Facile Synthesis of Magnetic Nitrogen-Doped Porous Carbon ... - MDPI

nanomaterials Article

Facile Synthesis of Magnetic Nitrogen-Doped Porous Carbon from Bimetallic Metal–Organic Frameworks for Efficient Norfloxacin Removal Hui Wang 1 , Xi Zhang 2 , Yan Wang 1 , Guixiang Quan 1 , Xiangyun Han 1 and Jinlong Yan 1, * 1 2

*

School of Environmental Science and Engineering, Yancheng Institute of Technology, Yancheng 224051, China; [email protected] (H.W.); [email protected] (Y.W.); [email protected] (G.Q.); [email protected] (X.H.) College of Life and Environmental Science, Shanghai Normal University, Shanghai 200234, China; [email protected] Correspondence: [email protected]; Tel.: +86-515-8829-8805

Received: 6 August 2018; Accepted: 23 August 2018; Published: 26 August 2018

Abstract: Magnetic nitrogen-doped porous carbon (MNPC) has been prepared via self-catalytic pyrolysis of bimetallic metal-organic frameworks (MOFs). The as-obtained MNPC showed favorable features for antibiotics adsorption such as high specific surface area (871 m2 g−1 ), high pore volume (0.75 cm3 g−1 ), porous structure, good graphitization degree, and rich N-doping. Moreover, the MNPC has magnetic properties due to the Co species, which is embedded with a high dispersion, so the absorbent can be easily separated. Based on the above excellent characteristics, the MNPC was used as the absorbent for norfloxacin (NOR) removal. The experimental maximum NOR adsorption capacity of MNPC was 55.12 mg g−1 at 298.15 K and a pH of 6.0 with an initial NOR concentration of 50 mg L−1 . The data analysis of the kinetics revealed that the experimental data of NOR uptakes versus time agreed with the pseudo-second order model. The isotherm data analysis revealed the favorable application of the Freundlich model. Based on the adsorption results over a wide range of conditions, the dominant adsorption mechanisms were found to be pore-filling, electrostatic interaction, and the H-bond. Keywords: self-catalytic pyrolysis; porous carbon; metal–organic frameworks; antibiotics; adsorption

1. Introduction Over the past few decades, the emission of pharmaceutical compounds into the environment has sharply increased due to fast population growth and the rapid expansion of the pharmaceuticals industry. Antibiotics are one of most important type of pharmaceuticals, and are usually used as drugs or feed additives [1–4]. However, large amounts of antibiotics are stable and cannot be easily degraded, thus they are persistent in the environment. In addition, antibiotics could generate antibiotic-resistance genes in microorganisms, which can proliferate and widely disseminate in ecosystems. Fluoroquinolones are a commonly used antibiotic and their concentration is relatively high in the environment [5,6]. Norfloxacin (NOR) is one of the most frequently used fluoroquinolone antibiotics, and is always used to treat infectious diseases. It has been detected in the surface water and found to be toxic to aquatic organisms and human beings [7,8]. Therefore, it is necessary to develop a cost-effective method to remove NOR from wastewater. Such methods as advanced oxidation, electrochemical methods, and biological treatments have been applied extensively to remove NOR from wastewaters. Among all of these methods, adsorption is the top priority owing to its simplicity, low operating cost, safety, and efficiency [9–11]. With such advantages as a large specific surface area and porous structure, carbon materials have been applied to remove NOR from water. For example, Xing et al. investigated the adsorption of norfloxacin (NOR) onto multiwall carbon nanotubes and activated carbon, and the results showed that activated carbon Nanomaterials 2018, 8, 664; doi:10.3390/nano8090664

www.mdpi.com/journal/nanomaterials

Nanomaterials 2018, 8, 664

2 of 15

(AC) has a better NOR sorption capacity due to its higher surface area [12]. Theydan et al. prepared AC from a lignocellulosic biomass to remove NOR from water, and a maximum removal percentage of 98.13% was achieved [13]. Although a significant amount of research has been expended on adsorbing materials for antibiotics removal during the past few decades, development of novel adsorbents with higher performance is still needed. As a new class of porous inorganic–organic materials, metal–organic frameworks (MOFs), have attracted wide attention owing to their high surface area and tunable pore size, which is widely used in areas of drug delivery, gas storage, and separation and catalysis [14–17]. Recently, MOFs have been used as templates or novel sources to prepare porous carbons through further carbonization. For example, Xu et al. applied MOFs as sacrificial templates to synthesis nanoporous carbons for the first time. They introduced furfuryl alcohol into the MOF-5 through a vapor phase protocol, which was then carbonized at 1000 ◦ C under an inert atmosphere to obtain porous carbon [18]. Park et al. presented hierarchically porous carbon from highly crystalline MOFs and used it as a hydrogen storage adsorbent [19]. Most recently, Huang et al. demonstrated the application of MOF-derived porous carbon as an adsorbent for antibiotics removal. They prepared porous carbon through a one-step carbonization of zeolitic imidazolate framework-8 (ZIF-8), which showed a larger specific surface area due to the vaporization of the center metal (Zn) of ZIF-8 during the pyrolysis process. They were further used for ciprofloxacin removal from water [20]. Although porous carbon derived from ZIF-8 has a high specific surface area for antibiotics adsorption, it has limitations in terms of adsorption capacity and ease of separation. Recently, the sustainability of the adsorption process has been advocated, such as: green adsorbent and green separation methods [21–23]. It has been demonstrated that the incorporation of magnetic nanoparticles on the surface of adsorbents can be engineered to allow the magnetic separation and recovery of the absorbents [24,25]. Wang et al. synthesized reduced graphene oxide/magnetite composites through an in situ reaction and utilized it as an adsorbent with a magnetically separable property for fluoroquinolone antibiotics [26]. Cai et al. encapsulated magnetic nanoparticles into carbon with a well-constructed core-shell structure, and then used it as an adsorbent for organic pollutants isolation [27]. However, the preparation of magnetic adsorbents usually needs additional processes to load the magnetic metal oxide, and it is difficult to control the dispersion of loaded particles during the synthesis process. Herein, we report a simple but efficient solution process for the fabrication of a new form of magnetic nitrogen-doped porous carbon (MNPC) adsorbents for the NOR removal. The MNPC was directly synthesized by self-catalytic pyrolysis of bimetallic MOFs, which were prepared by using divalent Zn2+ and Co2+ as center metal ions and 2-methylimidazole as the ligand (Figure 1). In the carbonization process, the Zn, with a boiling point of around 900 ◦ C, was evaporated during the calcination process, and the porous structure was formed simultaneously. Furthermore, the Co species were embedded in the porous structure with a high dispersion due to the coordination structure of the MOF’s precursor, and so the MNPC had magnetic properties. Moreover, the Co species can act as catalyst to improve the graphitization degree of MNPC, which can enhance the adhesion between antibiotics and adsorbents through π–π conjugation. More importantly, by the development of such multiple structures, the adsorption performance was significantly enhanced.

Nanomaterials 2018, 8, 664 Nanomaterials 2018, 8, x FOR PEER REVIEW

3 of 15 3 of 14

Figure 1. illustration of of the the construction construction process process for for the the magnetic Figure 1. Schematic Schematic illustration magnetic nitrogen-doped nitrogen-doped porous porous carbon (MNPC). carbon (MNPC).

2. Materials and Methods

2.1. Synthesis Synthesis of of MNPC MNPC 2.1. Materials: The The zinc zinc nitrate 6H22O), O), cobaltous cobaltous nitrate nitrate hexahydrate Materials: nitrate hexahydrate hexahydrate (Zn(NO (Zn(NO33))22··6H hexahydrate (Co(NO ) · 6H O), 2-methylimidazole and norfloxacin were analytic grade provided by Aladdin Aladdin 3 2 2 (Co(NO3)2·6H2O), 2-methylimidazole and norfloxacin were analytic grade provided by Chemical Reagent Co., Ltd., Shanghai, China. Chemical Reagent Co., Ltd., Shanghai, China. Synthesis of O (0.27 (0.27 g, 6H22O) O) (1.40 (1.40 g, g, Synthesis of MNPC: MNPC: Typically, Typically,Co(NO Co(NO33))22··6H 6H22O g, 0.9 0.9 mmol) mmol) and and (Zn(NO (Zn(NO33))22··6H 4.7 mmol) were first dissolved in 100 mL of methanol. 2-Methylimidazole (3.70 g, 45.1 mmol) in 4.7 mmol) were first dissolved in 100 mL of methanol. 2-Methylimidazole (3.70 g, 45.1 mmol) in 100 100 mL methanol added toabove the above solution. quickly stirring 24 h, the products mL methanol was was thenthen added to the solution. AfterAfter quickly stirring for 24for h, the products were were separated by centrifugation and washed thoroughly with methanol. The obtained bimetallic separated by centrifugation and washed thoroughly with methanol. The obtained bimetallic MOFs ◦ for 24 h under a vacuum before MOFsdried wereat dried at overnight, 50 ◦ C overnight, and further activated were 50 °C and further activated at 200at°C200 for C 24 h under a vacuum before use. ◦ C with the ramp rate use. as-synthesized The as-synthesized bimetallic nanocrystals heated 950with The bimetallic MOFsMOFs nanocrystals were were heated to 950to°C the ramp rate of 3 ◦ ◦ of 3 C/min under a N atmosphere and carbonized at 950 C for 2 h, and then cooled to room 2 °C/min under a N2 atmosphere and carbonized at 950 °C for 2 h, and then cooled to room temperature temperature naturally. Finally, the MNPC was entirely fabricated. The magnetic carbon (MC) prepared naturally. Finally, the MNPC was entirely fabricated. The magnetic carbon (MC) prepared by MOFs by MOFs only as Coaions as a central wasfor used for comparison. The porous carbon prepared with onlywith Co ions central ion wasion used comparison. The porous carbon (PC)(PC) prepared by by MOFs with only Zn ions as a central ion was also prepared. In fact, the methanol used in this MOFs with only Zn ions as a central ion was also prepared. In fact, the methanol used in this process process be recycled membranes tosustainable realize sustainable fabrication could becould recycled throughthrough membranes to realize fabrication [28,29]. [28,29]. 2.2. Adsorption Performance of MNPC 2.2. Adsorption Performance of MNPC The adsorption experiments on NOR were conducted in 250 mL stopper conical flasks, and then The adsorption experiments on NOR were conducted in 250 mL stopper conical flasks, and then placed in a thermostatic shaker with a speed of 200 rpm. In the adsorption experiment, 80 mg placed in a thermostatic shaker with a speed of 200 rpm. In the adsorption experiment, 80 mg of of adsorbent was added to 100 mL of adsorbate solution. The influence of initial concentrations adsorbent was added to 100 mL of adsorbate solution. The influence of initial concentrations (5–50 (5–50−1mg L−1 ), pH (2–10), temperature and ionic strength on the adsorption of NOR were also mg L ), pH (2–10), temperature and ionic strength on the adsorption of NOR were also investigated. investigated. The solution pH was adjusted by dilute HCl or NaOH solution. The concentration The solution pH was adjusted by dilute HCl or NaOH solution. The concentration of NOR was of NOR was measured with a UV–Visible spectrophotometer (TU-1810, Beijing Purkinje General measured with a UV–Visible spectrophotometer (TU-1810, Beijing Purkinje General Instrument Co. Instrument Co. Ltd., Beijing, China) at 272 nm. The adsorbed capacity (q) and removal rate (η) were Ltd., Beijing, China) at 272 nm. The adsorbed capacity (q) and removal rate (η) were calculated calculated according to the following equations: according to the following equations:

−t)V/m Ct )V/m q q= = (C(C 0 −0 C

(1) (1)

−t)/C Ct0)/C0 η η= = (C(C 0 −0 C

(2) (2)

−)1 )represent where C00 and Ctt (mg L−1 representthe theinitial initialand andfinal finalconcentrations concentrationsof ofNOR NORin inthe the feed feed solution, solution, respectively, V respectively, V is the volume of NOR solution (L), and m is is the the dry dry mass mass of of MNPC MNPC (g). (g).

3. Results


4 of 15

Nanomaterials 2018, 8, x FOR PEER REVIEW

4 of 14

3. Results 3.1. Characterization 3.1. Characterization The X-Ray diffraction (XRD, Rigaku D/Max-RB, Rigaku Corporation, Tokyo, Japan) The X-ray diffraction (XRD, Rigaku D/Max-RB, Rigaku Corporation, Tokyo, Japan) measurements is measurements is usually conducted to evaluate the structure of materials, and the XRD pattern of the usually conducted to evaluate the structure of materials, and the XRD pattern of the MNPC is presented MNPC is presented in Figure 2a. The MNPC shows an obvious diffraction peak at the 2θ = 26°, in Figure 2a. The MNPC shows an obvious diffraction peak at the 2θ = 26◦ , corresponding to the (002) corresponding to the (002) plane of the graphitic carbon [30]. The diffraction peaks located at around plane of the graphitic carbon [30]. The diffraction peaks located at around 44◦ and 51◦ are ascribed to fcc 44° and 51° are ascribed to fcc Co, which is embedded in the carbon shell [31,32]. There was no Co, which is embedded in the carbon shell [31,32]. There was no characteristic peak of Zn in the XRD characteristic peak of Zn in the XRD patterns due to the effective evaporation during the highpatterns due to the effective evaporation during the high-temperature calcination. The XRD pattern temperature calcination. The XRD pattern of MC was similar to that of MNPC (Figure S1). The of MC was similar to that of MNPC (Figure S1). The graphitization degree of MNPC was further graphitization degree of MNPC was further detected using Raman spectra (JY H800UV, Jobin-Yvon detected using Raman spectra (JY H800UV, Jobin-Yvon Corporation, Longjumeau, France), and the Corporation, Longjumeau, France), and the result is shown in Figure 2b. Two broad peaks at 1330 result is shown in Figure 2b. Two broad peaks at 1330 and 1583 cm−1 are obvious, and are related to and 1583 cm−1 are obvious, and are related to the D-band and G-band, respectively. The D-band is the D-band and G-band, respectively. The D-band is associated with defects in the carbon structure,2 associated with defects in the carbon structure, while the G-band is attributed to the vibration of sp while the G-band is attributed to the vibration of sp2 carbon atoms in both the rings and chain [25]. carbon atoms in both the rings and chain [25]. The graphitization degree of MNPC can be found by The graphitization degree of MNPC can be found by calculating the ratios of the integrated intensities calculating the ratios of the integrated intensities of the graphitic G-band to that of the disorderof the graphitic G-band to that of the disorder-induced D-band. As calculated, the value of IG /ID was induced D-band. As calculated, the value of IG/ID was 1.07, which was close to the MC (1.02, Figure 1.07, which was close to the MC (1.02, Figure S2a). However, the values of IG /ID were higher than that S2a). However, the values of IG/ID were higher than that of PC (0.94, Figure S2b) due to the catalytic of PC (0.94, Figure S2b) due to the catalytic action of Co. action of Co.

Figure diffraction (XRD) (XRD) pattern pattern and and (b) (b)Raman Ramanspectrum spectrumof ofthe theMNPC. MNPC. Figure 2. 2. (a) (a) X-Ray X-ray diffraction

To further probe for the chemical identification of elements in the MNPC, the X-ray To further probe for the chemical identification of elements in the MNPC, the X-ray photoelectron photoelectron spectroscopy (XPS, PHI-5000C ESCA system, Perkin–Elmer, Hopkinton, MA, USA) spectroscopy (XPS, PHI-5000C ESCA system, Perkin–Elmer, Hopkinton, MA, USA) measurements measurements were performed. According to the results, the elemental content of C, N, O, and Co were performed. According to the results, the elemental content of C, N, O, and Co was 80.29, 10.58, was 80.29, 10.58, 7.19, 1.95 at %, respectively. The C, N, O, and Co contents of MC are 90.67, 3.6, 4.38, 7.19, 1.95 at %, respectively. The C, N, O, and Co contents of MC are 90.67, 3.6, 4.38, and 1.34 at %, and 1.34 at %, respectively (Figure S3). The high-resolution C1s spectra (Figure 3a) could be fitted respectively (Figure S3). The high-resolution C1s spectra (Figure 3a) could be fitted with three peaks at with three peaks at 284.6, 286.4, and 287.8 eV, corresponding to the sp2 aromatic rings, C–O, and C=O, 2 284.6, 286.4, and 287.8 eV, corresponding to the sp aromatic rings, C–O, and C=O, respectively [33]. respectively [33]. The peak of sp2 carbon showed the strongest intensity, indicating that the MNPC The peak of sp2 carbon showed the strongest intensity, indicating that the MNPC predominantly predominantly 2consisted of sp2-hybridized carbon due to the effective catalytic graphitization. As consisted of sp -hybridized carbon due to the effective catalytic graphitization. As seen from the seen from the high-resolution N1s (Figure 3b), three different types of nitrogen species were well high-resolution N1s (Figure 3b), three different types of nitrogen species were well deconvoluted. deconvoluted. The N-6 atoms were located at 398.5 eV, and were bonded with two carbon atoms in The N-6 atoms were located at 398.5 eV, and were bonded with two carbon atoms in a C6 ring, so a a C6 ring, so a pair of lone electrons could be introduced simultaneously. This was beneficial to the pair of lone electrons could be introduced simultaneously. This was beneficial to the formation of a formation of a hydrogen bond with the NOR molecule. The N-5 was centered at 400.4 eV, associated hydrogen bond with the NOR molecule. The N-5 was centered at 400.4 eV, associated with the adjacent with the adjacent phenolic or carbonyl group. The N–Q atoms bond with three carbon atoms in the phenolic or carbonyl group. The N–Q atoms bond with three carbon atoms in the center of graphitic center of graphitic plane [34,35]. The additional N-doping can increase the adsorption sites for NOR, plane [34,35]. The additional N-doping can increase the adsorption sites for NOR, and further improve and further improve the adsorption performance of MNPC. the adsorption performance of MNPC.


5 of 15

Nanomaterials 2018, 8, x FOR PEER REVIEW Nanomaterials 2018, 8, x FOR PEER REVIEW

5 of 14 5 of 14

Figure Figure 3. 3. (a) (a) C C 1s 1s spectra spectra and and (b) (b) N N 1s 1s spectra spectra of of the the MNPC. MNPC. Figure 3. (a) C 1s spectra and (b) N 1s spectra of the MNPC.

The N2 sorption isothermal (ASAP 2020, Micromeritics Inc., Norcross, GA, USA) was further The N N22 sorption sorption isothermal (ASAP (ASAP 2020, Micromeritics Micromeritics Inc., Inc., Norcross, Norcross, GA, GA, USA) was was further further The examined to analyze isothermal the pore structure 2020, of MNPC. As seen from Figure 4a, the USA) MNPC showed a examined to analyze the pore structure of MNPC. As seen from Figure 4a, the MNPC showed a examined to analyze the pore structure of MNPC. As seen from Figure 4a, the MNPC showed typical IV-type isotherm with a hysteresis loop at p/p0 = 0.4–1.0 (inset), indicating the mesoporousa typical IV-type IV-type isotherm with loop at 0.4–1.0 (inset), (inset), indicating indicating the the mesoporous mesoporous typical isotherm with aa hysteresis hysteresis loop atp/p p/p00 == 0.4–1.0 structure of MNPC [36]. Figure 4b shows the Barrett–Joyner–Halenda (BJH) pore size distribution structure of MNPC [36]. Figure 4b shows the Barrett–Joyner–Halenda (BJH) pore size distribution structure of MNPC [36]. Figure 4b shows the Barrett–Joyner–Halenda (BJH) pore size distribution profile of MNPC derived from desorption branches of isotherms. Clearly, the MNPC pores’ radii was profile of MNPC derived from desorption branches of isotherms. Clearly, the MNPC pores’ radii was was profile MNPC derived branches of isotherms. Clearly, MNPC pores’ mainly of concentrated at 2.0from nm,desorption further indicating that the mesopores werethe dominant in the radii structure mainly concentrated concentrated at at 2.0 2.0 nm, nm, further further indicating indicating that that the2 mesopores mesopores were were dominant dominant in in the the structure mainly of MNPC. The specific surface area of MNPC was 871the m2 g−−11, much larger than that of MC structure obtained of MNPC. The specific surface area of MNPC was 871 m g , much larger than that of MC obtained 2 −1 of MNPC. The specific surface area of MNPC was 871 m g , much larger than that of MC obtained with the absence of a Zn ion (Figure S4). With Zn coordination, the ZnO would be formed during the with the absence of a Zn ion (Figure S4). With Zn coordination, the ZnO would be formed during the with the absence of a Znwhich ion (Figure With Zn coordination, the ZnO would be formed during the carbonization process, can actS4). as sacrificial template accelerating the formation of the porous carbonization process, process, which which can can act act as as sacrificial sacrificial template template accelerating the the formation formation of the the porous porous carbonization structure of MNPC [37,38]. Furthermore, the pore volume accelerating of MNPC was 0.76 cm33 g−−1of 4b), 1 (Figure structure of MNPC [37,38]. Furthermore, the pore volume of MNPC was 0.76 cm g (Figure 4b), 3 −1 structure of MNPC [37,38]. Furthermore, the pore volume of MNPC was 0.76 cm g (Figure 4b), 3 −1 which is much larger than that of MC (0.16 cm3 g−1). The increased specific surface area and pore which is much larger that (0.16 cm Theincreased increased specific specific surface surface area area and and pore which much larger than thanincreasing that of of MC MC cm3 gg−1surface ).). The volumeisare favorable for the(0.16 accessible area for NOR accumulation duringpore the volume are favorable for increasing the accessible surface area for NOR accumulation during the volume areprocess favorable increasing area for NOR accumulation during the adsorption andfor then enhance the the accessible adsorptionsurface capacity. adsorption process and then enhance the adsorption capacity. adsorption process and then enhance the adsorption capacity.

Figure 4. (a) N2 sorption isotherm and (b) pore size distribution of the MNPC. Figure Figure 4. 4. (a) (a) N N22 sorption sorptionisotherm isotherm and and (b) (b) pore pore size size distribution distribution of of the the MNPC. MNPC.

The surface morphology of MNPC was investigated with scanning electron microscopy (SEM, surface morphology of MNPC was investigated withmicroscopy scanning electron microscopy (SEM, JEOLThe JSM-6700F, Tokyo, Japan) and transmission electron (TEM, JEOL JEM-200CX, The surface morphology of MNPC was investigated with scanning electron microscopy (SEM, JEOL JSM-6700F, Tokyo, Japan) and transmission electron microscopy (TEM, JEOL JEM-200CX, Tokyo, Japan). As seen from the SEM image in Figure 5a, the bimetallic MOFs precursor shows a JEOL JSM-6700F, Tokyo, Japan) and transmission electron microscopy (TEM, JEOL JEM-200CX, Tokyo, Tokyo, Japan). As seen from the SEM image in Figure 5a, the bimetallic MOFs precursor shows a cubic-like an average size of 505a, nm. pyrolysis at 950 °C, theshows Zn species were Japan). Asstructure seen fromwith the SEM image in Figure theAfter bimetallic MOFs precursor a cubic-like cubic-like structure with an average size of 50 nm. After pyrolysis at 950 °C, the Zn species were volatilized, andan theaverage pores left simultaneously [38]. Moreover, the◦MNPC retained thewere morphology of structure with size of 50 nm. After pyrolysis at 950 C, the Zn species volatilized, volatilized, and the pores left simultaneously [38]. Moreover, the MNPCthat retained the morphology of the MOF’s precursor with a good dispersion (Figure 5b). This indicates the structure kept well, and the pores left simultaneously [38]. Moreover, the MNPC retained the morphology of the MOF’s the MOF’s precursor with a good dispersionThe (Figure 5b). This indicates the structure kept well, even after the calcination. TEM image inthat Figure 5cthat reveals that well, the MNPC had precursor withhigh-temperature a good dispersion (Figure 5b). This indicates the structure kept even after even after the high-temperature calcination. The TEM image in Figure 5c reveals that the MNPC had athe uniform morphology with an interconnected porous structure, and the Co nanopaticals (NPs) were high-temperature calcination. The TEM image in Figure 5c reveals that the MNPC had a uniform a uniform morphology with an interconnected porous structure, and the Co nanopaticals (NPs) were embedded the an porous carbon. The high-resolution electron(NPs) microscopy (HRTEM) morphologyinwith interconnected porous structure, andtransmission the Co nanopaticals were embedded in embedded in the porous carbon. The high-resolution transmission electron microscopy (HRTEM) image (Figure 5d) shows further that MNPC exhibitselectron an obvious core-shell structure. The(Figure graphitic the porous carbon. The high-resolution transmission microscopy (HRTEM) image 5d) image (Figure 5d) were showsthe further MNPC exhibits an obvious core-shell structure. TheÅgraphitic carbon structures shellsthat with an interplane spacing of (002) lattice (3.4 ), which shows further that MNPC exhibits an obvious core-shell structure. Thecrystal graphitic carbon structures carbon were thegraphitization shells with an interplane spacing of (002) crystal (3.4were Å ), tightly which resultedstructures from the catalytic behavior of Co NPs [31]. Moreover, thelattice Co NPs resulted from the catalytic graphitization behavior of Co NPs [31]. Moreover, the Co NPs were tightly wrapped by graphitic carbon shells due to the coordinating structures of bimetallic MOFs as a wrapped by graphitic carbon shells due to the coordinating structures of bimetallic MOFs as a


6 of 15

were the shells with an interplane spacing of (002) crystal lattice (3.4 Å), which resulted from the catalytic graphitization behavior of Co NPs [31]. Moreover, the Co NPs were tightly wrapped by Nanomaterials 2018, 8, x FOR PEER REVIEW 6 of 14 graphitic carbon shells due to the coordinating structures of bimetallic MOFs as a precursor. In addition, the HRTEM shows distinct lattice an interplanar spacing of an 0.2interplanar nm, which matched precursor. Inimage addition, theaHRTEM imagefringe showswith a distinct lattice fringe with spacing well with the spacing of (111) planes of the Co phase. Besides, the MC showed a dodecahedron-like of 0.2 nm, which matched well with the spacing of (111) planes of the Co phase. Besides, the MC structure with a particle sizestructure around with 250 nm, and the also embedded the structure showed a dodecahedron-like a particle sizeCo around 250 nm, andinthe Cocarbon also embedded (Figure S5). in the carbon structure (Figure S5).

Figure 5. SEM images of (a) Bimetallic metal-organic frameworks (MOFs) precursor; and (b) the Figure 5. SEM images of (a) Bimetallic metal-organic frameworks (MOFs) precursor; and (b) the MNPC; (c) TEM and (d) HRTEM images of the MNPC. MNPC; (c) TEM and (d) HRTEM images of the MNPC.

As seen from the high-angle annular dark field-scanning electron microscopy (HAADF-STEM) As6a), seen from annular dark field-scanning electron microscopy (HAADF-STEM) (Figure The Co the NPshigh-angle were embedded into graphitic carbon structure. The elemental mapping was (Figure 6a), The Co NPs were embedded into graphitic carbon structure. The elemental mapping was performed to illustrate the spatial distribution of C, N, O, and Co in the structure of MNPC in Figure performed to illustrate the spatial of C,mapping N, O, andresults Co in the structure of MNPCthe in Figure 6b. 6b. As revealed in Figure 6c–f, distribution the elemental further confirmed uniform As revealed in Figure 6c–f, the elemental mapping results further confirmed the uniform distribution distribution of Co and N species within the MNPC structure. Besides, the MC also showed a of Co and N species within the MNPC thethat MCthe also showedcan a homogeneous homogeneous distribution (Figure S5). It structure. is generallyBesides, accepted N species promote the distribution S5). It is and generally accepted that the Nperformance. species can promote the formation of formation of (Figure hydrogen bonds accelerate the adsorption Moreover, the Co species hydrogen bonds and accelerate the adsorption performance. Moreover, the Co species with a good with a good dispersion within the carbon structure is beneficial to the further separation of dispersion within the carbon structure is beneficial to the further separation of adsorbents. adsorbents.

Nanomaterials 2018, 8, 664 Nanomaterials 2018, 8, x FOR PEER REVIEW

7 of 15 7 of 14

Figure Figure6.6.The Thehigh-angle high-angleannular annulardark darkfield-scanning field-scanningelectron electronmicroscopy microscopy(HAADF-STEM): (HAADF-STEM): (a) (a) image, image, and (b–f) mapping images of MNPC. and (b–f) mapping images of MNPC.

3.2. Performance 3.2.Adsorption Adsorption Performance The Theadsorption adsorptionbehavior behaviorof ofthe theMNPC MNPCon onthe theNOR NORwas wasinvestigated investigatedby bybatch batchmode modeexperiments experiments −1 aqueous solution. As seen from the adsorption curves in Figure 7, the adsorption capacity in 10 mg L − 1 in 10 mg L aqueous solution. As seen from the adsorption curves in Figure 7, the adsorption capacity sharply sharplyincreased increasedwith withthe theadsorption adsorptiontime, time,suggesting suggestingthat thatthe theNOR NORin inthe theaqueous aqueoussolution solutioncould could be quickly and easily removed by the adsorbents [39]. As time goes on, the change of adsorption be quickly and easily removed by the adsorbents [39]. As time goes on, the change of adsorption capacity reaching an an adsorption adsorptionequilibrium equilibriumowing owingtotothe thefact factthat thatthe the number capacity became became slower slower until until reaching number of of adsorption sites decreased as the adsorption time increased. Obviously, the adsorption capacity of adsorption sites decreased as the adsorption time increased. Obviously, the adsorption capacity of the theMNPC MNPCadsorbents adsorbentswas wasmuch muchlarger largerthan thanthat thatof of MC, MC, indicating indicatingthat thatthe theMNPC MNPCexhibited exhibitedmuch much better adsorption performance. After 150 min, the final adsorption capacity of MNPC adsorbents was better adsorption performance. After 150 min, the final adsorption capacity of MNPC adsorbents −1 −1 8.84 , larger of that MC (7.98 mg g ).mg Asgseen the from Figure 7b,Figure the MNPC inMNPC the aqueous −1 ). from wasmg 8.84g mg g−1 , than largerthat than of MC (7.98 As seen the 7b, the in the solution could be easily separated under an external magnetic field. aqueous solution could be easily separated under an external magnetic field. The Thebetter betteradsorption adsorptionperformance performance of of MNPC MNPC adsorbents adsorbents can can be be ascribed ascribed to tothe thefollowing following aspects: aspects: First, the larger specific surface area of MNPC can provide more adsorption sites for the First, the larger specific surface area of MNPC can provide more adsorption sites for the NOR NOR adsorption. NOR molecules moleculescan canbe beeasily easilytransported transportedbetween between smooth channels in adsorption. Second, Second, the the NOR thethe smooth channels in the the MNPC due to the interconnected porous structure. Moreover, the MNPC has a good MNPC due to the interconnected porous structure. Moreover, the MNPC has a good graphitization graphitization which is beneficial of π–π interactions NORwhich and degree, which degree, is beneficial to the formationtoofthe π–πformation interactions between NOR andbetween absorbents, absorbents, which then further improves the adsorption capacity [8,40]. Furthermore, hydrogen then further improves the adsorption capacity [8,40]. Furthermore, hydrogen bonding is easy to form bonding is easy to form between MNPC adsorbents and NOR due to the effect of nitrogen doping, between MNPC adsorbents and NOR due to the effect of nitrogen doping, which further promotes which further promotes adsorption performance [38]. Inthe conclusion, combining the pore structure, adsorption performance [38]. In conclusion, combining pore structure, large specific surface area, large specific surface area, good graphitization degree, and the effective nitrogen doping, MNPC can good graphitization degree, and the effective nitrogen doping, MNPC can be considered as an excellent be considered as an for excellent candidate material for NOR adsorption application. candidate material NOR adsorption application. As is well known, the structure and surface properties of adsorbents have important influence As is well known, the structure and surface properties of adsorbents have important influence on the adsorption performance. Generally, the adsorbent’s structure has a great effect on the physical on the adsorption performance. Generally, the adsorbent’s structure has a great effect on the physical adsorption, and the chemical adsorption is usually related to the functional groups on the surface of adsorption, and the chemical adsorption is usually related to the functional groups on the surface of adsorbents [41]. The MNPC has large specific surface area, which can provide abundant adsorption adsorbents [41]. The MNPC has large specific surface area, which can provide abundant adsorption sites for NOR adsorption. The porous structure is beneficial to the NOR molecules’ penetration. In sites for NOR adsorption. The porous structure is beneficial to the NOR molecules’ penetration. addition, such oxygen-containing functional groups as –COOH and –OH and N-doping are on the In addition, such oxygen-containing functional groups as –COOH and –OH and N-doping are on surface of MNPC, so the hydrogen bonding can be easily formed between the NOR molecules the the surface of MNPC, so the hydrogen bonding can be easily formed between the NOR molecules MNPC, which then promotes the adsorption capacity [8]. Moreover, the aromatic structures and C=C the MNPC, which then promotes the adsorption capacity [8]. Moreover, the aromatic structures and double bonds in NOR can contribute to the affinity between MNPC and NOR through the π–π C=C double bonds in NOR can contribute to the affinity between MNPC and NOR through the π–π interactions and then increase the adsorption capacity [42]. interactions and then increase the adsorption capacity [42].


8 of 15

Nanomaterials 2018, 8, x FOR PEER REVIEW Nanomaterials 2018, 8, x FOR PEER REVIEW

8 of 14 8 of 14

Figure 7. 7. (a) (a) Plots Plots of of adsorption adsorption capacity capacity vs vs adsorption adsorption time time of of MNPC MNPC and and MC MC in in the the NOR NOR aqueous aqueous Figure −1; and（b）the Figure 7. (a) Plots of adsorption capacity vs adsorption time of MNPC and MC in the NOR aqueous solutions at concentrations of 10 mg L photo of MNPC separated under an external 1 ; and (b) the photo of MNPC separated under an external solutions at concentrations of 10 mg L− −1; and（b）the photo of MNPC separated under an external solutions at concentrations of 10 mg L magnetic field. field. magnetic magnetic field.

As is well known, the amount of adsorbents has a critical effect on the adsorption performance. As known, the of has critical effect effect on on the the adsorption adsorption performance. performance. As is is well wellof known, the amount amount of adsorbents adsorbents has aa by critical The influence the adsorbent’s dosage was explored adding various amounts of MNPC to 100 The influence of the adsorbent’s dosage was explored by adding various amounts of MNPC to 100 The influence of the adsorbent’s dosage was explored by adding various amounts of MNPC to mL 100 −1 mL of a 10 mg L NOR solution. As seen from Figure 8a, the adsorption capacity decreased with the − 1 of a of 10amg L LNOR solution. AsAs seen from Figure 8a, the adsorption capacity decreased with the −1 NOR mL 10 mg solution. seen from Figure 8a, the adsorption capacity decreased with the increase of the absorbent’s dosage due to the completely exposed adsorption sites at the low dosage. increase of the absorbent’s dosage due to the completely exposed adsorption sites at the low dosage. increaseatofhigher the absorbent’s dosage due to theadsorption completelysites exposed sites at the low While dosage, the unoccupied wereadsorption excess and resulted in adosage. lower While atathigher dosage, the the unoccupied adsorption sites were excess and resulted in a lower in adsorption While higher dosage, unoccupied adsorption sites were excess and resulted lower adsorption capacity [43]. Considering the adsorbent amounts and adsorption capacity, ana0.8 g/L capacity [43]. Considering the adsorbent amounts and adsorption capacity, an 0.8 g/L dosage of MNPC adsorption capacity Considering thestudies. adsorbent amounts and adsorption capacity, an 0.8 g/L dosage of MNPC was[43]. selected for further was selected for further studies.for further studies. dosage of MNPC was selected

Figure 8. (a) Plots of adsorption capacity vs dosage; and (b) plots of removal rates vs. pH value with −1. adsorption 8. of (a)0.8 Plots of capacity vsdosage; dosage; and (b) plots ofremoval removal rates pH value with aFigure dosage g Lof All the curves werevs obtained inand a 10 mg L −1 of NOR aqueous solution at 30 °C. Figure 8. (a) Plots adsorption capacity (b) plots rates vs.vs.pH value with a −1. All the curves were obtained in a 10 mg L −1 − 1 − 1 ◦ a dosage of 0.8 g L NOR aqueous solution at 30 °C. dosage of 0.8 g L . All the curves were obtained in a 10 mg L NOR aqueous solution at 30 C.

pH is another important factor affecting adsorption performance. As seen in Figure 8b, the NOR pH is another important factor affecting adsorption performance. As seen in Figure 8b, the NOR adsorption on MNPC initially increased the pHperformance. value ranging to 6.0, then pH is another important factor affectingwith adsorption As from seen in2.0 Figure 8b, and the NOR adsorptionwhen on MNPC initiallypHincreased with the pH value ranging from 2.0 toa 6.0, and then decreased the solution value was higher than 6.0. The NOR contained carboxyl and adsorption on MNPC initially increased with the pH value ranging from 2.0 to 6.0, and then decreased decreased when the solution pH value was higher than 6.0. The NOR contained a carboxyl and piperazinyl group,pH which shows two proton-binding two acida dissociation pKa when the solution value was higher than 6.0. The sites. NORIts contained carboxyl andconstant piperazinyl piperazinyl group, which shows two proton-binding sites. Its two acid dissociation constant pKa values which were 6.22 andtwo 8.51,proton-binding respectively. Insites. the solution, the dissociation protonation–deprotonation reaction of group, shows Its two acid constant pKa values were valueswould were occur. 6.22 and 8.51, respectively. In the solution, the protonation–deprotonation reaction of NOR The NOR can exist in cationic form (pH < 6.2), zwitterionic/neutral form (6.2 < pH 6.22 and 8.51, respectively. In the solution, the protonation–deprotonation reaction of NOR would NOR would occur. The NOR can exist in cationic form (pH < 6.2), zwitterionic/neutral form (6.2 < pH + < 8.5), or anionic (pH >in8.5) [8,12].form In acidic a large amount ofform H ions the occur. The NOR form can exist cationic (pH conditions, < 6.2), zwitterionic/neutral (6.2surrounds < pH < 8.5), + ions surrounds the < 8.5), orofanionic form (pH > 8.5)compete [8,12]. In acidic conditions, a large amount of Hcationic surface MNPC, which could with the NOR molecule existing in the form, and so + or anionic form (pH > 8.5) [8,12]. In acidic conditions, a large amount of H ions surrounds the surface surface of MNPC, which could compete with the NOR molecule existing in the cationic form, and so theMNPC, bindingwhich of NOR to adsorbent restricted. When the pHexisting value ranges 6.0 to 8.7, and the ratio of of could compete is with the NOR molecule in the from cationic form, so the the binding of NOR to adsorbent is restricted. When the pH value ranges from 6.0 to 8.7, the ratio of + the zwitterion form is increased, so the competition between H and NOR ions for surface adsorption binding of NOR to adsorbent is restricted. When the pH value +ranges from 6.0 to 8.7, the ratio of the the zwitterion form is increased, so the competition between H adsorption and NOR ions for surface adsorption sites is decreased in an improved capacity. However, when zwitterion form is correspondingly, increased, so the resulting competition between H+ and NOR ions for surface adsorption sites is decreased correspondingly, resulting in an improved adsorption capacity. However, when the pH is higher than the pKa2 of NOR, the anionic form dominates and the repulsion between the sites is decreased correspondingly, resulting in an improved adsorption capacity. However, when the pH is higher than the pKa 2 of NOR, the anionic form dominates and the repulsion between the NOR molecule and the negatively charged MNPC is increased, and so the adsorption capacity is the pH is higher than the pKa2 of NOR, the anionic form dominates and the repulsion between the NOR molecule and the The negatively MNPC is increased, so thepH adsorption capacity is significantly decreased. further charged studies were conducted at the and optimum value of 6.0. significantly decreased. The further is studies were at the optimum pH value of 6.0. The initial NOR concentration another keyconducted factor controlling the adsorption performance of The initial NOR concentration is another key factor controlling the adsorption performance of MNPC, as shown in Figure 9a. The initial NOR concentration ranged from 1.0 to 100 mg L−1 at a pH −1 MNPC, as shown in 9a. Thecapacity initial NOR concentration ranged fromconcentration, 1.0 to 100 mg which L at awas pH of 6.0. Obviously, theFigure adsorption is increased with the solution of 6.0. Obviously, the adsorption capacity is increased with the solution concentration, which was


9 of 15

NOR molecule and the negatively charged MNPC is increased, and so the adsorption capacity is significantly decreased. The further studies were conducted at the optimum pH value of 6.0. The initial NOR concentration is another key factor controlling the adsorption performance of MNPC, as shown in Figure 9a. The initial NOR concentration ranged from 1.0 to 100 mg L−1 at a pH Nanomaterials 2018, 8, x FOR PEER REVIEW of 14 Nanomaterials 2018, 8, xthe FORadsorption PEER REVIEW of 14 of 6.0. Obviously, capacity is increased with the solution concentration, which99was − 1 increased from 3.04 to 55.12 mg g −1 . A higher initial NOR concentration meant a higher concentration increased from toto55.12 mg gg−1. .AAhigher initial NOR concentration meant concentration increasedwhich from3.04 3.04 higherand initial NOR concentration meantaquickly ahigher highertransfer concentration gradient, led to a55.12 highmg driving force, so the NOR molecules could to the gradient, which led to a high driving force, and so the NOR molecules could quickly transfer totothe gradient, which led to a high driving force, and so the NOR molecules could quickly transfer the pores of the MNPC. As shown in Figure 9b, the adsorption capacity of MNPC for NOR increased pores ofofthe MNPC. As shown ininFigure 9b, the adsorption capacity ofofMNPC for NOR increased ◦ pores the MNPC. As shown Figure 9b, the adsorption capacity MNPC for NOR increased with increased temperature (20–40 C), suggesting that the higher temperature was beneficial to the with temperature withincreased increased temperature(20–40 (20–40°C), °C),suggesting suggestingthat thatthe thehigher highertemperature temperaturewas wasbeneficial beneficialtotothe the adsorption process. adsorption process. adsorption process.

Figure (a)change The change of adsorption capacity vs initial concentration; and(b) (b)the thechange changeof ofadsorption adsorption Figure 9. (a) 9. The of adsorption capacity vs initial concentration; and Figure 9. (a) The change of adsorption capacity vs initial concentration; and (b) the change of − adsorption −1 rates vs. temperature. Allcurves the curves obtained in NOR solution with a dosage a pH rates vs. temperature. All the werewere obtained in NOR solution with a dosage of of 0.80.8 g Lg Lat1 aatpH of rates vs. temperature. All the curves were obtained in NOR solution with a dosage of 0.8 g L −1 at a pH of 7.0. of 7.0. 7.0.

The the adsorption The influence of ionic strength on performance was also investigated and the Theinfluence influenceof ofionic ionicstrength strengthon onthe theadsorption adsorptionperformance performancewas wasalso alsoinvestigated investigatedand andthe the results are shown in Figure 10. When the salt concentration increased from 0.0 to 0.1 M, the results in Figure 10. When the salt increased from 0.0 from to 0.1 M, resultsare areshown shown in Figure 10. When theconcentration salt concentration increased 0.0 the to adsorption 0.1 M, the adsorption capacity decreased slightly. the had no effect capacity decreased slightly. Generally, theGenerally, salt concentration had no significant on the adsorption adsorption capacity decreased slightly. Generally, thesalt saltconcentration concentration hadeffect nosignificant significant effecton on the adsorption capacity of NOR on the MNPC, which indicates that the interaction between NOR capacity of NOR on the MNPC, which indicates that the interaction between NOR and MNPC was the adsorption capacity of NOR on the MNPC, which indicates that the interaction between NOR and quite certain range quite stable was in a certain rangeinin ofaasalt concentration. andMNPC MNPC was quitestable stable certain rangeofofsalt saltconcentration. concentration.

Figure 10. Plots of adsorption capacity vs ionic strength. All the curves were obtained in 10 mg L−1−1−1 Figure10. 10.Plots Plotsof ofadsorption adsorption capacity capacity vs vs ionic ionic strength. strength. All Figure All the the curves curves were were obtained obtained in in10 10mg mgLL −1 at a pH of 7.0. NOR aqueous solution with a dosage of 0.8 g L− −1 1 NOR aqueous solution with a dosage of 0.8 g L at a pH of 7.0. NOR aqueous solution with a dosage of 0.8 g L at a pH of 7.0.

3.3. 3.3.Recyclability Recyclability In In practical practical application, application, the the recyclability recyclability isis another another critical critical factor factor for for the the adsorbents. adsorbents. The The recyclability of the MNPC was investigated using a methanol solution (containing 10% ammonia) recyclability of the MNPC was investigated using a methanol solution (containing 10% ammonia)as as the theeffluents effluentsand andthe theresults resultsare areshown shownininFigure Figure11. 11.The Theadsorption adsorptioncapacity capacityofofthe theMNPC MNPCremained remained −1 atat 12.0 12.0 mg mg gg−1 after after five five cycles, cycles, and and was was slightly slightly decreased, decreased, indicating indicating the the good good regeneration regeneration performance of the absorbents in the NOR solution. Hence, the MNPC adsorbent could performance of the absorbents in the NOR solution. Hence, the MNPC adsorbent couldbe bereused reused


10 of 15

3.3. Recyclability In practical application, the recyclability is another critical factor for the adsorbents. The recyclability of the MNPC was investigated using a methanol solution (containing 10% ammonia) as the effluents and the results are shown in Figure 11. The adsorption capacity of the MNPC remained at 12.0 mg g−1 after five cycles, and was slightly decreased, indicating the good regeneration performance of the absorbents in the NOR solution. Hence, the MNPC adsorbent could be reused Nanomaterials which 2018, 8, xisFOR PEER for REVIEW 10 of 14 effectively, helpful reducing the cost of adsorption.

11. Regeneration adsorbents in in the the 10 10 mg mg LL−−11 norfloxacin (NOR) Figure 11. Regeneration property property of the MNPC adsorbents − 1 −1 dosage of of 0.8 0.8 gg LL atataapH aqueous solution with aa dosage pHofof7.7.

3.4. Adsorption AdsorptionKinetics Kinetics 3.4. Adsorption kinetic kinetic models models are are usually usually used used to to evaluate evaluate the the variation variation of of adsorption capacity with with Adsorption adsorption capacity adsorption time, which can further reflect the relationship between adsorption performance and the adsorption time, which can further reflect the relationship between adsorption performance and the structure of of adsorbent. adsorbent. In In this this work, work, the the pseudo-first-order pseudo-first-order and and pseudo-second-order pseudo-second-order models models were were structure employed to analyze the experimental data. These two models are shown as follows: employed to analyze the experimental data. These two models are shown as follows: ln( qe -qt ) = ln qe − k1 t ln(q e −qt ) = ln qe − k1 t

(3) (3)

t 1 t t= t +1 2 = qt k2 qe 2qe+ qt qe k2 qe

(4) (4)

−1 are NOR uptakes −1 where equilibrium, tt is 1 1(min where qqee and and qqtt (mg (mg gg−1))are NOR uptakes at equilibrium, is the theadsorption adsorptiontime, time,and andkk (min−1) and − 1 − 1 −1) are k2 (g mg−1min min ) are rate constants two modes [44,45]. thethe rate constants of of two modes [44,45]. Figure 12 shows shows the the results resultsofoffitting fittingthe thetwo twokinetic kineticmodels. models. seen from Figure As As seen from the the Figure 11a,11a, the 2 value of the pseudo-first order 2 the experimental data severely deviates from the fitted data, and the R experimental data severely deviates from the fitted data, and the R value of the pseudo-first model was relatively low, low, indicating indicating a low low correlation correlation of NOR NOR adsorption kinetics data, and so the pseudo-first-order model was inconsistent with the the experimental experimental data. data. However, the the experimental experimental adsorption capacity values were in agreement with the theoretical adsorption capacity values according adsorption capacity values were in agreement with the theoretical adsorption capacity values 2 2 to the pseudo-second-order model (Figuremodel 11b) with a corresponding of 0.9996, illustrating the according to the pseudo-second-order (Figure 11b) with R a corresponding R of that 0.9996, adsorption fit well to the pseudo-second-order kinetic model. illustrating data that the adsorption data fit well to the pseudo-second-order kinetic model.

Figure 12. Adsorption kinetics of NOR on MNPC (a) pseudo-first-order and (b) pseudo-second-order models.

experimental data severely deviates from the fitted data, and the R2 value of the pseudo-first order model was relatively low, indicating a low correlation of NOR adsorption kinetics data, and so the pseudo-first-order model was inconsistent with the experimental data. However, the experimental adsorption capacity values were in agreement with the theoretical adsorption capacity values according the pseudo-second-order model (Figure 11b) with a corresponding R2 of 0.9996, Nanomaterials to 2018, 8, 664 11 of 15 illustrating that the adsorption data fit well to the pseudo-second-order kinetic model.

Figure Adsorption kinetics of NOR on MNPC (a) pseudo-first-order andpseudo-second-order (b) pseudo-second-order Figure 12.12. Adsorption kinetics of NOR on MNPC (a) pseudo-first-order and (b) models. models.

3.5. Adsorption Isotherm 3.5. Adsorption Isotherm To further study how the adsorbate interacts with the adsorbent, the adsorption models have To further study how the adsorbate interacts with the adsorbent, the adsorption models have been applied to8,understand the adsorption mechanism. Thus the Freundlich and Langmuir adsorption Nanomaterials 2018, x FOR PEER REVIEW 11 of 14 been applied to understand the adsorption mechanism. Thus the Freundlich and Langmuir isotherm models were used according to: adsorption isotherm models were used according to: qm KL Ce qm KL Ce (5) (5) qeq= = e 1+KL1C+ e K C L e 1/n 1/n qe =qeK= F CeK F Ce

(6) (6)

where where qqee, , qqmm, ,and and CCe e (mg (mg gg−1−)1 )are arethe theadsorption adsorptioncapacity, capacity, equilibrium equilibrium concentration, concentration, and and the the maximum maximumadsorption adsorptioncapacity, capacity, respectively, respectively, and and KKLL, ,KKF,F ,and andnnare arethe theLangmuir Langmuirand and Freundlich Freundlich parameters parameters[13]. [13].Figure Figure13 13displays displaysboth boththe theexperimental experimentaldata dataand andthe thefitting fittingisotherms isothermsof ofthe theabove above two twoisotherm isothermmodels. models.According Accordingtotothe theresults, results,the theadsorption adsorptionisotherm isothermofofNOR NORonto ontothe theMNPC MNPCfits fits Freundlich Freundlichisotherm isothermmodel modelwith withhigher highercorrelation correlationcoefficients coefficientsRR22values values(0.9988) (0.9988)compared comparedwith withthe the Langmuir Langmuirisotherm isothermmodel model(0.9841, (0.9841,Figure FigureS6), S6),indicating indicatingthat thatthe theadsorption adsorptionprocess processpredominantly predominantly features featuresmultilayer multilayeradsorption. adsorption.

◦ C. Figure13. 13.Freundlich Freundlichisotherms isothermsfor forthe theadsorption adsorptionofofNOR NORby byMNPC MNPCatat30 30°C. Figure

3.6.Adsorption Adsorption Thermodynamics 3.6. Thermodynamics Theadsorption adsorptionof ofNOR NORon onthe theMNPC MNPCwas wasfurther furtherdemonstrated demonstratedby byevaluation evaluationof ofchanges changesin inthe the The θ ), enthalpy (4Hθθ ), and entropy (4S θ ) as follows: θ θ Gibbs free energy ( 4 G Gibbs free energy (△G ), enthalpy (△H ), and entropy (△S ) as follows: θ θ ∆G∆G = -RTlnK c = −RTlnK c

CA C Kc K=c = A CS CS

(7)(7) (8)(8)

where R is the ideal gas constant, T represents absolute temperature (K), CA and CS (mg L−1) are the amount of NOR adsorbed and remained in the solution at equilibrium, respectively. After making the substituting of △Gθ = △Hθ–T△Sθ into Equation (9): ln(Kc ) = θ

θ

∆Gθ ∆Hθ ∆Sθ =+ RT RT R

(9)


12 of 15

where R is the ideal gas constant, T represents absolute temperature (K), CA and CS (mg L−1 ) are the amount of NOR adsorbed and remained in the solution at equilibrium, respectively. After making the substituting of 4Gθ = 4Hθ –T 4Sθ into Equation (9): ln(Kc ) = −

∆Gθ ∆H θ ∆Sθ = − + RT RT R

(9)

The values of 4Hθ and 4Sθ were then calculated from the slope and intercept of the linear regression of ln(Kc ) versus 1/T [41]. As calculated, the value of 4Hθ was 70.08 kJ·mol−1 , indicating that sorption of NOR on the MNPC was an exothermic process. Moreover, the value of 4Hθ was higher than 20 kJ·mol−1 , indicating the NOR sorption onto MNPC could be mainly attributed to chemisorption. Another important thermodynamic parameter is entropy 4Gθ . As calculated, the 4Gθ value was negative, indicating that the adsorption could occur spontaneously. 4. Conclusions The MNPC was successfully prepared by self-catalytic pyrolysis of bimetallic MOF with Zn and Co as metal ions and 2-methylimidazole as a ligand. The resultant MNPC possessed a large surface area, porous structure, good graphitization, and highly dispersed N species, simultaneously. The synergistic effect of the above characteristics offered MNPC excellent adsorption performances. The MNPC exhibited a dramatic enhancement in the adsorption to NOR compared with the MC derived from the MOF with only Co as the metal ion. The adsorption capacity was 55.12 mg g−1 with an initial concentration of 50 mg L−1 at 30 ◦ C. The pseudo-second-order and Freundlich models were a good fit for adsorption kinetics and isotherm for NOR adsorption. In the process of NOR adsorption onto the MNPC, the π–π interaction, hydrogen bonding, and pore-filling significantly improved the adsorption capability. Overall, this material is a potential adsorbent for the NOR and is expected to be used for removal of other pollutants in waste water. Supplementary Materials: The following are available online at http://www.mdpi.com/2079-4991/8/9/664/s1, Figure S1: XRD pattern of MC, Figure S2: Raman spectra of (a) MC and (b) PC, Figure S3: The XPS spectra of MC and MNPC, C1s and N1s of MC, Figure S4: N2 sorption isotherm of MC, Figure S5: (a) SEM, (b) TEM, (c) HRTEM and mapping images of MC, Figure S6: Langmuir isotherm of NOR adsorption on the MNPC at 30 ◦ C. Author Contributions: Conceptualization, Hui, Wang and Jinlong, Yan; Methodology, Xi, Zhang and Yan, Wang; Validation, Xiangyun Han; Formal Analysis, Hui, Wang; Investigation, Xiangyun Han; Resources, Hui, Wang; Data Curation, Hui, Wang and Xiangyun Han; Writing-Original Draft Preparation, Hui, Wang.; Writing-Review & Editing, Jinlong, Yan; Visualization, Guixiang Quan; Supervision, Jinlong Yan; Project Administration, Hui, Wang; Funding Acquisition, Hui Wang. Funding: This research was funded by the Natural Science Foundation of Jiangsu Province—Youth Fund Project (BK20170475) and National Natural Science Foundation of China (21677119). Acknowledgments: The authors would like to thank C. J. Ma from Analysis and Test Center of Yancheng Institute of Technology, for help with the TEM measurements. Conflicts of Interest: The authors declare no conflict of interest

References 1.

2.

3.

Zhang, H.; Jia, Y.; Khanal, S.K.; Lu, H.; Fang, H.; Zhao, Q. Understanding the role of extracellular polymeric substances (EPS) on ciprofloxacin (CIP) adsorption in aerobic sludge, anaerobic sludge and sulfate-reducing bacteria (SRB) sludge systems. Environ. Sci. Technol. 2018, 52, 6476–6486. [CrossRef] [PubMed] Xiao, T.; Tang, Z.; Yang, Y.; Tang, L.; Zhou, Y.; Zou, Z. In situ construction of hierarchical WO3 /g-C3 N4 composite hollow microspheres as a Z-scheme photocatalyst for the degradation of antibiotics. Appl. Catal. B 2018, 220, 417–428. [CrossRef] Wang, S.; Li, X.; Zhao, N.; Quan, X.; Chen, S.; Yu, H. Enhanced adsorption of ionizable antibiotics on activated carbon fiber under electrochemical assistance in continuous-flow modes. Water Res. 2018, 134, 162–169. [CrossRef] [PubMed]


4.

5.

6.

7. 8.

9.

10. 11.

12. 13.

14.

15.

16. 17. 18. 19. 20.

21.

22. 23.

13 of 15

Wu, H.; Li, C.; Che, H.; Hu, H.; Hu, W.; Liu, C.; Ai, J.; Dong, H. Decoration of mesoporous Co3 O4 nanospheres assembled by monocrystal nanodots on g-C3 N4 to construct Z-scheme system for improving photocatalytic performance. Appl. Surf. Sci. 2018, 440, 308–319. [CrossRef] Wang, B.; Lv, X.L.; Feng, D.; Xie, L.H.; Zhang, J.; Li, M.; Xie, Y.; Li, J.R.; Zhou, H.C. Highly stable Zr(IV)-based metal-organic frameworks for the detection and removal of antibiotics and organic explosives in water. J. Am. Chem. Soc. 2016, 138, 6204–6216. [CrossRef] [PubMed] Gao, J.; Lu, Y.; Zhang, X.; Chen, J.; Xu, S.; Li, X.; Li, X.; Tan, F. Elucidating the electrostatic interaction of sulfonic acid functionalized SBA-15 for ciprofloxain adsorption. Appl. Surf. Sci. 2015, 349, 224–229. [CrossRef] Yan, B.; Niu, C.H.; Wang, J. Kinetics, electron-donor-acceptor interactions, and site energy distribution analyses of norfloxacin adsorption on pretreated barley straw. Chem. Eng. J. 2017, 330, 1211–1221. [CrossRef] Peng, X.; Hu, F.; Zhang, T.; Qiu, F.; Dai, H. Amine-functionalized magnetic bamboo-based activated carbon adsorptive removal of ciprofloxacin and norfloxacin: A batch and fixed-bed column study. Bioresour. Technol. 2018, 249, 924–934. [CrossRef] [PubMed] Jabbari, V.; Veleta, J.M.; Zarei-Chaleshtori, M.; Gardea-Torresdey, J.; Villagrán, D. Green synthesis of magnetic MOF@GO and MOF@CNT hybrid nanocomposites with high adsorption capacity towards organic pollutants. Chem. Eng. J. 2016, 304, 774–783. [CrossRef] Seo, P.W.; Khan, N.A.; Jhung, S.H. Removal of nitroimidazole antibiotics from water by adsorption over metal–organic frameworks modified with urea or melamine. Chem. Eng. J. 2017, 315, 92–100. [CrossRef] Lu, H.; Wang, J.; Li, F.; Huang, X.; Tian, B.; Hao, H. Highly efficient and reusable montmorillonite/Fe3 O4 /humic acid nanocomposites for simultaneous removal of Cr(VI) and aniline. Nanomaterials 2018, 8, 537. [CrossRef] [PubMed] Wang, Z.Y.; Yu, X.D.; Pan, B.; Xing, B.S. Norfloxacin sorption and its thermodynamics on surface-modified carbon nanotubes. Environ. Sci. Technol. 2010, 44, 978–984. [CrossRef] [PubMed] Turco, A.; Monteduro, A.G.; Mazzotta, E.; Maruccio, G.; Malitesta, C. An innovative porous nanocomposite material for the removal of phenolic compounds from aqueous solutions. Nanomaterials 2018, 8, 350. [CrossRef] [PubMed] Mo, Z.W.; Zhou, H.L.; Zhou, D.D.; Lin, R.B.; Liao, P.Q.; He, C.T.; Zhang, W.X.; Chen, X.M.; Zhang, J.P. Mesoporous metal-organic frameworks with exceptionally high working capacities for adsorption heat transformation. Adv. Mater. 2018, 30, 1704350. [CrossRef] [PubMed] Oveisi, M.; Asli, M.A.; Mahmoodi, N.M. MIL-Ti metal-organic frameworks (MOFs) nanomaterials as superior adsorbents: Synthesis and ultrasound-aided dye adsorption from multicomponent wastewater systems. J. Hazard. Mater. 2018, 347, 123–140. [CrossRef] [PubMed] Zheng, F.; Yang, Y.; Chen, Q. High lithium anodic performance of highly nitrogen-doped porous carbon prepared from a metal-organic framework. Nat. Comm. 2014, 5, 5261. [CrossRef] [PubMed] Guan, B.Y.; Yu, L.; Lou, X.W. A dual-metal-organic-frameworks derived electrocatalyst for oxygen reduction. Energy Environ. Sci. 2016, 9, 3092–3096. [CrossRef] Liu, B.; Shioyama, H.; Akita, T.; Xu, Q. Metal-organic framework as a template for porous carbon synthesis. J. Am. Chem. Soc. 2008, 130, 5390–5391. [CrossRef] [PubMed] Yang, S.J.; Kim, T.; Im, J.H.; Kim, Y.S.; Lee, K.; Jung, H.; Park, C.R. MOF-derived hierarchically porous carbon with exceptional porosity and hydrogen storage capacity. Chem. Mater. 2012, 24, 464–470. [CrossRef] Li, S.; Zhang, X.; Huang, Y. Zeolitic imidazolate framework-8 derived nanoporous carbon as an effective and recyclable adsorbent for removal of ciprofloxacin antibiotics from water. J. Hazard. Mater. 2017, 321, 711–719. [CrossRef] [PubMed] Didaskalou, C.; Buyuktiryaki, S.; Kecili, R.; Fonte, C.P.; Szekely, G. Valorisation of agricultural waste with an adsorption/nanofiltration hybrid process: from materials to sustainable process design. Green Chem. 2017, 19, 3116–3125. [CrossRef] Razali, M.; Kim, J.F.; Attfield, M.; Budd, P.M.; Drioli, E.; Lee, Y.M.; Szekely, G. Sustainable wastewater treatment and recycling in membrane manufacturing. Green Chem. 2015, 17, 5196–5205. [CrossRef] Likon, M.; Cernec, F.; Svegl, F.; Saarela, J.; Zimmie, T.F. Papermill industrial waste as a sustainable source for high efficiency absorbent production. Waste Manage. 2011, 31, 1350–1356. [CrossRef] [PubMed]


24.

25.

26.

27.

28.

29. 30. 31.

32.

33.

34.

35. 36.

37. 38.

39.

40.

41.

42.

14 of 15

Liu, Y.; Gao, Z.; Wu, R.; Wang, Z.; Chen, X.; Chan, T.D. Magnetic porous carbon derived from a bimetallic metal-organic framework for magnetic solid-phase extraction of organochlorine pesticides from drinking and environmental water samples. J. Chromatogr. A 2017, 1479, 55–61. [CrossRef] [PubMed] Jin, L.; Zhao, X.; Qian, X.; Dong, M. Nickel nanoparticles encapsulated in porous carbon and carbon nanotube hybrids from bimetallic metal-organic-frameworks for highly efficient adsorption of dyes. J. Colloid Interface Sci. 2018, 509, 245–253. [CrossRef] [PubMed] Yuan, Y.; Yang, D.; Mei, G.; Hong, X.; Wu, J.; Zheng, J.; Pang, J.; Yan, Z. Preparation of konjac glucomannan-based zeolitic imidazolate framework-8 composite aerogels with high adsorptive capacity of ciprofloxacin from water. Colloid. Surface. A 2018, 544, 187–195. [CrossRef] Niu, H.; Wang, Y.; Zhang, X.; Meng, Z.; Cai, Y. Easy synthesis of surface-tunable carbon-encapsulated magnetic nanoparticles: adsorbents for selective isolation and preconcentration of organic pollutants. ACS Appl. Mater. Inter. 2012, 4, 286–295. [CrossRef] [PubMed] Fodi, T.; Didaskalou, C.; Kupai, J.; Balogh, G.T.; Huszthy, P.; Szekely, G. Nanofiltration-enabled in Situ solvent and reagent recycle for sustainable continuous-flow synthesis. ChemSusChem 2017, 10, 3435–3444. [CrossRef] [PubMed] Schaepertoens, M.; Didaskalou, C.; Kim, J.F.; Livingston, A.G.; Szekely, G. Solvent recycle with imperfect membranes: A semi-continuous workaround for diafiltration. J. Memb. Sci. 2016, 514, 646–658. [CrossRef] Li, N.; Yang, S.; Chen, J.; Gao, J.; He, H.; Sun, C. Electro-adsorption of tetracycline from aqueous solution by carbonized pomelo peel and composite with aniline. Appl. Surf. Sci. 2016, 386, 460–466. [CrossRef] Chen, Y.Z.; Wang, C.; Wu, Z.Y.; Xiong, Y.; Xu, Q.; Yu, S.H.; Jiang, H.L. From bimetallic metal-organic framework to porous carbon: high surface area and multicomponent active dopants for excellent electrocatalysis. Adv. Mater. 2015, 27, 5010–5016. [CrossRef] [PubMed] Torad, N.L.; Hu, M.; Ishihara, S.; Sukegawa, H.; Belik, A.A.; Imura, M.; Ariga, K.; Sakka, Y.; Yamauchi, Y. Direct synthesis of MOF-derived nanoporous carbon with magnetic Co nanoparticles toward efficient water treatment. Small 2014, 10, 2096–2107. Ling, L.L.; Liu, W.J.; Zhang, S.; Jiang, H. Magnesium Oxide Embedded nitrogen self-doped biochar composites: fast and high-efficiency adsorption of heavy metals in an aqueous solution. Environ. Sci. Technol. 2017, 51, 10081–10089. [CrossRef] [PubMed] Zhao, S.S.; Yan, T.T.; Wang, H.; Chen, G.R.; Huang, L.; Zhang, J.P.; Shi, L.Y.; Zhang, D.S. High capacity and high rate capability of nitrogen-doped porous hollow carbon spheres for capacitive deionization. Appl. Surf. Sci. 2016, 369, 460–469. [CrossRef] Xie, Y.; Yang, W.; Wang, M.; Ge, X. Fibrous N-doped hierarchical porous carbon microspheres: Synthesis and adsorption performance. Chem. Eng. J. 2017, 323, 224–232. [CrossRef] Wang, H.; Shi, L.; Yan, T.; Zhang, J.; Zhong, Q.; Zhang, D. Design of graphene-coated hollow mesoporous carbon spheres as high performance electrodes for capacitive deionization. J. Mater. Chem. A 2014, 2, 4739–4750. [CrossRef] Sun, L.; Tian, C.; Li, M.; Meng, X.; Wang, L.; Wang, R.; Yin, J.; Fu, H. From coconut shell to porous graphene-like nanosheets for high-power supercapacitors. J. Mater. Chem. A 2013, 1, 6462–6470. [CrossRef] Peng, H.; Ma, G.; Sun, K.; Mu, J.; Zhang, Z.; Lei, Z. Formation of carbon nanosheets via simultaneous activation and catalytic carbonization of macroporous anion-exchange resin for supercapacitors application. ACS Appl. Mater. Interfaces 2014, 6, 20795–20803. [CrossRef] [PubMed] Wang, H.; Zhang, D.S.; Yan, T.T.; Wen, X.R.; Zhang, J.P.; Shi, L.Y.; Zhong, Q.D. Three-dimensional macroporous graphene architectures as high performance electrodes for capacitive deionization. J. Mater. Chem. A 2013, 1, 11778–11789. [CrossRef] Peng, B.; Chen, L.; Que, C.; Yang, K.; Deng, F.; Deng, X.; Shi, G.; Xu, G.; Wu, M. Adsorption of antibiotics on graphene and biochar in aqueous solutions induced by π-π interactions. Sci. Rep. 2016, 6, 31920. [CrossRef] [PubMed] Peiris, C.; Gunatilake, S.R.; Mlsna, T.E.; Mohan, D.; Vithanage, M. Biochar based removal of antibiotic sulfonamides and tetracyclines in aquatic environments: A critical review. Bioresour. Technol. 2017, 246, 150–159. [CrossRef] [PubMed] Ji, L.; Chen, W.; Duan, L.; Zhu, D. Mechanisms for strong adsorption of tetracycline to carbon nanotubes: A comparative study using activated carbon and graphite as adsorbents. Environ. Sci. Technol. 2009, 43, 2322–2327. [CrossRef] [PubMed]


43. 44. 45.

15 of 15

Zhu, H.; Chen, T.; Liu, J.; Li, D. Adsorption of tetracycline antibiotics from an aqueous solution onto graphene oxide/calcium alginate composite fibers. RSC Adv. 2018, 8, 2616–2621. [CrossRef] Liang, C.; Zhang, X.; Feng, P.; Chai, H.; Huang, Y. ZIF-67 derived hollow cobalt sulfide as superior adsorbent for effective adsorption removal of ciprofloxacin antibiotics. Chem. Eng. J. 2018, 344, 95–104. [CrossRef] Han, X.; Wang, H.; Zhang, L. Efficient removal of methyl blue using nanoporous carbon from the waste biomass. Water Air Soil Poll. 2018, 229, 26. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).