Adsorption performance and mechanism of antibiotics from aqueous solutions on porous boron nitride–carbon nanosheets

Gang Wang *ab, Yunqi Zhang b, Shiyong Wang ab, Yuwei Wang a, Haoran Song a, Sihao Lv a and Changping Li *a
aSchool of Environment and Civil Engineering, Research Center for Eco-Environmental Engineering, Dongguan University of Technology, Dongguan 523106, Guangdong, China. E-mail: wghy1979@163.com; licpbit@hotmail.com
bSchool of Chemical Engineering, Dalian University of Technology, Dalian 116024, Liaoning, China

Received 12th February 2020 , Accepted 3rd May 2020

First published on 4th May 2020


Antibiotics are a class of emerging contaminants with a potential threat to human and animal health. Chloramphenicol (CAP) and roxithromycin (ROX), which are extensively used, have attracted attention recently. In this work, two kinds of porous boron nitride–carbon nanosheets (BCNs and MBCNs) with different pore distributions were synthesized for the effective adsorption of antibiotics. The maximum adsorption capacity for CAP was 546.15 mg g−1, and the adsorption capacity for ROX was 575.68 mg g−1. It is worth noting that with the increase of molecular weight of antibiotics, huge molecules are blocked from entering micropores, and the adsorption capacity of BCNs decreases, indicating that a micropore-filling effect is the main mechanism of antibiotic adsorption on BCNs. In addition, hydrophobic interaction in chemical adsorption and electrostatic interaction were also important factors affecting the adsorption capacity of MBCN and BCN. The adsorbents were regenerated by simple high temperature calcination and excellent reusability was achieved after 6 cycles. Overall, the boron nitride–carbon nanosheets should be a good choice for the next generation of high performance adsorption materials.



Water impact

The abuse and uninterrupted emission of antibiotics have become a global problem. Benefiting from a high specific surface area and thermal stability, BCN adsorbents exhibit excellent adsorption performance. The maximum adsorption capacity for chloramphenicol was 546.15 mg g−1, and the adsorption capacity for roxithromycin was 575.68 mg g−1. More importantly, the adsorbents were regenerated by simple high temperature calcination and excellent reusability was achieved.

1. Introduction

The abuse and uninterrupted emission of antibiotics have become a global problem. About 46% of antibiotics are eventually discharged into rivers through sewage, and the rest are spread through feces and soil sludge.1 Chloramphenicol (CAP) is a broad-spectrum antibiotic, which has been widely used to treat bacterial infections in animal productions. CAP has been frequently detected in surface water, ground water and even drinking water of China. CAP can cause multiple diseases including bone marrow depression, aplastic anemia and so on.2 Roxithromycin (ROX) is a semi-synthetic macrolide widely used to treat respiratory tract, urinary and soft tissue infections. ROX poses adverse effects on organisms and human health by biomagnification through food chains due to its widespread application and ineffective removal of wastewater treatment plants.3 Therefore, more attention should be paid to the treatment of antibiotic pollutants like CAP and ROX.

The common methods to treat antibiotics in water include chemical methods,4 physical methods5–7 and biological methods.8,9 Physical techniques are still the best choice for the treatment of antibiotic wastewater. Among which, adsorption is one of the most promising methods with the advantages of high efficiency, simple to design, inexpensive and no highly toxic by-products. Carbon-based adsorbents have been widely used in various adsorption fields due to their high surface area, extensive source and low cost. The commonly used carbon-based adsorbents are mainly activated carbon, graphene, carbon nanotubes and biological carbon materials.10–14 However, these adsorbents are usually difficult to regenerate and are easily lost with wastewater, which increase the process steps and cost of adsorption operation. To effectively recycle and separate adsorbents, Bach15 reported the synthesis of mesoporous carbon from Cu3(BTC)2 for the adsorption of CAP. A high maximum adsorption capacity of 37.2 mg g−1 was obtained. Li16 prepared biochar from peanut shells using ammonium polyphosphate via pyrolysis, which exhibited excellent adsorption performance with a monolayer chloramphenicol adsorption capacity of 423.7 mg g−1. This was due to the richness of N- and P-containing functional groups and its high surface area. Liao17 demonstrated that sediments played significant roles in ROX adsorption and its capacity was affected by the physicochemical properties of sediments. Recently, Song18 synthesized novel boron nitride bundles and achieved a high adsorption capacity of 391.59[thin space (1/6-em)]mg g−1 for oxytetracycline (OTC). The BN bundles were regenerated by a simple high-temperature calcination method and showed a stable removal efficiency of 94.6% after 5 cycles.

Boron nitride–carbon nanosheets, as graphite analogues with a few layers, have attracted more and more attention due to their unique structure and properties, such as high specific surface areas, structural defects, and excellent thermal stability. As ideal adsorbents for wastewater treatment systems, BCNs have exhibited excellent adsorption performance on various pollutants, such as organic dyes (rhodamine)19 and gases (NOx, CO, HCN, CO2).20–22 In the application for water purification, compared with commercial carbon-containing adsorbents, BCNs have many significant advantages, such as high saturated adsorption capacity, fast adsorption rates and excellent reusability. However, to the best of our knowledge, there have been no reports on the antibiotic adsorption performance of BCNs to date.

In this work, boron nitride–carbon nanosheets (BCNs) and mesoporous boron nitride–carbon nanosheets (MBCNs) were synthesized by a high-temperature solid-state method20,23 and used as adsorbents to adsorb CAP and ROX in water, respectively. The equilibrium adsorption isotherm and kinetics of different adsorbents and antibiotics were investigated. The effects of pH values and salt ions on antibiotic adsorption behavior were considered. In addition, the materials before and after adsorption were tested and analyzed to explore the mechanism of the adsorption process. As-made BCN with special pore structure and hydrophobic surface owns superior adsorption capacity comparing with activated carbon (AC), multi-walled carbon nanotubes (MWCNTs) and graphene oxide (GO). Remarkable cycling stability and high antibiotic adsorption capacity were achieved, due to their excellent oxidation resistance, high specific surface area, hydrophobic surface properties and positive surface charges.

2. Materials and methods

2.1 Preparation of the BCN and MBCN

BCN and MBCN were prepared by solid state synthesis reported previously.20,23 Typically, 0.3 g H3BO3, 1 g activated carbon (LD2000, Nanjing Linda Activated Carbon Co., Ltd, China) or ordered mesoporous carbon (OMC) and 3.6 g urea were mixed in 40 mL water and dried at 60 °C. The resultant composite was pyrolyzed in a tubular oven at 900 °C for 10 h under a nitrogen atmosphere. After cooling down, the sample was further doped in NH3 at 930 °C for 3 h. The BCN is derived from activated carbon (AC), while MBCN is derived from ordered mesoporous carbon.

2.2 Material characterization

The samples prepared in this work were characterized by transmission electron microscopy (TEM, JEM-2010F) and field emission scanning electron microscopy (FE-SEM, FEI Sirion 200). Fourier transform infrared (FT-IR) spectroscopy was carried out on a Bruker Vertex 70 FTIR spectrometer in a range of 4000–500 cm−1. N2 adsorption–desorption isotherms were obtained at 77 K using an Autosorb 6B instrument. Thermogravimetric analysis (TGA) was performed on a TG209 (NETZSCH Co). The thickness was analyzed by atomic force microscopy (AFM) on a Park NX10. The surface wettability was proven using an automated contact angle tester (DSA-X, Betop Scientific). The surface characteristics of the samples were analyzed by using X-ray photoelectron spectroscopy (XPS, Thermo ESCALAB 250, USA) with Al Kα ( = 1486.6 eV) radiation.

2.3 Antibiotic concentration measurements

The concentration of ROX was detected with a semi-preparative liquid chromatograph (LC-6 AD, Shimadzu, Japan). An Agilent C18 150 mm × 2.1 mm column (at a packing particle size of 5 μm) was used. ROX was dissolved in a 10% methanol aqueous solution with pH adjusted to about 7.0 using a phosphorus–ammonia buffer solution. As shown in Fig. 1a, this method exhibits a linear correlation coefficient of 0.9997 (R2).
image file: d0ew00117a-f1.tif
Fig. 1 Relationships between the concentration and absorbance of ROX (a) and CAP (b) solutions.

The concentration of CAP was measured using a UV-vis spectrophotometer (TU-1810DASPC, Purkinje, China). CAP was dissolved directly in deionized water that had a pH of about 7.0. The absorbance values of the CAP solutions with different concentrations of 1–60 mg L−1 were determined and the standard curves of the chloramphenicol solutions were obtained, as shown in Fig. 1b. This method exhibits a linear correlation coefficient of 0.9994 (R2).

Adsorption experiments were conducted in 100 mL conical flasks with magnetic stirring at a certain solution temperature (T). Adsorption was carried out by adding 5 mg of sorbent into 20 mL aqueous solutions of CAP/ROX at initial concentrations (C0) from 50 to 1000 mg L−1. The pH values of the CAP/ROX solutions were adjusted using phosphate/ammonia buffer solutions. The BCN, MBCN and carbon materials after adsorption experiments are denoted as BCN-T-CAP/ROX, MBCN-T-CAP/ROX and GO/AC/MWCNTs-T-CAP/ROX, respectively.

3. Results and discussion

3.1 Characterization of materials

Fig. 2 shows the morphology and structure of the samples. The lattice fringes of BCN and MBCN can be seen in Fig. 2a and d. The TEM images in Fig. 2b and e show respectively the porous BCN and MBCN materials with a few-layer flaky microstructure. Similarly, the porous structure can be clearly observed in the SEM image of BCN (Fig. 2c). The size of BCN and MBCN is about 1–2 μm and the thickness is about a few nanometers (Fig. S1). It is noted that the nanosheets are heavily agglomerated. The reaction of carbon with urea and boric acid is employed because they yield BC4N.20,24,25
image file: d0ew00117a-f2.tif
Fig. 2 Morphology and structure of the samples. TEM images of (a and b) BCN and (d and e) MBCN; SEM images of (c) BCN and (f) MBCN.

In order to identify the hydrophilic and hydrophobic differences of the adsorbent materials, contact angle tests were performed. Fig. 3a, c and e show that the contact angles of BCN, MBCN and AC are 105.9°, 147.3° and 22.5° at the moment that a water droplet is on the materials, respectively. After 10 seconds, the contact angles changed to 53.3°, 54.1° and 8.1°, respectively (Fig. 3b, d and f). The results of contact angle tests reflect the difference in the hydrophilic and hydrophobic properties of the materials. It can be seen that MBCN is more hydrophobic, BCN is next, and AC is the most hydrophilic material. After excluding the influence of other factors on the adsorption, it can be inferred that MBCN prefers to adsorb more hydrophobic antibiotics, followed by BCN and AC.


image file: d0ew00117a-f3.tif
Fig. 3 Contact angle measurements. Photographs of the water droplet on BCN (a and b), MBCN (c and d) and AC (e and f) at t = 0 and 10 s, respectively.

The functional groups and chemical bonds of ROX, BCN and MBCN before and after the adsorption of ROX and CAP were determined by FT-IR spectrometry (Fig. 4a and b). Before the adsorption of antibiotics, the peaks at the wavenumbers of ∼780 cm−1 and ∼1391 cm−1 correspond to the bending vibration of B–N–B and the stretching vibration of B–N, respectively. Meanwhile, the peak at ∼1104 cm−1 corresponds to the C[double bond, length as m-dash]N bond,26 and the peak for –OH near the wavenumber of ∼3434 cm−1 nearly disappears, which proves that the oxygen-containing functional groups on the surface of MBCN are significantly reduced, which may be the direct cause of its increased hydrophobicity.27,28 XPS spectra confirmed not only the presence of B, N, C and O but also their combination in the form of O (Fig. S2). The O1s spectrum was also deconvoluted into three peaks at 531.3, 532.2 and 533.8 eV, which were assigned to N–O, C–O, and B–O, respectively. After the adsorption of the antibiotics, the typical C–O peak corresponding to antibiotics was found at the wavenumber of ∼1091 cm−1. In addition, the peak strength for –OH was significantly increased at the wavenumber of ∼3436 cm−1, indicating the successful adsorption of the antibiotics on BCN and MBCN.


image file: d0ew00117a-f4.tif
Fig. 4 (a) FT-IR spectra of ROX, BCN and MBCN before and after the adsorption of ROX; (b) FT-IR spectra of BCN and MBCN before and after the adsorption of CAP.

The specific surface area and pore structure directly affect the adsorption property of materials, so it is very necessary to conduct Brunauer–Emmett–Teller (BET) tests on the materials. Nitrogen adsorption/desorption isotherms were obtained to determine the specific surface area and pore diameter distribution of the samples. The BET test results of BCN, MBCN and AC are shown in Fig. 5 and Table S1. The nitrogen adsorption/desorption isotherms of BCN and AC at 77 K are type I curves, characteristic of microporous materials. MBCN has a type IV curve, showing the characteristics of mesoporosity.29 Due to the presence of micropores, the surface areas of AC and BCN calculated by the BET equation are 1509 m2 g−1 and 1098 m2 g−1, respectively, which are composed of micropores (82.9% and 77.8%) and mesoporous pores (17.1% and 22.2%), and the total pore volumes are 1.124 and 0.534 cm3 g−1, respectively. It is worth noting that the average pore diameter of BCN is 2.4 nm, and the main characteristic pore diameter is located at 0.59 nm, 1.18 nm and 2.73 nm. In contrast, the surface area of MBCN is only 539 m2 g−1, in which the specific surface area of mesopores is 422 m2 g−1 and the specific surface area of macropores is 117 m2 g−1. The average pore diameter is 5.4 nm, and the main characteristic pore diameter is 9.31 nm, showing a relatively wide pore diameter distribution. Obviously, AC > BCN > MBCN was obtained in terms of their specific surface area and pore volume, and MBCN > AC > BCN for their average pore diameter. These results indicate that pore diameter and specific surface area will have an important effect on the adsorption behavior of antibiotics.


image file: d0ew00117a-f5.tif
Fig. 5 (a) N2 adsorption/desorption isotherms and (b) pore size distribution curves of BCN, MBCN and AC.

3.2 Equilibrium adsorption isotherm

Fig. 6 shows the adsorption isotherms of CAP and ROX on BCN, MBCN and carbon materials, which are fitted by Langmuir and Freundlich models (ESI). Table 1 lists the corresponding isotherm fitting parameters of the two antibiotics. It can be seen that the Langmuir fitting model has a high correlation coefficient, which is more suitable to describe the antibiotic adsorption isotherm process on the BCN and MBCN. It can be inferred that the adsorption process of the antibiotics is in the single molecular layer of a uniform surface.30 According to Fig. 6a and d, the maximum adsorption capacity of MBCN is 370.60 mg g−1 for CAP and 575.68 mg g−1 for ROX. The maximum adsorption capacity for CAP of BCN and AC is 546.15 and 439.90 mg g−1, while the maximum adsorption capacity for ROX is 460.09 and 360.24 mg g−1, respectively. The summary of the antibiotic adsorption capacity of several adsorbents in recent report is listed in Table S3. It is shown that the boron nitride–carbon nanosheets should be a good choice for the next generation of high performance adsorption materials. In addition, for both kinds of antibiotics, the BCN and MBCN show a significantly higher adsorption amount than MWCNTs and GO, suggesting that BCN and MBCN are very promising materials for antibiotic adsorption. The major reasons for the large difference in antibiotic adsorption on the two kinds of boron nitride–carbon nanosheets and AC are the differences in surface properties and pore structures. The main factors affecting the adsorption process of antibiotics are pore structure, surface chemistry and heteroatom content.31,32 The adsorption capacity of adsorbent materials is not simply correlated with their specific surface area and pore volume. Adsorption capacity also depends on the accessibility of organic molecules to the inner surface of the adsorbent, that is, on the adsorbent pore size. Since the CAP molecule size is smaller than that of ROX, it can enter into smaller pores and adsorb on active sites, while ROX is restricted to enter into micropores, which is defined as the micropore-filling effect.33 Combined with BET tests, we know that the mesopores contribute 78.3% to the specific surface area of MBCN, while micropores account for 77.8% and 82.9% of the total specific surface area of BCN and AC, respectively. This difference in the pore structure is the main reason why MBCN has a higher adsorption capacity for ROX, while BCN and AC have a higher adsorption capacity for CAP.
image file: d0ew00117a-f6.tif
Fig. 6 Adsorption isotherm plots and non-linear fitting curves of CAP adsorption onto (a) MBCN, (b) BCN and (c) carbon materials at 25 °C; adsorption experimental plots and non-linear fitting curves of ROX adsorption onto (d) MBCN, (e) BCN and (f) carbon materials at 25 °C.
Table 1 Adsorption isotherm parameters describing the adsorption of the antibiotics based on the Langmuir and Freundlich models and separation factor (KL) value
Antibiotics Adsorbents Langmuir model Freundlich model
Q max (mg g−1) K L (L mg−1) R 2 R L 1/n K F R 2
CAP BCN 546.15 0.42 0.998 0.02–0.34 0.199 172.45 0.809
MBCN 420.54 1.00 0.998 0.05–0.80 0.237 211.94 0.989
ROX BCN 460.09 0.04 0.991 0.004–0.03 0.149 202.73 0.664
MBCN 575.68 0.20 0.990 0.02–0.16 0.189 203.57 0.832


Meanwhile, the surface charges and hydrophilic/hydrophobic properties of the adsorbents determined by the heteratom content, surface chemistry and roughness34 also play a key role in the antibiotic adsorption performance. The contact angle test demonstrates that the order of hydrophobicity is MBCN > BCN > AC from high to low. Moreover, according to Table S2, the solubility of ROX in water is less than that of CAP, and the octanol/water partitioning coefficient (X[thin space (1/6-em)]log[thin space (1/6-em)]P) is greater than that of CAP, indicating that roxithromycin is more hydrophobic. The strong attraction between hydrophobic molecules and hydrophobic carbon surfaces is called hydrophobic interaction (or hydrophobic bonding).31 The larger adsorption capacity for ROX of MBCN than that of BCN is essentially due to the increasing hydrophobic interaction.35 This also explains why BCN and AC have a similar pore structure, but the adsorption capacity of BCN for the two antibiotics is greater than that of AC.

In general, the value of RL indicates the adsorption property of the adsorbent, RL > 1 means unfavorable to adsorption, RL = 1 means a linear relationship with initial concentration, and 0 < RL <1 means favorable to adsorption. As shown in Table 1, for the two antibiotics, the separation factor RL is in the range of 0–1, which means that the antibiotics are easily adsorbed onto BCN and MBCN.36 In addition, 1/n of the Freundlich model is between 0 and 0.5, indicating its easy adsorption,37 which is consistent with the RL results.

3.3 Adsorption kinetics

In order to investigate the effect of contact time on the adsorption behavior of antibiotics, adsorption kinetics was tested and fitted by pseudo-first-order and pseudo-second-order kinetic models. It can be seen from Fig. 7a and b and 8a and b that, in the initial stage, the adsorption rate of the two antibiotics on BCN and MBCN was very fast, and could reach equilibrium within 60 minutes, and then the adsorption rate gradually decreased to reach the adsorption equilibrium. This may be because that in the initial stage, CAP and ROX molecules can easily obtain a large number of empty adsorption sites, and the remaining vacancies are difficult to occupy due to the repulsive force between the bulk phase and the antibiotics on the adsorbent.38 The adsorption kinetic parameters of the two models are summarized in Table 2. According to the correlation coefficient (R2) and the normalized standard deviation (ΔQ), the pseudo-second-order kinetic model achieves a better match with the experiment plots, suggesting that there are many factors affecting the adsorption behavior of adsorbents. Further studies on the adsorption mechanism of antibiotics on the boron nitride–carbon nanosheets are needed.
image file: d0ew00117a-f7.tif
Fig. 7 Adsorption experimental plots and adsorption kinetics of CAP onto (a) MBCN and (b) BCN at 25 °C; adsorption experimental plots of CAP adsorption onto (c) MBCN and (d) BCN at 25, 35, and 45 °C.

image file: d0ew00117a-f8.tif
Fig. 8 Adsorption experimental plots and adsorption kinetics of ROX onto (a) MBCN and (b) BCN at 25 °C; adsorption experimental plots of ROX adsorption onto (c) MBCN and (d) BCN at 25, 35, and 45 °C.
Table 2 Adsorption kinetic model parameters of CAP and ROX adsorbed on BCN and MBCN
Antibiotics Adsorbents Pseudo-first-order model Pseudo-second-order model
q exp K 1 q e R 2 ΔQ (%) K 2 × 10−2 q e R 2 ΔQ (%)
CAP BCN 508.17 0.40 492.43 0.99 6.58 0.21 508.49 0.999 1.54
MBCN 297.66 0.17 298.16 0.993 0.87 0.14 313.55 0.999 3.21
ROX BCN 447.62 0.64 440.84 0.997 3.78 0.70 446.70 0.998 2.55
MBCN 536.42 0.036 553.49 0.974 6.22 0.75 608.03 0.990 1.44


Fig. 7c and d and 8c and d show the influence of temperature on the adsorption of the antibiotics, in which the temperature range was 25–45 °C. The adsorption capacity for the two antibiotics of BCN and MBCN increases with the increase of temperature, indicating that the adsorption is an endothermic process, and the increase of temperature is beneficial to the adsorption process. It is well known that physical adsorption is exothermic and chemical adsorption is endothermic. Therefore, it can be speculated that chemisorption exists in the adsorption process.

3.4 The effects of pH and salinity

Salt ions are commonly found in various water bodies, so it is necessary to study the effect of salt ions on the adsorbent adsorption of antibiotics. As shown in Fig. 9a and b, 0–1.0 M NaCl was added to CAP and ROX solutions to study the effect of salinity on the adsorption capacity. It can be seen that the adsorption capacity decreases with the increase of NaCl concentration. The decrease of adsorption capacity is mainly due to the competitive adsorption between salt ions and antibiotic molecules at the adsorption site of the adsorbent.38 On the other hand, despite the increase of competitive adsorption, the adsorption capacity for the two kinds of antibiotics of BCN and MBCN is still as high as 300–500 mg g−1, indicating that the synthesized boron nitride–carbon nanosheets have a high potential application value in antibiotic wastewater treatment.
image file: d0ew00117a-f9.tif
Fig. 9 (a and b) Effects of NaCl ion concentration; (c) zero charge potential curves of the electrodes prepared from BCN, AC and MBCN; (d and e) effects of solution pH on the CAP and ROX adsorption onto MBCN and BCN (t = 25 °C); (f) zeta potential of BCN, AC and MBCN.

To further determine the factors affecting the adsorption process, we studied the effect of solution pH for the adsorption of CAP and ROX on BCN and MBCN, in the ranges of 2.0–10.0 and 2.0–7.0, respectively. As illustrated in Fig. 9d, the adsorption capacity for ROX increases with the increase of pH. The pKa of ROX is 9.2, while that of CAP is 5.5. According to the zero charge potential (Epzc) and zeta potential tests of the samples (Fig. 9c and f), BCN and MBCN have significant positive charges on the surface in the neutral solution, while AC has negative charges. The zero charge point (pHpzc) of BCN and MBCN is about pH = 7.7 and 6.7, respectively. This means that the surface charges of BCN and MBCN remain positive in the pH range of 2.0–7.7 and 2.0–6.7. When pHpzc < pH < 9.2, ROX exists in the form of a cation, and with the increase of pH, the surface charges of BCN and MBCN change to zero (or negative), the electrostatic repulsion between the adsorbents and ROX gradually weakens, and the adsorption capacity gradually increases. Similarly, as shown in Fig. 9e, the adsorption capacity for CAP shows an upward trend in the range of 2.0 < pH < 7.0, and then decreases sharply in the range of 7.0 to 8.0, and then increases rapidly. Obviously, solution pH is one of the important operational parameters that affect the adsorption process. The specific explanation is as follows: when the pH value is less than 5.5, CAP is mainly in the form of a cation and has a strong electrostatic repulsion with positively charged adsorbents. When 5.5 < pH < 6.7 (or 7.7), CAP is mainly in an anion form, while the MBCN (or BCN) surface has a positive charge, and the adsorption capacity increases with pH. When pKa < pH < 8.0, the decrease in the adsorption capacity can be explained by the repulsive effect between CAP and negatively charged adsorbents. Subsequently, the increase in the adsorption capacity for CAP of BCN and MBCN can be attributed to the following fact: the hydroxyl dissociation of CAP leaves the CAP with positive charges38 that are attached to the negatively charged adsorbents, resulting in a high adsorption capacity.

It is generally believed that the adsorption method has been favorable in wastewater treatment due to its advantages of low cost, simple to design and high efficiency. However, the poor reusability of adsorbents is one of the serious drawbacks that restrict its industrial application. Due to their high thermal stability, a simple high-temperature calcination method was used (10 min at 500 °C) to regenerate BCN and MBCN (Fig. S3), and the adsorption capacities of the recycled adsorbents after 6 cycles are shown in Fig. 10a and b. After 6 times of regeneration, the equilibrium adsorption capacity can still remain above 91.9% of the initial adsorption capacity, which proves that after thermal treatment in air, the prepared BCN and MBCN can be used as effective adsorbents for multiple cycles. Fig. S4a and b display the XRD patterns of BCN and MBCN before and after 2 cycles. It can be observed that the main characteristic diffraction peaks of BCN and MBCN after regeneration are still detectable. The SEM images (Fig. S4c–f) and EDS analysis (inset) of the samples show that BCN and MBCN still retain the original morphology of the nanosheets but the ball-like architecture collapsed during regeneration. In order to compare the effects of different regeneration conditions on the cycling performance of BCN, MBCN and AC, ROX and CAP were hydrolyzed with 0.2 M HCl and 0.2 M NaOH to regenerate the adsorbents (Fig. 10c and d). The experiment showed that the cycling stability of the adsorbents treated by high-temperature calcination was far better than that by acid and base hydrolysis. We believe that acid/base hydrolysis regeneration methods cannot completely achieve the desorption of antibiotics, and the residual antibiotics still occupy the adsorption sites of adsorbents, leading to the loss of adsorption capacity, poor cycle stability and regeneration performance.


image file: d0ew00117a-f10.tif
Fig. 10 (a and b) Cycle stability of BCN and MBCN after regeneration at 500 °C for 10 min in air; (c and d) cycle stability of BCN, MBCN and AC under different regeneration conditions.

4. Conclusion

Hydrophobic BCN and MBCN materials with different pore distributions and high specific surface areas were synthesized for the effective adsorption of antibiotics. The maximum adsorption capacity for CAP of BCN was 546.15 mg g−1, and the adsorption capacity for ROX was only 447.62 mg g−1. However, the adsorption capacity of MBCN for ROX was relatively high, reaching 575.68 mg g−1, while that for CAP was only 297.66 mg g−1. In addition, the Langmuir model fits better with the adsorption isotherm, and the pseudo-second-order kinetic model can well describe the antibiotic adsorption kinetic curve. It is worth noting that with the increase of the molecular weight of antibiotics, huge antibiotic molecules are blocked from entering the micropores, and the adsorption capacity of BCN decreases, indicating that the micropore-filling effect is the main mechanism of the antibiotic adsorption on BCN. In addition, we found that the hydrophobic interaction in chemical adsorption was also an important factor affecting the adsorption capacity of MBCN and BCN. Finally, the experiments on the effects of pH and salt ions on adsorption indicate that electrostatic interaction was also involved in the antibiotic adsorption process of BCN and MBCN. The as-prepared BCN and MBCN can be recycled by simple high-temperature regeneration in air. After 6 cycles, the equilibrium adsorption capacity can still remain above 91.9% of the initial adsorption capacity, which proves that the boron nitride–carbon materials are a promising candidate for wastewater antibiotic adsorption.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 21878049), the Dongguan Introduction Program of Leading Innovative and Entrepreneurial Talents, and the Dongguan Academician Workstation Project (DGYSZ-2018-06).

Notes and references

  1. Q. Q. Zhang, G. G. Ying, C. G. Pan, Y. S. Liu and J. L. Zhao, Comprehensive evaluation of antibiotics emission and fate in the river basins of China: source analysis, multimedia modeling, and linkage to bacterial resistance, Environ. Sci. Technol., 2015, 49, 6772–6782 CrossRef CAS PubMed.
  2. H. D. Xu, Y. C. Zhang, J. J. Li, Q. Q. Hao, X. Li and F. H. Liu, Heterogeneous activation of peroxymonosulfate by a biochar-supported Co3O4 composite for efficient degradation of chloramphenicols, Environ. Pollut., 2020, 257, 113610–113618 CrossRef CAS PubMed.
  3. J. P. Li, Z. F. Min, W. Li, L. J. Xu, J. G. Han and P. P. Li, Interactive effects of roxithromycin and freshewater microalgae, Chlorella pyrenoidosa: Toxicity and removal mechanism, Ecotoxicol. Environ. Saf., 2020, 191, 110156–110165 CrossRef CAS PubMed.
  4. H. Tekin, O. Bilkay, S. S. Ataberk, T. H. Balta, I. H. Ceribasi, F. D. Sanin, F. B. Dilek and U. Yetis, Use of Fenton oxidation to improve the biodegradability of a pharmaceutical wastewater, J. Hazard. Mater., 2006, 136, 258–265 CrossRef CAS PubMed.
  5. T. A. K. Sandhya Babel, Low-cost adsorbents for heavy metals uptake from contaminated water: a review, J. Hazard. Mater., 2003, 97, 219–243 CrossRef.
  6. V. Homem and L. Santos, Degradation and removal methods of antibiotics from aqueous matrices--a review, J. Environ. Manage., 2011, 92, 2304–2347 CrossRef CAS PubMed.
  7. P. A. Neale and A. I. Schäfer, Magnetic ion exchange: Is there potential for international development?, Desalination, 2009, 248, 160–168 CrossRef CAS.
  8. J. Y. Ji, Y. J. Xing, Z. T. Ma, J. Cai, P. Zheng and H. F. Lu, Toxicity assessment of anaerobic digestion intermediates and antibiotics in pharmaceutical wastewater by luminescent bacterium, J. Hazard. Mater., 2013, 246–247, 319–323 CrossRef CAS PubMed.
  9. B. Demirel and P. Scherer, The roles of acetotrophic and hydrogenotrophic methanogens during anaerobic conversion of biomass to methane: a review, Rev. Environ. Sci. Bio/Technol., 2008, 7, 173–190 CrossRef CAS.
  10. J. Ma, F. Yu, L. Zhou, L. Jin, M. Yang, J. Luan, Y. Tang, H. Fan, Z. Yuan and J. Chen, Enhanced adsorptive removal of methyl orange and methylene blue from aqueous solution by alkali-activated multiwalled carbon nanotubes, ACS Appl. Mater. Interfaces, 2012, 4, 5749–5760 CrossRef CAS PubMed.
  11. F. Yu, J. Ma and D. S. Bi, Enhanced adsorptive removal of selected pharmaceutical antibiotics from aqueous solution by activated graphene, Environ. Sci. Pollut. Res., 2015, 22, 4715–4724 CrossRef CAS PubMed.
  12. Y. S. L. L. Ji, Z. Y. Xu, S. R. Zheng and D. Q. Zhu, Adsorption of monoaromatic compounds and pharmaceutical antibiotics on carbon nanotubes activated by KOH etching, Environ. Sci. Technol., 2010, 44, 6429–6436 CrossRef PubMed.
  13. K. C. Bedin, A. C. Martins, A. L. Cazetta, O. Pezoti and V. C. Almeida, KOH-activated carbon prepared from sucrose spherical carbon: Adsorption equilibrium, kinetic and thermodynamic studies for Methylene Blue removal, Chem. Eng. J., 2016, 286, 476–484 CrossRef CAS.
  14. P. Liao, Z. Zhan, J. Dai, X. Wu, W. Zhang, K. Wang and S. Yuan, Adsorption of tetracycline and chloramphenicol in aqueous solutions by bamboo charcoal: A batch and fixed-bed column study, Chem. Eng. J., 2013, 228, 496–505 CrossRef CAS.
  15. T. V. Tran, D. T. C. Nguyen, H. T. N. Le, H. L. Ho, T. T. Nguyen, V. D. Doan, T. D. Nguyen and L. G. Bach, Response surface methodology-optimized removal of chloramphenicol pharmaceutical from wastewater using Cu3(BTC)2-derived porous carbon as an efficient adsorbent, C. R. Chim., 2019, 22, 794–803 CrossRef CAS.
  16. J. W. Yang, G. Z. Ji, Y. Gao, W. Fu, M. Irfan, L. Mu, Y. L. Zhang and A. M. Li, High-yield and high-performance porous biochar produced from pyrolysis of peanut shell with low-dose ammonium polyphosphate for chloramphenicol adsorption, J. Cleaner Prod., 2020, 8, 121516–121525 CrossRef.
  17. Y. Q. Huang, Y. W. Wang, Y. Z. Huang, L. X. Zhang, F. Ye, J. L. Wang, J. G. Shang and Q. J. H. Liao, Impact of sediment characteristics on adsorption behavior of typical antibiotics in Lake Taihu, China, Sci. Total Environ., 2020, 718, 137329 CrossRef CAS PubMed.
  18. Q. Song, J. Liang, Y. Fang, C. Cao, Z. Liu, L. Li, Y. Huang, J. Lin and C. Tang, Selective adsorption behavior/mechanism of antibiotic contaminants on novel boron nitride bundles, J. Hazard. Mater., 2019, 364, 654–662 CrossRef CAS PubMed.
  19. D. Tu, H. Liao, Q. Deng, X. Liu, R. Shang and X. Zhang, Renewable biomass derived porous BCN nanosheets and their adsorption and photocatalytic activities for the decontamination of organic pollutants, RSC Adv., 2018, 8, 21905–21914 RSC.
  20. K. Raidongia, A. Nag, K. Hembram, U. V. Waghmare, R. Datta and C. N. R. Rao, BCN: A Graphene Analogue with Remarkable Adsorptive Properties, Chem. – Eur. J., 2010, 16, 149–157 CrossRef CAS PubMed.
  21. J. Xiong, W. S. Zhu, H. P. Li, W. J. Ding, Y. H. Chao, P. W. Wu, S. H. Xun, M. Zhang and H. M. Li, Graphene-like Boron nitride induced remarkable adsorption capacity for dibenzothiophene in fuels, Green Chem., 2015, 17, 1647–1656 RSC.
  22. A. Ahmadi Peyghan, N. L. Hadipour and Z. Bagheri, Effects of Al doping and double-Antisite defect on the adsorption of HCN on a BC2N nanotube: density functional theory studies, J. Phys. Chem. C, 2013, 117, 2427–2432 CrossRef CAS.
  23. Y. Zhang, G. Wang, S. Wang, J. Wang and J. Qiu, Boron-nitride-carbon nanosheets with different pore structure and surface properties for capacitive deionization, J. Colloid Interface Sci., 2019, 552, 604–612 CrossRef CAS PubMed.
  24. A. Nag, K. Raidongia, K. P. S. S. Hembram, R. Datta, U. V. Waghmare and C. N. R. Rao, Graphene analogues of BN: novel synthesis and properties, ACS Nano, 2010, 4, 1539–1544 CrossRef CAS PubMed.
  25. P. He, C. Zhou, S. Tian, J. Sun, S. Yang, G. Ding, X. Xie and M. Jiang, Urea-assisted aqueous exfoliation of graphite for obtaining high-quality graphene, Chem. Commun., 2015, 51, 4651–4654 RSC.
  26. M. Chhetri, S. Maitra, H. Chakraborty, U. V. Waghmare and C. N. R. Rao, Superior performance of borocarbonitrides, BxCyNz, as stable, low-cost metal-free electrocatalysts for the hydrogen evolution reaction, Energy Environ. Sci., 2016, 9, 95–101 RSC.
  27. O. G. Apul and T. Karanfil, Adsorption of synthetic organic contaminants by carbon nanotubes: a critical review, Water Res., 2015, 68, 34–55 CrossRef CAS PubMed.
  28. B. Pan and B. S. Xing, Adsorption mechanisms of organic chemicals on carbon nanotubes, Environ. Sci. Technol., 2008, 42, 9005–9013 CrossRef CAS PubMed.
  29. W. S. Zhu, X. Gao, Q. Li, H. P. Li, Y. H. Chao, M. J. Li, S. M. Mahurin, H. M. Li, H. Y. Zhu and S. Dai, Controlledgas exfoliation of boron nitride into few-Layered nanosheets, Angew. Chem., Int. Ed., 2016, 55, 10766–10770 CrossRef CAS PubMed.
  30. I. Langmuir, The adsorption of gases on plane surfaces of glass, mica and platinum, J. Am. Chem. Soc., 1918, 40, 1361–1403 CrossRef CAS.
  31. C. Moreno-Castilla, Adsorption of organic molecules from aqueous solutions on carbon materials, Carbon, 2004, 42, 83–94 CrossRef CAS.
  32. X. Peng, F. Hu, F. L. Lam, Y. Wang, Z. Liu and H. Dai, Adsorption behavior and mechanisms of ciprofloxacin from aqueous solution by ordered mesoporous carbon and bamboo-based carbon, J. Colloid Interface Sci., 2015, 460, 349–360 CrossRef CAS PubMed.
  33. L. Ji, Y. Wan, S. Zheng and D. Zhu, Adsorption of tetracycline and sulfamethoxazole on crop residue-derived ashes: implication for the relative importance of black carbon to soil sorption, Environ. Sci. Technol., 2011, 45(13), 5580–5586 CrossRef CAS PubMed.
  34. X. M. Li, D. Reinhoudt and M. Crego-Calama, What do we need for a superhydrophobic surface? A review on the recent progress in the preparation of superhydrophobic surfaces, Chem. Soc. Rev., 2007, 36, 1350–1368 RSC.
  35. P. Pendleton, S. H. Wu and A. Badalyan, Activated carbon oxygen content influence on water and surfactant adsorption, J. Colloid Interface Sci., 2002, 246, 235–240 CrossRef CAS PubMed.
  36. M. R. Azhar, H. R. Abid, V. Periasamy, H. Sun, M. O. Tade and S. Wang, Adsorptive removal of antibiotic sulfonamide by UiO-66 and ZIF-67 for wastewater treatment, J. Colloid Interface Sci., 2017, 500, 88–95 CrossRef CAS PubMed.
  37. C. H. Hou, S. C. Huang, P. H. Chou and W. Den, Removal of bisphenol A from aqueous solutions by electrochemical polymerization on a carbon aerogel electrode, J. Taiwan Inst. Chem. Eng., 2015, 51, 103–108 CrossRef CAS.
  38. S. Tian, J. Dai, Y. Jiang, Z. Chang, A. Xie, J. He, R. Zhang and Y. Yan, Facile preparation of intercrossed-stacked porous carbon originated from potassium citrate and their highly effective adsorption performance for chloramphenicol, J. Colloid Interface Sci., 2017, 505, 858–869 CrossRef CAS PubMed.

Footnote

Electronic supplementary information (ESI) available: AFM images of BCN and MBCN; adsorption kinetics; TGA data; XPS analysis of the samples; SEM images of the samples before and after cycling. See DOI: 10.1039/d0ew00117a

This journal is © The Royal Society of Chemistry 2020