Converting obsolete copy paper to porous carbon materials with preeminent adsorption performance for tetracycline antibiotic

Abstract

To date, the employment of developed adsorbents in antibiotic wastewater treatment practices is still limited due to their low adsorption capacity, slow kinetics and especially high cost. Thus, in this work, we first report the conversion of obsolete copy paper to highly porous carbon materials (PCMs) via a two-step method: low-temperature carbonization and alkali activation. The influence of activation temperature and KOH content on the porosity and adsorption capacity of PCMs was also studied. Notably, copy paper (CP) is consumed in extremely large quantities with a high proportion being obsoleted; it mainly consists of micro- and nano-cellulose fibers that can be used as an ideal carbon source. The PCMs were characterized by several techniques and methodologies. The PCMs-850-4 exhibited an ultrahigh BET surface area of 3598.95 m² g⁻¹ and total pore volume of 1.887 cm³ g⁻¹. The influence of temperature, initial concentration, contact time, pH and ionic strength on the adsorption of tetracycline (TC) from water to PCMs-850-4 was investigated through batch adsorption studies. Analyses of adsorption isotherm, kinetics and thermodynamics property were conducted to understand the adsorption behavior. The equilibrium experimental data was well fitted to the Langmuir model, and the kinetic data was best described by the pseudo-second-order rate model. Importantly, the PCMs-850-4 displayed an ultrahigh adsorption of 1437.76 mg g⁻¹ at 298 K, and increasing temperature benefited the adsorption. Also, their fast kinetics and great regeneration make the PCMs-850-4 a promising adsorbent for the low-cost, highly efficient and fast removal of organic pollutants from water environments.

---

1. Introduction

Antibiotics discharged into the environment have received increasing attention because they have been proven to be a class of potent pollutants.¹ Most antibiotics cannot be completely metabolized and absorbed by organisms, and a large proportion excreted through urine and feces still maintain biological activity.² Tetracycline (TC) is utilized extensively for human therapy and in the farming industry, and it is also used as food additive to prevent disease. As shown in Fig. 1, tetracycline contains tricarbonylamide, phenolic diketone and dimethylamine groups.³ The widespread use of TC has become an urgent issue as it poses a variety of undesirable effects, such as cancer, acute and chronic toxicity, and antibiotic resistance of microorganisms. Residues of TC have frequently been detected in soil, surface water, groundwater, and even drinking water.^4,5 There are many treatment methods for the removal of pharmaceutical antibiotics from wastewaters, such as chemical precipitation,⁶ degradation,⁷ bioremoval,⁸ adsorption,⁹ chemical oxidation/reduction,¹⁰ and membrane filtration.¹¹ Adsorption has the advantages of easy handling, high efficiency and economic feasibility, etc. The excellent properties and high efficiency of adsorbents is important for the removal of TC. Activated carbons are usually utilized as adsorbent to remove contaminants from wastewater via π–π conjugation, van der Waals force, electrostatic interaction, or chemical bonds.^12,13 However, large-scale application of commercial activated carbon is hindered due to its high cost and poor reusability. Hence, the development of economic, eco-friendly and more efficient adsorbents to remove TC is an urgent research subject.^14–16

Fig. 1 The molecular structure of TC.

Porous carbon materials (PCMs) are considered promising materials for applications in many fields, especially as special adsorbents in wastewater purification,¹⁷ owing to their large specific surface area, high physicochemical stability, high adsorption capacity, high mechanical strength and high degree of surface reactivity.¹⁸ Mass consumption of fossil fuels and increasing environmental problems have brought to people's attention the importance of exploiting the sustainable production of fuels, chemicals and materials,^19–21 and of striving to minimize contaminations such as toxic gases emissions²² and solid waste.^23,24 With the advent of the information era, photocopying has gradually replaced handwriting, so that the amount of copy paper (CP) is on the rise and enormous. Nevertheless, how to deal with the waste CP has become a serious problem. The principal component of CP is plant fiber, including cellulose, hemicelluloses and lignin, which are ideal carbon precursors for the preparation of PCMs as adsorbents to remove pollutants. In a deeper sense, renewable cellulose, hemicelluloses, and lignin are the most abundant biomass materials with excellent properties, such as easy availability, biodegradability and non-polluting. The microstructure of CP is composed of a large number of belts and filamentous fibers in micro and nano size, which benefit the alkali activation process to produce a large number of micropores. Therefore, CP is a more ideal precursor for the preparation of PCMs.

Currently, there is no doubt that biomass is the cost-optimal precursor for the preparation of PCMs. For example, Martins et al. reported that NaOH-activated carbon produced from macadamia nut shells showed the maximum monolayer adsorption capacity (Q_m) of 455.33 mg g⁻¹ for TC.²⁵ Maneerung et al. reported activated carbon derived from carbon residue from biomass gasification to remove Rhodamine B with a maximum monolayer adsorption capability of 189.83 mg g⁻¹.²⁶ Altenor et al. prepared activated carbon using vetiver roots as precursor by chemical activation for the adsorption of methylene blue and phenol.²⁷ However, to the best of our knowledge, the use of obsolete CP as carbon precursor to prepare PCMs has not been reported.

Herein, we used obsolete CP as a renewable biomass precursor to fabricate high-performance PCMs, which are expected to display fast kinetics, good regeneration capacity and excellent environmental adaptability for TC removal, with a huge potential for large-scale application in wastewater treatment. More importantly, this work provides new insight into the preparation of advanced adsorbents and the utilization of biomass resources, which is of great significance for the environment and energy fields. Additionally, the PCMs are promising materials not only for wastewater treatment, but also for energy storage and conversion and catalysis applications.

2. Experimental

2.1. Materials and chemicals

Tetracycline hydrochloride (TC, 98%) was obtained from Aladdin Industrial Corporation (Shanghai, China), and stored in the dark at 2–8 °C. Waste copy papers were obtained daily from an office. KOH (AR, 96%), KBr (AR, 99%) and HCl (36–38%) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All chemicals were of analytical reagent grade and used directly without any further purification. The water used in all experiments was deionized water.

2.2. Instruments and characterization methods

The carbonization and activation was carried out in a tube furnace (SK-GO6123K, Tianjin, China). Elemental analysis on the samples was conducted using an element analyzer (FLASH1112A, CE, Italy). The microtexture and morphology were analyzed by transmission electron microscopy (TEM; IEM-200CX, JEOL, Japan) and scanning electron microscopy (SEM; JSM-7001F, JEOL, Japan). Fourier transform infrared spectra were recorded on a Tensor FTIR spectrometer (FT-IR; Nicolet Nexus 470, USA). X-ray diffraction (XRD) analysis was taken on an X-ray diffractometer (Bruker D8 Advance, Bruker AXS, Germany) using Cu Kα radiation (λ = 1.5406 Å, 40 kV, 40 mA), and the data were collected from 2θ = 10–70° at a scan rate of 7° min⁻¹ for phase identification. Raman spectra were analyzed using a Laser Raman spectrometer (DXR, Thermo Fisher, USA) with a 532 nm wavelength incident laser light and 10 mW power. The N₂ adsorption–desorption isotherms were measured using a BELSORP instrument (BEL, Japan, Inc.) at 77 K. X-ray photoelectron spectroscopy (XPS) analysis was conducted in a Kratos Axis Ultra DLD spectrometer with X-ray excitation provided by a monochromatic Al Kα source (1486.6 eV).

2.3. Preparation of PCMs

In brief, the CP was carbonated in a tube furnace at a ramp of 5 °C min⁻¹ from RT to 500 °C and maintained at 500 °C for 2 h in nitrogen atmosphere. The carbonized CP (CCP) was obtained and stored for further use. Subsequently, the CCP was sufficiently ground with activation agent KOH at a mass ratio of 1 : 4 (CCP : KOH), and the obtained mixture was subject to thermal treatment at desirable temperatures varying from 750 to 900 °C for 1 h in a tube furnace under the protection of nitrogen, with a heating ramp of 5 °C min⁻¹. Finally, the resulting products were soaked in 2 M HCl to eliminate impurity. The PCMs were collected via vacuum filtration, washed several times with deionized water to neutral, and then dried in oven at 80 °C for 12 h. The influence of KOH content for PCMs was also studied. The preparation method was the same as above-mentioned, except that the mass ratio of CCP and KOH was varied from 1 : 1 to 1 : 5 at 850 °C (labeled as PCMs-T-x; T and x represent temperature and mass ratios, respectively).

2.4. Batch adsorption experiments

All batch experiments were carried out using a 10 mL centrifuge tube containing 2 mg of PCMs-850-4 and 10 mL of TC solution in the dark, and all of the tests were carried out in triplicate. Supernatants were obtained through a 0.45 μm membrane filter and determined by an UV-vis spectrophotometer (Agilent Cary 8454 UV-vis) at a fixed wavelength of 357 nm. Our preliminary study (adsorbent-free) indicated that the loss of TC owing to filter interception was far less than 4% under the applied experimental condition (data not shown).

The adsorption isotherms were performed using TC solutions with the initial concentrations ranging from 50 to 350 mg L⁻¹ at 298, 308 and 318 K for 12 h to reach equilibrium. The equilibrium adsorption amount Q_e (mg g⁻¹) was calculated by the following equation:

The kinetic studies were investigated at appropriate time intervals of 1 to 90 min using 100, 150 and 200 mg L⁻¹ TC solutions at 298 K, respectively. The adsorption amount at t time Q_t (mg g⁻¹) was calculated by the following equation:

herein, C₀ and C_t (mg L⁻¹) are the beginning and at any time (t) concentrations of TC, respectively. C_e (mg L⁻¹) is concentration of TC at equilibrium condition. V (L) is volume of the TC solution, and M (g) is mass of the PCMs-850-4.

To further study the adsorption of TC onto PCMs-850-4, we investigated the effect of solution pH value and ionic strength on the adsorption. The experiments were conducted using 250 mg L⁻¹ TC solution. The solution pH values were in a range from 2 to 6, which were adjusted using 0.1 mol L⁻¹ NaOH and 0.1 mol L⁻¹ HCl solutions. Moreover, different concentrations of NaCl contents were 0.05, 0.2, 0.4, 0.6, 0.8 and 1.0 M, respectively. The above solutions were used for adsorption at 298 K for 12 h to reach equilibrium.

3. Results and discussion

Fig. 2 shows the influence of activation temperature and KOH content on porosity and adsorption capacity of PCMs, and the different parameters of porosity characteristics are listed in Table 1. The result shown in Fig. 2a indicates that activation temperature significantly influenced the porosity of PCMs. BET surface area increased with the increase of temperature from 2978.44 m² g⁻¹ to 3598.95 m² g⁻¹, but decreased to 3344.58 m² g⁻¹ when the temperature reached 900 °C. Carbon and KOH might overreact at high temperature and result in the increase of pore size and decrease of BET surface area, naturally leading to the decline in TC adsorption capacity. As shown in Fig. 2a, the PCMs-900-4 had a large number of mesopores. From the inset of Fig. 2a, the pore size increased with increasing temperature, which confirmed the above-mentioned speculation. Meanwhile, BET surface area increased with the increase of KOH content (N₂ adsorption–desorption isotherm of PCMs-850-5 was not provided because of an ultra-low yield of about 2%). The pore size also showed a rising trend with the increase of the KOH content (from inset of Fig. 2b). The above results show that the activation temperature and KOH content had a significant influence on the porosity of PCMs. Fig. 2c and d indicate that the PCMs-850-4 displayed the optimal adsorption capacity. Combining all above results, we find that the porosity of PCMs was crucial for the TC adsorption performance. Therefore, PCMs-850-4 was chosen for further investigation in the following adsorption experiments.

Fig. 2 (a and b) N₂ adsorption–desorption isotherms and pore size distributions (the DFT method) of PCMs-T-x, (c) equilibrium adsorption capacity of PCMs-850-x for TC, (d) equilibrium adsorption capacity of PCMs-T-4 for TC.

Table 1 The porosity characteristics of PCMs-T-x synthesized at different parameters

Parameters	Specific surface area (m^2 g^−1)	Micropore surface area (m^2 g^−1)	Total pore volume (cm^3 g^−1)	Micropore volume (cm^3 g^−1)	Microporosity (%)	Average pore size (nm)
PCMs-750-4	2978.44	2661.32	1.259	1.390	89.35	1.645
PCMs-800-4	3053.65	2752.88	1.321	1.442	90.15	1.982
PCMs-850-1	2606.63	2187.60	1.232	0.966	83.94	1.718
PCMs-850-2	2866.05	2695.57	1.281	1.096	92.79	1.867
PCMs-850-3	3051.30	2790.76	1.612	1.499	91.46	1.916
PCMs-850-4	3598.95	3302.89	1.887	1.632	91.77	2.002
PCMs-900-4	3344.58	1153.53	2.486	0.779	34.49	2.973

---

We conducted scanning electron microscopy (SEM) study for the synthesized PCMs-850-4. SEM images of CP, CCP and PCMs-850-4 are shown in Fig. 3. It was observed that the CP was made up of many stacked ribbons with some impurity, and it possessed a relatively smooth surface. From Fig. 3b, it is clear that the stack was not sealed; ribbons mutually crisscross with the abundant space. Fig. 3c and d reveal that CCP still maintained its original morphology, but the surface became rougher with wrinkled and lamellar structures. The structure of PCMs-850-4 undergoes tremendous changes. The ribbons become fragmented in the activation process. Moreover, the surface of the bulk greatly changed, and the layer structure was more obvious. The results imply that PCMs-850-4 possesses an abundantly porous structure. The morphology of PCMs-850-4 was further investigated by TEM. PCMs-850-4 (in Fig. 4a) had the presence of a large number of pores, and meanwhile belt-like and some layered structures also could be observed in Fig. 4b, which matched well with the results of SEM images.

Fig. 3 SEM images of CP (a and b), CCP (c and d) and PCMs-850-4 (e and f).

Fig. 4 TEM images of PCMs-850-4.

To further analyze the porosity of PCMs-850-4, the N₂ sorption isotherm was measured. Strong N₂ adsorption of PCMs-850-4 exhibited a combined type-I/IV adsorption isotherm (Fig. 5a). This behavior was widely different from the I-type of traditional activated carbon according to N₂ adsorption–desorption isotherms,²⁸ which is representative of a combination of microporous and mesoporous structures. We observed a blurry capillary condensation phenomenon in a relative pressure range of 0.15–0.30, which was typically mesoporous. The specific surface area of PCMs-850-4 calculated by the BET method was 3598.95 m² g⁻¹ with a microporosity of 91.77%, which could provide a large amount of active sites for TC molecules. The inset of Fig. 5a exhibits the pore size distributions, which were analyzed by DFT method. The pore size had a concentrated distribution range of 1.766–2.313 nm, and the micropores were concentrated at 1.76 nm. The total pore volume of PCMs-850-4 was calculated to be 1.887 cm³ g⁻¹, and the mean pore diameter was 2.002 nm. The large volume of the PCMs-850-4 is significantly favorable for the rapid diffusion of TC molecules, resulting in an exceptionally fast rate of adsorption. Furthermore, the high proportion of micropores is extremely beneficial for micropore filling.²⁹ However, as shown in Fig. 5b, the BET surface areas (<25 m² g⁻¹) of CP and CCP were negligible relative to PCMs-850-4. Correspondingly, the adsorption capacities (<30 mg g⁻¹) of CP and CCP for TC were very little (as shown in Fig. S1†). The results indicate that KOH was crucial for the porosity and adsorption capacity of PCMs-850-4.

Fig. 5 N₂ adsorption–desorption isotherms of (a) PCMs-850-4, (b) CP and CCP.

The phase structure of CCP and PCMs-850-4 was characterized by XRD and Raman spectroscopy. Fig. 6a shows the XRD diffraction patterns of CCP and PCMs-850-4. The XRD patterns revealed CCP and PCMs-850-4 with very low crystalline order. We can find two wide diffraction peaks in the patterns of PCMs-850-4. One of them is located at 22–26° and correspond to the (002) crystal plane.^30,31 The other feature band with weak intensity is from the combined action of diffraction peaks of the (100) and (101) crystal planes. Abundant pore structures (primarily micropores) continuously scattered the X-ray radiation and led to the increasing background in the 2θ range between 10 and 39°,³² which was consistent with textural data of the porosity produced in the activation process. However, the patterns of CCP had only one wide diffraction peak located at 22–26° corresponding to the (002) crystal plane, indicating only the existence of amorphous carbon structure. The results demonstrate that PCMs-850-4 had mostly amorphous carbon structure and a certain trend of graphitization.

Fig. 6 XRD patterns (a) and Raman spectra (b) of CCP and PCMs-850-4.

The Raman spectra of CCP and PCMs-850-4 are shown in Fig. 6b. Two prominent broad bands are located around 1580 (G band) and 1350 cm⁻¹ (D band), which correspond to the graphitic lattice vibration mode and disorder in the graphitic structure, respectively. We could conveniently estimate the degree of graphitization by the intensity ratio of the G/D peak (I_G/I_D).³³ The higher the I_G/I_D ratio value, the higher the degree of ordering. The I_G/I_D ratios were 0.69 for CCP and 0.97 for PCMs-850-4, revealing that CCP treatment with KOH results in increasing disorder, which is caused by potassium vapor intercalating into the carbon layer in the activation process. The activation reaction is expressed as follows:³⁴

6KOH + 2C = 2K + 3H₂ + 2K₂CO₃

XPS test was conducted to probe the chemical identification of the elements in the CCP and PCMs-850-4. The results agree well with the elemental analysis (Table S1†). It can be seen that there were C 1s and O 1s peaks from the survey scan spectra (Fig. S2†). The spectra for C 1s spectrum (Fig. 7a and c) have three peaks at 284.5, 284.83 and 286.8 eV, which were attributed to C–C, C=C, C–O and C=O,^35,36 respectively. The three peaks fitted well to the O 1s spectra (Fig. 7b and d) of both samples, located at 531.3, 532.4 and 533.7 eV, which correspond to C=O, H–O and C–O–C,^37–39 respectively. It is noteworthy that the present forms of carbon have no evident change, while obvious changes took place in the oxygen after carbonization (in Table S2†). These results strongly prove that the PCMs-850-4 possesses oxygen-rich groups, benefiting the interactions with TC molecules.

Fig. 7 High-resolution XPS spectra of CCP (a and b) and PCMs-850-4 (c and d).

FT-IR spectra of CCP and PCMs-850-4 are shown in Fig. S3.† As can be seen in the spectrum of CCP, the broad absorbance between 500 and 1650 cm⁻¹ is attributed to the C–C, C=C, C–O and O–H bending vibrations, respectively.⁴⁰ We also observed an extremely weak peak at 1780 cm⁻¹ (C=O). The weak absorbance peak at 2930 cm⁻¹ is ascribed to –CH₂– stretching vibrations. Obviously, the band at 3440 cm⁻¹ is from the O–H stretching vibrations. However, the spectrum of PCMs-850-4 did not have an obvious change, and the peak intensity of bound O–H slightly decreased, indicating the decrease of hydrogen and oxygen elements. These results agreed well with the XPS analysis and elemental analysis.

3.1. Adsorption of TC

The adsorption isotherm describes how adsorbate molecules are distributed between the solid and liquid phases when the adsorption system reaches equilibrium state.⁴¹ The Langmuir isotherm model hypothesizes that adsorption onto the adsorbent's surface is via monolayer adsorption, while the Freundlich isotherm model differs from the Langmuir isotherm, assuming multilayer adsorption.^42,43 Basically, adsorption isotherms indicate how interaction occurs between the adsorbate and adsorbents, being important for optimizing the use of adsorbents.

Adsorption isotherm. Langmuir model was employed to analyze the adsorption data of TC onto the PCMs-850-4, with the linear equation described as follows:herein, Q_m (mg g⁻¹) is the maximum adsorption capacity of Langmuir monolayer adsorption. K_L (L mg⁻¹) is the Langmuir constant. The parameters of Langmuir together with coefficients are recorded in Table 2. Obviously, the data obtained from experiment are well fitted with the Langmuir model (R² > 0.99), indicating that the adsorption process of TC onto PCMs-850-4 occurs upon a homogeneous surface, on which the TC is distributed as a monolayer. The linear and non-linear fitting of Langmuir at different temperatures are displayed in Fig. 8. Langmuir model could be better fitted to adsorption isotherm data via linear fitting (Fig. 8b). Non-linear fitting suggested that Langmuir fitting values were more consistent with the experimental values. Furthermore, the adsorption capacity (Q_e,exp) of TC onto PCMs-850-4 in aqueous solutions was 1437.76 mg g⁻¹ at 298 K. Here, the obtained carbon in this work showed a higher adsorption amount of TC than that of other reported adsorbents^25,44–48 (listed in Table 5).

Table 2 Langmuir isotherm model parameters for TC

Langmuir isotherm model
T (K)	Qe,exp (mg g^−1)	KL (L mg^−1)	Qe,cal (mg g^−1)	R^2
298	1437.76	0.502	1497.01	0.9987
308	1575.88	0.843	1607.72	0.9990
318	1644.82	0.893	1709.4	0.9987

---

Fig. 8 (a) The non-linear fitting and (b) linear fitting curves of TC adsorption onto the PCMs-850-4 by Langmuir model at three different temperatures.

Table 3 Adsorption thermodynamics parameters for TC

T (K)	K0	ΔG^θ (kJ mol^−1)	ΔH^θ (kJ mol^−1)	ΔS^θ (kJ mol^−1)
298	7.2635	−4.913	1.631	0.022
308	7.5468	−5.175
318	7.5733	−5.353

---

Table 4 Adsorption kinetics parameters for TC

C0 (mg L^−1)	Qe,exp (mg g^−1)	Pseudo-first-order model			Pseudo-second-order model
		Qe,cal (mg g^−1)	K1 × 10^−3 (min)^−1	R^2	Qe,cal (mg g^−1)	K2 × 10^−5 (g mg^−1 min^−1)	R^2
100	495.83	14.49	52.5	0.9570	496.03	2.229	0.9999
150	740.73	51.89	55.7	0.9816	741.84	9.296	0.9999
200	982.65	181.16	56.7	0.9921	988.14	62.61	0.9999

---

Table 5 Adsorption capacity for TC of PCMs-850-4 compared with other adsorbents

Adsorbents	Qm (mg g^−1)	Ref.
NaOH-activated carbon	455.33 (298 K)	25
Clay	800 (298 K)	44
Smectite	462 (298 K)	45
C2	672.0 (298 K)	46
ITAC-Fe	769.23 (298 K)	47
GO	313 (298 K)	48
PCMs-850-4	1437.76 (298 K)	Present work
	1575.88 (308 K)
	1644.82 (318 K)

---

In order to estimate the thermodynamic parameters, experiments measuring TC adsorption performance onto PCMs-850-4 were performed at different temperatures. From Fig. 8a, we could easily find that the adsorption capacity increased from 1437.76 to 1644.82 mg g⁻¹ when temperature was elevated from 298 to 318 K. The change of Gibbs free energy (ΔG^θ), standard enthalpy (ΔH^θ) and entropy (ΔS^θ) were calculated using the following formulas:

ΔG^θ = −RT ln(K₀) (4)

ΔG^θ = ΔH^θ − TΔS^θ (5)

K₀ is defined as follows:

where R (8.314 J mol⁻¹ K⁻¹) is the universal gas constant, and T (K) is the temperature of the solution. a_s is the activity of adsorbed TC, and a_e is the activity of TC in solution at equilibrium. Q_e (mg g⁻¹) is the amount of TC adsorbed per unit mass of PCMs-850-4; vs. is the activity coefficient of the adsorbed TC; and v_e is the activity coefficient of TC in solution. With TC concentration decreasing to zero gradually in the solution, as shown in Fig. 9a, K₀ is obtained by fitting ln(Q_e/C_e) vs. Q_e from the straight line intercept with the vertical axis. ΔG^θ was plotted against T to calculate ΔH^θ and ΔS^θ from the slope and intercept,⁴⁹ and the results are shown in Fig. 9b. The values of K₀, ΔH^θ, ΔS^θ and ΔG^θ at different temperatures are given in Table 3. The positive value obtained for the standard enthalpy change of 1.631 kJ mol⁻¹ indicates the adsorption process was endothermic, which is consistent with the results of batch experiments at different temperatures. The positive standard entropy change indicates increased randomness at the solid–liquid interface during the adsorption, and the negative value of Gibbs free energy confirmed that the adsorption of TC onto PCMs-850-4 was spontaneous at the studied temperatures.⁵⁰

Fig. 9 (a) Fitting ln(Q_e/C_e) vs. Q_e and (b) ΔG^θ plotted against T.

Adsorption kinetics. The adsorption kinetics at the different concentrations is illustrated in Fig. 10a–c. Adsorption rapidly increased within 5 min; subsequently, the adsorption rate declined and reached an equilibrium within 90 min. The adsorption rate was an important parameter to understand the adsorption dynamics. To investigate the adsorption kinetics of TC onto PCMs-850-4, pseudo-first-order and pseudo-second-order kinetic models were used to fit the experimental data.⁵¹

Fig. 10 The adsorption kinetics (a–c) and the linear-fitting kinetics curves for TC adsorption onto PCMs-850-4 by the pseudo-first-order (d) and pseudo-second-order (e) rate model with different initial TC concentrations (C₀ = 100, 150 and 200 mg g⁻¹) at 298 K.

The linear expression of pseudo-first-order kinetic model is given as:

ln(Q_e − Q_t) = ln Q_e − K₁t (7)

where K₁ (min⁻¹) is the rate constant of the pseudo-first-order equation. Relevant data for adsorption kinetics are shown in Fig 10d and e, and kinetic parameters are recorded in Table 4. Obviously, linearity of the pseudo-first-order mechanism is acceptable but not ideal due to the defective correlation coefficient (R² < 0.999). Thus, the pseudo-first-order kinetics was not enough to explain the rate processes.

The experimental data were further interpreted by pseudo-second-order kinetics model. The linear form could be expressed as follows:

here, K₂ (g mg⁻¹ min⁻¹) is the rate constant of the pseudo-second-order. The parameters are presented in Table 4. From Fig. 10e, the pseudo-second-order model could describe adsorption behavior over the entire range of adsorption (R² > 0.999), suggesting that the adsorption rate of TC onto PCMs-850-4 is mainly controlled by the chemisorption.⁵²

The kinetic results were further analyzed using the intra-particle diffusion model to study the adsorption process.⁵³ The intra-particle diffusion model equation is written as:

Q_t = K_i^1/2 + C (9)

where K_i [mg g⁻¹ min^−1/2] and C are the intra-particle diffusion coefficient and intercept for the intra-particle diffusion model, respectively. The plots of analysis results obtained for Q_t versus t^1/2 are shown in Fig. S4,† and the values of K_i, C_i, and R² at different concentrations of TC are listed in Table S3.† According to the intra-particle diffusion model, the plot of Q_t versus t^1/2 is a straight line through the origin, indicating the rate limiting step is intra-particle diffusion alone. However, the plots are multilinear, demonstrating two or more rate-limiting steps involved in the sorption process. As shown in Fig. S4,† the plots displayed three straight lines, which fit well to the intra-particle diffusion model (R² > 0.91). The first straight portion expresses the diffusion of TC from the bulk solution to the solid PCMs-850-4; the second linear portion indicates the intra-particle diffusion of TC into the mesopores and macropores of PCMs-850-4; the third linear portion with smaller K_i value represents the very slow rate of adsorption, indicating equilibrium adsorption. Piecewise-linear pattern findings indicate that the adsorption of TC onto PCMs-850-4 involves complex mechanisms.

3.2. Effect of pH

The adsorption of TC onto the PCMs-850-4 was investigated by conducting the experiments with solution pH value ranging from 2 to 6 (Fig. 11a). The highest adsorption capacity was found at pH = 3, and it decreased from pH 3–6. The solution pH is an important parameter impacting the adsorption process owing to the change in structure of TC molecules and in the surface charge of the adsorbent.⁵⁴ According to the fractions of TC (Fig. 11b), before the pH of solution reached the pK_a1 (3.30) value of TC, the main species was TC⁺, but the dominant species was TC⁰ when solution pH was between pK_a1 to pK_a2 (3.30–7.70), while the surface of PCMs-850-4 was rich in oxygen-containing functional groups, which mainly existed in the negatively charged form. As a rule, with the increasing pH, the TC⁺ decreases and the adsorption capacity of PCMs-850-4 should decrease. From Fig. 11a, we find the highest adsorption capacity at pH = 3; the adsorption capacity did not decrease significantly and maintained a considerable amount, implying that not only electrostatic interaction was involved in the adsorption process.

Fig. 11 (a) Influence of pH on TC adsorption onto PCMs-850-4, (b) fractions of TC at different pH values.

3.3. Effect of ionic strength

Ionic strength is also an impact parameter for adsorption capacity. Herein, different amounts of NaCl were added to the TC solutions to study the effect of ionic strength on the adsorption capacity.⁵⁵ As depicted in Fig. 12a, increasing NaCl concentration from 0.05 to 1.0 M led to the decrease of adsorption capacity. Competitive effects among Na⁺, Cl⁻ and the TC⁺ with the functional surface of the PCMs-850-4 might explain this result. The shielding effects of ions for TC⁺ molecules were enhanced in the higher NaCl concentration, resulting in the decrease of adsorption amount. Thus, the adsorption process partially involves electrostatic interaction.

Fig. 12 (a) Effect of ionic strength on TC adsorption onto PCMs-850-4, (b) reusability of PCMs-850-4.

3.4. Reusability of PCMs-850-4

The regeneration of adsorbents is crucial to assess the feasibility of practical and large-scale application. The regeneration test on the PCMs-850-4 was carried out by immersing the saturated PCMs-850-4 into NaOH (0.2 M) solution for 12 h at 318 K in a constant temperature vibrator. The adsorption–desorption test was conducted continually five times, and the results are depicted in Fig. 12b. The regenerated PCMs-850-4 displayed good reusability, and adsorption capacities decreased slightly, but still maintained the high adsorption capacity of 901.69 mg g⁻¹ after five cycles. The PCMs-850-4 had excellent reusability for the removal of TC. The potential of the PCMs-850-4 in large-scale application of antibiotics removal will be further investigated in our following study.

3.5. Proposed mechanism

In general, the adsorption mechanism was affected by many factors, including the physicochemical properties of PCMs-850-4 (e.g. surface functional groups and pore size) and the mass-transfer process in solution. The graphite surface of PCMs-850-4 possesses a strong van der Waals force,⁵⁶ which was demonstrated by the results of XRD and Raman spectroscopy. The strong van der Waals force could happen easily between TC and the graphite surface of PCMs-850-4. Because TC is a planar ring molecule, hence, TC could be easily adsorbed onto PCMs-850-4 via the π–π stacking interaction between the aromatic structure of TC and the skeleton of PCMs-850-4. However, in addition to van der Waals force and π–π stacking, other interactions could contribute to the adsorption. It has been proven that the main species are TC⁺ and TC⁰ in acidic TC solution; thus, TC⁺ could be easily adsorbed onto the positively charged surface of PCMs-850-4, suggesting that electrostatic attraction partially contributes to TC adsorption onto PCMs-850-4. More importantly, there are many functional groups in the surface of PCMs-850-4, such as –OH, C=O, as demonstrated by the results of FT-IR and XPS spectroscopy; therefore, hydrogen binding and chemical reaction may occur. Remarkably, ultrahigh BET surface area provided a large number of sites for the abovementioned interaction of TC adsorbed onto PCMs-850-4. In conclusion, TC adsorption onto PCMs-850-4 is a complicated process, involving physical and chemical processes. More detailed adsorption mechanisms need further in-depth study.

4. Conclusions

In summary, we report the synthesis of PCMs via a two-step method, namely carbonization and alkali activation. Owing to their large surface area, highly porous structure, and rich oxygen-containing functional groups, the as-prepared PCMs-850-4 exhibited the high adsorption capacity (Q_m) of 1437.76 mg g⁻¹ at 298 K. Batch adsorption experiments indicated that the PCMs-850-4 were effective and rapid in removing TC. Moreover, the PCMs-850-4 displayed excellent reusability after extended regeneration cycles, demonstrating that the adsorbent could be used long-term, and large-scale application is practical. Considering the facile synthetic process, accessible raw materials, the high adsorption capacity, ultrafast adsorption rate, and good reusability, the PCMs have great promise as highly efficient and environmental friendly adsorbents in the removal of antibiotics and other organic contaminants.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 21277063 and 21407058, 21446015, 21546013, U1510126), the National Basic Research Program of China (973 Program, 2012CB821500), Natural Science Foundation of Jiangsu Province (BK20140534), Research Fund for the Doctoral Program of Higher Education of China (20133227110022 and 20133227110010) and Jiangsu Planned Projects for Postdoctoral Research Funds (1102119C).

References

1. M. B. Ahmed, J. L. Zhou, H. H. Ngo and W. Guo, Sci. Total Environ., 2015, 532, 112–126.
2. X. D. Zhu, Y. J. Wang, R. J. Sun and D. M. Zhou, Chemosphere, 2013, 92, 925–932.
3. Z. Zhang, H. Lan, H. Liu, H. Li and J. Qu, RSC Adv., 2015, 5, 42407–42413.
4. L. Ji, W. Chen, L. Duan and D. Zhu, Environ. Sci. Technol., 2009, 43, 2322–2327.
5. Y. J. Wang, D. A. Jia, R. J. Sun, B. J. Zhu and D. M. Zhou, Environ. Sci. Technol., 2008, 42, 3254–3259.
6. C. Li, Z. Zhuang, H. Feng, Z. Wu, Y. Hong and L. Zhang, ACS Appl. Mater. Interfaces, 2013, 5, 9719–9725.
7. Z. Zhang, J. Hao, W. Yang, B. Lu, X. Ke, B. Zhang and J. Tang, ACS Appl. Mater. Interfaces, 2013, 5, 3809–3815.
8. B. E. Taştan, S. E. Karatay and G. Dönmez, Water Sci. Technol., 2012, 66(10), 2177–2184.
9. J. Dai, J. Pan, L. Xu, X. Li, Z. Zhou, R. Zhang and Y. Yan, J. Hazard. Mater., 2012, 205–206, 179–188.
10. B. Li, Y. Dong, C. Zou and Y. Xu, Ind. Eng. Chem. Res., 2014, 53, 4199–4206.
11. X. Wei, K. Xin, S. Wang, X. Hai, J. Wang and J. Chen, Ind. Eng. Chem. Res., 2013, 52, 17583–17590.
12. D. J. D. Ridder, L. Villacorte, A. R. D. Verliefde, J. Q. J. C. Verberk, S. G. J. Heijman, G. L. Amy and J. C. V. Dijk, Water Res., 2010, 44, 3077–3086.
13. G. K. Ramesha, A. V. Kumara, H. B. Muralidhara and S. Sampath, J. Colloid Interface Sci., 2011, 361, 270–277.
14. G. Magnacca, A. Allera, E. Montoneri, L. Celi, D. E. Benito, L. G. Gagliardi, M. C. Gonzalez, D. O. Mártire and L. Carlos, ACS Sustainable Chem. Eng., 2014, 2, 1518–1524.
15. S. Nayab, A. Farrukh, Z. Oluz, E. Tuncel, S. R. Tariq, R. H. Ur, K. Kirchhoff, H. Duran and B. Yameen, ACS Appl. Mater. Interfaces, 2014, 6, 4408–4417.
16. Y. Kuwahara, S. Tamagawa, T. Fujitani and H. Yamashita, J. Mater. Chem. A, 2013, 1, 7199–7210.
17. B. Wang, W. Chen, H. Fu, X. Qu, S. Zheng, Z. Xu and D. Zhu, Carbon, 2014, 79, 203–212.
18. M. A. Yahya, Z. Al-Qodah and C. W. Z. Ngah, Renewable Sustainable Energy Rev., 2015, 46, 218–235.
19. Y. Q. Li, Y. A. Samad, K. Polychronopoulou, S. M. Alhassan and K. Liao, ACS Sustainable Chem. Eng., 2014, 2, 1492–1497.
20. C. Wu, L. Wang, P. T. Williams, J. Shi and J. Huang, Appl. Catal., B, 2011, 108, 6–13.
21. M. M. Titirici, R. J. White, C. Falco and M. Sevilla, Energy Environ. Sci., 2012, 5, 6796–6822.
22. R. Zhou, G. Hu, R. Yu, C. Pan and Z. L. Wang, Nano Energy, 2015, 588–596.
23. S. Hu and Y. L. Hsieh, ACS Sustainable Chem. Eng., 2014, 2, 726–734.
24. T. A. Hassan, V. K. Rangari and S. Jeelani, ACS Sustainable Chem. Eng., 2014, 2, 706–717.
25. A. C. Martins, O. Pezoti, A. L. Cazetta, K. C. Bedin, D. A. S. Yamazaki, G. F. G. Bandoch, T. Asefa, J. V. Visentainer and V. C. Almeida, Chem. Eng. J., 2015, 260, 291–299.
26. T. Maneerung, J. Liew, Y. Dai, S. Kawi, C. Chong and C.-H. Wang, Bioresour. Technol., 2016, 200, 350–359.
27. S. Altenor, B. Carene, E. Emmanuel, J. Lambert, J. J. Ehrhardt and S. Gaspard, J. Hazard. Mater., 2009, 165, 1029–1039.
28. Y. Young Soo, C. Se Youn, S. Jinyong, K. Byung Hoon, C. Sung-Jin, B. Seung Jae, H. Y. Suk, T. Yongsug, P. Yung Woo and P. Sungjin, Adv. Mater., 2013, 25, 1993–1998.
29. Y. Zhang, X. Ma and R. Yong, Environ. Pollut., 2014, 185, 213–218.
30. R. H. Bragg and M. L. Hammond, Carbon, 1965, 3, 340.
31. B. E. Warren and P. Bodenstein, Acta Crystallogr., 1965, 18, 282–286.
32. R. Diduszko, A. Swiatkowski and B. J. Trznadel, Carbon, 2000, 38, 1153–1162.
33. A. C. Ferrari, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 14095–14107.
34. E. Raymundo-Piñero, P. Azaïs, T. Cacciaguerra, D. Cazorla-Amorós, A. Linares-Solano and F. Béguin, Carbon, 2005, 43, 786–795.
35. C. Hinnen, D. Imbert, J. M. Siffre and P. Marcus, Appl. Surf. Sci., 1994, 78, 219–231.
36. T. R. Gengenbach, R. C. Chatelier and H. J. Griesser, Surf. Interface Anal., 1996, 24, 271–281.
37. L. T. Weng, C. Poleunis, P. Bertrand, V. Carlier, M. Sclavons, P. Franquinet and R. Legras, J. Adhes. Sci. Technol., 1995, 9, 859–871.
38. M. J. Goldberg, J. G. Clabes and C. A. Kovac, J. Vac. Sci. Technol., A, 1988, 6, 991–996.
39. C. Jones and E. Sammann, Carbon, 1990, 28, 509–514.
40. C. Zaiming, C. Baoliang and C. T. Chiou, Environ. Sci. Technol., 2012, 46, 11104–11111.
41. I. A. W. Tan, A. L. Ahmad and B. H. Hameed, J. Hazard. Mater., 2008, 154, 337–346.
42. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361–1403.
43. H. M. F. Freundlich, J. Phys. Chem., 1906, 57, 385–471.
44. K. Pankaj, R. F. Giese and D. S. Aga, Environ. Sci. Technol., 2004, 38, 4097–4105.
45. Z. Li, P. H. Chang, J. S. Jean, W. T. Jiang and C. J. Wang, J. Colloid Interface Sci., 2010, 341, 311–319.
46. J. Rivera Utrilla, C. V. Gómez Pacheco, M. Sánchez Polo and R. Ocampo Pérez, J. Environ. Manage., 2013, 131, 16–24.
47. G. Li, D. Zhang, M. Wang, J. Huang and L. Huang, Ecotoxicol. Environ. Saf., 2013, 98, 273–282.
48. Y. Gao, Y. Li, L. Zhang, H. Huang, J. Hu, S. M. Shah and X. Su, J. Colloid Interface Sci., 2012, 368, 540–546.
49. J. Shah, M. R. Jan, A. U. Haq and M. Zeeshan, J. Saudi Chem. Soc., 2012, 19, 301–310.
50. H. T. Fan, X. Fan, J. Li, M. Guo, D. Zhang, F. Yan and T. Sun, Ind. Eng. Chem. Res., 2012, 51, 5216–5223.
51. Y. S. Ho, J. C. Y. Ng and G. Mckay, Sep. Purif. Rev., 2011, 29, 189–232.
52. Y. Ge, X. Cui, D. Xiao and Z. Li, J. Mater. Chem., 2014, 2, 2136–2145.
53. W. J. Weber and J. C. Morris, J. Sanit. Eng. Div., 1963, 89, 31–60.
54. G. Cheng and K. G. Karthikeyan, Environ. Sci. Technol., 2005, 39, 2660–2667.
55. J. Ma, F. Yu, L. Zhou, L. Jin, M. Yang, J. Luan, Y. Tang, H. Fan, Z. Yuan and J. Chen, ACS Appl. Mater. Interfaces, 2012, 4, 5749–5760.
56. L. Ji, W. Chen, L. Duan and D. Zhu, Environ. Sci. Technol., 2009, 43, 2322–2327.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra24707a

---