First, the concentration of the leached iron in the degradation of MB was measured under optimized catalytic conditions and it was 0.22 mg/L. This confirms that the AgNPs-Fe3O4@PDA catalyst is very stable for MB degradation reaction. Second, we soaked 10 mg of the AgNPs-Fe3O4@PDA in 1 M HNO3 for different time ranging from 6 h to 24 h. Then nitric acid treated adsorbents were used for catalytic reduction of MB. The AgNPs-Fe3O4@PDA still keep over 95% catalytic activity even treatment in 1 M HNO3 environment for 24 h (Supplementary Figure 6). The XRD pattern (Supplementary Figure 7a) shows that AgNPs are still on the surface of Fe3O4@PDA particles after acid treatment. Further, the sphere-like structure of AgNPs-Fe3O4@PDA is retained (Supplementary Figure 7b). Such a good stability may be attributed to the presence of PDA layer, which is helpful to protect the Fe3O4 cores and bind the AgNPs due to high affinity between PDA and AgNPs. These results suggest that the AgNPs-Fe3O4@PDA is acid resistant and stable in the experimental conditions used.

Conclusion

In conclusion, we demonstrate that the Fe3O4@PDA is a promising adsorbent for the extraction and recovery of AgNPs with a maximum adsorptive capacity of 169.5 mg/g. The magnetically separated AgNPs-Fe3O4@PDA holds highly catalytic activity, good stability and cyclic performance for MB reduction by NaBH4. It is potentially useful for the water treatment applications. Notably, present study provides a new way to recover noble metallic NPs for the catalytic removal of contaminants in water.

Experimental Section

Chemicals

GA powder, PVA, PVP, HA, citrate, silver nitrate (AgNO3), ammonia (NH3·H2O), sodium borohydride (NaBH4), dopamine hydrochloride, six hydrated ferric chloride (FeCl3·6H2O), and seven ferrous sulfate hydrate (FeSO4·7H2O) were obtained from Chongqing Taixin Chemical Reagents Company (Chongqing, China). PEI with a molecular weight of 25 000 was purchased from Sigma–Aldrich (Shanghai, China). All chemicals were of analytical grade. The ultra-pure water was used for preparing all solutions.

Synthesis of GA capped silver nanoparticles

GA stabilized AgNPs were synthesized by reducing AgNO3 in water with NaBH431. Briefly, 0.2430 g GA powder was added to a flask containing 800 mL of ultra-pure water with vigorous stirring for about 10 min, and a proper amount of 0.02 M AgNO3 was added and with constant stirring. Finally, 0.1360 g NaBH4 was quickly added with vigorously stirring for 12 h to obtain a yellow GA-AgNPs suspension. The obtained GA-AgNPs was stored at 4 °C for further use, and no aggregation was found for the GA-AgNPs during storage. The AgNPs and AuNPs with other different capping agents were prepared based on the modified procedures previously reported in the literature, which were given in Supporting Information.

Synthesis of Fe3O4@PDA core-shell microspheres

The Fe3O4 microspheres were prepared with a modified coprecipitation method. FeCl3·6H2O (6.1 g) and FeSO4·7H2O (4.2 g) were dissolved in 100 mL H2O. After being heated to 90 °C with stirring, 25% NH3·H2O was quickly added with stirring for 30 min. Finally, the black precipitation were washed with water and then dried in vacuum at 60 °C. To coat Fe3O4 cores with the polydopamine shell, 80 mg of Fe3O4 and 240 mg of dopamine hydrochloride were dissolved in 50 mL of PBS solution (pH 8.0). After shaking for 24 h at room temperature, the products were separated by using an external magnet, washed with ultra-pure water and ethanol several times, and then dried in an oven at 60 °C for overnight.

Instrumentation

A XS-105 Mettler Toledo analytical balance (Mettler-Toledo, Switzerland) was used to accurately weigh the amount of the adsorbents. UV-vis spectra were measured with a type UV-2450 spectrophotometer (Shimadzu, Suzhou, China). Fourier transform infrared (FT-IR) spectra were recorded on the Nicolet 170SX instrument (Madison, WI) in the transmission mode using KBr pellets of the sample. Thermogravimetric (TG) data were obtained by the TA-SDTQ 600 (Texas Instruments, Inc., New Castle, DE) in the temperature range from 25 °C to 800 °C at a heating rate of 10 °C/min. The X-ray diffraction (XRD) patterns were recorded using an XD-3 X-ray diffractometer (PuXi, Beijing, China) under the conditions of nickel-filtered Cu Kα radiation (λ = 0.15406 nm) at a current of 20 mA and a voltage of 36 kV. The magnetic property was determined by vibrating sample magnetometry (VSM, HH-15). The zeta potential was measured using a Malvern Instruments Zetasizer Nano-ZS90 (Malvern, UK) instrument. TEM images were obtained on a Tecnai G2 20 (FEI, USA).

Adsorption experiments and data treatment

The adsorption experiments were carried out in 100 mL stoppered conical flasks. 5 mg portion of Fe3O4@PDA was added to flask containing 50 mL of GA-AgNPs solution. Subsequently, flasks were shaken at 180 rpm in a thermostatic shaker at 25 °C. After adsorption, the solid adsorbents were separated from the solution with an external magnet. The concentration of the remaining AgNPs suspension was determined using UV–vis spectrophotometer by measuring the changes in the absorbance after adsorption.

For the adsorption kinetic studies, the pseudo-first-order48 (equation 2) and pseudo-second-order49 (equation 3) models were used to fit the experimental data.

Where qe is the equilibrium adsorption capacity (mg/g), qt is the adsorption capacity at time t, k1 (1/min) is the pseudo-first-order adsorption rate constant, and v0 is the initial adsorption rate [(mg/(g·min)].

Two adsorption isotherms, namely, Langmuir model13 (equation 4), Freundlich model14 (equation 5), and Dubinin-Radushkevich model50 (equation 6) were used to analyze the obtained adsorption data.

where Ce is the equilibrium concentration of the studied target solute (mg/L), qmax (mg/g) is the maximum adsorption capacity, b (L/mg) is constant related to energy of adsorption. k and n are the constants of Freundlich adsorption. β is the constant related to adsorption energy (mol2/kJ2); ε is the Polanyi potential that is equal to RT ln (1 + 1/Ce). R is the gas constant and T is the absolute temperature (K). Dubinin-Radushkevich isotherm is used to distinguish the physical and chemical adsorption of target in terms of its mean free energy E (kJ/mol), which can be calculated by E = 1/(2β)0.5.

Catalytic reduction experiment

The catalytic reduction experiment was carried out by adding 5 mg of the AgNPs-Fe3O4@PDA into 20 mL of 7.5 mg/L MB aqueous solution, followed by addition of 0.5 mL of fresh NaBH4 aqueous solution (0.1 M). After the mixture was shaken at 180 rpm in a thermostatic shaker at 25 °C for 30 min, the solid AgNPs-Fe3O4@PDA catalysts were separated from the solution with an external magnet. The concentration of remaining MB solution was determined by measuring the absorbance of the solution at 663 nm. The recovered AgNPs-Fe3O4@PDA catalysts were washed with 0.1 M HNO3 solution, ethanol and ultra-pure water several times, and then used for the next cycle process.

Additional Information

How to cite this article: Wu, M. et al. Removal of silver nanoparticles by mussel-inspired Fe3O4@ polydopamine core-shell microspheres and its use as efficient catalyst for methylene blue reduction. Sci. Rep. 7, 42773; doi: 10.1038/srep42773 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.