Calcium alginate/activated carbon/humic acid tri-system porous fibers for removing tetracycline from aqueous solution

Abstract In this study, activated carbon and humic acid powder were fixed by the cross-linking reaction of sodium alginate. Calcium alginate/activated carbon/humic acid (CAH) tri-system porous fibers were prepared by the wet spinning method and freeze-dried for the removal of tetracycline in aqueous solution. Subsequently, the morphology and structure of CAH fibers were measured by scanning electron microscopy (SEM) and the Brunauer-Emmett-Teller (BET) method. The effect of pH, contact time, temperature and other factors on adsorption behavior were analyzed. The Langmuir and Freundlich isotherm models were used to fit tetracycline adsorption equilibrium data. The dynamics data were evaluated by the pseudo-second-order model, the pseudo-second-order model and the intraparticle diffusion model. Thermodynamic study confirmed that the adsorption of tetracycline on CAH fibers was a spontaneous process.


INTRODUCTION
Antibiotics are not only commonly used in the treatment and prevention of human diseases 1 , but also widely used in animal husbandry and aquaculture to promote livestock growth 2, 3 .
The safety and effi cacy of tetracycline (TC) makes it one of the most widely used antibiotics in the world. However, tetracycline is poorly absorbed from the gastrointenstinal tract of animals 4 , with up to 50-80% excreted in animal feces. As a result, signifi cant amounts of tetracycline enter the ecosystem, leading ultimately to the emergence of resistance genes 5-7 . The presence of tetracycline in the water environment can affect the activity and composition of microorganism communities and change their ecological structure 8 . Experimental results indicate that allergic reactions in some susceptible individuals are caused by antibiotic residues in food, and overuse of antibiotics can lead to super infection and drug resistance 9 . Therefore, it is a matter of urgency that effi cient and clean methods are used to reduce and control environmental pollution caused by tetracycline.
Photocatalytic degradation 10 , ozonation 11 and adsorption are important methods for tetracycline removal. Among these methods, adsorption is particularly promising due to its low cost, non-toxicity and relative ease of use. Recent research shows that an adsorbent made by fi xing copper ions with sodium alginate has high removal effi ciency for tetracycline 12 , but the introduction of metal ions in the preparation process causes pollution to the environment. Experimental results show that activated carbon and graphene oxide have an adsorption effect on tetracycline 13, 14 , however the powder form is diffi cult to separate from the water solution, which means, once again, that its use causes environmental pollution. To overcome this problem, porous fi bers are prepared by encapsulating activated carbon and humic acid powder with sodium alginate. Activated carbon (AC), a form of graphite microcrystalline amorphous carbon with high degree of microporosity and huge specifi c surface area, is one of the most widely used materials for adsorbing dyes and pharmaceuticals 15-17 . Adding a large quantity of AC to the fi bers can improve their adsorption capacity to TC, but when the quantity of AC is too large, it is diffi cult to continue to improve the adsorption capacity of the fi bers. At this point, if a small amount of humic acid (HA) is added, the adsorption capacity of the fi bers can be improved. HA is a type of complex natural macromolecular organic polymers that exist widely in nature 18 , formed by a series of processes such as microbial decomposition and the transformation of animal and plant remains 19, 20 . HA contains a plurality of functional groups, which allows for considerable complexation and ion exchange capacity 21 . However, its low dispersibility in aqueous solution limits its practical applications in environmental protection.
Sodium alginate (SA) is a kind of natural polysaccharide polymers 22, 23 . Through the exchange of divalent cations and sodium ions, a gel with three-dimensional network structure can be obtained, which can successfully encapsulate AC and HA powders, and facilitate the recovery of materials to improve the effi ciency of adsorbents 24 . Experiments have shown that a large number of hydroxyl groups and carboxyl groups make the adsorption of dyes and other contaminants including methylene blue and metal ions possible 25-27 .
In this paper, aerogel adsorbent of calcium alginate / activated carbon/humic acid (CAH) porous fi bers were prepared by wet spinning and vacuum freeze-drying methods 28 . The chemical structure and surface morphology of the fi bers were characterized by scanning electron microscopy (SEM) and the Brunauer-Emmett-Teller (BET) method. The adsorption properties of the fi bers for tetracycline removal were investigated by varying the temperature, amount of adsorbent, reaction time and the initial pH of the solution.

Materials and Chemicals
TC (analytical reagent) was produced by Wuhan Lana White Pharmaceutical Factory and stored at 273 K . HA (No. 53680) was purchased from Tianjin guangfu Fine Chemical Research Institute. Powdered AC was purchased from Hairuicheng Co., Ltd., Qingdao, China. Sodium alginate was obtained from Shanghai Aibi Chemical Reagent Co. All solutions were prepared with deionized water.

Preparation of CAH fi bers
The HA powder and sodium alginate powder were mixed in beaker so that the HA could be distributed more evenly in the sodium alginate fl uid matrix. After adding deionized water to the beaker and stirring with the magnetic stirring device for 5 hours, activated carbon was added to the beaker and stirred for another 12 hours to obtain, the black fl uid mixture. The wet-spinning technique method is as follows: as soon as the mixture was injected into calcium chloride solution, ion exchange occurred rapidly between calcium ions and sodium alginate to form CAH fi bers. The fi bers were fully molded by leaving them to stand for 5 hours, followed by rinsing repeatedly with deionized water. After freezing the obtained fi bers, the samples were obtained by freeze-drying. In order to determine the best proportion of SA, AC and HA, two experiments were conducted. SA/AC fi bers were prepared with different mass ratios, and the optimum proportion of activated carbon was determined from the experiment. Subsequently, HA with different mass ratios was added to obtain the fi nal ratio of SA, AC and HA.

Characterization
The surface morphologies of CAH fi bers were characterized by a high resolution scanning electron microscope (Quanta250 FEG, FEI, America). The BET equation (ASAP2460, micromeritics, Shanghai, China) was used to measure the specifi c surface area of CAH fi bers by N 2 adsorption isotherm.

Adsorption studies
All experiments were carried out in a Vapour-bathing Constant Temperature Vibrator (SHZ-82A) with a rotational speed of 150 rpm. TC stock solution with a concentration of 1000 mg/L was stored at a temperature below 283 K. During the experiment, different concentrations of diluents were obtained by diluting the TC stock solution. Add 10 mg CAH fi bers to 20 ml TC solution at different temperatures. After adsorption and equilibrium, CAH fi bers were separated from TC solution. The equilibrium concentration of TC was measured by an ultraviolet-visible spectrophotometer. (TU-1810, Beijing Purkinje General Instrument Co., Ltd., Beijing) at wavelength of 360 nm.
The effects of TC concentration and ambient temperature on adsorption were simultaneously investigated. The initial concentration of TC solution ranged from 50 mg/L to 170 mg/L, with a total of 7 groups at temperatures of 303 K, 313 K and 323 K, respectively. In order to investigate the effect of initial pH on adsorption, the initial pH value of TC solution (90 mg/L) was set from 2.3 to 7. Adsorbents with different doses (5-35 mg) were added to the TC solution (110 mg/L) to determine the effect of the amount of adsorbent on adsorption. The effect of contact time on TC adsorption was investigated by adding 300 mL TC solution (90 mg/L) to the beaker and the quantity of adsorbent was 150 mg. The residual TC concentration was determined by taking out the supernatant at a specifi c time interval. The adsorption capacity q e (mg/g) and q t (mg/g) were calculated based on the following equations: Where c 0 and c e are the initial and equilibrium concentrations (mg/L) of TC solution, respectively; c t is the residual concentration of TC at time t; W is the mass of adsorbent (g) used in the corresponding adsorption experiments, and V is the volume of the solutions (L).
In order to obtain the optimal ratio of sodium alginate, activated carbon and humic acid, two groups of adsorption tests were conducted using 50 mg/ L TC solution. In the fi rst set of experiments: porous fi bers with a mass ratio of SA to AC of 1:1-1:5 were prepared, and the best ratio was deemed to be 1:5 following the adsorption experiments. It was noteworthy that when the mass ratio of SA to AC is greater than 1:5, SA cannot completely encapsulate the activated carbon, and the formability of the fi ber deteriorates. In the second set of experiments: preparation of SA, AC and HA mass ratio of 1:5:1 to 1:5:5 porous fi bers led to an optimal ratio of 1:5:1. Then, the humic acid ratio was lowered to prepare fi bers with a mass ratio of 1:5:0.5 and 1:5:0.25 for SA:AC:HA. The adsorption results showed that the optimal ratio of the three materials was 1:5:0.5. Figure 1 shows that the entire surface of CAH fi bers is wrinkled, whilst the humic acid particles are dotted. The activated carbon particles are embedded on the fi ber surface, piercing through several through-holes with large cross section. In particular, there are a number of porous hillocks distributed on the surface of CAH fi bers. The surface of the hillocks is evenly distributed with the micropores. When the diameters between two adjacent holes become larger, the two holes combine and form a larger hole, which increases the surface area of the fi ber.

Characterization of the prepared adsorbents
The properties of adsorbents are closely correlated with pore volume and surface area. The N 2 adsorption was performed on a CAH sample to evaluate its permanent porosity. Figure 2 A shows the adsorption hysteresis ring. It can be seen that CAH fi bers have reversible type IV isotherm, which is one of the main characteristics of mesoporous materials 29 . The surface area of CAH fi bers is 622 m 2 /g, as calculated by the brunauer-emmett-teller (BET) model. Figure 2 B shows the pore size distribu-tion of CAH fi bers. Most holes are mesopores, making up a proportion of 93.6%, whereas the proportions of micropores and macropores are 3.9% and 2.5%, respectively. CAH fi bers are mainly distributed with narrow mesoporous pores, generally smaller than 10 nm in size, and the peak center of the pore size is mainly 3.79 and 7.71 nm. The addition of activated carbon can increase the surface area of CAH fi bers. However, when the mass ratio of sodium alginate to activated carbon exceeds 1:5, the formability of the fi ber deteriorates.
Eff ects of AC/HA dose and pH value on adsorption Figure 3 A shows that activated carbon plays a major role in the adsorption of tetracycline by CAH fi bers, while a small amount of humic acid can greatly improve the adsorption effect of the adsorbent. Figure 3 B shows that when the proportion of humic acid exceeds 0.5, the adsorption effi ciency decreases. This may be attributed to an excessive HA content bringing about a decrease in the amount of AC in fi bers of the same weight. On the other hand, when the proportion is less than 0.5, the amount of humic acid is too small to play such a role.
Humic acid has strong complexation with tetracycline and shows pH dependence 30 , so the adsorption effi ciency of CAH fi bers on tetracycline is affected by pH, as shown in the Figure 4 A. There are ionizable groups including -COOH, -OH and -NH 2 31 in TC molecule,

Eff ects of temperature
Temperature has a direct effect on the diffusion rate of tetracycline and adsorbent performance, so it is an important factor in the adsorption process. The infl uence of temperature and TC initial concentrations on the adsorption effect is shown in Figure 4 B, which was carried out at 303, 313 and 323 K, respectively. The results showed that when the temperature was increased from 303 K to 323 K, the adsorption capacity of the adsorbent decreased from 266.78 mg/g to 230.92 mg/g. This may be due to a reduction in the number of available binding sites of activated carbon and humic acid to tetracycline. Also, the adsorption capacity of CAH fi bers to TC solution of the same concentration decreased, as temperature rose, indicating that the adsorption reaction is an exothermic process. For the initial concentration which are dissociated under different conditions, so that TC fi nally comes to form cations, zwitterions and anionic 32, 33 . When the solution environment is acidic, the main forms of tetracycline are cations and zwitterions, which can interact with the humic acid in the fi bers through hydrogen bond or electrostatic attraction 34 . The adsorption capacity increased as pH decreased. When pH ≤ 3, the trend of increasing adsorption capacity slowed, due to the competition between H + and tetracycline in the solution for the binding sites of humic acid. When pH ranged from 4.3 to 6.5, tetracycline still present in the form of zwitterion, but the adsorption effi ciency decreased. As the carboxylic groups in humic acid became progressively deprotonated, there was decreased carboxylic acid hydrogen bonding, which led to a reduction in adsorption effi ciency. of TC, as the concentration gradient increased, the adsorption capacity increased from 94.07 mg/g to 266.78 mg/g at 303 K.

Eff ects of adsorbent dosage
The amount of adsorbent is an important parameter affecting adsorption effi ciency. Figure 4 C shows the infl uence of different adsorbent dosages on the adsorption capacity and removal rate. It can be seen that as the amount of adsorbent increases, the adsorption capacity decreases rapidly, but the removal rate increases, reaching a maximum of 98.8%.This may be due to an increase in the number of sites capable of binding to TC. However, even when the number of available binding sites increases, a reduction in the utilization of binding sites would result in a decrease in adsorption capacity. Once the binding of the adsorption sites to TC reaches saturation, removal effi ciency no longer increases 34 .

Adsorpti on isotherm
Adsorption isotherm can be used to explain the essence of the adsorption behavior of CAH fi bers towards tetracycline to better understand the relationship between the adsorption amount and the residual concentration. The most commonly used models for studying adsorption equilibrium are the Langmuir model and the Freundlich model. The Langmuir model assumes the adsorption process takes place on a uniform surface and a single layer. c e (mg/L) represents the equilibrium concentration, that is, the concentration of the solution after adsorption equilibrium, and q e (mg/g) is the equilibrium adsorption capacity, which refl ects the change of solution concentrations caused by the adsorbent per unit mass. After fi tting the models, the maximum adsorption q max (mg/g) and Langmuir constant k L (L/g) can be calculated. The Langmuir model equation is expressed as follows 35 : At 303, 313 and 323 K, the maximum single-layer adsorption capacity is similar. The dimensionless constant R L can be used to evaluate whether the adsorption process is favorable. C 0 is the initial concentration of the solution, and the equation is as follows: (4) As listed in Table 1 , the R L values are all between 0 and 1, indicating that the adsorption process is favorable, with CAH fi bers having a good adsorption ability towards TC 36 . The Freundlich model can be applied to describe both single layer adsorption and heterogeneous surface adsorption. k F and n are Freundlich constants, where k F is related to the adsorption affi nity of the adsorbent, and n indicates the supporting force of the adsorption process, respectively. The Freundlich isothermal model is shown in the equation below (9) 37 : As shown in Figure 5, the R 2 values obtained by fi tting the adsorption isotherm with the two models are quite high, indicating that both models could fi t the adsorption process well. The 1/n value calculated by the Freundlich model equation is less than 1, indicating that the adsorption process was effective.

Adsorption kinetic study
The study of adsorption kinetics is important for describing the adsorption rate and the control mechanism of adsorption. 150 mg CAH fi bers were added to 300 mL TC solution with an initial concentration of 90 mg/L at 303 K. As shown in Figure 4 D, the adsorption rate is very fast in the fi rst 8 minutes, which could be explained by the large number of adsorption sites on CAH fi bers in the initial stage. As time goes on, however, the adsorption rate gradually slows until equilibrium is  In order to further study the adsorption process of TC on CAH fi bers, the diffusion mechanism is analyzed by the intra-particle diffusion model. Generally, the fi tting line does not pass through the origin, and the adsorption process is controlled by multiple stages, generally divided into three stages: (1) external particle diffusion stage; (2) pore diffusion stage; (3) adsorption reaction stage. The intra-particle diffusion model is used to describe the rate control step of the whole process, and the formula is 41 : Where, k id is the particle internal diffusion constant (mg/g min 1/2 ), q t is the adsorption capacity at time t (min), and C i is a parameter related to boundary layer and thickness. Figure 6 C shows that the adsorption process is divided into two stages. In the fi rst stage, the fi tting line is steep and there are a large number of adsorption sites on the surface of CAH fi ber with a high adsorption rate, which is mainly attributable to macroporous diffusion. The second stage is the gradual adsorption stage with a small slope. Intra-particle diffusion is a step that controls the adsorption rate, which is mainly attributable to microporous diffusion 42, 43 . And k 1d (10.6713 mg/g min 1/2 ) is signifi cantly higher than k 2d (2.3977 mg/g min 1/2 ), indicating that external diffusion plays a more important role in adsorption kinetics.

Adsorption thermodynamic study
In order to study the energy change of CAH fi bers on TC adsorption, the thermodynamic data of the adsorption process are calculated at different temperatures. Enthalpy (ΔH) and entropy (ΔS) are obtained using the following equations 44 : Where R is the universal gas constant (8.314 J/K) and T represents the absolute temperature in Kelvin (K). A straight line can be obtained by plotting 1/T and ln(q e /c e ), ΔH and ΔS can be calculated according to the linear slope and intercept, respectively. The specifi c thermodynamic parameters are listed in Table 3. The value of ΔH is −29.32 kJ/mol which means that exothermic reactions take place between the adsorbent and TC, with low temperatures more favorable for adsorption. reached, because the number of adsorption sites and the amount of TC in the solution are both dropping, and the remaining TC needs to spend more time fi nding binding sites that can form hydrogen bonds.
In this paper, the pseudo-fi rst-order model, the pseudo-second-order model and the intra-particle diffusion model are used to analyze and fi t the experimental data, respectively. The mathematical expression of the pseudo-fi rst-order model and pseudo-second-order model is as follows 38, 39 : Where k 1 (1/min) is the adsorption rate constant for the pseudo-fi rst -order adsorption process, k 2 (g/mg·min) i s the corresponding kinetic constant, q e and q t are the adsorption amount of TC (mg/g) under equilibrium conditions and the adsorption amount of TC (mg/g) at a specifi c time t (min), respectively. The values of k 1 , k 2 , q e are determined by fi tting the slope and intercept of the line. All the above data are listed in Table 2, and the results showed that the quasi-second order model describes the experimental data well 40 . The correlation coeffi cient of the pseudo-second order model is 0.98, which is much higher than that of the pseudo-fi rst-order model. Compared with the equilibrium adsorption calculated by the pseudo-fi rst-order model (135.80 mg/g), the equilibrium adsorption calculated by the pseudo--second-order model (156.98 mg/g) was closer to the experimental adsorption (157.19 mg/g). It can be seen from Figure 6 that the pseudo-second-order model can better fi t the data. intra-particle diffusion model Table 2. The kinetic constants of adsorption of TC on CAH fi bers ΔG less than zero indicates that the adsorption process is spontaneous.

CONCLUSIONS
The experimental results in this paper showed that CAH fi bers can effectively remove TC from aqueous solution. The optimal ratio of SA, AC and HA was 1:5:0.5, and the maximum adsorption capacity of the porous fi bers was up to 95.948 mg/g after adsorption of the TC solution with a concentration of 50 mg/L. Adsorption was more favorable at a lower temperature (303 K) and in an acidic solution environment. At 303 K, the maximum adsorption capacity of CAH fi bers for TC was 333.33 mg/g. Thermodynamic parameters showed that the adsorption process of TC on CAH fi bers was spontaneous and exothermic. The isothermal parameters and kinetics were well described by the Langmuir isotherm model and the pseudo-second-order model, respectively. Based on these analyses, CAH fi ber is an environmentally friendly material that can effectively remove TC from aqueous solution and is easily separated from the water environment, which has broad application prospects.

ACKNOWLEDGMENTS
This work was supported by the National Natural Scienc e Foundation of China (51672140), Natural Scie nce Foundat ion of Shandong Province (ZR2015EM038), Taishan S cholar Program of Shandong Province (201511029).