Abstract

Activated carbon made from agricultural waste (walnut shells) was investigated as a suitable adsorbent for effectively removing quinoline from industrial wastewater. The activated carbon was treated with phosphoric acid and oxidized by ammonium persulfate and its ability to adsorb pyridine and quinoline in aqueous solution was investigated. Kinetic parameters for the adsorption process were determined through pseudo-first-order and pseudo-second-order kinetic models and intraparticle diffusion models. Equilibrium experiments and adsorption isotherms were analyzed using Langmuir and Freundlich adsorption isotherms. After reaching equilibrium, the activated carbon adsorbed quinoline in preference to pyridine: the equilibrium adsorptions from individual aqueous solutions (200 μL L−1) of quinoline and pyridine were 166.907 mg g−1 and 72.165 mg g−1, respectively. Thermodynamic studies of quinoline adsorption were conducted at different temperatures and indicated that quinoline adsorption was an endothermic and spontaneous process. The column-adsorption of quinoline and pyridine was consistent with the Thomas model and the Yoon-Nelson model. The removal efficiency of quinoline reached more than 97% for a velocity of 6 mL min−1 at the initial adsorption stage.

INTRODUCTION

Removal of nitrogen heterocyclic compounds (NHCs), which are generated ubiquitously by coking plants, pharmaceutical factories, and related industries, has been extensively studied because of their high priority as environmental contaminants with toxic, carcinogenic and teratogenic properties (Shi et al. 2019). Due to the low biodegradability, the malodorous compound quinoline, as a typical example of an NHC, can be stable in water and soil for long periods of time (Lu et al. 2019). Consequently, it is essential that it is removed from wastewater before it is discharged into the environment.

Quinoline cannot be readily removed by conventional water treatment methods; therefore advanced treatment processes are required, such as microbiological degradation (Liu et al. 2016; Dinamarca et al. 2018), physico-chemical adsorption (Rameshraja & Srivastava 2012), catalytic degradation (Chang et al. 2014; Lovjeet et al. 2018) and biosorption–adsorption of clay minerals or activated carbon with attached microorganisms (Bai et al. 2010; Buchtmann et al. 2015). Among these, the method of immobilization of quinoline-degrading bacteria on different sorption carriers is an environmentally friendly alternative to achieve complete removal of contaminants. Accordingly, the choice of the sorption carriers will play an indispensable and important role.

The main disadvantage of taking clay mineral materials as sorption carriers, such as zeolite, bentonite and montmorillonite, is that competitive interactions of monovalent cations, in particular Na+ and K+, can significantly block the adsorption of NHCs. It has been found that in ion exchange processes, inorganic ions are easier to exchange than organic ions (Ding et al. 2013).

Activated carbon has been used as a biofilm carrier,whose role in the physical adsorption in association with biodegradation was proved in the wastewater process (Scott & Karanjkar 1998; Klimenko et al. 2002). The attached microorganisms breed faster with an increased metabolic activity.

In the present study, low-cost biosorbents for quinoline in water were prepared from walnut shells derived from agricultural waste. The process involved converting the walnut shells into activated carbon, using phosphoric acid as an activating agent, and modification by ammonium persulfate oxidation. The focus of the work was to understand the properties of the activated carbon adsorbent in the absence of a biofilm. The ability of the activated carbon to adsorb quinoline in aqueous solution was investigated with individual solutions of quinoline and pyridine, and with a mixture of the two, in kinetic, equilibrium, thermodynamic and column-adsorption studies.

MATERIALS AND METHODS

Preparation of high-performance walnut shell activated carbon adsorbent

Walnut shells were obtained from a local market. Walnut shell activated carbon was prepared by treatment of the shells with phosphoric acid followed by mild oxidation (Lim et al. 2010; Li et al. 2011). First, the walnut shells were boiled to remove impurities, oven-dried, crushed, and sieved to a particle size of 0.3–0.5 mm. Next, the shells were mixed with 60% phosphoric acid (1:2.5 w/w) and maintained at 60 °C for 48 h. The shells were then carbonized for 80 min at 300 °C and then activated for 80 min at 600 °C in an SX2-4-13 type muffle furnace. The activated carbon obtained was thoroughly washed with 0.1 M hydrochloric acid and distilled water and dried at 105 °C for 24 h. Finally, the resulting activated carbon was oxidized by mixing 1.0 g of walnut shell activated carbon (WSAC) with 20 mL of a 2.0 mol L−1 solution of ammonium persulfate, (NH4)2S2O8, in a glass conical flask and keeping it at 60 °C for 3 h. The oxidized material was filtered, washed with the distilled water, dried in a vacuum oven at 110 °C for 24 h, and then sealed in a vial before use. The oxidized WSAC samples were denoted as OWSAC.

Adsorption kinetic studies of the OWSAC surface

OWSAC (0.1 g) was added to a 100 mL aqueous solution containing 200 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v) in a 250 mL glass flask. The three flasks were shaken simultaneously at the room temperature by a double shaker (TAITEC NR-30, Japan) at 160 rpm for 600 min. Samples of the supernatant (about 0.5 mL) were withdrawn at regular intervals to determine the concentration of the pyridine and/or quinoline. In order to investigate the mechanism of adsorption, data were fitted to Lagergren's pseudo-first-order kinetic model (Karamanis & Assimakopoulos 2007) and pseudo-second-order kinetic model (Parab & Sudersanan 2010) and analyzed with the intraparticle diffusion equation.

Lagergren's pseudo-first-order kinetic model: 
formula
and pseudo-second-order kinetic model: 
formula
where t (min) is the contact time, k1 (min−1) and k2 (g mg−1 min−1) are the adsorption rate constants, and qe and qt (mg g−1) represent the uptake amount of pyridine or quinoline by the adsorbent at equilibrium and time t, respectively.
The limiting step of the adsorption process can be predicting from the diffusion coefficient using a diffusion-based model. The possibility of intraparticle diffusion resistance affecting the adsorption was explored in this study by using the intraparticle diffusion equation (Delchet et al. 2012): 
formula
where t (min) is the contact time, qt (mg g−1) is the quinoline or pyridine uptake amount at time t, Kt (mg g−1 min−1/2) is the intraparticle diffusion rate constant and can be determined from the slopes of linear plots, and C is a constant that indicates the thickness of the boundary layer, i.e. the larger the value of C the greater is the boundary layer effect.

Equilibrium studies and adsorption isotherms

A series of experiments were carried out with different concentrations of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v). Ten different volumes of NHC were used (5 μL increments in the range 5–50 μL). For each volume, the NHC was added to 100 mL of water and 0.1 g of OWSAC in a 250 mL Erlenmeyer flask and the three flasks shaken at 160 rpm. After 12 h the resultant supernatants were withdrawn and the concentration of the NHCs determined.

It was important to establish the most appropriate correlation for the equilibrium adsorption state. The Langmuir and Freundlich adsorption isotherms were used to fit the experimental adsorption results (Ding et al. 2013). The nonlinear forms of these isotherms are given as follows.

Langmuir isotherm: 
formula
Freundlich isotherm: 
formula
where qe (mg g−1) is the amount of NHC adsorbed at equilibrium, Ce (mg L−1) is the equilibrium concentration of NHC. KL (L mg−1) is a constant related to the free energy or net enthalpy of adsorption (KLeΔG/RT) and qm (mg g−1) is the adsorption capacity at the isotherm temperature. KF and n are equilibrium constants indicative of adsorption capacity and adsorption intensity, respectively.

Thermodynamic studies

OWSAC (0.01 g) was added to 10 mL of an aqueous solution containing 50 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline 2:3 v/v) at one of three different temperatures (293, 303, and 313 K) and the temperature maintained for 24 h. Three thermodynamic parameters – standard Gibbs free energy (ΔGo), standard enthalpy (ΔHo) and standard entropy (ΔSo) changes – were determined in order to investigate the thermodynamic nature of the adsorption process. The quantities ΔHo and ΔSo were calculated from the slope and intercept, respectively, of the straight line obtained from plotting lnKd versus 1/T using the following equation (Mohan & Singh 2002): 
formula
where Kd (mL g−1) is the distribution coefficient, R (8.314 J mol−1 K−1) is the gas constant and T (K) is the absolute temperature of the aqueous solution.
After obtaining ΔHo and ΔSo values for the adsorption, ΔGo at each temperature was calculated using the well-known equation: 
formula

Column adsorption model

In order to improve efficiency, practical wastewater treatment usually uses a column packed with adsorbent. Therefore, it was imperative to investigate the dynamic adsorption of quinoline by OWSAC in a packed column. In this study, Thomas and Yoon-Nelson models were employed to estimate the quinoline adsorption efficiency with different velocities of an aqueous solution (200 μL L−1) from a mixture solution (pyridine:quinoline 2:3 v/v). Experiments used 2.5 g of OWSAC and a column with diameter 1.2 cm and height 4.5 cm.

The theoretical models were defined as follows.

Thomas model: 
formula
Yoon-Nelson model: 
formula
where Kth is the Thomas velocity constant (mL min−1 mg−1); q0 is the adsorbance per unit mass of adsorbent to adsorbate at equilibrium (mg g−1); x is adsorbent quantity used to fill the column (g); C0 is the initial concentration of adsorbate (mg L−1); Ct is the effluent concentration of adsorbate (mg L−1); ν is the velocity of the adsorbate through the column (mL min−1); k is the velocity constant (min−1); τ is time needed for 50% of the adsorbate to be adsorbed (min); and t is time of adsorption.

Analysis

In order to ensure reliability and improve accuracy of the experimental data in this study, static and mobile adsorption of pyridine and quinoline were investigated in duplicate and the mean standard deviation determined. All samples were collected by filtering the supernatant through a 0.22 μm mixed cellulose ester membrane (Millipore, Ireland). The filtered supernatant was then analyzed with a double beam visible spectrophotometer (RUNQEE G10, China) and reverse phase high-pressure liquid chromatography (HPLC, ichrom5100, China), using a C18 column (250 mm length and 4.6 mm internal diameter) and UV detection at 254 and 275 nm. The mobile phase consisted of methanol and water in the ratio 75:25 v/v with a flow rate of 1 mL min−1. All of the figures and the data fitting presented in this paper were obtained using the Origin 7.5 program (OriginLab, USA).

RESULTS AND DISCUSSION

The adsorption process and kinetic study

The adsorption mechanism of quinoline and pyridine by OWSAC was investigated by using dynamic adsorption models of pseudo-first-order kinetics, pseudo-second-order kinetics, and intraparticle diffusion. Fitting the experimental data for quinoline and pyridine showed that the adsorption processes were more consistent with pseudo-second-order kinetics rather than pseudo-first-order kinetics (see Table 1 and Figure 1). This finding indicated that the adsorption processes tended to involve chemical reaction accompanied by physical adsorption (Aksu 2002). After 60 minutes, the equilibrium adsorptions of individual solutions of quinoline and pyridine by OWSAC were 166.9 and 72.2 mg g−1 respectively. This suggested that the OWSAC adsorbed quinoline more readily than pyridine in aqueous solution. The corresponding total adsorption from a mixed solution of quinoline and pyridine was 124.29 mg g−1; the individual absorbances of quinoline and pyridine from the mixed solution were 98.6 and 26.75 mg g−1 respectively. The adsorption of quinoline was about 2.5 times that of pyridine. This indicated that there was no significant difference in the adsorptions of quinoline and pyridine by OWSAC in the individual and mixed solutions.

Table 1

The sorption rate constants associated with pseudo-first- and pseudo-second-order kinetic and intraparticle diffusion models from individual solutions of quinoline and pyridine, and a mixed solution of the two: OWSAC (0.1 g) added to a 100 mL aqueous solution containing 200 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v) in a 250 mL glass flask at room temperature under 160 rpm shaking, detected by spectrophotometer and HPLC

Adsorption dynamic equations and parametersQuinolinePyridineMixture of quinoline and pyridine (3:2)
Pseudo-first-order 
k1(min−10.0064 0.0064 0.0076 
 qeexp(mg g−1166.907 ± 0.0658 72.1655 ± 0.0589 124.2910 ± 0.0784 
 qecal(mg g−125.7929 ± 0.0457 17.5914 ± 0.0279 37.446 ± 0.0368 
 R2 0.9086 0.9829 0.9421 
Pseudo-second-order kinetic model 
k2(g·mg−1 min−10.0001 0.0014 0.0242 
 qeexp(mg g−1166.907 ± 0.075 72.1655 ± 0.0653 124.2910 ± 0.0358 
 qecal(mg g−1166.667 ± 0.0684 72.4638 ± 0.0569 125 ± 0.0469 
 R2 0.9998 0.9996 0.9992 
Intraparticle diffusion 
Kt1 127.53 44.0076 44.4518 
 Pseudo-first-order 3.342 1.6 4.4043 
 Pseudo-first-order 0.6741 0.429 0.8835 
Adsorption dynamic equations and parametersQuinolinePyridineMixture of quinoline and pyridine (3:2)
Pseudo-first-order 
k1(min−10.0064 0.0064 0.0076 
 qeexp(mg g−1166.907 ± 0.0658 72.1655 ± 0.0589 124.2910 ± 0.0784 
 qecal(mg g−125.7929 ± 0.0457 17.5914 ± 0.0279 37.446 ± 0.0368 
 R2 0.9086 0.9829 0.9421 
Pseudo-second-order kinetic model 
k2(g·mg−1 min−10.0001 0.0014 0.0242 
 qeexp(mg g−1166.907 ± 0.075 72.1655 ± 0.0653 124.2910 ± 0.0358 
 qecal(mg g−1166.667 ± 0.0684 72.4638 ± 0.0569 125 ± 0.0469 
 R2 0.9998 0.9996 0.9992 
Intraparticle diffusion 
Kt1 127.53 44.0076 44.4518 
 Pseudo-first-order 3.342 1.6 4.4043 
 Pseudo-first-order 0.6741 0.429 0.8835 
Figure 1

The pseudo-second-order kinetic curve of individual solutions of quinoline and pyridine, and a mixed solution of the two: OWSAC (0.1 g) added to a 100 mL aqueous solution containing 200 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v) in a 250 mL glass flask at room temperature under 160 rpm shaking, detected by spectrophotometer and HPLC.

Figure 1

The pseudo-second-order kinetic curve of individual solutions of quinoline and pyridine, and a mixed solution of the two: OWSAC (0.1 g) added to a 100 mL aqueous solution containing 200 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v) in a 250 mL glass flask at room temperature under 160 rpm shaking, detected by spectrophotometer and HPLC.

The above results show that the adsorption property of OWSAC for quinoline displayed a stronger advantage than that of a commercial-grade coconut-based granular activated carbon with a determined quinoline saturated adsorption capacity of 77.82 mg g−1 in an individual quinoline solution under the similar condition (Rameshraja & Srivastava 2012).

Figure 2 shows plots of qt versus t1/2 for the adsorption of individual solutions of quinoline and pyridine, and a mixed solution of the two, by OWSAC at room temperature. These plots showed similar trends, with each adsorption process containing three regions. None of the curves passed through the origin of coordinates for times between 2 and 120 minutes, which indicated that intraparticle diffusion was not the only controlling step for adsorption. Rather, it was the combination of surface adsorption and intraparticle transport within the pores of OWSAC that controlled the adsorption process.

Figure 2

Intraparticle diffusion model analysis of individual solutions of quinoline and pyridine, and a mixed solution of the two: OWSAC (0.1 g) added to a 100 mL aqueous solution containing 200 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v) in a 250 mL glass flask at room temperature under 160 rpm shaking, detected by spectrophotometer and HPLC.

Figure 2

Intraparticle diffusion model analysis of individual solutions of quinoline and pyridine, and a mixed solution of the two: OWSAC (0.1 g) added to a 100 mL aqueous solution containing 200 μL L−1 of either pyridine, quinoline or a mixture of the two (pyridine:quinoline = 2:3 v/v) in a 250 mL glass flask at room temperature under 160 rpm shaking, detected by spectrophotometer and HPLC.

The slope of the linear portion of the plots shown in Figure 2 is the rate constant Kt of intraparticle diffusion (Lorenc-Grabowska & Gryglewicz 2005). The slope increases with the adsorption rate, where Kt1, Kt2, and Kt3 represent the diffusion rate constant for the three stages. The rate constant in the first stage (Kt1) was the biggest, which indicated that external diffusion adsorption was very fast. After saturating the outer surface of the adsorbent particles, quinoline and pyridine entered the inner pores of the OWSAC and were adsorbed by particles of the inner surface. Meanwhile, the diffusional resistance gradually increased as quinoline and pyridine diffused into the inner surface of the OWSAC, and the intraparticle diffusion speed decreased until it reached the adsorption equilibrium. Consequently, the rate constant in the second stage (Kt2) was smaller than in the first stage, and smallest of all in the third stage (Kt3).

Equilibrium studies and adsorption isotherms

In order to obtain the equilibrium isotherm, 1 g of adsorbant was used and the initial concentrations of NHC were varied from 50 to 500 μL L−1. In each case, the amounts of adsorbed quinoline and pyridine were determined.

Figure 3 shows the application of nonlinear Langmuir and Freundlich isotherms for the adsorption of quinoline and pyridine, from separate solutions, at 313 K. Figure 4 shows the quinoline adsorption for a mixed solution of quinoline and pyridine at 293, 303, and 313 K. It can be seen that the adsorption processes were in agreement with the Freundlich isotherm rather than Langmuir isotherm.

Figure 3

Nonlinear Langmuir (dashed line) and Freundlich (solid line) isotherms analysis of quinoline and pyridine adsorption, from separate solutions, shaken under 160 rpm at 313 K for 12 h: 1 g of adsorbant used and the initial concentrations of NHC varied from 50 to 500 μL L−1. In each case, the amounts of adsorbed quinoline and pyridine were determined at 275 nm and 254 nm wavelength by spectrophotometer.

Figure 3

Nonlinear Langmuir (dashed line) and Freundlich (solid line) isotherms analysis of quinoline and pyridine adsorption, from separate solutions, shaken under 160 rpm at 313 K for 12 h: 1 g of adsorbant used and the initial concentrations of NHC varied from 50 to 500 μL L−1. In each case, the amounts of adsorbed quinoline and pyridine were determined at 275 nm and 254 nm wavelength by spectrophotometer.

Figure 4

Nonlinear Langmuir (dashed line) and Freundlich (solid line) isotherms analysis of the quinoline adsorption for a mixed solution of quinoline and pyridine (pyridine:quinoline = 2:3 v/v) shaken at 293, 303, and 313 K for 12 h to reach adsorption equilibrium: 1 g of adsorbant used and the initial concentrations of NHC varied from 50 to 500 μL L−1; then quinoline concentration was measured at 275 nm wavelength by HPLC.

Figure 4

Nonlinear Langmuir (dashed line) and Freundlich (solid line) isotherms analysis of the quinoline adsorption for a mixed solution of quinoline and pyridine (pyridine:quinoline = 2:3 v/v) shaken at 293, 303, and 313 K for 12 h to reach adsorption equilibrium: 1 g of adsorbant used and the initial concentrations of NHC varied from 50 to 500 μL L−1; then quinoline concentration was measured at 275 nm wavelength by HPLC.

The values of the monolayer saturation capacity qm, calculated from the Langmuir isotherm, were approximately 294 mg g−1 for both quinoline and pyridine adsorptions from separate solutions. This result shows superiority compared with the maximum quinoline adsorption capacity of 56.6 mg g−1 obtained by the Langmuir model for the sulfuric acid-treated activated carbon from coconut shell (Ferreira et al. 2019).

Subsequently, in order to assess whether the two single adsorption processes were consistent with the Langmuir or the Freundlich isotherms, a chi-square analysis was used to estimate the degree of difference (χ2) between the experimental and theoretical isotherms. The equation used for chi-square analysis (Mirmohseni et al. 2012) was: 
formula
where qecal (mg g−1) is the equilibrium uptake amount calculated from the isotherm and qeexp (mg g−1) is the experimental equilibrium uptake. A smaller χ2 value indicated a better fitting isotherm.

From Table 2, it is clear that not only were the r2 values closer to unity for the Freundlich isotherm, but also the χ2 values for the Freundlich isotherm were much smaller, implying that the individual adsorption processes of quinoline and pyridine on OWSAC were multilayer adsorption processes rather than monolayer adsorption.

Table 2

Langmuir and Freundlich adsorption isotherm constants analysis of individual solutions of quinoline and pyridine shaken under 160 rpm at 313 K for 12 h : 1 g of OWSAC used and the initial concentrations of the individual solution varied from 50 to 500 μL L−1. In each case, the amounts of adsorbed quinoline and pyridine were determined at 275 nm and 254 nm wavelength by spectrophotometry respectively

QuinolinePyridine
Langmuir qm(mg g−1294.4007 ± 0.5000 294.3438 ± 0.4000 
KL 0.0180 ± 0.0048 0.0019 ± 0.0002 
R2 0.8683 0.9961 
χ2 152.0799 1.5533 
Freundlich KF 38.0589 ± 0.42 1.8107 ± 0.48 
n 2.89 ± 0.039 1.399 ± 0.013 
R2 0.9987 0.9995 
χ2 0.4029 0.1263 
QuinolinePyridine
Langmuir qm(mg g−1294.4007 ± 0.5000 294.3438 ± 0.4000 
KL 0.0180 ± 0.0048 0.0019 ± 0.0002 
R2 0.8683 0.9961 
χ2 152.0799 1.5533 
Freundlich KF 38.0589 ± 0.42 1.8107 ± 0.48 
n 2.89 ± 0.039 1.399 ± 0.013 
R2 0.9987 0.9995 
χ2 0.4029 0.1263 

Adsorption thermodynamic parameter

The thermodynamic parameters of quinoline adsorbed from an aqueous quinoline solution and from an aqueous solution of quinoline and pyridine (pyridine:quinoline = 2:3 v/v) are shown in Table 3. Similar thermodynamic parameters for quinoline adsorption were found for both solutions, indicating that the adsorption processes for quinoline were not affected by the presence of pyridine. The endothermic nature of the adsorption process was shown by the positive value of ΔH°; however, increased randomness at the solid/solution interface, with some structural changes in the adsorbate and adsorbent system for quinoline, was indicated by the positive value of ΔS°. The negative value of ΔG° suggested that the adsorption process was spontaneous.

Table 3

Thermodynamic parameters for quinoline adsorption from individual quinoline solution and a mixture solution of quinoline and pyridine (pyridine:quinoline 2:3 v/v) onto OWSAC shaken at three different temperatures (293, 303, and 313 K) for 24 h (C0individual = 54.65 mg L−1, C0mixed = 32.79 mg L−1, mOWSAC=1 g L−1)

T(K)Kd (L g−1)
ΔG° (kJ mol−1)
ΔH° (kJ mol−1)
ΔS° (J mol−1 K−1)
individualmixedindividualmixedindividualmixedindividualmixed
293 17.3824 ± 0.061 14.9912 ± 0.055 −6.9632 ± 0.054 −6.9513 ± 0.018 5.9612 ± 0.0537 14.1801 ± 0.026 44.0623 ± 0.034 71.0244 ± 0.037 
303 18.7012 ± 0.052 18.8804 ± 0.061 −7.3847 ± 0.062 −7.0245 ± 0.028     
313 20.3125 ± 0.018 21.7636 ± 0.064 −7.8441 ± 0.057 −8.1536 ± 0.048     
T(K)Kd (L g−1)
ΔG° (kJ mol−1)
ΔH° (kJ mol−1)
ΔS° (J mol−1 K−1)
individualmixedindividualmixedindividualmixedindividualmixed
293 17.3824 ± 0.061 14.9912 ± 0.055 −6.9632 ± 0.054 −6.9513 ± 0.018 5.9612 ± 0.0537 14.1801 ± 0.026 44.0623 ± 0.034 71.0244 ± 0.037 
303 18.7012 ± 0.052 18.8804 ± 0.061 −7.3847 ± 0.062 −7.0245 ± 0.028     
313 20.3125 ± 0.018 21.7636 ± 0.064 −7.8441 ± 0.057 −8.1536 ± 0.048     

Thomas and Yoon-Nelson models analysis of dynamic adsorption

In this study, the column-adsorption of quinoline and pyridine from a mixed solution (quinoline:pyridine = 3:2 v/v) with an initial concentration of 200 μL L−1 using 2.5 g of OWSAC with column diameter 1.2 cm and height 4.5 cm was consistent with the Thomas and Yoon-Nelson models (Qu et al. 2019). Linear plots were obtained by plotting ln[C0/(CtC0)] and ln[Ct/(C0Ct)] against t, according to the two models respectively (see Figure 5). Figure 6 shows that the removal efficiency of quinoline from the mixed solution decreased with an increase in time and velocity of flow.

Figure 5

(a) Thomas and (b) Yoon-Nelson models analysis of the quinoline adsorption with different velocities of an aqueous solution (200 μL L−1) from a mixture solution (pyridine:quinoline 2:3 v/v). Experiments used 2.5 g of OWSAC and a column with diameter 1.2 cm and height 4.5 cm.

Figure 5

(a) Thomas and (b) Yoon-Nelson models analysis of the quinoline adsorption with different velocities of an aqueous solution (200 μL L−1) from a mixture solution (pyridine:quinoline 2:3 v/v). Experiments used 2.5 g of OWSAC and a column with diameter 1.2 cm and height 4.5 cm.

Figure 6

The removal efficiency of quinoline and pyridine with different velocities of an aqueous solution (200 μL L−1) from a mixture solution (pyridine:quinoline 2:3 v/v). Experiments used 2.5 g of OWSAC and a column with diameter 1.2 cm and height 4.5 cm.

Figure 6

The removal efficiency of quinoline and pyridine with different velocities of an aqueous solution (200 μL L−1) from a mixture solution (pyridine:quinoline 2:3 v/v). Experiments used 2.5 g of OWSAC and a column with diameter 1.2 cm and height 4.5 cm.

The removal efficiency of quinoline reached more than 97% for a velocity of 6 mL min−1 and 93% for 8 mL min−1 in the initial adsorption stage. In contrast, the corresponding efficiencies for pyridine were much lower: approximately 65% and 58%, respectively (data not shown). Benefield et al. (1982) proposed the idea of dynamic adsorption, with saturated adsorption being reached when the concentration of adsorbate in the effluent was 0.95 times that of the import solution. In our study, it took about 6 h for quinoline adsorption to reach equilibrium at a velocity 8 mL min−1; this was about 2.6 times the adsorption rate for pyridine. These findings suggested that OWSAC adsorbed quinoline more effectively than pyridine.

CONCLUSION

In this study, OWSAC (prepared from walnut shells as agricultural waste, modified with a weak acid activator and a weak oxidizer) exhibited significant adsorption capacity for quinoline in aqueous solution in a physicochemical multilayer adsorption process. The rapid adsorption process fitted well with the pseudo-second-order kinetic model with the equilibrium uptake amount for quinoline and pyridine of 166.9 mg g−1 and 72.2 mg g−1 respectively. The good correlation coefficients with low χ2 suggest that the adsorption of quinoline and pyridine on OWSAC could be best described by the Freundlich adsorption isotherm.

Intraparticle diffusion studies showed that the surface adsorption and intraparticle transport within the pores of OWSAC controlled the adsorption process. Thermodynamic parameters were evaluated using equilibrium constants determined at different temperatures. The value of ΔG° was found to be negative at all temperatures, indicating the spontaneity of the adsorption process. Positive values of ΔH° and ΔS° showed the endothermic nature and increase in disorder associated with quinoline adsorption. The column-adsorption process of quinoline and pyridine was consistent with the Thomas and Yoon-Nelson models. The removal efficiency of quinoline reached more than 97% for a velocity of 6 mL min−1.

REFERENCES

REFERENCES
Bai
Y. H.
,
Sun
Q. H.
,
Xing
R.
,
Wen
D. H.
&
Tang
X. Y.
2010
Removal of pyridine and quinoline by bio-zeolite composed of mixed degrading bacteria and modified zeolite
.
Journal of Hazardous Materials
181
(
1
),
916
922
.
Benefield
L. D.
,
Judkins
J. F.
&
Weand
B. L.
1982
Process Chemistry for Water and Wastewater Treatment
.
Prentice-Hall, Englewood Cliffs
,
NJ, USA
.
Chang
L.
,
Zhang
Y. M.
,
Gan
L.
,
Xu
H.
,
Yan
N.
,
Liu
R.
&
Rittmann
B. E.
2014
Internal loop photo-biodegradation reactor used for accelerated quinoline degradation and mineralization
.
Biodegradation
25
(
4
),
587
594
.
Delchet
C.
,
Tokarev
A.
,
Dumail
X.
,
Toquer
G.
,
Barre
Y.
,
Guari
Y.
,
Guerin
C.
,
Larionova
J.
&
Grandjean
A.
2012
Extraction of radioactive cesium using innovative functionalized porous materials
.
RSC Advances
2
(
13
),
5707
5716
.
Ding
D. H.
,
Zhao
Y. X.
,
Yang
S. J.
,
Shi
W. S.
,
Zhang
Z. Y.
,
Lei
Z. F.
&
Yang
Y. N.
2013
Adsorption of cesium from aqueous solution using agricultural residue – walnut shell: equilibrium, kinetic and thermodynamic modeling studies
.
Water Research
47
(
7
),
2563
2571
.
Ferreira
M. E. O.
,
Vaz
B. G.
,
Borba
C. E.
,
Alonsoa
C. G.
&
Ostroski
I. C.
2019
Modified activated carbon as a promising adsorbent for quinoline removal
.
Microporous and Mesoporous Materials
277
,
208
216
.
Klimenko
N.
,
Winther-Nielsen
M.
,
Smolin
S.
,
Nevynna
L.
&
Sydorenko
J.
2002
Role of the physico-chemical factors in the purification process of water from surface-active matter by biosorption
.
Water Research
36
,
5132
5140
.
Lim
W. C.
,
Srinivasakannan
C.
&
Balasubramanian
N.
2010
Activation of palm shells by phosphoric acid impregnation for high yielding activated carbon
.
Journal of Analytical and Applied Pyrolysis
88
(
2
),
181
186
.
Liu
J.
,
Yu
Y.
,
Chang
Y.
,
Li
B. K.
,
Bian
D. J.
,
Yang
W.
,
Huo
H. L.
,
Huo
M. X.
&
Zhu
S. Y.
2016
Enhancing quinoline and phenol removal by adding Comamonas testosteroni bdq06 in treatment of an accidental dye wastewater
.
International Biodeterioration & Biodegradation
115
,
72
84
.
Lorenc-Grabowska
E.
&
Gryglewicz
G.
2005
Adsorption of lignite-derived humic acids on coal-based mesoporous activated carbons
.
Journal of Colloid & Interface Science
284
(
2
),
416
423
.
Lovjeet
S.
,
Pawan
R.
&
Shri
C.
2018
Comparative evaluation of synthesis routes of Cu/zeolite Y catalysts for catalytic wet peroxide oxidation of quinoline in fixed-bed reactor
.
Journal of Environmental Management
215
,
1
12
.
Lu
Q. Y.
,
Zhang
C. Y.
,
Wang
W. Y.
,
Yuan
B. Y.
,
Zhang
Y. M.
&
Rittmann Bruce
E.
2019
Bioavailable electron donors leached from leaves accelerate biodegradation of pyridine and quinoline
.
Science of the Total Environment
654
,
473
479
.
Mirmohseni
A.
,
Seyed Dorraji
M. S.
,
Figoli
A.
&
Tasselli
F.
2012
Chitosan hollow fibers as effective biosorbent toward dye: preparation and modeling
.
Bioresource Technology
121
,
212
220
.
Qu
J. J.
,
Song
T.
,
Liang
J. S.
,
Bai
X.
,
Li
Y.
,
Wei
Y. N.
,
Huang
S. Q.
,
Dong
L. Y.
&
Jin
Y.
2019
Adsorption of lead (П) from aqueous solution by modified Auricularia matrix waste: a fixed-bed column study
.
Ecotoxicology and Environmental Safety
169
,
722
729
.
Rameshraja
D.
&
Srivastava
V. C.
2012
Quinoline adsorption onto granular activated carbon and bagasse fly ash
.
Chemical Engineering Journal
181–182
(
182
),
343
351
.