Removal of tetracyclines from aqueous solutions by electrocoagulation/pecan nutshell coupling processes: synergistic effect and mechanism

Antibiotics have been extensively used for human and veterinary therapeutic purposes (Javid et al. 2016; Song 2016). The amount of antibiotics used each year has reached up to 100,000–200,000 tons globally, with more than 25,000 tons in China annually (Yan et al. 2013). The most commonly used tetracyclines (TCs) include tetracycline hydrochloride (TC), oxytetracycline (OTC) and chlorotetracycline (CTC) (Ji et al. 2016). However, only partially TCs can be metabolized by human and animals and unabsorbed residues will be released into the environment. Hence, the overuse of these compounds results in the pollution of the environment eventually. In China, tetracycline antibiotic residues have been detected in many water bodies, such as the Yangtze River Basin, Songhua River Basin, Poyang Lake Basin, Haihe River Basin and Three Gorges Reservoir area (Luo et al. 2011; Wang et al. 2017a, 2017b). Residues of TCs are also found in the aquatic environment in the United States (Karthikeyan & Meyer 2006), Iran (Javid et al. 2016) and other countries. Such antibiotics can cause chronic poisoning of aquatic organisms, and will be partly absorbed and then accumulated by plants when they enter the soil. Finally, they will cause harm to human body through the food chain, and enhance the antibiotic resistance of microorganisms via generating resistance genes (Wan et al. 2010).

At present, TC antibiotics effluent is commonly purified by physicochemical and biological treatment, and anaerobic process in biological methods is preferred in general (Shi et al. 2017). However, considering the relatively high residues in TCs wastewater, physicochemical method should be used for pretreatment before biological treatment. There are various physical, chemical or biological processes for TCs contaminated waters, such as coagulation-sedimentation, adsorption, flotation, osmosis, micro electrolysis and photo-Fenton process, etc. (Kakavandi et al. 2016; Antonherrero et al. 2017; Barthoumi et al. 2017; Saitoh et al. 2017; Zhang et al. 2017; Khanday & Hameed 2018; Zhang et al. 2018). Other than the above mentioned processes, electrocoagulation (EC), electrooxidation and electrofloatation in electrochemistry area are also available as relatively new technologies. Electrocoagulation combines the advantages of coagulation, flotation and electrochemistry, becoming an emerging technology in water and wastewater treatment. EC is one kind of treatment techniques that is capable of being clean, easyhandling and high efficiency in the removal of organic pollutants (Moussa et al. 2017). In consideration of the shortcomings of traditional electrocoagulation, including easy passivation of electrodes, low efficiency, high energy consumption, it has become a hot research area to combine electrocoagulation technology with other technologies to improve the treatment efficiency and reduce the treatment cost. E. Hernández-Francisco combined electroflocculation-Fenton/photo-Fenton technologies to treat phenolic pollutants in refinery effluents, and the final total organic carbon (TOC) removal rate reached 88% (Hernández–Francisco et al. 2017); Marius Sebastian Secula removed acid dyes from wastewater by electroflocculation/biochar adsorption coupling, which improved the removal rate of indigo carmine in aqueous solutions and reduced the adsorption time and energy consumption (Secula et al. 2012).

The main aim of this work is to investigate the efficiency and the mechanism of the electrocoagulation technique coupling with PS for removing TCs from aqueous solution. It may provide a highly effective technology to treat high concentration TCs effluents.

PS powder was purchased from Meihekou Tiancheng Food Company, dried at 60 °C for 24 h and sieved with a 80 mesh. Tetracycline hydrochloride (C₂₂H₂₄N₂O₈•HCl, MW = 480.9 gmol⁻¹, USP grade, ≥98%), oxytetracycline hydrochloride (C₂₂H₂₄N₂O₉•HCl, MW = 496.89 gmol⁻¹, USP grade, ≥95%), chlortetracycline hydrochloride (C₂₂H₂₃ClN₂O₈)•HCl, MW = 515.34 gmol⁻¹, USP grade, ≥80.0%) and sodium sulphate (NaSO_4, AR) were purchased from Shanghai Aladdin Bio-Chem Technology Co.

Scanning electron microscope (SEM) measurements were performed by a field emission scanning electron microscope (Hitachi, S-4800). Fourier transform infrared spectrum (FTIR) was used to verify the change of the bonds on a SHIMADZU Iraffinity-1S spectrophotometer using KBr as a diluent. The method of the point of zero charge (pzc) in this work is as follows. In each of the 50 mL bottles, 40 mL of water is added, and the pH is adjusted with 0.1 mol/L HNO₃ ranging from 2 to 10. Then 100 mg of PS powder is added and the final pH of the solution is measured after vibrating at the constant temperature for 24 h. When the initial and final pH values are equal, the initial pH is the point of zero charge. After flocculation, flocs are carefully withdrawn from the bottom of jars, and then vacuum freeze-dried to prevent the change of functional groups in flocs.

As shown in Figure 1, the experiment was conducted in an electrocoagulation cell (15 cm × 11.9 cm × 15.6 cm). The bottom of the cell was equipped with six perforated tubes which connected to an air pump to provide agitating aeration. The electro-chemical setup was constituted with an aluminum plate as the anode and a stainless steel plate as the cathode. Aluminum and stainless steel are widely used in the process of electric flocculation, because the multivalent ion coagulation effect produced by the multivalent electrode used as the anode is economical and common. Aluminum and stainless steel plates can be used as a single electrode or in combination, and the flocculation produced in combination has a higher flocculation effect. The electrodes were connected to a digital DC power supply (WYJ, 0–30 V; 0–5 A; DC Regulated Power Supply Double Way Output, TESTMART, Shanghai, China). The effective electrode area was 36 cm².

For every experimental run, 800 mL of TCs solution (250 mg/L) was placed into the reactor and 1 g/L Na₂SO₄ solution was added to adjust the conductivity. The current density and electrode space were set to a desired value and then a certain quantity of PS was added to the solution. To assess the treatment process, 5 mL water sample was drawn every 5 min during the experiment. The samples were filtered through Whatman 0.45 μm filters, and then measured with a UV-vis spectrophotometer (UVmini-1240, Shimadzu). The maximum absorption wavelength of TC, OTC, and CTC were 356, 354 and 365 nm, respectively. The influencing factors were also investigated, containing the plate spacing, current density, walnut shell dose and initial concentration of TCs.

Figure 2 reveals the SEM images of PS morphology; the PS appears to have a rough surface with irregular pores. The distinct gullies on the surface could increase the specific surface area, which facilitates the contact between PS and target pollutant molecules.

FTIR spectrum of PS was illustrated in Figure 3. For PS, the major intense bands were at 3,415.03 cm⁻¹ (-OH stretching); 2,933.78 cm⁻¹ (attributed to C-H stretching vibration in aliphatics); 1,735 cm⁻¹ (C = O stretching vibration in hemicellulose); 1,264.36 cm⁻¹ (C-O-C stretching vibration in cellulose) and 1,047 cm⁻¹ (C-O stretching vibration). The specific bands at 1,500–1,300 cm⁻¹ was owing to O-H stretching vibration, the deformation of a methoxy group in lignin structure and aliphatic CH₂ group in cellulose structure (Xu et al. 2013). Hence, it can be concluded that there are characteristic functional groups on the surface of PS, which may interact with the substances in the EC + PS system via charge or hydrogen bond.

pHpzc is the pH at which the surface exhibits net zero charge. The results show that the pHpzc of PS is approximately 5.0 (Figure 4), meaning that when the pH is less than 5.0, PS has a positive charge, and opposite charge would be obtained at pH > 5.0.

Figure 5 shows the effects of EC + PS coupling method on the removal of TC (a), OTC (b) and CTC (c). PS adsorption process and EC were used as controls, respectively. The order of removal rate of TCs for different systems is EC + PS > EC≫PS. Individual EC system can also reach a relatively high removal rate if the reaction time is long enough. Therefore, compared to EC system, the enhanced treatment effect of EC + PS system was mainly reflected in the decrease of reaction time, meaning a higher reaction rate. The following relationship of the removal rate on TCs at 40 min has been acquired: TC:PS (3.02%) + EC (36.32%) < EC + PS (89.90%); OTC:PS (4.32%) + EC (35.87%) < EC + PS (76.00%); CTC:PS (5.73%) + EC (13.13%) < EC + PS (50.78%). Thus the removal of TCs by EC + PS coupling process is not a simple addition of EC process and PS adsorption process, while there is a synergistic effect existing during the whole EC + PS coupling procedure. According to the comparison in Figure 5(a)–5(c), the order of removal rate of tetracyclines is CTC > TC > OTC, which is attributed to their different molecular structure. Yuanzhu Li studied dissipation kinetics of OTC, TC, and CTC residues in soil, and found the stability order of the three TCs was OTC > TC > CTC, which was consistent with the result of our experiments (Li et al. 2016). Wanru Chen investigated the adsorption and transformation of tetracycline antibiotics with aluminum oxide, and the reactivity towards Al₂O₃ followed the order of CTC > TC > OTC (Chen & Huang 2010).

It can be seen from the reaction equations that metal ions are generated by anodic dissolution (Equation (3)), and then form a series of mononuclear or multinuclear metal hydroxy complexes in the electrolytic cell, such as Al(OH)²⁺, Al(OH)₂⁺, Al₂(OH)₂⁴⁺, Al(OH)₄⁻, Al₆(OH)₁₅³⁺, Al₇(OH)₁₇⁴⁺, Al₈(OH)₂₀⁴⁺, Al₁₃O₄(OH)₂₄⁷⁺, Al₁₃(OH)₃₄⁵⁺, and finally Al(OH)₃ (s) would be formed through a complex precipitation kinetic procedure (Equations (5) and (6)). All these Al containing components have a large surface area, and is conducive to the rapid adsorption of TCs and the capture of colloidal particles in aqueous, which is finally removed by precipitation. The mechanism of coupling methods is just the addition of electrocoagulation and adsorption of biomass materials in electrocoagulation coupled with banana peel, rice husk and other materials (De Carvalho et al. 2015; Wang et al. 2017a). But in our coupling systems, there is also a synergistic effect throughout the TC antibiotics removal process.

In EC + PS coupling system, PS would have different kind of charge at different pH value (Figure 4). According to the fact that the initial pH of the solution is between 3.5 and 3.6, yet the final pH is between 6.7 and 6.9, the PS was first positively charged and then negatively charged during the whole treatment process, which led to the enhanced TCs removal because charge neutralization would occur between the PS with negative charge and the TCs ion with positive charge.

The chemical state and composition of elements of the flocculate acquired from different systems was analyzed using X-ray photoelectron spectroscope (XPS). Figure 6 shows the results of XPS analysis for PS and the flocculates obtained in EC + PS and EC + PS + TC systems. C1s peak for PS can be deconvoluted into two peaks at binding energies of ∼284.57 and 285.79 eV, respectively. The former can be associated with C = C bonding (Chen & Huang 2010), and the latter is assigned to C-N bonding (Wei et al. 2016). In EC + PS system, the peak at 284.57 eV does not shift, but the peak at 285.79 eV has transferred to 285.92 eV (C-N binding), because Al³⁺ produced in EC + PS process could coordinate with C-N functional group. Compared to EC + PS system, the two peaks generated by the flocculate in EC + PS + TC system shift more obviously. The peak representing C = C bond shifts a little from 284.57 to 284.51 eV, probably attributed to the weak interaction between TC and C = C. The peak representing C-N moves more largely from 285.79 to 285.56 eV, which is due to the coordination between Al³⁺, C-N and TC. For PS, there is a peak of O1s at the binding energy 532.32 eV, while it has transferred to 532.24 eV in PS + EC system and 532.09 eV in EC + PS + TC system, respectively. It can be inferred that the peak shift results from the coordination of Al³⁺ with C = O firstly and then, with TC molecules. Therefore, according to XPS analysis, the Al³⁺ produced in the EC process could coordinate with C-N and C = O on the surface of biomass materials, which would further coordinate with TC to form larger particles, so as to improve the flocculation performance and accelerate the removal of pollutants. PS plays a role of ‘bridging’ throughout the whole removal procedure. Shuying Jia et al. believed that Cu (II) can further interact with tetracycline after coordination with C-N and C = O of the flocculant and enhance the flocculation effect, when they studied the removal of coexisting Cu (II) and TC using aromatic functionalized chitosan flocculant (Jia et al. 2016). This conclusion was in accordance with our work. Therefore, charge neutralization and coordination adsorption are the main reasons for the synergistic effect of EC + PS + TC system.

The effect of current density (2.5, 5, 7.5 and 10 mA/cm², respectively) on the removal rate of TC in water was shown in Figure 7. The removal efficiency of TC increased with the current density increasing in EC and EC + PS process. When CD became higher, the anodic dissolution of Al increased, which leads to an increase in the amount of metal hydroxides as well as an improvement of removal efficiency. But if the reaction time is enough, increasing the current density could speed up the removal rate and has little effect on the final removal rate. Compared Figure 7(a) with 7b, the highest efficiency enhanced of TC removal by PS was obtained at low current density values of 2.5 mA/cm². According to Equation (1), the energy consumption in condition of 2.5 mA/cm² with PS after 40 is lowest among all conditions. Taking energy saving into consideration, the current density of 2.5 mA/cm² was chosen as ideal condition in subsequent experiments.

Figure 8 shows the influence of plate spacing (3, 6, 9 and 12 cm) on TC removal in EC and EC + PS processes. In the EC process for 40 min (Figure 8), the change of plate spacing has little effect on the removal process, yet between 40 and 60 min, the removal rate is highest when the plate spacing is 12 cm. It is mainly because the floc volume of aluminum hydroxides produced gradually increases with the reaction going on, and the large plate spacing is conducive to the full contact between the flocculate and TC. In EC + PS process (Figure 8(b)), the plate spacing has little effect on the removal rate. This is attributed to the existence of PS in the coupling system which promotes the uniform distribution of Al in the system, makes Al³⁺ fully contact with TC and reduces the influence of changing plate spacing. The larger the plate spacing is, the greater the voltage in the reaction tank is, and the more energy would be consumed in the treatment process. Therefore, 3 cm is selected as the best plate spacing in the experiment.

Figure 9 compared the removal efficiencies of TCs for simple EC and EC-PS coupling system at different PS dosages. With the PS dosage increasing from 1 to 8 g/L at 40 min, the removal rate of TC was 66.70%, 77.80%, 89.90%, 91.30% and 92.20%, respectively. The greatest enhancement of PS-EC process over TCs occurred at 40 min (Figure 9(a)), because the removal rate of TC in EC-PS system could be increased by 55.90%, compared to the pure EC treatment (36.30%). The influence of PS dosage on the removal of OTC and CTC was similar to that of TC (Figure 9(b) and 9(c)). Taking into the operating cost of electrocoagulation process, 5 g/L was selected as the best dosage of pecan shell.

The effect of initial concentration of TCs on the removal efficiency of the coupling method was investigated in Figure 10. The removal rate of TC decreased with the increase of TC concentration. Such results were attributed to the ability of the whole system to treat pollutants was limited under specific current density, PS dosage and reaction time. With the increase of the initial concentration, the removal rate of TC decreases beyond the range of the treatment ability. For the three kinds of TCs (TC, OTC and CTC) with the initial concentration of 250 mg/L, the removal rate could reach 89.90%, 82.92% and 93.85% at 40 min, respectively.

Figure 11 shows the effect of different doses of PS on the UED and UEMD in PS + EC system at different reaction time, 20 and 40 min are selected, respectively. At 20 and 40 min, the UED and UEMD decreased gradually with the increase of PS dosage, and decreased much more at 20 min than 40 min which indicated that PS-EC system had more prominent energy-saving influence in a short time. At the dosage of 5 g/L and reaction time of 40 min, the removal rate reached a high level and UED and UEMD of TC, OTC and CTC decreased obviously: UED decreased by 49.62%, 53.24%, 26.35% and UEMD decreased by 60.60%, 36.68%, 6.62%, respectively. Therefore, in PS-EC coupling treatment system, TCs can be effectively removed, and the UED and UEMD could also be significantly reduced, which means lower consumption of energy and electrode materials.

The optimal conditions for EC + PS coupling process were as follows: current density was 2.5 mA/cm², plate spacing was 3 cm and reaction time was 40 min. Under the best experiment condition, the removal efficiency of TC, OTC and CTC with initial concentration of 250 mg/L could reach 89.90%, 82.92% and 93.85%, respectively. Compared with EC process, the removal efficiency in EC-PS system could be increased by 55.90%, 45.10% and 14.98%, respectively; the UED of the treatment process could be reduced by 49.62%, 53.24% and 26.35%, and the UEMD could be reduced by 49.35%, 80%, 85.65%, 44.37%, respectively. The enhancement of EC + PS system mainly accounts for the synergistic effect of charge neutralization and coordination adsorption. In addition, the sedimentation of floccules produced in EC + PS system could be promoted by PS, which is conductive to the subsequent solid-liquid separation. This study clearly demonstrated that the EC + PS coupling process is an efficient and cheaper technique for TCs removal.

We gratefully thank the National Natural Science Foundation of China (No. 41073063).

All relevant data are included in the paper or its Supplementary Information.