Prussian blue analogues derived porous nitrogen-doped carbon microspheres as high-performance metal-free peroxymonosulfate activators for non-radical-dominated degradation of organic pollutants

Na Wang , Wenjie Ma , Ziqiu Ren , Yunchen Du *, Ping Xu and Xijiang Han *
MIIT Key Laboratory of Critical Materials Technology for New Energy Conversion and Storage, School of Chemistry and Chemical Engineering, Harbin Institute of Technology, Harbin 150001, China. E-mail:;

Received 25th September 2017 , Accepted 14th November 2017

First published on 14th November 2017

Nitrogen-doped carbon materials are becoming a new type of metal-free heterogeneous catalysts in advanced oxidation processes (AOPs) for wastewater treatment and environmental remediation. In this study, porous nitrogen-doped carbon (PNC) microspheres derived from Zn–Co Prussian blue analogues (Zn–Co PBAs) are employed as heterogeneous catalysts in peroxymonosulfate (PMS) activation. The unique configuration of the metal centers/clusters bound by cyanide groups (–C[triple bond, length as m-dash]N) of Zn–Co PBAs offers the PNC microspheres abundant porosity, a high graphitization degree, and rich nitrogen substitution, which lead to improvements in the catalytic performance. PNC-800 (pyrolyzed at 800 °C) exhibits better performance than other common carbon materials and homologous nitrogen-doped carbocatalysts derived from ZIF-8/ZIF-67. Based on radical quenching and trapping experiments, a non-radical pathway is proposed to dominate methylene blue (MB) degradation; and the high graphitization degree and rich surface graphitic N sites of PNC-800 are two key factors that induce the non-radical pathway. Several influential factors, including catalyst dosage, PMS concentration, pH value and reaction temperature, are investigated in detail. Notably, MB degradation over PNC-800 is almost completely insusceptible to common ions and natural organic matter, and maintains its catalytic efficiency under the background conditions of several real water samples. More interestingly, this non-radical pathway and the good catalytic performance of the PNC-800/PMS system are universal in the degradation of other typical organic pollutants. We believe that these PNC microspheres may be a promising green heterogeneous catalyst for the degradation of organic pollutants, and this study can be used for the design of high-performance carbocatalysts in non-radical systems in the future.


The water deterioration that accompanies flourishing urbanization and industrialization has always been a serious issue that motivates worldwide concern. During the past few decades, considerable efforts have been focused on alleviating the water crisis by purifying wastewater through various conventional physical, chemical and biological technologies.1–3 Advanced oxidation processes (AOPs) are regarded as one of the most efficient approaches for the disposal of various toxic organic contaminants, owing to their high efficiency and nearly non-selective degradation.4,5 Hydrogen peroxide (H2O2), ozone (O3), persulfate (S2O82−, PS) and peroxymonosulfate (HSO5, PMS) are generally used as powerful oxidants to decompose diverse recalcitrant organics into innocuous carbon dioxide, water and mineralized acids.6–9 In these cases, hydroxyl radicals (˙OH) generated in Fenton or Fenton-like processes always suffer from disadvantages such as pH restriction (pH = 3–4), peroxide instability and coagulation of sludge.6 Sulfate radical (SO4˙)-based AOPs have been popularized as a promising alternative, owing to a series of merits: higher standard redox potential (2.5–3.1 V) compared with hydroxyl radicals (1.8–2.7 V), flexibility in a broad pH range (2.0–8.0), and a longer half-life period (30–40 μs).10,11 In addition, some previous studies have proposed that a new reaction pathway based on a non-radical mechanism is also involved in PMS or/and PS activation processes.12–15 For example, Zhang and co-workers discovered that copper oxide (CuO) could efficiently activate PS to remove various pollutants without the generation of sulfate radicals. The outer-sphere interaction between the CuO surface and PS was found to cause electron rearrangement in the PS molecule, which subsequently increased the oxidative ability of PS.12 Lee et al. reported that graphited nano-diamond (G-ND) could activate PS to oxidize phenolic compounds through a non-radical process, in which both PS and phenol could effectively bind to the G-ND surface and form a charge transfer complex, benefiting from the mediating effect of G-ND.13 For a long time, metal-based materials have been employed extensively as heterogeneous catalysts for PMS or/and PS activation in sulfate radical-dominated systems. Although they alleviate the drawbacks of homogeneous activation to some extent, secondary contamination caused by metal leaching cannot be fully avoided. Therefore, metal-free catalysts with good catalytic efficiency are highly desirable to supersede the conventional metal-based catalysts for wastewater remediation.

Recently, carbon materials, such as activated carbon (AC), nano-diamonds, carbon nanotubes (CNTs), graphene and reduced graphene oxide (GO and rGO) and graphitic carbon nitride (g-C3N4), are being applied as new types of green catalysts due to their large surface area, unique electronic properties, sp2-hybridized carbon configuration, and non-secondary contamination.16–21 In particular, it has been well reported that the surface engineering of pristine carbon materials by the substitution of some carbon atoms with heteroatoms (e.g. N, P, S and B) is an effective way to tailor their electron-donor properties and consequently enhance their catalytic activities.22–25 For instance, Liu et al. designed nitrogen (N) and sulfur (S) co-doped multi-walled carbon nanotubes (CNTs) and validated their enhanced PMS activation performance, which was ascribed to more active sites generated by the introduction of N and S atoms.22 Duan and co-workers demonstrated that N-doped graphene (NG) nanomaterials (melamine as nitrogen precursor) could exhibit superior PMS activation properties compared to undoped rGO, other metal-based catalysts and carbon allotropes in phenol degradation.24 Furthermore, Duan et al. also investigated the modification of carbon nanotubes with N atoms and discussed the roles of nitrogen heteroatoms in PMS activation, and more importantly, they proposed a novel non-radical-based oxidation pathway accompanied by radical generation in this PMS activation process.25 Although significant improvements have been achieved in these successful examples, most of them are still stuck with tedious preparation procedures, low heteroatom content and additional heteroatom sources with different degrees of toxicity.

The in situ pyrolysis of metal–organic frameworks (MOFs) is becoming an attractive strategy for the preparation of porous carbon materials, since the periodic arrangement of different atoms in highly ordered MOFs offers fundamental advantages for homogeneous chemical distribution and heteroatom substitution.26–28 Zeolitic imidazolate frameworks (ZIFs), such as ZIF-67 [Co(MeIm)2] and ZIF-8 [Zn(MeIm)2] (MeIm = 2-methylimidazole), have been exemplified as ideal self-sacrificing templates for preparing porous N-doped carbon materials, benefitting from their highly microporous structure, thermally stable carbon frameworks and the nitrogenous ligand 2-methylimidazole. Lin et al. fabricated a magnetic cobalt/carbon nanocomposite (CCN) with immobilized cobalt and increased the porosity through the one-step carbonization of ZIF-67, and validated the significantly enhanced PMS activation performance of CCN in the degradation of caffeine in water.27 Wang and co-workers reported that N-doped porous carbons (NPCs) derived from ZIF-8 could display better performance than nitrogen-free porous carbon in PMS activation, owing to the excellent capability of graphitic N in activating adjacent carbon atoms.28 Prussian blue and its analogues (PBAs), represented as M3(II)[M(III)(CN)6]2 frameworks, are a type of coordination polymer composed of metal centers/clusters surrounded by six cyanide groups (–C[triple bond, length as m-dash]N) in an octahedral configuration.29 Similarly, PBAs can also be directly transformed into porous carbon frameworks while retaining their well-defined morphologies during high-temperature pyrolysis, and the obtained carbonaceous materials have a large surface area and interconnected pores, which are conducive to the improvement of catalytic activity.30–32 Furthermore, compared with ZIFs, which possess 2-methylimidazole groups, the –C[triple bond, length as m-dash]N units in PBAs may endow the resultant carbon frameworks with a richer nitrogen content due to their higher N/C ratio. Lin et al. previously converted cobalt hexacyanoferrate (Co3[Fe(CN)6]2) into magnetic nanocomposites consisting of carbon, cobalt and cobalt ferrite, as a durable and recyclable heterogeneous catalyst for activating PMS.31 Unfortunately, they only focused on the catalytic performance of metal-based active sites without paying attention to the contribution of the N-doped carbon matrix. In view of the potential merits of PBAs, it is promising and meaningful to investigate the catalytic effectiveness of PBA-derived carbocatalysts in the PMS activation process. However, there are few related papers available.

Herein, we employed Zn–Co PBAs as a self-sacrificial template, combined with acid etching to prepare porous nitrogen-doped carbon (PNC) microspheres and investigated their PMS activation performance in the degradation of methylene blue (MB) and other typical organic pollutants. It was found that the presence of Zn centers in PBAs is favorable for the complete removal of metallic species and the formation of rich porosity in the resultant carbon microspheres, while Co centers can significantly promote the graphitization degree of the carbon matrix under high temperature. The as-obtained PNC microspheres have abundant porosity, a high graphitization degree and a rich nitrogen content, which can modulate the electronegativity of adjacent carbon atoms, thus boosting their catalytic performance. Radical quenching and trapping experiments reveal that the radicals are less contributive to the degradation of MB, and a non-radical pathway is dominant in the PNC-800/PMS system. More importantly, the involved reaction mechanism proposes that the high graphitization degree and rich surface graphitic N sites may be responsible for the non-radical process. Some anions (Cl, H2PO42− and HCO3) and natural organic matter (humic acid) were taken into account in MB degradation, and the results indicate that the non-radical system can effectively resist the interference of radical scavengers, and even in real samples from actual bodies of water, the catalyst can still maintain comparable catalytic performance. The effects of PMS concentration, catalyst dosage, pH value and reaction temperature on PMS activation were investigated in detail. The comparable degradation efficiencies of PNC-800 for several other typical organic pollutants (rhodamine B, orange II, phenol, bisphenol and tetracycline) also demonstrate that the PNC microspheres are a promising candidate for heterogeneous carbocatalysts in PMS activation, and are expected to be used in industrial environmental remediation in the future.

Experimental section

Synthesis of Zn3[Co(CN)6]2·12H2O precursor

Zn3[Co(CN)6]2·12H2O microspheres were prepared according to a previous report.33 Typically, 6.0 mmol of zinc acetate dehydrate [Zn(Ac)2·2H2O] and 4.0 g of polyvinylpyrrolidone (PVP, K30) were dissolved in 200 mL of deionized water with mechanical stirring to form a pellucid solution (denoted as solution A). Then, 4.0 mmol of potassium hexacyanocobaltate (K3[Co(CN)6]) was dispersed into 200 mL of deionized water with ultrasonication to form a transparent aqueous solution (denoted as solution B). Solution B was then added into solution A dropwise under ultrasonication and vigorous stirring in an ice-water bath for 1 h. Finally, the above mixture was aged for 6 h at room temperature and the white precipitate was collected and washed with deionized water several times and dried at 60 °C in a vacuum oven for 12 h.

Synthesis of PNC-X microspheres

The as-prepared Zn3[Co(CN)6]2·12H2O microspheres were pyrolyzed in a tubular furnace under N2 atmosphere at 600 °C for 2 h, with a ramping rate of 2 °C min−1 from room temperature to the designated temperature. The products were etched using hydrofluoric acid (HF, 5 wt%) and hydrochloric acid (HCl, 3.0 M) solution successively to thoroughly remove the residual cobalt and zinc species. Then, the products were collected, purified using deionized water and absolute ethanol several times, and dried at 60 °C. To obtain the PNC microspheres, the above products were further treated with a continuous N2 flow at a designated temperature (600 °C, 700 °C, 800 °C and 900 °C) for 2 h. The final samples were denoted as PNC-X microspheres, where X refers to the secondary pyrolysis temperature.


Powder X-ray diffraction (XRD) patterns were measured on an X'PERT PRO MPD X-ray diffractometer with a Cu Kα radiation source (λ = 1.5406 Å) (PANalytical B.V.). Scanning electron microscopy (SEM) images were obtained on a HELIOS NanoLab 600i (FEI). Transmission electron microscopy (TEM) images and high-resolution TEM (HRTEM) images were recorded using a JEM-3000F from JEOL. A confocal Raman spectroscopic system (Renishaw, In Via) with a 633 nm laser was employed to obtain Raman spectra. Thermogravimetric (TG) analysis was determined on a SDT Q600 TGA (TA Instruments) within a temperature range of 20 °C to 800 °C at a heating rate of 10 °C min−1. Nitrogen adsorption isotherms were obtained on an ASAP 2020 (Micromeritics, USA). Samples were typically prepared for measurement by degassing at 120 °C for 12 h. X-ray photoelectron spectra (XPS) were obtained with a PHI 5700 ESCA system equipped with an A1 Kα radiation as the source (1486.6 eV). The DMPO (5,5-dimethyl-1-pyrroline-N-oxide) trapped EPR spectra were collected using a Bruker EPR A200 spectrometer at room temperature, which was operated at X-field with a centered field at 3350 G and a sweep width of 100 G.

Catalytic tests

The catalytic activity of the PNC-X microspheres for activating potassium peroxymonosulfate (Oxone®, 2KHSO5·KHSO4·K2SO4 from Sigma-Aldrich) was evaluated by oxidizing the toxic dye, MB. The effects of several factors such as catalyst dosage, PMS concentration, reaction temperature, initial pH value, various anions and background organic matter on the MB degradation by the PNC-X microspheres were further examined. The pH value was adjusted using either 0.1 M HCl or NaOH aqueous solution and was recorded with a Leici pH meter (model PHS-25). In a typical run, 5 mg of catalyst was dispersed into 50 mL MB solution (100 mg L−1) with the assistance of ultrasonication for 3 seconds. After that, a certain amount of Oxone was added to the solution to initiate the reaction, and the mixture was magnetically stirred simultaneously at room temperature. At given time intervals, a fixed volume of reaction mixture was withdrawn and then immediately mixed with excessive saturated Na2S2O3 solution to scavenge the reactive species and terminate the oxidation. After the reaction, the catalyst was removed from the mixture, collected by centrifugation (10[thin space (1/6-em)]000 rpm) and dried at 60 °C. Then, the used catalysts were calcined at 300 °C for 2 h in air to remove the oxidized functional groups for regeneration. The change in the maximum absorption peak at 664 nm in the UV-vis absorption spectrum (TU-1901, PERSEE Co., Ltd. China) reflected the degradation of MB, which could be calculated using the following equation:
Degradation efficiency (%) = [(C0Ct)/C0] × 100%(1)

Results and discussion

The fabrication procedure for the PNC-X microspheres is schematically shown in Scheme 1. The Zn3[Co(CN)6]2·12H2O precursor was synthesized by a simple solution-phase method in the presence of K3[Co(CN)6], [Zn(Ac)2·2H2O] and PVP as raw materials. The Zn3[Co(CN)6]2·12H2O precursor consists of well-defined microspheres with very smooth surfaces and a narrow size distribution of ca. 500–600 nm (Fig. S1a). The intensive diffraction peaks can be perfectly indexed to the standard pattern of Zn3[Co(CN)6]2·12H2O (JCPDS no. 32-1468), and no additional impurity peaks are detected (Fig. S1b), suggesting the formation of high-purity PBAs (Zn3[Co(CN)6]2·12H2O). After high-temperature pyrolysis, the smooth surfaces of the precursor microspheres become very rough with some external bulges and tentacles (Fig. S2a and b), and numerous nanoparticles can be observed in these spherical carbon frameworks (Fig. S2c). XRD was used to investigate the composition of the obtained carbon-based nanocomposites (Fig. S3), and all diffraction peaks can be well matched with the standard pattern of Co3ZnC (JCPDS no. 29-0524), indicating the in situ formation of metal carbides in the hybrid nanocomposite. When these metal carbide/carbon hybrid microspheres were etched by hydrofluoric acid (HF, 5 wt%) and hydrochloric acid (HCl, 3.0 M), most tentacles and the embedded nanoparticles disappeared (Fig. S2d–f), implying the successful removal of Co3ZnC nanoparticles. Moreover, TG analysis on these treated microspheres reveals a weight loss of more than 99% in air, which again verifies that Co and Zn species were almost completely removed during the acidic treatment processes (Fig. S4). Finally, the products were further carbonized at a designated temperature to obtain the PNC-X microspheres. The secondary carbonization not only eliminates the impurity atoms and functional groups generated in the acid etching processes, but also ensures that the final PNC-X samples are from the same parent.
image file: c7ta08472b-s1.tif
Scheme 1 Schematic illustration of the synthesis of PNC-X microspheres.

SEM images of the PNC-X microspheres (PNC-600, PNC-700, PNC-800 and PNC-900) are presented in Fig. 1a–d. It can be seen that the PNC-X samples inherit the morphologies of the precursor microspheres well. The slightly reduced particle size is attributed to the volume shrinkage of the PNC-X microspheres induced by the increased pyrolysis temperature.33 TEM images of the PNC-800 microspheres were obtained in order to acquire a comprehensive understanding of their microstructure. A typical view of a single PNC-800 microsphere (Fig. 1e) displays a shaggy and porous structure. A closer inspection indicates that the abundant pores result due to the etching of metal carbide nanoparticles embedded in the carbon matrix (Fig. 1f). The porous structure is believed to be favorable for both the diffusion of reaction substrates and the exposure of active sites.34 The high-resolution TEM image clearly reveals the existence of well-distributed mesopores (marked with red circles) that are typically 5–10 nm in diameter (Fig. 1g). Meanwhile, the inset identifies the presence of oriented multilayer graphitic sheets stacked in parallel with an adjacent interlayer distance of approximately 0.34 nm. Fig. 1h presents the selected-area electron diffraction (SAED) pattern of PNC-800, where concentric blurred rings can be observed, confirming the graphitic nature of PNC-800.35 The elemental mapping results also support the fact that C and N elements are uniformly dispersed in PNC-800 (Fig. 1j and k). The morphologies and structural characteristics of PNC-600, PNC-700 and PNC-900 were also investigated (Fig. S5). All samples present similar appearances and nanostructures to those of PNC-800.

image file: c7ta08472b-f1.tif
Fig. 1 SEM images (a–d) of the PNC-X microspheres; the magnified TEM images (e and f) of a single PNC-800 microsphere; HRTEM image (g) of PNC-800, and the inset of g shows the partial enlargement of the lattice fringe of PNC-800; selected-area electron diffraction (SAED) pattern (h) and scanning TEM (STEM) image (i) of PNC-800; and elemental mapping images (j and k) of C and N in a single PNC-800 microsphere.

The crystalline structures of the PNC-X microspheres were characterized using XRD. As shown in Fig. 2a, PNC-X microspheres at different temperatures show almost identical diffraction patterns with a broad diffraction peak centered at 2θ = 26.5° corresponding to the (002) plane, and a weak diffraction peak at 2θ = 43.5° corresponding to the (100) plane of graphitic carbon, which indicates that a high pyrolysis temperature can induce the formation of graphite in these carbon materials. Raman spectra were obtained to discern the structure and crystallization of the PNC-X microspheres by examining the bonding states of carbon atoms in these given carbon materials (Fig. 2b). All PNC-X microspheres display two distinguishable peaks, which can be assigned to D- and G-bands at about 1355 cm−1 and 1585 cm−1, respectively. It is well known that the D-band corresponds to the breathing mode of A1g symmetry involving phonons near the K zone boundary, which is not permitted in perfect graphite and becomes active in the presence of defects/disorders. The G-band corresponds to the E2g mode resulting from stretching vibrations of sp2 bonds, which is usually associated with sp2 sites in graphitic carbon.26,36 The intensity ratio of D- to G-bands (ID/IG) in a Raman spectrum is a common criterion to evaluate the ordered crystal structures of carbon atoms (the graphitization degree). As can be seen, the ID/IG values of PNC-600, PNC-700, PNC-800 and PNC-900 present a continuous decrease from 1.01 to 0.76, suggesting a low defect content and high graphitization degree generated during the high-temperature pyrolysis. It is generally accepted that the graphitic structure is conductive to the charge transfer process,37 and therefore, a high graphitization degree of a carbon material is helpful for promoting electron transfer between PMS and carbon catalysts, thus accelerating the catalytic reaction rate.

image file: c7ta08472b-f2.tif
Fig. 2 XRD patterns (a), Raman spectra (b), N2 adsorption–desorption isotherms (c), and the comparison of BET surface areas and pore volumes (d) of the PNC-X microspheres. The isotherms of PNC-700, PNC-800 and PNC-900 are removed upwards of 200, 400, and 700 cm3 g−1 at the beginning for clarity, respectively.

Fig. 2c exhibits the N2 adsorption–desorption isotherms of the PNC-X microspheres. All samples give standard IV-type isotherms according to the IUPAC classification, with long and broad hysteresis loops at a high relative pressure (P/P0 > 0.7), which means that many disordered mesopores are created in the PNC-X microspheres during the pyrolysis process.38 Based on these isotherms, the corresponding textural parameters, such as BET surface areas and pore volumes, were further obtained. As indicated in Fig. 2d, PNC-600, PNC-700 and PNC-800 exhibit quite similar textural parameters, and their BET surface areas and pore volumes are 441.8 m2 g−1 and 0.818 cm3 g−1, 406.7 m2 g−1 and 0.872 cm3 g−1, and 446.7 m2 g−1 and 0.997 cm3 g−1, respectively. This is because these samples were prepared through the secondary carbonization of the same parent, acid-etched Co3ZnC/C microspheres (Scheme 1). It is worth noting that a different secondary carbonization temperature leads to two side effects on the microstructure of PNC-X microspheres. On one hand, a higher secondary carbonization temperature can intensify the pyrolysis degree of the carbon framework, resulting in the release of more low molecular weight carbides and nitrides. This process favors the generation of richer porosity.39 On the other hand, an increased secondary carbonization temperature can also induce the shrinkage of the carbon framework, which leads to a negative impact on the porous microstructure. The similar textural parameters of PNC-600, PNC-700 and PNC-800 may be attributed to the balance between these two opposite effects. However, when the secondary carbonization temperature reaches 900 °C, the shrinkage of carbon framework plays a more progressive role in determining the microstructure. Therefore, PNC-900 shows decreased BET surface area and pore volume.

The chemical composition and nitrogen bonding configuration of the PNC-X microspheres were further characterized using XPS. A survey of the XPS spectra reveals that the PNC-X microspheres are mainly composed of elements such as carbon, nitrogen and oxygen, and negligible amounts of cobalt and zinc can be detected (total atom percentages of cobalt and zinc are less than 1.0%) (Fig. S6). The atom percentage parameters of C, N and O in the PNC-X microspheres are summarized in Table 1. The C atomic ratio presents a monotonous increase from 79.2% for PNC-600 to 89.8% for PNC-900, accompanied by a decreasing nitrogen content from 16.1% to 7.4%, indicating that a high pyrolysis temperature is unfavorable for the doping of nitrogen species. The low oxygen content in PNC-600 (4.3%), PNC-700 (2.3%), PNC-800 (1.6%) and PNC-900 (1.6%) demonstrates that oxygen-containing groups may be involved through the surface oxidation of the PNC-X microspheres. The bonding configurations of nitrogen atoms in the PNC-X microspheres were examined using high-resolution N 1s spectra, as shown in Fig. 3a–d. All N 1s spectra of the PNC-X microspheres can be separated into four peaks, centered at 398.3 eV, 399.3 eV, 400.9 eV and 404.1 eV, which can be assigned to pyridinic N, pyrrolic N, graphitic N and oxidized N, respectively. According to previous reports, pyridinic N contributes a p-electron to the π-conjugated system in a carbon framework, while pyrrolic N contributes two p-electrons to the π system due to contributions from pyridine and pyrrol functional groups, and graphitic N atoms are those that replace C atoms in the carbon framework.40 Obviously, it can be seen that the relative contents of graphitic N and oxidized N increase remarkably with the increase of the pyrolysis temperature, accompanied by the reduced content of pyridinic N and pyrrolic N. This is because the latter species possess relatively low thermal stability and are susceptible to being converted into more stable graphitic N.41 As reported, graphitic N atoms possess higher electronegativity and a smaller atomic radius, which are beneficial to promoting the electron transfer from neighboring carbon atoms, increasing the charge density of adjacent carbon atoms, and subsequently, dramatically enhancing the catalytic performance.42 In view of this fact, good catalytic performance of the PNC-700 and PNC-800 microspheres can be expected due to their relatively high density of surface graphitic N (total nitrogen content × graphitic N proportion).

Table 1 The XPS spectral parameters and the fitting results for the N 1s peaks of the PNC-X microspheres
Samples XPS (at%) Proportion of total N 1s
C N O Pyridinic N Pyrrolic N Graphitic N Nitric oxide
PNC-600 79.2 16.1 4.30 44.8 29.5 16.6 8.90
PNC-700 81.7 13.1 2.30 34.5 29.2 24.9 11.4
PNC-800 86.3 10.5 1.60 30.4 27.8 29.9 11.9
PNC-900 89.8 7.40 1.60 27.5 23.8 33.8 14.8

image file: c7ta08472b-f3.tif
Fig. 3 High-resolution N 1s XPS spectra of PNC-600 (a), PNC-700 (b), PNC-800 (c), and PNC-900 (d).

The catalytic activities of various carbocatalysts for PMS activation were evaluated for MB degradation (Fig. 4a). It can be clearly seen that PMS alone does not present impressive degradation efficiency, and only 19.8% of MB is removed within 30 min, implying the weak self-decomposition of PMS in the absence of a catalyst. In contrast, when the PNC-X microspheres are used for PMS activation in the reaction system, the degradation efficiency of MB is significantly enhanced, and it is found that the pyrolysis temperature has an obvious impact on the catalytic performance of the PNC-X microspheres. PNC-600 and PNC-700 present moderate activity for PMS activation, and 44.5% and 56.0% of MB is decomposed within 2 min, respectively. A degradation efficiency of about 85.0% can be achieved by PNC-800 within 2 min under the same conditions. PNC-900 shows a decreased efficiency of 53.0% for MB removal within 2 min. It can be observed that PNC-800 displays a significant superiority in the initial stage of the degradation reaction, and almost 100.0% of MB can be removed within 10 min. PNC-600, PNC-700 and PNC-900 exhibit inferior removal efficiencies of 85.4%, 88.9% and 94.2% for MB degradation within 10 min, respectively. In order to exclude the adsorption effect on the catalytic performance of the catalyst, PNC-800, which has the largest specific surface area and pore volume, was selected to remove MB alone. A weak MB adsorption efficiency (15.8%) within 30 min confirms that the contribution from physical adsorption is very poor, and the catalytic decomposition is primarily responsible for MB degradation. The specific surface area, graphitization degree, nitrogen content and active nitrogen sites (i.e. graphitic N) are usually considered to be four important factors that can affect the performance of carbocatalysts.17,18 In our case, PNC-600, PNC-700 and PNC-800 have quite similar BET surface areas and pore volumes (Fig. 2d), while their catalytic performances are greatly different (Fig. 4a), demonstrating that the textural parameters are not a crucial factor for PMS activation. Obviously, more attention should be paid to the graphitization degree and surface active graphitic N sites. Although PNC-700 shows a slightly higher density of surface graphitic N compared with PNC-800 (Table 1), its relatively weak graphitization degree (Fig. 2b) slows down the electron transfer, and thus PNC-700 fails to show comparable catalytic performance to PNC-800. By comparison, the significantly enhanced graphitization degree in PNC-900 cannot make up for the shortcomings of the lower nitrogen content, surface graphitic N sites and smaller surface area. The trade-off relationship between these four important factors makes it difficult to determine the performance of the catalyst from one aspect. Therefore, it is our opinion that the excellent catalytic performance of PNC-800 should be attributed to its high graphitization degree and graphitic N content, as well as large surface area. To further address the superiority of the carbocatalysts derived from PBAs in PMS activation, the degradation of MB over several common carbocatalysts was investigated, as shown in Fig. 4b. It can be observed that the commercial activated carbon (AC), carbon nanotubes (CNTs) and graphene oxide (GO) are able to activate PMS to degrade 52.5%, 56.8% and 56.6% of MB within 30 min, respectively. However, their effectiveness is far behind that of PNC-800 (Fig. 4b), indicating that pristine sp2 carbon systems are less active in PMS activation.24 In addition, the home-made N-doped carbocatalysts (HNC-8 and HNC-67 derived from ZIF-8 and ZIF-67) were also employed as heterogeneous catalysts for PMS activation (Fig. 4b). As observed, HNC-8 presents a moderately improved catalytic efficiency as compared with the common carbocatalysts, but its performance is still much less than that of PNC-800. A comparison of the elemental compositions of PNC-800, HNC-8 and HNC-67 is presented in Fig. S7, which validates that HNC-8 has a higher density of nitrogen dopants. However, the amorphous nature (i.e. low graphitization degree) of HNC-8 greatly limits its catalytic ability (Fig. S8), and only 61.0% of MB is removed within 30 min. As we know, metallic Co can act as a catalyst to promote the graphitization degree of carbon materials at high temperature.43 Therefore, the catalytic performance of HNC-67 is noticeably enhanced and a better degradation efficiency of 96.2% is achieved within 30 min due to the improved graphitization degree (ID/IG = 0.70) (Fig. S8). It is unfortunate that the enhanced graphitization is accompanied by a drastically decreased nitrogen content and surface graphitic N sites. As a result, HNC-67 cannot produce comparable catalytic effectiveness to PNC-800 in MB degradation. Based on these results, it can be easily concluded that PBAs are good self-sacrificing precursors for creating N-doped carbocatalysts that show exciting performance by reaching an advanced balance between the graphitization degree and surface graphitic N sites.

image file: c7ta08472b-f4.tif
Fig. 4 Degradation of MB under various systems (a), and comparison with other conventional carbocatalysts for PMS activation (b). Reaction conditions: [MB] = 100 mg L−1, [Oxone] = 1.0 g L−1, catalyst = 0.1 g L−1, T = 25 °C, pH = 6.30.

For ascertaining the primary reactive radicals responsible for MB degradation, radical quenching reactions were performed, as shown in Fig. 5a and b. It is well known that sulfate radicals (SO4˙), hydroxyl radicals (˙OH), superoxide anion radicals (O2˙), and singlet oxygen (1O2) are usually involved in the PMS activation process.44 Although 1O2 is not a typical kind of radical, it is sometimes derived from radical reactions,45 and thus, its contribution should also be evaluated. Tert-butyl alcohol (TBA) is taken as the typical scavenger for ˙OH (k˙OH = 3.8–7.6 × 108 M−1 s−1), and methanol (MeOH) is utilized to scavenge both SO4˙ (kSO4˙ = 3.2 × 106 M−1 s−1) and ˙OH (k˙OH = 9.7 × 108 M−1 s−1).46 Benzoquinone (BQ) and L-histidine are remarkable trapping agents for O2˙ (kO2˙ = 0.9–1.0 × 109 M−1 s−1) and 1O2 (kO21 = 3.2 × 107 M−1 s−1), respectively.47,48 When TBA is added into the reaction solution with a molar ratio of TBA to PMS at 1000[thin space (1/6-em)]:[thin space (1/6-em)]1, the degradation efficiency of MB reaches 91.4% within 5 min (Fig. 5a), which indicates that the contribution of ˙OH to MB degradation is rather limited. In contrast, when the MeOH concentration is 500 and 1000 times that of PMS in the degradation system (Fig. 5a), the MB removal efficiency is reduced to 66.3% and 59.9% within 5 min, respectively. However, 94.4% and 93.2% of MB is eventually degraded within 30 min. In order to completely eliminate the contributions of SO4˙ and ˙OH, MB degradation was further carried out in absolute MeOH solution (Fig. 5a). It was found that the removal efficiency of MB decreases from 91.0% to 52.7% within 5 min; however, over 90.0% of MB removal can still be achieved within 30 min. The relatively weak inhibiting effects of MeOH mean that SO4˙ and ˙OH do not contribute greatly to the degradation of MB. In addition, we also investigated MB degradation in quenching reaction systems containing different concentrations of BQ and L-histidine. As shown in Fig. 5b, a low-concentration of BQ still fails to significantly suppress the MB removal efficiency, and excessive BQ (100 mM) in the reaction system even results in a reverse promotion on the degradation of MB. This is because BQ can react with PMS to produce a dioxirane intermediate and then release highly active 1O2.10 Although L-histidine can induce a concentration-dependent inhibition during the initial stage, the degradation efficiency of MB ultimately exceeds 90% within 60 min under a higher concentration of L-histidine compared with other reports.47,48 This implies that 1O2 only makes a small contribution, and in this study, L-histidine is preferable to be used as the retardant rather than the quencher in MB degradation. Based on these systematic investigations, it can be concluded that various routine radicals are not the key factors that account for the rapid degradation of MB; that is, a non-radical pathway is dominant in the PNC-800/PMS system.

image file: c7ta08472b-f5.tif
Fig. 5 The effects of the radical quenching agents on MB degradation with PNC-800 (a and b); DMPO trapped EPR spectra of PNC-800, PMS and PNC-800/PMS systems (c); DMPO trapped EPR spectra of the PNC-800/PMS system at different reaction times (d).

EPR examinations using 5,5-dimethyl-1-pyrroline (DMPO) as a radical trapping agent were applied to further determine the inexistence of radicals involved in the PNC-800/PMS system. As shown in Fig. 5c, there is no obvious signals when only PNC-800 or PMS is present in the reaction solution, indicating that SO4˙ or ˙OH will not be produced under these conditions. It is very interesting that the EPR signals of DMPO-˙OH and DMPO-SO4˙ cannot be observed yet, even if PNC-800 and PMS coexist in the reaction system. Instead, a clear pattern (αN = 7.3 ± 0.1 G and αH = 3.9 ± 0.1 G) associated with the oxidation product of 5,5-dimethylpyrrolidone-2-(oxy)-(1) or 5,5-dimethyl-1-pyrrolidone-2-oxyl (DMPO-X) is identified.49,50 This abnormal phenomenon should be attributed to the fact that the extremely strong oxidative ability from the interaction of PNC-800 and PMS also accounts for the oxidation of DMPO. If DMPO is injected into the solution at different times (2 min and 5 min) during the reaction, the obtained DMPO-X signals become very weak and eventually disappear (Fig. 5d). This result exactly reflects that some special intermediates with strong oxidative properties may be created rapidly when PNC-800 interacts with PMS in the reaction system and are consumed quickly by the direct oxidation (non-radical pathway) of organic pollutants. The interaction between PNC-800 and PMS was captured by Raman spectroscopy (Fig. S9), where a newly formed peak at 834 cm−1 can be ascribed to the formation of peroxo species bonded to sp2-hybridized carbon.51

In previous studies, PMS activation over certain carbocatalysts has been considered to be a routine radical process, where zigzag edges with delocalized π-electrons and several oxygen-containing groups can facilitate electron transfer from the carbocatalyst to PMS.52–54 In these studies, the degradation of organic pollutants has involved three steps. PMS is first attached in close proximity to the redox active surface of the catalyst by inner-sphere complexation, and then the peroxide bond (–O–O–) of PMS is homolytically cleaved by accepting one electron from the carbocatalyst to produce SO4˙ and ˙OH. Finally, the organic pollutants are oxidized and decomposed into carbon dioxide, water and mineralized acids. The generation of SO4˙ and ˙OH from PMS can be summarized through the following redox reactions:55,56

C–π + HSO5 → C–π+ + OH + SO4˙(2)
C–π+ + HSO5 → C–π + H+ + SO5˙(3)
HSO5 → OH + SO4˙(4)
HSO5 → ˙OH + SO4(5)

Recently, some researchers discovered that N-doped carbocatalysts could induce a unique non-radical oxidation process in PMS/PS activation, and almost all of them appointed graphitic N atoms as the primary active sites.23,25,57–59 Based on experimental results and density functional theory (DFT) calculations, Duan et al. confirmed that graphitic N in carbon frameworks could effectively break the chemical inertness of the sp2-hybridized carbon configuration, inducing charge transfer from neighboring carbon atoms to graphitic nitrogen atoms, which gives rise to positively-charged sites.23 As a result, these positively charged sites effectively improve the adsorption capability of PMS and produce electron transfer intermediates. Unlike the conventional radical-based PMS activation process, the electron transfer intermediate acts as a favorable platform for direct electron transfer from organic molecules to PMS instead of generating SO4˙ as the reactive species through a redox reaction.13,20,58 However, it should be pointed out that this process is also highly related to the crystallinity of the carbocatalyst. If the carbocatalyst has a high graphitization degree, the transferred electron will be quickly transmitted through the conjugated π-network, and this non-radical process will continue, leading to further decomposition of organic pollutants (pathway I in Scheme 2). In contrast, if the carbocatalyst has poor crystallinity (e.g. amorphous carbon), the transferred electron from organic molecules cannot be rapidly spread out, and the accumulated charge will induce the cleavage of the peroxide bond (–O–O–) of PMS to generate SO4˙ and ˙OH, resulting in a conventional radical oxidation process (pathway II in Scheme 2). In view of these characteristics, the non-radical process usually occurs in the presence of highly graphitic carbocatalysts containing N heteroatoms, such as N-doped CNTs and N-doped graphene.20,23,25,59 In our case, PNC-800 is composed of highly graphitic carbon frameworks (Fig. 2a and b), and its rich graphitic N sites also ensure desirable integration with PMS (Fig. S9). Therefore, for the PNC-800/PMS system herein, a non-radical degradation process dominates.

image file: c7ta08472b-s2.tif
Scheme 2 Mechanism of PMS activation on PNC-X microspheres.

As we all know, certain common anions and natural organic matter (NOM) in wastewater compete with organic contaminants to react with radicals, subsequently changing the acidic/basic conditions of the reaction system and thus interfering with the catalytic efficiency in radical-based AOPs.11 Therefore, various anions (Cl, H2PO4 and HCO3) and natural organic matter (humic acid) were necessarily introduced in the reaction solution to evaluate their effects on MB degradation over PNC-800 (Fig. 6). In general, SO4˙ is vulnerable to being scavenged by halogen ions such as Cl, because SO4˙ can oxidize Cl to chlorine radicals (Cl˙), which presents a lower redox potential for organic mineralization and can react with organics to form refractory chloride.60Fig. 6a shows that MB degradation is slightly suppressed at 10 mM and 15 mM of Cl in the solution. However, this effect is very weak, and 92.0% of MB degradation can still be achieved within 15 min, suggesting that the shielding effect of Cl is almost noneffective in the current system. Previous studies have shown that H2PO4 is not only a typical quencher for ˙OH, but also a much stronger complex ion for the chelating reaction of binding the catalyst with the surface –OH and Me–O groups.61 However, H2PO4 with a concentration of between 5 mM and 15 mM has almost no impact on the MB removal efficiency in our system (Fig. 6b). Bicarbonate (HCO3) is a representative radical scavenger in conventional AOPs.51 In the PNC-800/PMS system, HCO3 shows a detrimental effect on the MB degradation at the lower concentration of 5 mM, and a reinforced MB degradation at the higher concentration of 15 mM (Fig. 6c). This result is the opposite of that observed in conventional identification,62 and can be explained by the fact that a low concentration of HCO3 is able to tune the acidic properties and surface charge of PNC-800, whereas when excessive HCO3 is present, the reaction solution changes to an alkaline buffer solution, which restricts the tendency for a pH decrease during the catalytic process.20 Humic acid (HA) is a kind of ubiquitous NOM in natural waters, and its abundant phenolic hydroxyl and carboxyl groups may absorb onto the surfaces of heterogeneous catalysts and block the most active sites.63 Herein, the implication of HA is also taken into account in MB degradation in the PNC-800/PMS system (Fig. 6d). Apparently, when the concentrations of HA are 10 mg L−1, 20 mg L−1 and 40 mg L−1, the degradation efficiency of MB is slightly inhibited to 79.3%, 76.3% and 70.0%, respectively, within 2 min. However, it should be noted that the inhibiting effects of the relatively higher HA concentration in our systems are still weaker than those of some other previously reported systems.8,13,20 According to the above results, one can see that the influence of these typical background species on MB degradation is very feeble, and the non-radical system can effectively resist the interference of radical scavengers. In addition, the effects of catalyst dosage, Oxone concentration, initial pH, and reaction temperature on MB degradation with PNC-800 were further evaluated (Fig. 7). The degradation efficiency of MB increases from 66.1% with 0.05 g L−1 of catalyst to 85.0% with 0.10 g L−1 of catalyst within 2 min (Fig. 7a), suggesting that the increased loading of catalyst provides more active sites for PMS to interact with the sp2-hybridized system of PNC-800. However, further increasing the catalyst loading does not noticeably improve the MB removal efficiency under the limited dye concentration. Similarly, the degradation efficiency of MB is promoted slightly within 2 min with the increase of PMS concentration (Fig. 7b), but this improvement is indistinguishable, which is mainly because the limited active sites on the surface of PNC-800 cannot be occupied by excessive PMS.62 It is well known that PMS activation in radical systems is favored under neutral or alkaline conditions,64 and the conventional Fenton reaction is only limited within a narrow pH range (3–4).65 In our case, when the initial pH value of the reaction system increases from 6.3 to 13.2, almost all of the MB is removed within 10 min, and even at a pH of 3.4, 91.5% of MB is also degraded within 15 min (Fig. 7c). This fact demonstrates that the non-radical process can effectively extend the applicable pH range of the catalyst. The effect of reaction temperature on PMS activation by PNC-800 is demonstrated in Fig. 7d. When the temperature is increased from 283 K to 308 K, the MB degradation efficiency is enhanced slightly. The activation energy (Ea) of MB degradation on the surface of PNC-800 was calculated by plotting ln[thin space (1/6-em)]k against 1/T based on the Arrhenius equation, and was determined to be about 15.15 kJ mol−1 (the inset of Fig. 7d). Although it is difficult to compare the catalytic performances of various heterogeneous catalysts under different experimental conditions, this low Ea is significantly less than those of previously reported high-performance heterogeneous catalysts [CoMoO4 (69.89 kJ mol−1), Fe3O4/Mn3O4/rGO (25.49 kJ mol−1), Mn3O4/rGO (49.50 kJ mol−1)].66–68 Moreover, it should be mentioned that the relatively low Ea is slightly higher but close to that of diffusion controlled reactions, which usually ranges from 10 to 13 kJ mol−1.69 This result confirms that the apparent reaction rate for this process is still dominated by the rate of intrinsic chemical reactions on the surface of PNC-800; however, this reaction rate is very fast and almost comparable to the rate of mass transfer. The stability of the catalyst was assessed by three cycling tests of MB degradation on PNC-800 (Fig. S10). The comparable MB removal efficiencies in repeated batch experiments demonstrate the excellent catalytic stability of PNC-800. In order to testify the universal applicability of the catalyst, PNC-800 was also employed as a heterogeneous catalyst to deal with the degradations of two other organic dyes (rhodamine B (RhB) and orange II), two phenolic compounds (phenol and bisphenol A (BPA)) and an antibiotic compound (tetracycline (TC)) (Fig. 8a). The degradation efficiencies of RhB, orange II, phenol, BPA and TC over PNC-800 reach up to 92.8%, 93.2%, 89.8%, 87.3% and 92.7% within 10 min, respectively, and almost all pollutants can be ultimately removed, implying that PNC-800 is indeed an excellent heterogeneous catalyst for the disposal of different organic pollutants. In addition, MeOH was also utilized as a universal scavenger for both SO4˙ and ˙OH in the degradation of BPA and TC (Fig. 8b and c). As can be seen, when BPA and TC degradation is performed in absolute MeOH solution, the removal efficiencies of BPA and TC are reduced to 79.1% and 87.0% within 10 min, respectively. However, removal efficiencies of over 90% for BPA and TC can be still achieved within 30 min. Therefore, it can be inferred that the non-radical pathway also works for the degradation of these different types of organic pollutants. Real samples from three bodies of water were then used to investigate the integrated effects of background ions and organic matter on the degradation of these different pollutants (Fig. 8d–f). The degradation efficiencies of MB in surface water, mining sewage and tap water are all above 90.0% within 10 min, which is close to that obtained for Milli-Q water (Fig. 8d). Similar phenomena can also be observed in the degradation of BPA and TC, and their degradation efficiencies in surface water, mining sewage and tap water are 84.5%, 82.9%, 83.8% and 90.7%, 89.8%, 88.9% within 10 min, respectively, and the corresponding decreases in removal efficiency are less than 5.0% (Fig. 8e and f). These results further verify that this non-radical degradation process in the PNC-800/PMS system is favorable for eliminating the negative effects of various impurities, such as NOMs and other trace ions in actual bodies of water. Furthermore, the key step in the practical application of heterogeneous catalysts is to mold them into particulates before filling them into a fixed bed reactor.70 This process effectively reduces the loss of the catalyst in suspension systems and facilitates the recovery and reuse of the catalyst. The PNC microspheres herein have a stable microstructure, and can survive the conventional molding process to ensure good mass transfer of organic pollutants and oxidants. Therefore, PNC microspheres are expected to be used in industrial degradation of organic pollutants in the future.

image file: c7ta08472b-f6.tif
Fig. 6 The influence of radical scavengers (a) Cl, (b) H2PO4, (c) HCO3, and (d) humic acid (HA) on MB degradation with PNC-800. Reaction conditions: [MB] = 100 mg L−1, [Oxone] = 1.0 g L−1, catalyst = 0.1 g L−1, T = 25 °C, pH = 6.30.

image file: c7ta08472b-f7.tif
Fig. 7 The influence of catalyst dosage (a), Oxone concentration (b), pH value (c), and reaction temperature (d) on MB degradation with PNC-800. Reaction conditions: [MB] = 100 mg L−1, [Oxone] = 1.0 g L−1, catalyst = 0.1 g L−1, T = 25 °C, pH = 6.30.

image file: c7ta08472b-f8.tif
Fig. 8 The degradation of different organic pollutants in PNC-800/PMS systems (a); the effects of methanol quenching on BPA (b) and TC (c) degradation over PNC-800; the influence of samples from real bodies of water on the degradation of MB (d), BPA (e) and TC (f) over PNC-800. Reaction conditions: [MB, RhB, orange II, TC] = 100 mg L−1, [phenol, BPA] = 25 mg L−1, [Oxone] = 1.0 g L−1, catalyst = 0.1 g L−1, T = 25 °C, pH = 6.30.


Porous nitrogen-doped carbon (PNC) microspheres constructed through the in situ pyrolysis of Zn–Co PBAs were employed as a heterogeneous catalyst for PMS activation. The architecture of the metal centers/clusters bound by cyanide groups (–C[triple bond, length as m-dash]N) in the Zn–Co PBAs endows the PNC microspheres with abundant porosity, a high graphitization degree, and rich nitrogen substitution. PNC-800 displays better PMS activation performance toward MB compared with other common carbon materials and homologous N-doped carbocatalysts derived from ZIF-8 and ZIF-67. The degradation efficiency of MB is almost completely uninhibited by the introduction of various anions, natural organic matter and oxidant scavengers. Radical quenching and trapping experiments further prove the absence of reactive radicals, and a non-radical mechanism is proposed to be responsible for the MB degradation process. This non-radical pathway and good catalytic performance of the PNC-800/PMS system are also universal in the degradation of other typical organic pollutants. Furthermore, the high graphitization degree and rich surface graphitic N sites of PNC-800 are considered to be the critical factors behind this non-radical pathway. It is believed that this work may provide new insights into the use of nitrogen-rich nanocarbons as heterogeneous catalysts for PMS activation, opening up a new avenue for the design and preparation of various high-performance carbocatalysts in the future.

Conflicts of interest

There are no conflicts to declare.


This work is supported by the financial support from Natural Science Foundation of China (21676065, 21371039, and 21571043).

Notes and references

  1. A. Y. Hoekstra, Nat. Clim. Change, 2014, 4, 318–320 CrossRef .
  2. C. Barrera-Díaz, I. Linares-Hernández, G. Roa-Morales, B. Bilyeu and P. Balderas-Hernández, Ind. Eng. Chem. Res., 2008, 48, 1253–1258 CrossRef .
  3. P. A. Neale, A. Antony, M. E. Bartkow, M. J. Farre, A. Heitz, I. Kristiana, J. Y. Tang and B. I. Escher, Environ. Sci. Technol., 2012, 46, 10317–10325 CAS .
  4. J. Giménez, B. Bayarri, Ó. González, S. Malato, J. Peral and S. Esplugas, ACS Sustainable Chem. Eng., 2015, 3, 3188–3196 CrossRef .
  5. I. Velo-Gala, J. A. Pirán-Montaño, J. Rivera-Utrilla, M. Sánchez-Polo and A. J. Mota, Chem. Eng. J., 2017, 323, 605–617 CrossRef CAS .
  6. X. Xue, K. Hanna and N. Deng, J. Hazard. Mater., 2009, 166, 407–414 CrossRef CAS PubMed .
  7. R. B. P. Marcelino, M. M. D. Leao, R. M. Lago and C. C. Amorim, J. Environ. Manage., 2017, 195, 110–116 CrossRef CAS PubMed .
  8. H. Lee, H.-J. Lee, J. Jeong, J. Lee, N.-B. Park and C. Lee, Chem. Eng. J., 2015, 266, 28–33 CrossRef CAS .
  9. N. Wang, Y. Du, W. Ma, P. Xu and X. Han, Appl. Catal., B, 2017, 210, 23–33 CrossRef CAS .
  10. Y. Zhou, J. Jiang, Y. Gao, J. Ma, S. Y. Pang, J. Li, X. T. Lu and L. P. Yuan, Environ. Sci. Technol., 2015, 49, 12941–12950 CrossRef CAS PubMed .
  11. P. Hu and M. Long, Appl. Catal., B, 2016, 181, 103–117 CrossRef CAS .
  12. T. Zhang, Y. Chen, Y. Wang, J. Le Roux, Y. Yang and J. P. Croue, Environ. Sci. Technol., 2014, 48, 5868–5875 CrossRef CAS PubMed .
  13. H. Lee, H. I. Kim, S. Weon, W. Choi, Y. S. Hwang, J. Seo, C. Lee and J. H. Kim, Environ. Sci. Technol., 2016, 50, 10134–10142 CrossRef CAS PubMed .
  14. B. Jiang, D. Dai, Y. Yao, T. Xu, R. Li, R. Xie, L. Chen and W. Chen, Chem. Commun., 2016, 52, 9566–9569 RSC .
  15. J. Chen, L. Zhang, T. Huang, W. Li, Y. Wang and Z. Wang, J. Hazard. Mater., 2016, 320, 571–580 CrossRef CAS PubMed .
  16. E. Saputra, S. Muhammad, H. Sun and S. Wang, RSC Adv., 2013, 3, 21905–21910 RSC .
  17. H. Sun, Y. Wang, S. Liu, L. Ge, L. Wang, Z. Zhu and S. Wang, Chem. Commun., 2013, 49, 9914–9916 RSC .
  18. M. Wei, L. Gao, J. Li, J. Fang, W. Cai, X. Li and A. Xu, J. Hazard. Mater., 2016, 316, 60–68 CrossRef CAS PubMed .
  19. X. Duan, C. Su, L. Zhou, H. Sun, A. Suvorova, T. Odedairo, Z. Zhu, Z. Shao and S. Wang, Appl. Catal., B, 2016, 194, 7–15 CrossRef CAS .
  20. X. Duan, Z. Ao, L. Zhou, H. Sun, G. Wang and S. Wang, Appl. Catal., B, 2016, 188, 98–105 CrossRef CAS .
  21. Y. Li, Y. Jiang, Z. Ruan, K. Lin, Z. Yu, Z. Zheng, X. Xu and Y. Yuan, J. Mater. Chem. A, 2017, 5, 21300–21312 CAS .
  22. H. Liu, P. Sun, M. Feng, H. Liu, S. Yang, L. Wang and Z. Wang, Appl. Catal., B, 2016, 187, 1–10 CrossRef CAS .
  23. X. Duan, Z. Ao, H. Sun, L. Zhou, G. Wang and S. Wang, Chem. Commun., 2015, 51, 15249–15252 RSC .
  24. X. Duan, Z. Ao, H. Sun, S. Indrawirawan, Y. Wang, J. Kang, F. Liang, Z. H. Zhu and S. Wang, ACS Appl. Mater. Interfaces, 2015, 7, 4169–4178 CAS .
  25. X. Duan, H. Sun, Y. Wang, J. Kang and S. Wang, ACS Catal., 2015, 5, 553–559 CrossRef CAS .
  26. R. Qiang, Y. Du, D. Chen, W. Ma, Y. Wang, P. Xu, J. Ma, H. Zhao and X. Han, J. Alloys Compd., 2016, 681, 384–393 CrossRef CAS .
  27. K. Y. Lin and B. C. Chen, Dalton Trans., 2016, 45, 3541–3551 RSC .
  28. G. Wang, S. Chen, X. Quan, H. Yu and Y. Zhang, Carbon, 2017, 115, 730–739 CrossRef CAS .
  29. C. Lundgren and R. W. Murray, Inorg. Chem., 1988, 27, 933–939 CrossRef CAS .
  30. B. K. Barman and K. K. Nanda, Green Chem., 2016, 18, 427–432 RSC .
  31. K. A. Lin and B. J. Chen, Chemosphere, 2017, 166, 146–156 CrossRef CAS PubMed .
  32. R. Qiang, Y. Du, H. Zhao, Y. Wang, C. Tian, Z. Li, X. Han and P. Xu, J. Mater. Chem. A, 2015, 3, 13426–13434 CAS .
  33. L. Ma, R. Chen, Y. Hu, G. Zhu, T. Chen, H. Lu, J. Liang, Z. Tie, Z. Jin and J. Liu, Nanoscale, 2016, 8, 17911–17918 RSC .
  34. J. Deng, Y. J. Chen, Y. A. Lu, X. Y. Ma, S. F. Feng, N. Gao and J. Li, Environ. Sci. pollut. Res. Int., 2017, 24, 14396–14408 CrossRef CAS PubMed .
  35. P. Parent, C. Laffon, I. Marhaba, D. Ferry, T. Z. Regier, I. K. Ortega, B. Chazallon, Y. Carpentier and C. Focsa, Carbon, 2016, 101, 86–100 CrossRef CAS .
  36. A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14095 CrossRef CAS .
  37. X. Yue, S. Huang, Y. Jin and P. K. Shen, Catal. Sci. Technol., 2017, 7, 2228–2235 CAS .
  38. K. S. Sing, Pure Appl. Chem., 1985, 57, 603–619 CrossRef CAS .
  39. I. Sierra, U. Iriarte-Velasco, M. Gamero and A. T. Aguayo, Microporous Mesoporous Mater., 2017, 250, 88–99 CrossRef CAS .
  40. S. Li, Z. Yu, Y. Yang, Y. Liu, H. Zou, H. Yang, J. Jin and J. Ma, J. Mater. Chem. A, 2017, 5, 6405–6410 CAS .
  41. Y. He, X. Han, Y. Du, B. Song, P. Xu and B. Zhang, ACS Appl. Mater. Interfaces, 2016, 8, 3601–3608 CAS .
  42. W. Ma, Y. Du, N. Wang and P. Miao, Environ. Sci. Pollut. Res. Int., 2017, 24, 16276–16288 CrossRef CAS PubMed .
  43. Z. Yan, Q. Hu, G. Yan, H. Li, K. Shih, Z. Yang, X. Li, Z. Wang and J. Wang, Chem. Eng. J., 2017, 321, 495–501 CrossRef CAS .
  44. X. Tian, P. Gao, Y. Nie, C. Yang, Z. Zhou, Y. Li and Y. Wang, Chem. Commun., 2017, 53, 6589–6592 RSC .
  45. X. Li, J. Liu, A. I. Rykov, H. Han, C. Jin, X. Liu and J. Wang, Appl. Catal., B, 2015, 179, 196–205 CrossRef CAS .
  46. K. Zhu, J. Wang, Y. Wang, C. Jin and A. S. Ganeshraja, Catal. Sci. Technol., 2016, 6, 2296–2304 CAS .
  47. Y. Wang, D. Cao, M. Liu and X. Zhao, Catal. Commun., 2017, 102, 85–88 CrossRef CAS .
  48. Y. Wang, D. Cao and X. Zhao, Chem. Eng. J., 2017, 328, 1112–1121 CrossRef CAS .
  49. S. V. Verstraeten, S. Lucangioli and M. Galleano, Inorg. Chim. Acta, 2009, 362, 2305–2310 CrossRef CAS .
  50. R. A. Floyd and L. M. Soong, Biochem. Biophys. Res. Commun., 1977, 74, 79–84 CrossRef CAS PubMed .
  51. Z. Wang, Y. Du, Y. Liu, B. Zou, J. Xiao and J. Ma, RSC Adv., 2016, 6, 11040–11048 RSC .
  52. S. Yang, X. Yang, X. Shao, R. Niu and L. Wang, J. Hazard. Mater., 2011, 186, 659–666 CrossRef CAS PubMed .
  53. W. Peng, S. Liu, H. Sun, Y. Yao, L. Zhi and S. Wang, J. Mater. Chem. A, 2013, 1, 5854–5859 CAS .
  54. H. Sun, S. Liu, G. Zhou, H. M. Ang, M. O. Tade and S. Wang, ACS Appl. Mater. Interfaces, 2012, 4, 5466–5471 CAS .
  55. K.-Y. A. Lin and Z.-Y. Zhang, Chem. Eng. J., 2017, 313, 1320–1327 CrossRef CAS .
  56. W.-D. Oh, Z. Dong and T.-T. Lim, Appl. Catal., B, 2016, 194, 169–201 CrossRef CAS .
  57. P. Hu, H. Su, Z. Chen, C. Yu, Q. Li, B. Zhou, P. J. J. Alvarez and M. Long, Environ. Sci. Technol., 2017, 51, 11288–11296 CrossRef CAS PubMed .
  58. N. Liu, L. Zhang, Y. Xue, J. Lv, Q. Yu and X. Yuan, Sep. Purif. Technol., 2017, 184, 213–219 CrossRef CAS .
  59. D. Li, X. Duan, H. Sun, J. Kang, H. Zhang, M. O. Tade and S. Wang, Carbon, 2017, 115, 649–658 CrossRef CAS .
  60. M. G. Antoniou, A. A. de la Cruz and D. D. Dionysiou, Appl. Catal., B, 2010, 96, 290–298 CrossRef CAS .
  61. F. Qi, W. Chu and B. Xu, Chem. Eng. J., 2014, 235, 10–18 CrossRef CAS .
  62. Y. H. Guan, J. Ma, Y. M. Ren, Y. L. Liu, J. Y. Xiao, L. Q. Lin and C. Zhang, Water Res., 2013, 47, 5431–5438 CrossRef CAS PubMed .
  63. Y. Yao, Y. Cai, G. Wu, F. Wei, X. Li, H. Chen and S. Wang, J. Hazard. Mater., 2015, 296, 128–137 CrossRef CAS PubMed .
  64. C. Gong, F. Chen, Q. Yang, K. Luo, F. Yao, S. Wang, X. Wang, J. Wu, X. Li, D. Wang and G. Zeng, Chem. Eng. J., 2017, 321, 222–232 CrossRef CAS .
  65. L. Clarizia, D. Russo, I. Di Somma, R. Marotta and R. Andreozzi, Appl. Catal., B, 2017, 209, 358–371 CrossRef CAS .
  66. Y. Fan, W. Ma, J. He and Y. Du, RSC Adv., 2017, 7, 36193–36200 Search PubMed .
  67. B. Yang, Z. Tian, B. Wang, Z. Sun, L. Zhang, Y. Guo, H. Li and S. Yan, RSC Adv., 2015, 5, 20674–20683 RSC .
  68. Y. Yao, C. Xu, S. Yu, D. Zhang and S. Wang, Ind. Eng. Chem. Res., 2013, 53, 3637–3645 CrossRef .
  69. Y. Du, W. Ma, P. Liu, B. Zou and J. Ma, J. Hazard. Mater., 2016, 308, 58–66 CrossRef CAS PubMed .
  70. D. D. Phan, F. Babick, M. T. Nguyen, B. Wessely and M. Stintz, Chem. Eng. Sci., 2017, 173, 242–252 CrossRef CAS .


Electronic supplementary information (ESI) available: SEM and XRD image of Zn3[Co(CN)6]2·12H2O; SEM, TEM and XRD of Co3ZnC; TEM and SAED patterns of PNC-X; XPS survey of PNC-X; N 1s XPS, XRD and Raman spectra of PNC-800, HNC-8 and HNC-67; Raman spectra of PMS and PNC-800 in PMS solution; stability tests of PNC-800. See DOI: 10.1039/c7ta08472b

This journal is © The Royal Society of Chemistry 2018