Abstract

Preparation of sludge-derived mesoporous carbon materials (SMCs) through pyrolysis of excess activated sludge from urban municipal sewage plants is an effective means of reducing pollution and utilizing a waste resource. This paper presented a method of SMC preparation in which calcium oxide (CaO), polyacrylamide (PAM), and chitosan (CAS) flocculating agents were used as pore-forming additives. Physical and chemical characterizations of the prepared SMCs were conducted by scanning electron microscopy (SEM), Brunauer–Emmett–Teller (BET), Fourier transform infrared (FTIR), and X-ray photoelectron spectroscopy (XPS). The prepared SMCs were used to adsorb a tetracycline (TC) antibiotic pollutant. The influences of pH, adsorption time, temperature, and pollutant concentration on TC adsorption capacity were determined. The experiments demonstrated that weakly acidic conditions were conducive to TC adsorption, which mainly occurs via electrostatic and π-π interactions. The TC adsorption process by SMCs conformed better to the pseudo-second-order models, indicating that chemical adsorption was the dominant adsorption process. The isothermal adsorption of TC by the SMCs conformed to the Freundlich model. This implied that TC easily adhered onto the SMC surfaces via multilayer homogeneous adsorption. Thermodynamic studies revealed that the adsorption of TC onto SMCs was spontaneous and endothermic.

INTRODUCTION

Activated sludge is a semi-solid material produced after sewage is subjected to a biological treatment process. Recently, amounts of activated sludge have increased alongside growth in urban sewage treatment capacity. The disposal of activated sludge has implications for water, soil, and atmospheric pollution and is, therefore, of great concern to researchers (Hadi et al. 2015). However, sludge processing is always overlooked in water treatment processes, generally resulting in low sludge treatment rates. Without safe and proper processing, sludge can cause ‘secondary pollution’ to the surrounding environment. Currently, widely used sludge disposal methods include drying/incineration, microbial fermentation, sanitary landfilling and land utilization. Nevertheless, most of these methods are less than ideal due to their high costs and risks of environmental pollution (Xu et al. 2015). Compared to other disposal techniques, sewage sludge pyrolysis has the following advantages. (1) The treated pollutants are harmless and the organic components of the sludge are decomposed into micromolecular substances and CO2 under high-temperature anoxic conditions (Fernández et al. 2014). Meanwhile, heavy metal elements, Si, and O in sludge can form inorganic compounds which are difficult to effuse. Hence, pyrolysis can release nutrients and polluting heavy metals from sludge (Sánchez et al. 2009). (2) Sludge pyrolysis has high resource utilization – the charcoal gained from pyrolysis is a porous material with high adsorption capacity and can be used in products such as water treatment agents and soil conditioners (Jaria et al. 2016). Sludge-derived activated carbon is often applied to remove organic pollutants and heavy metals from water due to the excessive amount of activated sludge, high production as well as low matrix costs.

According to the provisions of International Union of Pure and Applied Chemistry (IUPAC), porous materials with a pore size distribution of 2–50 nm are called mesoporous materials (Khalili et al. 2000). Although sludge-derived mesoporous carbon materials (SMCs) have a smaller specific surface area than commercial activated carbon, SMCs show equivalent capacities to adsorb methylene blue, rhodamine B, phenol and antibiotics (Ding et al. 2012; Li et al. 2013; Zaini et al. 2013; Zou et al. 2013; Takdastan et al. 2016). Generally, SMCs prepared by the activation method has higher adsorption capacity than that prepared by the direct carbonization approach. The former related to the development of porous structures and abundant surface acidity groups (Liu et al. 2010; Pan et al. 2011), and the latter often loses some oxygen-containing functional groups on the surface, especially carbonyl groups. Additionally, basic groups or alkali metals on the surface of sludge-derived activated carbon facilitate reactions with the acidic functional groups of pollutants. As a result, complex or metal salts are formed and deposited in the pores. In addition to the chemical properties of their surfaces, the pore structure and polarity of SMCs also have important impacts on pollutant adsorption (Nielsen et al. 2015). For example, a pore diameter similar to the size of adsorbate molecules is conducive to adsorption and separation. A hydrophilic porous structure enhances the scattering of polar inorganic mineral phases, and the corresponding polar surface is suitable for the adsorption of polar molecules. Kacan (2016) carried out activation and pyrolysis of textile sludge by using KOH to prepare sludge-derived activated carbon material. The adsorption capacities of the dyes Synozol Blue reactive (RSB) and Setapers Yellow-Brown (P2RFL) reached 8.54 mg/g and 5.4 mg/g, respectively. The main forces acting between sludge-derived activated carbon and dye molecules is electrostatic attraction and π-π bonding. Gupta & Garg (2015) prepared sludge-derived activated carbon by ZnCl2 activation to adsorb lignin and amoxicillin. They found that the interactions between the adsorbent and adsorbate mainly included electrostatic interaction, complexation between electron donors and electron acceptors, and hydrogen bonding. Acids, alkalis, and salts are commonly used agents in chemical activation, including ZnCl2, KOH, and H3PO4, etc. These agents cause catalytic dehydration condensation with oxygen-containing functional groups, thus leaving more carbon skeletons in the material after aromatization. Hence, porous activated carbon with a high specific surface area is obtained (Wen et al. 2011; Yang et al. 2014; Mohammadi & Mirghaffari 2015). However, chemical activation methods have many disadvantages, such as high energy consumption, equipment corrosion, complicated post-processing techniques, and environmental pollution (Pietrzak et al. 2014).

Excess activated sludge from sewage treatment plants generally contains a certain amount of flocculating agents and stabilizer components. Since flocculating agents are generally macromolecular polymers with high molecular volumes, they can provide large molecular cavities after pyrolysis, which is conducive to pore formation. On the other hand, flocculating agents make flocs with sludge particles in the water phase. The adsorption bridging effect facilitates the formation of flocculating agent-sludge flocs. Inspired by this, two kinds of flocculating agents, polyacrylamide (PAM), chitosan (CAS) and a sludge stabilizing agent (CaO) were used as pore-forming activators in this research to prepare SMCs with a highly porous structure. The prepared SMCs were used to adsorb an emerging pollutant – that of tetracycline (TC) antibiotics. The influences of pH, adsorption time and pollutant concentration on adsorption performance are discussed. The adsorption performance and mechanism by which SMCs adsorb antibodies are also revealed.

MATERIALS AND METHODS

Preparation of SMCs

In this work, SMCs were prepared according to the previous researches (Kacan 2016; Xin et al. 2016). A certain mass of pre-screened sludge powder was added in water and stirred continuously until an emulsified sludge mixture was formed. Next, the additives PAM, CAS, and CaO were added to the sludge at a proportion of 1% by mass and stirred to achieve dissolution. The emulsified solution was poured into a culture dish and then put into a drying oven overnight. The dried sludge products were put into a high-temperature tube furnace and the N2 was controlled at 100 mL/min. After the air in the furnace was exhausted completely, the temperature was increased at 15 °C/min. Pyrolysis was conducted for 1 h after the temperature reached 900 °C to eliminate the additives. After combustion and cooling to room temperature, the solid residues in the furnace were collected, ground and screened by 40-mesh film (diameter <0.5 mm). The original dried sludge powder was labeled as SS, while the adsorption materials added with CaO, CAS and PAM were labeled MC-CaO, MC-CAS, and MC-PAM, respectively.

Characterization of SMCs

The surface morphologies of SMCs were examined by scanning electron microscopy (SEM) (QUANTA 450, FEI). The specific surface areas and pore diameter distributions of the SMCs were analyzed by a 3H-2000PS2 specific surface area and pore diameter analyzer (Bayside Company). The liquid nitrogen low-temperature adsorption–desorption method was applied. The specific surface area was calculated according to the Brunauer–Emmett–Teller (BET) model. Pore distribution was calculated by the Barrett–Joyner–Halenda (BJH) model. Fourier transform infrared spectroscopy (FTIR) was used to examine the chemical compositions of adsorbent surface before and after combustion. The samples were completely dried at 353 K using a vacuum oven for 12 h to remove the absorbed water before tests, and then evenly mixed with potassium bromide and made into semitransparent slices. The slices were then put on the analysis bench of EQUINOX55 FT-IR (Bruker, Germany) to obtain the infrared spectrogram over a range of 500 cm−1 to 4,000 cm−1. X-ray photoelectron spectroscopy (XPS) analysis (ESCALAB MK II) was used to characterize the chemical composition of adsorption material surface. The radiation source used was Al Kα (hv = 1,486.6 eV) with an energy spectral range of 0–1,201 eV. Scanning results were used after peak fitting. The peak width at half the height of the Gaussian peaks of all characteristic spectral elements remained stable.

Adsorption experiment of SMCs

The adsorption characteristics of SMCs were tested by static adsorption experiments. The initial concentration of TC solution was set at 5–50 mg·L−1. Appropriate concentrations of HCl and NaOH were used to adjust TC solutions to different initial pHs of 3.0 and 10.0. The background solution used was 0.02 mol·L−1 NaCl. Next, 15 mg quantities of adsorbent were added into brown reaction flasks. They were mixed evenly and put in a constant-temperature water bath oscillator (25–45 °C). Various contact times (20 min–24 h) were set and the equilibrium concentration was tested by an ultraviolet spectrophotometer (U-3900, Hitachi) at 357 nm after the supernatant was filtered through a 0.45 μm membrane. Each experimental data was obtained by the average values of triple parallel samples and ∼5% relative errors were provided in some measurements.

A group of static adsorption experiments were carried out. Adsorption quantity Qe (mg·g−1) was used as parameter that reflected the performance of the adsorbent. A calculation formula of Qe was:  
formula
(1)
where C0 and Ce were the initial TC concentration and the TC concentration at adsorption equilibrium (mg·L−1), V was solution volume (L), and m was mass of the adsorbent (g).

The adsorption process of TC on SMCs was simulated by the pseudo-first-order kinetic model, and pseudo-second-order kinetic model, and intra-particle diffusion model. Different models could be expressed by the following equations (Zhu et al. 2015).

Pseudo-first-order kinetic model:  
formula
(2)
Pseudo-second-order kinetic model:  
formula
(3)
Intra-particle diffusion:  
formula
(4)
where Qe and Qt were equilibrium adsorption capacity and the adsorption capacity at time t (mg·g−1), k1 was the pseudo-first-order rate constant (min−1), k2 was the pseudo-second-order rate constant (g·(mg·min)−1), k3 was the intra-particle diffusion constant (mg·g−1·min−1/2), and C was diffusion constant.

The Langmuir and Freundlich isothermal models were applied for fitting the adsorption experimental data. The Langmuir model is an ideal model, as it hypothesizes that the solution is the ideal solution and the surface is the monolayer adsorption process. The Freundlich model is an empirical model and is generally used to describe that the surface adsorption is multilayer adsorption. The linear isothermal adsorption equations of two models were as follows (Babaei et al. 2016).

Langmuir equation:  
formula
(5)
Freundlich equation:  
formula
(6)
where Ce was the TC concentration in solution at adsorption equilibrium (mg·L−1), Qm was the maximum adsorption capacity (mg·g−1), b was the Langmuir constant (L·mg−1), and Kf and n were the Freundlich constants.
The essential characteristics of the Langmuir isotherm could be described in terms of dimensionless constant separation factor or equilibrium parameter, RL which was defined by Equation (7) (Hadi et al. 2010):  
formula
(7)
where b and C0 (mg·L−1) were the Langmuir constant and the highest initial TC concentration, respectively. Value of RL reflects the difficulties of heterogeneous adsorption process. RL = 1 means linear adsorption process, 0< RL< 1 reflects easy implementation of the adsorption process, and RL > 1 reflects difficult implementation.
Adsorption thermodynamics generally involve different parameters, including enthalpy change (ΔH), entropy change (ΔS), Gibbs free energy (ΔG) and solid–liquid distribution coefficient (Kd). Calculation equations of these parameters were:  
formula
(8)
 
formula
(9)
 
formula
(10)
where Kd was the solid–liquid distribution coefficient at equilibrium, ΔS was the entropy of reaction (J·mol−1·K−1), ΔH was the enthalpy of reaction (kJ·mol−1), R was gas constant (R = 8.314 J·mol−1·K−1), T was absolute temperature (K), and ΔG was the free energy of reaction (kJ·mol−1). Values of Kd could be calculated from the intercept on the ordinate of the plot ln (Qe/Ce) against Qe (Yamaguchi et al. 2016). After that another linear data fitting was performed by plotting lnKd as a function of 1/T. ΔH was determined by slope of straight line and ΔS was determined by intercept. Finally, ΔG under corresponding temperature was calculated according to Equation (10).

RESULTS AND DISCUSSION

Surface and pore structure morphologies

Microstructural observation is an effective way to determine the surface characteristics of materials. SEM images of SS and SMCs are shown in Figure 1. In SS, particles were scattered around, and the SS had a smooth surface with dense structures. Moreover, non-obvious microstructures were observed (Figure 1(b)). Compared to SS, the MC-CaO, MC-CAS, and MC-PAM samples prepared by high-temperature (900 °C) carbonization presented greater surface roughness. It was found that there were many irregular cylindrical fractures or pores in these structures, and their quantity was significantly higher than that in SS. Hence, MC-CaO, MC-CAS, and MC-PAM were the more porous carbon adsorption materials. It can be seen from the SEM images that the SMCs prepared from these three substances had similar surfaces. The MC-CaO sample had the highest surface roughness. It was dominated by an irregular particulate structure accompanied by a great growth of pits and pores. Meanwhile, the pore structure became more developed and complicated, and numerous porous structures developed. There were white particulate substances on the surface, which might be residual calcium salts. The MC-CAS and MC-PAM samples had similar surface morphologies, which were covered by uneven microscopic particles, accompanied with deep ‘karst cave’ structures and diversified macroporous structures (Figure 1(f) and 1(h)). The porous structures of SMCs might be attributed to the elimination of additives as well as the release of volatiles from the sludge. The rich porous structure enhanced the surface structural performance and the adsorption capacity of the material. Flocculating agents were used as additives as the ‘assemble’ stable flocculating structures with sludge through the adsorption bridging effect. After that the obvious difference in the thermodynamics of flocculating molecules and sludge resulted in a pore-forming effect (Górka et al. 2009).

Figure 1

SEM images of SS (a, b), MC-CaO (c, d), MC-CAS (e, f), and MC-PAM (g, h) samples (from top to bottom, respectively), at magnifications of × 2,000 (left column) and × 50,000 (right column).

Figure 1

SEM images of SS (a, b), MC-CaO (c, d), MC-CAS (e, f), and MC-PAM (g, h) samples (from top to bottom, respectively), at magnifications of × 2,000 (left column) and × 50,000 (right column).

Information on the internal structure of materials is listed in Table 1. By introducing the flocculating agent, the prepared SMCs gained relatively high specific surface areas (111.76–131.55 m2/g) and pore volumes (0.143–0.175 cm3/g). The MC-CAS sample acquired the highest specific surface area and pore volume of the three SMCs. This might be related to the distribution and nucleation of templates in the flocculating agents. The MC-CAS sample exhibited the highest mesoporous specific surface area and mesoporous volume, reaching 118.44 m2/g and 0.17 cm3/g, respectively. The average pore diameter of the SMCs ranged from 4.89 to 5.79 nm with mesopores predominating. Such rich mesoporous structures can effectively enhance the adsorption capacity of carbon materials.

Table 1

Structural feature of the prepared SMCs

Material Specific surface area (m2/g) Microporous specific surface area (m2/g) Total pore volume (cm3/g) Microporous volume (cm3/g) Average pore diameter (nm) Mesoporous specific surface area (m2/g) Mesoporous volume (cm3/g) 
SS 1.25 0.00248 181.21 0.042 0.00248 
MC-CaO 111.76 30.62 0.162 0.014 5.79 81.14 0.148 
MC-PAM 116.42 52.76 0.143 0.024 4.89 63.66 0.119 
MC-CAS 131.55 13.11 0.175 0.005 5.31 118.44 0.17 
Material Specific surface area (m2/g) Microporous specific surface area (m2/g) Total pore volume (cm3/g) Microporous volume (cm3/g) Average pore diameter (nm) Mesoporous specific surface area (m2/g) Mesoporous volume (cm3/g) 
SS 1.25 0.00248 181.21 0.042 0.00248 
MC-CaO 111.76 30.62 0.162 0.014 5.79 81.14 0.148 
MC-PAM 116.42 52.76 0.143 0.024 4.89 63.66 0.119 
MC-CAS 131.55 13.11 0.175 0.005 5.31 118.44 0.17 

Analysis of surface chemical properties

The SS and SMC samples were characterized by FTIR. It can be seen from Figure 2 that the most prominent feature of SS spectrum was the adsorption peak at 3,420 cm−1 which corresponded to the O-H stretching vibration, indicating that sludge might contain water, alcohol and phenol substances (Peng et al. 2016). There were two absorption peaks at 2,920 cm−1 and 2,850 cm−1, which were consistent with the stretching vibration absorption peak of the aliphatic series –CH2, indicating that the sludge contained aliphatic hydrocarbons or alkanes (Lin et al. 2012). There are also two absorption peaks at 1,720 cm−1 and 1,648 cm−1, which are related with C = O. This implies the existence of acid and lipid substances (Cao et al. 2009). The peak at 1,540 cm−1 is the characteristic absorption peak of –CONH, indicating that the original sludge contained amides. This further verifies the existence of proteins in the sludge (Peng et al. 2016). The peak at 1,384 cm−1 is the C = O vibration peak of carboxyl or carbonyl, which implies the existence of carbohydrate substances (Chia et al. 2012). Obvious C-O and C-O-C stretching vibration peaks are observed at 1,084 cm−1 and 1,030 cm−1, which might be related with ether, alcohol, phenol and glycosidic bonds (Kacuráková et al. 2000; Chia et al. 2012). After high-temperature processing, the functional groups on MC-CaO, MC-CAS, and MC-PAM changed significantly. Firstly, the –OH peak was attenuated, mainly due to dehydration and deoxygenation during the high-temperature carbonization process. The disappearance of peaks at 1,720 cm−1, 2,920 cm−1, and 2,850 cm−1 was mainly attributed to the pyrolysis of fatty substances. The attenuation of absorption peaks at 1,540 cm−1 and 1,384 cm−1 demonstrated that proteins and carbohydrate substances in the sludge were also decreased by decomposition reactions. The aromatic hydrogen bond at 780 cm−1 (bending vibration peaks of C-H) was enhanced differently (Zhang et al. 2011). This indicated that the degree of aromatization increased after high-temperature carbonization of the sludge. In brief, the organic substances mainly came from proteins, carbohydrates, oils/fats and other non-digested organic matter within microorganisms. Accordingly, functional groups on the material surfaces mainly included carboxyl, carbonyl, phenolic hydroxyl groups, and acylamino, amino acid, methyl, methylene and aromatic groups.

Figure 2

FTIR spectra of the SS and SMC samples.

Figure 2

FTIR spectra of the SS and SMC samples.

Results of XPS characterization of the SS and SMC samples are shown in Figure S1 (available with the online version of this paper) and Table 2. Obviously, all adsorption material surfaces contained Si, Cl, C, Ca, N, O, and Fe. It could be seen from the peak intensity that MC-CaO showed the strongest absorption peak of Ca 2p, which reflected the abundant calcium salt residues on the material surface. This was in accordance with the results of the SEM analysis. Based on XPS analysis of element contents (Table 2), the C content increased sharply after high-temperature carbonization, which proved that the adsorption materials had a strengthened carbon skeleton. The highest Ca content was detected in the MC-CaO sample, which might be related to the added CaO. The N and O contents dropped sharply, revealing that high-temperature carbonization destroyed the bound water, protein, carbohydrate, and grease in the sludge. However, the carbon material surfaces still contained 15–20% oxygen, which indicated that the material still had a certain amount of oxygen functional groups. These functional groups may improve the adsorption performance of the materials.

Table 2

XPS characterization of the SS and SMC samples

Peak SS MC-CaO MC-CAS MC-PAM 
  PP At. % 
C 1s 57.85 68.32 74.31 72.99 
Ca 2p 1.65 6.37 1.52 1.93 
N 1s 13.08 2.93 3.71 3.49 
O 1s 24.49 19.34 16.53 18.63 
Other 2.93 3.04 3.93 2.96 
Total 100 100 100 100 
Peak SS MC-CaO MC-CAS MC-PAM 
  PP At. % 
C 1s 57.85 68.32 74.31 72.99 
Ca 2p 1.65 6.37 1.52 1.93 
N 1s 13.08 2.93 3.71 3.49 
O 1s 24.49 19.34 16.53 18.63 
Other 2.93 3.04 3.93 2.96 
Total 100 100 100 100 

The XPS technique can make a qualitative description of the types and properties of the functional groups on a material's surface. The C content of the material was explored further. As shown in Figure S2 (available online), the SS surface contained five chemical forms of C. The binding energies for these forms were 284.7 eV (C-H), 285.3 eV (C-N), 286.4 eV (C-O), 287.4 eV (C = O) and 288.5 eV (O-C = O) (Zhao et al. 2013). According to our calculations (Table S1, available online), the C-N group content on the SS surface was 16.30% and the total content of oxygen-containing functional groups was 27.78%. However, the number of carbonyl groups in the SMCs dropped dramatically after high-temperature carbonization. At the same time, the oxygen functional groups were mainly phenolic hydroxyl groups and carboxyl groups. The functional groups on the SMC surfaces were calculated next. It can be seen from Table S1 that the percentage of C-N groups on the material surface was lower than 11%, but the total percentage of oxygen functional groups remained at about 25%. These oxygen functional groups can be used to maintain the hydrophilic performance of the material surface and are conducive to the adsorption of pollutants with polar groups.

Experimental analysis of static adsorption

Effect of pH

A solution's pH will influence the existing state and charge of TC, and thereby affect its adsorption by SMCs. Overall, the adsorption capacities (Qe) of all SMCs were greater than that of SS, especially under acidic conditions (Figure 3). Compared to SS (Qe = 5.50 mg·g−1, pH = 2.5), MC-CaO, MC-CAS and MC-PAM exhibited adsorption capacities that were 462%, 389% and 326% higher, reaching 30.95 mg·g−1, 26.94 mg·g−1 and 23.45 mg·g−1, respectively. Under various solution pHs (2.5–10.5), carbon materials showed different adsorption effects. With increasing pH, the adsorption capacity of SS increased gradually, while that of MC-CaO declined. Moreover, pH changes could slightly influence the adsorption capacities of MC-CAS and MC-PAM. The differences in adsorption capacity could be related to the charge of TC and the potential changes of the SMCs. There are three ionization equilibrium constants in an aqueous solution of TC, which are pKa1 = 3.30, pKa2 = 7.68 and pKa3 = 9.68 (Zhang et al. 2016). Therefore, TC can be decomposed into four different forms under different pHs, namely, a kation (TC+), zwitter-ion (TC±), anion (TC), and dianion (TC2−). At relatively low pH, TC loses electrons completely and presents as TC+. The sludge surface was negatively charged under neutral conditions, due to the existence of anionic functional groups (e.g. carboxyl, hydroxyl, and phosphate) on the SS surface. However, the negative charge declined sharply when the pH was lower than 4.6. The sludge was positively charged at a pH of 3.0 (Zheng et al. 2007). In such conditions, SS had a repulsive interaction with TC+. However, the negative charge of the sludge increased with increasing pH. Therefore, the adsorption improved gradually as the pH increased. After high-temperature pyrolysis, the aromatic ring structures (sp2) of the carbon materials were enhanced, and stable π-π bonds were formed with the TC ring structures (Peiris et al. 2017). Therefore, adsorption of carbon materials was dominated by π-π bonds and Van der Waals' forces, and was slightly influenced by pH. Since MC-CaO contained a lot of CaO residue, Ca2+ ions were formed under acidic conditions, which would facilitate ion exchange and complexation with TC+. Consequently, the adsorption effect was strengthened. As the pH increased to neutral, the content of Ca2+ ions dropped dramatically. Then, the TC adsorption capacity decreased due to the absence of ion exchange and complexation, although there was still π-π interaction with TC.

Figure 3

Effect of pH on the adsorption capacities (Qe) of the materials.

Figure 3

Effect of pH on the adsorption capacities (Qe) of the materials.

Adsorption kinetics

The effect of adsorption time on the adsorption of TC by the SMCs is shown in Figure 4. The SS material reached TC adsorption equilibrium quickly (180 min). In contrast, the adsorption capacities of MC-CaO, MC-CAS, and MC-PAM increased quickly at first and then increased more slowly with time. The carbon materials reached adsorption saturation after 600 min. Meanwhile, the SMCs showed stronger TC adsorption performance compared to SS. This was related to SMCs' more porous structures and specific surface areas. To evaluate the rate control step and adsorption mechanism in the adsorption process, pseudo-first-order and pseudo-second-order models were fitted to the adsorption data. The fitting results are given in Figure S3 (available online) and Table 3, which showed that under these experimental conditions, the TC adsorption processes of the SMCs were best described by pseudo-second-order kinematic models. The correlation coefficients of the SMCs ranged from R2 = 0.9477–0.9757, and the calculated equilibrium adsorption capacities of the pollutants were as close to the experimental results. These results demonstrated that the TC adsorption processes of SS, MC-CaO, MC-CAS, and MC-PAM were controlled by chemical reactions rather than simple physical diffusion (Gao et al. 2011). It could be ascertained from comparison of the pseudo-second-order rate constant (k2) that the pollutant adsorption rate of SS was higher than those of MC-CAS and MC-PAM. This was because the electrostatic interactions between SS and pollutants were quicker than the π-π interaction forces between the carbon material and the TC. The MC-CaO material achieved the highest adsorption rate, which was attributed to the ion exchange and complexation between Ca2+ and pollutants (Parolo et al. 2012). Additionally, the intra-particle diffusion model was used for fitting the adsorption experimental data in order to determine intraparticle diffusion process.

Table 3

Parameters of kinetic models of SMCs adsorbing antibiotics

Material Pseudo-first-order
 
Pseudo-second-order
 
Intra-particle diffusion model
 
Qe (mg·g−1k1 × 10−2 (min−1R2 Qe (mg·g−1k2 × 10−4 (g·mg−1·min−1R2 k3 (mg·g−1·min−1/2C (mg·g−1R2 
SS 15.27 1.34 0.9461 16.09 9.38 0.9749 3.96 −6.34 0.9589 
MC-CaO 24.98 2.14 0.8523 27.81 12.5 0.9477 6.33 −0.82 0.939 
MC-CAS 24.85 0.95 0.9469 28.04 4.25 0.9757 8.67 −12.39 0.9358 
MC-PAM 25.64 0.24 0.9525 28.81 0.58 0.9607 11.51 −19.52 0.9553 
Material Pseudo-first-order
 
Pseudo-second-order
 
Intra-particle diffusion model
 
Qe (mg·g−1k1 × 10−2 (min−1R2 Qe (mg·g−1k2 × 10−4 (g·mg−1·min−1R2 k3 (mg·g−1·min−1/2C (mg·g−1R2 
SS 15.27 1.34 0.9461 16.09 9.38 0.9749 3.96 −6.34 0.9589 
MC-CaO 24.98 2.14 0.8523 27.81 12.5 0.9477 6.33 −0.82 0.939 
MC-CAS 24.85 0.95 0.9469 28.04 4.25 0.9757 8.67 −12.39 0.9358 
MC-PAM 25.64 0.24 0.9525 28.81 0.58 0.9607 11.51 −19.52 0.9553 
Figure 4

Effect of time on the adsorption capacities (Qe) of the materials.

Figure 4

Effect of time on the adsorption capacities (Qe) of the materials.

The diffusion constant C obtained from the intra-particle model indicates if the intraparticle diffusion is a controlling step or not. When C is zero, the adsorption process is controlled by intraparticle diffusion only, while the adsorption mechanism is complex. The values of correlation coefficient were relatively good for the experimental data, indicating that intra-particle diffusion was one of TC adsorption controlling steps of SMCs. However, the diffusion constant C in this experiment presented a non-zero value, indicating that the adsorption process was a complex pathway and the intra-particle diffusion was not the sole rate controlling factor. Membrane diffusion, pore diffusion and surface diffusion, which occurred simultaneously during reaction, might also control reaction rate (Yang et al. 2015).

Adsorption isotherms modeling

The effects of initial pollutant concentration and temperature on TC adsorption by the SMCs are shown in Figure 5. In this experiment, two isothermal adsorption models (Langmuir and Freundlich) were chosen for fitting to the equilibrium adsorption data. The fitting results are exhibited in Table 4. With an increase of the initial concentration, the TC adsorption capacity of the SMCs increased gradually. This was because, given the low pollutant concentration, there were many adsorption sites on the SMC surfaces that adsorbed and combined with the abundant TC molecules. As the TC concentration increased, the adsorption sites on the adsorbent surface reduced accordingly and adsorption saturation was reached quickly. The correlation coefficient (R2) of the Freundlich isothermal adsorption model ranged from 0.9112–0.9847 for the various materials, indicating good fits in all cases. This suggested that the Freundlich isothermal adsorption model was more suitable for fitting the adsorption data in this experiment than the Langmuir model, and TC was mainly adsorbed onto the carbon material surfaces in a multilayer homogeneous manner. Moreover, the theoretical maximum saturation adsorption capacities (Qm) calculated from the Langmuir models showed that the SMCs had higher Qm values than that of SS. This was consistent with the experimental results. Besides, Qm increased continuously with temperature, which demonstrated that temperature increases were conducive to TC adsorption by SMCs. The equilibrium parameter RL ranged between 0 and 1 throughout the adsorption process, indicating that TC was easily adsorbed onto the surfaces of the SMCs. The parameter n was Freundlich constant, indicating the surface heterogeneity of adsorption materials as well as the binding strength of the objective pollutants (Feng et al. 2012). It can be seen from Table 5 that the n values of the SMCs were higher than those of SS, which showed a stronger affinity and binding strength of carbonized materials to TC.

Table 4

Parameters of Langmuir and Freundlich isothermic models

Material Temperature (°C) Langmuir model
 
Freundlich model
 
Qm (mg·g−1b (L·mg−1RL R2 Kf (mg·g−1n R2 
SS 25 33.63 0.051 0.27–0.79 0.9099 2.33 1.35 0.9151 
35 36.44 0.063 0.24–0.76 0.8827 2.37 1.19 0.9281 
45 43.35 0.05 0.28–0.80 0.8466 1.49 0.87 0.9442 
MC-CaO 25 38.56 0.18 0.09–0.52 0.8599 9.85 2.66 0.9614 
35 41.19 0.28 0.06–0.42 0.8722 13.81 3.19 0.9439 
45 60.28 0.12 0.14–0.62 0.9089 9.65 1.93 0.9839 
MC-CAS 25 37.22 0.27 0.06–0.43 0.9046 12.59 3.29 0.9388 
35 47.95 0.46 0.04–0.30 0.9701 16.78 4.01 0.9711 
45 69.24 0.27 0.06–0.42 0.9771 16.86 1.89 0.9847 
MC-PAM 25 35.53 0.25 0.07–0.44 0.8642 11.79 3.31 0.9327 
35 41.64 0.38 0.04–0.34 0.9211 14.40 3.11 0.9112 
45 69.08 0.23 0.08–0.46 0.8754 15.41 1.89 0.9379 
Material Temperature (°C) Langmuir model
 
Freundlich model
 
Qm (mg·g−1b (L·mg−1RL R2 Kf (mg·g−1n R2 
SS 25 33.63 0.051 0.27–0.79 0.9099 2.33 1.35 0.9151 
35 36.44 0.063 0.24–0.76 0.8827 2.37 1.19 0.9281 
45 43.35 0.05 0.28–0.80 0.8466 1.49 0.87 0.9442 
MC-CaO 25 38.56 0.18 0.09–0.52 0.8599 9.85 2.66 0.9614 
35 41.19 0.28 0.06–0.42 0.8722 13.81 3.19 0.9439 
45 60.28 0.12 0.14–0.62 0.9089 9.65 1.93 0.9839 
MC-CAS 25 37.22 0.27 0.06–0.43 0.9046 12.59 3.29 0.9388 
35 47.95 0.46 0.04–0.30 0.9701 16.78 4.01 0.9711 
45 69.24 0.27 0.06–0.42 0.9771 16.86 1.89 0.9847 
MC-PAM 25 35.53 0.25 0.07–0.44 0.8642 11.79 3.31 0.9327 
35 41.64 0.38 0.04–0.34 0.9211 14.40 3.11 0.9112 
45 69.08 0.23 0.08–0.46 0.8754 15.41 1.89 0.9379 
Table 5

Thermodynamic experimental parameters of the materials

Material ΔG (kJ·mol−1ΔS (J·mol−1 K−1ΔH (kJ·mol−1
SS 298 −0.55 36.41 10.29 
313 −0.92 
328 −1.28 
MC-CaO 298 −1.80 46.58 12.08 
313 −2.26 
328 −2.73 
MC-CAS 298 −2.09 56.74 14.82 
313 −2.66 
328 −3.23 
MC-PAM 298 −2.11 39.23 9.58 
313 −2.50 
328 −2.89 
Material ΔG (kJ·mol−1ΔS (J·mol−1 K−1ΔH (kJ·mol−1
SS 298 −0.55 36.41 10.29 
313 −0.92 
328 −1.28 
MC-CaO 298 −1.80 46.58 12.08 
313 −2.26 
328 −2.73 
MC-CAS 298 −2.09 56.74 14.82 
313 −2.66 
328 −3.23 
MC-PAM 298 −2.11 39.23 9.58 
313 −2.50 
328 −2.89 
Figure 5

Effects of initial pollutant concentration and temperature on adsorption capacity: SS (a), MC-CaO (b), MC-CAS (c), and MC-PAM (d).

Figure 5

Effects of initial pollutant concentration and temperature on adsorption capacity: SS (a), MC-CaO (b), MC-CAS (c), and MC-PAM (d).

Thermodynamic parameters

The effects of temperature and initial temperature on the TC adsorption performance of the SMCs were tested next, with the results shown in Table 5. The negative standard ΔH implied that the adsorption of TC by SMCs was an endothermic reaction and higher temperatures were beneficial to adsorption. This conformed to the conclusion of the isothermal models. In the adsorption process, ΔS >0, which reflected that the adsorption process was irreversible and the reaction continued towards increased system chaos. ΔG was negative at 298 K, 313 K and 328 K, which implied that the adsorption of TC by the SMCs was a spontaneous process. Moreover, the absolute value of ΔG was positively related to temperature, so temperature rises could facilitate the reaction. This conformed to the conclusion from ΔH >0. The molecular thermal reaction, energy transmission and mass transfer at the absorbent-adsorbate interface slowed down under excessively low temperature, which worked against the mass transfer of TC. With increased system temperature, the contact probability and frequency between TC in solution and the particulate surface of the carbon materials increased dramatically. As a result, the adsorption efficiency was increased.

CONCLUSIONS

The flocculating agents tested could be used as an additive and had a nucleation pore-forming effect, based on the evident difference in the thermodynamic stability of the flocculating molecules and the sludge. A porous structure was the main structural pattern of the SMCs. The specific surface areas and pore volumes ranged from 111.76–131.55 m2/g and 0.143–0.175 cm3/g, respectively. There were many types of polar groups on the SMC material surfaces, mainly carboxyl, carbonyl, phenolic hydroxyl and acylamino groups. All SMCs had good adsorption capacity for TC. Compared to dry sludge, the adsorption capacities of the SMCs can be significantly improved. The TC adsorption process of the SMCs conformed to the pseudo-second-order kinematic models and Freundlich isothermal adsorption model. The porous structure of the SMCs played the main role in pollutant adsorption, which was a spontaneous endothermic reaction. Acidic solutions and high temperatures were conducive to TC adsorption, and MC-CaO achieved the highest TC adsorption rate. The technique presented in this paper provides a new method of restoration of antibiotic-polluted water resources.

ACKNOWLEDGEMENTS

This work was supported financially by National Natural Science Foundation of China (Grant No. 41703120), Jilin University Postdoctoral Research Start-up Funds (801171050425), and Liaoning Innovation Team Project (No. LT2015017).

REFERENCES

REFERENCES
Babaei
A. A.
,
Lima
E. C.
,
Takdastan
A.
,
Alavi
N.
,
Goudarzi
G.
,
Vosoughi
M.
,
Hassani
G.
&
Shirmardi
M.
2016
Removal of tetracycline antibiotic from contaminated water media by multi-walled carbon nanotubes: operational variables, kinetics, and equilibrium studies
.
Water Science & Technology
74
(
5
),
1202
1216
.
Cao
X.
,
Ma
L.
,
Gao
B.
&
Harris
W.
2009
Dairy-manure derived biochar effectively sorbs lead and atrazine
.
Environmental Science & Technology
43
(
9
),
3285
3291
.
Chia
C. H.
,
Gong
B.
,
Joseph
S. D.
,
Marjo
C. E.
,
Munroe
P.
&
Rich
A. M.
2012
Imaging of mineral-enriched biochar by FTIR, Raman and SEM–EDX
.
Vibrational Spectroscopy
62
(
9
),
248
257
.
Ding
R.
,
Zhang
P.
,
Seredych
M.
&
Bandosz
T. J.
2012
Removal of antibiotics from water using sewage sludge- and waste oil sludge-derived adsorbents
.
Water Research
46
(
13
),
4081
4090
.
Feng
Y.
,
Zhou
H.
,
Liu
G.
,
Qiao
J.
,
Wang
J.
,
Lu
H.
,
Yang
L.
&
Wu
Y.
2012
Methylene blue adsorption onto Swede rape straw (Brassica napus L.) modified by tartaric acid: equilibrium, kinetic and adsorption mechanisms
.
Bioresource Technology
125
(
12
),
138
144
.
Fernández
J. M.
,
Nieto
M. A.
,
López-De-Sá
E. G.
,
Gascó
G.
,
Méndez
A.
&
Plaza
C.
2014
Carbon dioxide emissions from semi-arid soils amended with biochar alone or combined with mineral and organic fertilizers
.
Science of the Total Environment
482–483
(
1
),
1
.
Gao
Y.
,
Li
Y.
,
Zhang
L.
,
Huang
H.
,
Hu
J.
,
Shah
S. M.
&
Su
X.
2011
Adsorption and removal of tetracycline antibiotics from aqueous solution by graphene oxide
.
Journal of Colloid & Interface Science
368
(
1
),
540
546
.
Hadi
M.
,
Samarghandi
M. R.
&
McKay
G.
2010
Equilibrium two-parameter isotherms of acid dyes sorption by activated carbons: study of residual errors
.
Chemical Engineering Journal
160
(
2
),
408
416
.
Jaria
G.
,
Silva
C. P.
,
Ferreira
C. I.
,
Otero
M.
&
Calisto
V.
2016
Sludge from paper mill effluent treatment as raw material to produce carbon adsorbents: an alternative waste management strategy
.
Journal of Environmental Management
188
,
203
.
Kacuráková
M.
,
Capek
P.
,
Sasinková
V.
,
Wellner
N.
&
Ebringerová
A.
2000
FT-IR study of plant cell wall model compounds: pectic polysaccharides and hemicelluloses
.
Carbohydrate Polymers
43
(
2
),
195
203
.
Li
K.
,
Ji
F.
,
Liu
Y.
,
Tong
Z.
,
Zhan
X.
&
Hu
Z.
2013
Adsorption removal of tetracycline from aqueous solution by anaerobic granular sludge: equilibrium and kinetic studies
.
Water Science & Technology
67
(
7
),
1490
1496
.
Liu
C.
,
Tang
Z. G.
,
Chen
Y.
,
Su
S. J.
&
Jiang
W. J.
2010
Characterization of mesoporous activated carbons prepared by pyrolysis of sewage sludge with pyrolusite
.
Bioresource Technology
101
(
3
),
1097
1101
.
Parolo
M. E.
,
Avena
M. J.
,
Pettinari
G. R.
&
Baschini
M. T.
2012
Influence of Ca2+ on tetracycline adsorption on montmorillonite
.
Journal of Colloid and Interface Science
368
(
1
),
420
426
.
Peiris
C.
,
Gunatilake
S. R.
,
Mlsna
T. E.
,
Mohan
D.
&
Vithanage
M.
2017
Biochar based removal of antibiotic sulfonamides and tetracyclines in aquatic environments: a critical review
.
Bioresource Technology
246
,
150
159
.
Pietrzak
R.
,
Nowicki
P.
,
Kaźmierczak
J.
,
Kuszyńska
I.
,
Goscianska
J.
&
Przepiórski
J.
2014
Comparison of the effects of different chemical activation methods on properties of carbonaceous adsorbents obtained from cherry stones
.
Chemical Engineering Research & Design
92
(
6
),
1187
1191
.
Sánchez
M. E.
,
Menéndez
J. A.
,
Domínguez
A.
,
Pis
J. J.
,
Martínez
O.
,
Calvo
L. F.
&
Bernad
P. L.
2009
Effect of pyrolysis temperature on the composition of the oils obtained from sewage sludge
.
Biomass & Bioenergy
33
(
6–7
),
933
940
.
Takdastan
A.
,
Mahvi
A. H.
,
Lima
E. C.
,
Shirmardi
M.
,
Babaei
A. A.
,
Goudarzi
G.
,
Neisi
A.
,
Heidari
F. M.
&
Vosoughi
M.
2016
Preparation, characterization, and application of activated carbon from low-cost material for the adsorption of tetracycline antibiotic from aqueous solutions
.
Water Science & Technology
74
(
10
),
2349
.
Wen
Q.
,
Li
C.
,
Cai
Z.
,
Zhang
W.
,
Gao
H.
,
Chen
L.
,
Zeng
G.
,
Shu
X.
&
Zhao
Y.
2011
Study on activated carbon derived from sewage sludge for adsorption of gaseous formaldehyde
.
Bioresource Technology
102
(
2
),
942
947
.
Xin
W.
,
Song
Y.
,
Peng
J.
,
Liu
R.
&
Han
L.
2016
Characterization and adsorption properties of low-cost sewage sludge-derived carbon by using poly-aluminium chloride as pore-forming agent
.
Advances in Engineering Research
63
,
666
678
.
Yamaguchi
N. U.
,
Bergamasco
R.
&
Hamoudi
S.
2016
Magnetic MnFe2O4–graphene hybrid composite for efficient removal of glyphosate from water
.
Chemical Engineering Journal
295
,
391
402
.
Yang
H.
,
Li
S.
,
Chen
J.
,
Zhang
X.
&
Chen
Y.
2014
Adsorption of Pb(II) on mesoporous activated carbons fabricated from water hyacinth using H3PO4 activation: adsorption capacity, kinetic and isotherm studies
.
Applied Surface Science
293
(
293
),
160
168
.
Zaini
M. A. A.
,
Zakaria
M.
,
Mohd-Setapar
S. H.
&
Che-Yunus
M. A.
2013
Sludge-adsorbents from palm oil mill effluent for methylene blue removal
.
Journal of Environmental Chemical Engineering
1
(
4
),
1091
1098
.
Zhang
B.
,
Xiong
S.
,
Xiao
B.
,
Yu
D.
&
Jia
X.
2011
Mechanism of wet sewage sludge pyrolysis in a tubular furnace
.
International Journal of Hydrogen Energy
36
(
1
),
355
363
.
Zhang
X.
,
Shen
J.
,
Zhuo
N.
,
Tian
Z.
,
Xu
P.
,
Yang
Z.
&
Yang
W.
2016
Interactions between antibiotics and graphene-based materials in water: a comparative experimental and theoretical investigation
.
ACS Applied Materials & Interfaces
8
(
36
),
24273
24280
.
Zhao
C.
,
Xu
X.
,
Chen
J.
&
Yang
F.
2013
Effect of graphene oxide concentration on the morphologies and antifouling properties of PVDF ultrafiltration membranes
.
Journal of Environmental Chemical Engineering
1
(
3
),
349
354
.
Zheng
L.
,
Tian
Y.
&
Sun
D. Z.
2007
Effects of pH on the surface characteristics and molecular structure of extracellular polymeric substances from activated sludge
.
Environmental Science
28
(
7
),
1507
.
Zou
J.
,
Dai
Y.
,
Wang
X.
,
Ren
Z.
,
Tian
C.
,
Pan
K.
,
Li
S.
,
Abuobeidah
M.
&
Fu
H.
2013
Structure and adsorption properties of sewage sludge-derived carbon with removal of inorganic impurities and high porosity
.
Bioresource Technology
142
(
8
),
209
217
.

Supplementary data