Adsorption of Nitrate by a Novel Polyacrylic Anion Exchange Resin from Water with Dissolved Organic Matters: Batch and Column Study

Featured Application: The novelty of this work is that the prepared resin AEE-3 has a potential application in the removal of nitrate from real secondary treated wastewater containing dissolved organic matters. Abstract: A novel anion exchange resin AEE-3 was synthesized by N-alkylation of a weakly basic polyacrylic anion exchanger D311 with 1-bromopropane to e ﬀ ectively remove nitrate (NO 3 − -N) from aqueous solution. The related ﬁnding revealed that its adsorption isotherm obeyed the Langmuir model well, and the second-order model was more validated for the NO 3 − -N adsorption kinetics study. Compared to commercially-available polystyrene-based nitrate specialty resin Purolite A 520E (A520E), AEE-3 resin has a higher adsorbed amount and better regeneration performance toward NO 3 − -N in the existence of dissolved organic matter (DOM) using static and dynamic methods. Notably, a real secondary treated wastewater (STWW) obtained from a local municipal wastewater treatment plant was also assessed for NO 3 − -N removal in ﬁxed-bed columns. Observations from this study indicated that AEE-3 could e ﬀ ectively remove NO 3 − -N from contaminated surface water.


Introduction
With the rapid development of industry and agriculture, excessively-used nutrients such as nitrogen (N) and phosphorous (P) in crop production have been infiltrated into rivers, lakes and ponds, leading to the eutrophication of aquatic systems [1,2]. Nowadays, nitrate (NO 3 − -N) pollution in surface and ground water is a severe environmental problem due to its carcinogenicity, toxicity and potential hazards to human health and livestock. Therefore, the World Health Organization (WHO) set the NO 3 and can engender a strong affinity for resins via electrostatic attraction, hydrophobic interaction and so on [12][13][14]. Hence, AERs are more sensitive to fouling with DOMs, which would compete with NO 3 − -N for the active sites and cause the decrease in NO 3 − -N adsorption. More seriously, coexisting DOM could lead to an irreversible fixation to anion exchangers. Polyacrylic AERs with hydrophilic structures reveal a stronger resistance to organic fouling performance than polystyrene AERs [15]. Moreover, acrylic-based resins possess unique physiochemical properties, rapid adsorption rates and high ion-exchange capacities [16,17]. Accordingly, polyacrylic AERs could serve as a promising absorbent to effectively remove NO 3 − -N from polluted water in the presence of DOMs. Nevertheless, few works have shown the feasibility of polyacrylic resins for NO 3 − -N removal in real STWW containing NO 3 − -N at high levels.
However, competitive anions such as chloride (Cl − ) or sulfate (SO 4 2− ) exist widely in water, which can weaken NO 3 − -N uptake efficacy of AERs [18]. To circumvent this limitation, studies have been devoted to preparing a novel anion exchanger with prominent selectivity toward NO 3 − -N.
The long alkyl chain addition at the exchange sites of resins could be of great benefit for the preferable selectivity toward monovalent anions because of its steric hindrance and hydrophobic effect [19,20]. Hence, it is intriguing for us to modify the weakly basic resin by introducing long alkyl groups around the active amine groups, which can easily form strongly basic functional groups (quaternary ammonium) which interact with NO 3 − -N through strong electrostatic attraction.
Herein, a weakly basic polyacrylic anion exchanger D311 was modified with 1-bromopropane (C 3 H 7 Br) to prepare a strongly basic anion exchanger AEE-3 with high efficiency and brilliant selectivity for NO 3 − -N uptake. The adsorption kinetics, isotherms, influence of pH value and competitive anions were investigated. In addition, commercial nitrate-selective polystyrene resin Purolite A 520E (A520E) was chosen to make a comparison with AEE-3 in both batch and fixed-bed column experiments, further demonstrating that AEE-3 is more resistant to fouling by DOMs (humic acid, tannic acid and sodium dodecyl benzene sulfonate). Also, a dynamic experiment was studied to evaluate the adsorption performance of AEE-3 toward NO 3 − -N in real STWW.

Material
A commercially-available weakly basic anion exchanger D311 and strongly basic anion exchanger Purolite A 520E (A520E) were obtained from Huizhu Resin Co., Ltd. (Shanghai, China) and Purolite Co., Ltd. (Huzhou, China), respectively. Tannic acid (TA), humic acid (HA) and sodium dodecyl benzene sulfonate (SDBS) were purchased from Sinopharm Chemical Reagent Company (Shanghai, China). Other chemicals in this work were of analytical reagent acquired from Nanjing Chemical Reagent Co., Ltd. (Nanjing, China). STWW samples with a high NO 3 − -N concentration were obtained from a local municipal wastewater treatment plant in Nanjing, China.

Preparation and Characterization of AEE-3
D311 resin was firstly rinsed successively with 1 mol/L HCl and 1 mol/L NaOH in turn to remove impurities, and then washed with deionized water until neutral. Finally, resin was washed by ethyl alcohol and then dried at 323 K for 8 h.
The synthesis method of AEE-3 is the following (Scheme 1): 10 g of dried D311 resin was swollen in 100 mL of acetonitrile for 2 h. Afterwards, 20 mmol of KI and 20 mmol C 3 H 7 Br were added separately and the mixture maintained with stirring at 333 K for 48 h. After alkylation, the obtained resin beads were rinsed successively with dilute HCl, distilled water and dilute NaOH to attain a neutral pH. Finally, the product was extracted with anhydrous ethanol and dried at 323 K for 8 h. The obtained resin AEE-3 was characterized by Fourier transform infrared spectrometer (Nicolet 5700, Madison, USA) and N 2 adsorption analysis (ASAP-2010C, Micromeritics, Norcross, GA, USA). Appl. Sci. 2019, 9,

Batch Adsorption Experiments
Isothermal adsorption experiments of NO3 − -N on adsorbents were performed when 0.05 g resin was introduced to each 150 mL conical flasks containing 50 mL of NO3 − -N in solution (10 to 60 mg/L concentration). To investigate the adsorption kinetics, 0.2 g resin was immersed into 200 mL NO3 − -N solution with the initial concentration of 60 mg/L, and samples were withdrawn at different time intervals. The pH (4.0-10.0) with 50 mg/L NO3 − -N solution in the presence of Cl − (200 mg/L) was adjusted using 0.1 mol/L HCl or 0.1 mol/L NaOH solution, respectively, as required. A competing anion (SO4 2− or Cl − ) was added to the NO3 − -N solution to study the effect of negative ions on adsorption capacities. The static tests of D311 and AEE-3 were conducted in a bi-solute system (HA/NO3 − , SDBS/NO3 − or TA/NO3 − ) for four adsorption-desorption cycles. Analyses of NO3 − -N were obtained by measuring the UV1800 spectrophotometer at a wavelength of 220 nm. The adsorption amount Qe (mg/g) was calculated using Equation (1): where C0 and Ce are the initial and equilibrium concentrations (mg/L) of NO3 − -N, V is the volume of solution (L) and W represents the weight of dry resin (g).

Column Adsorption and Desorption
The experiments were carried out using a glass column (Φ18 × 200 mm) with 5 mL of dry fresh resin and a peristaltic pump to ensure the constant flow rate under 298 K. The initial concentration of NO3 − -N was 20 mg/L, containing 1 mg/L TA, 1 mg/L HA and 2 mg/L SDBS. The superficial liquid velocity (SLV) was 30 bed volume per hour (BV/h), and the breakthrough curve was obtained by analysis of different time intervals of the effluent. After adsorption, the elution of NO3 − -N from the resin was performed using 0.6 mol/L NaCl solution. The adsorption-desorption cycle was repeated three times to evaluate the regeneration efficiency of resins. The desorption rate (D) of NO3 − -N was defined by Equation (2): where Ca is the amount of NO3 − -N adsorbed on resins and Cd is the amount of NO3 − -N desorbed from resins by NaCl solution.

Characterization
The adsorption-desorption isotherms and the pore size distribution curve of D311 and AEE-3 are displayed in Figure 1. The isotherm plot of two resins can be considered type IV with a hysteresis loop in the relative pressure range of 0.8-1.0, hinting at the existence of both mesopores and macropores in these resins. Table 1 lists the essential properties of D311 and AEE-3.
The FT-IR spectra of D311 and AEE-3 samples are shown in Figure 2. Notably, two bands with peaks at 1366 cm −1 and 2930 cm −1 are ascribed to asymmetric bands of the alkyl groups (-CH2-and-CH3), and the peak at 1478 cm −1 is assigned to C-N group. The appearance of a new band at 1407 cm −1 Scheme 1. The alkylation reaction of C 3 H 7 Br onto D311 resin. by measuring the UV1800 spectrophotometer at a wavelength of 220 nm. The adsorption amount Q e (mg/g) was calculated using Equation (1):

Batch Adsorption Experiments
where C 0 and C e are the initial and equilibrium concentrations (mg/L) of NO 3 − -N, V is the volume of solution (L) and W represents the weight of dry resin (g).

Column Adsorption and Desorption
The experiments were carried out using a glass column (Φ18 × 200 mm) with 5 mL of dry fresh resin and a peristaltic pump to ensure the constant flow rate under 298 K. The initial concentration of NO 3 − -N was 20 mg/L, containing 1 mg/L TA, 1 mg/L HA and 2 mg/L SDBS. The superficial liquid velocity (SLV) was 30 bed volume per hour (BV/h), and the breakthrough curve was obtained by analysis of different time intervals of the effluent. After adsorption, the elution of NO 3 − -N from the resin was performed using 0.6 mol/L NaCl solution. The adsorption-desorption cycle was repeated three times to evaluate the regeneration efficiency of resins. The desorption rate (D) of NO 3 − -N was defined by Equation (2): where C a is the amount of NO 3 − -N adsorbed on resins and C d is the amount of NO 3 − -N desorbed from resins by NaCl solution.

Characterization
The adsorption-desorption isotherms and the pore size distribution curve of D311 and AEE-3 are displayed in Figure 1. The isotherm plot of two resins can be considered type IV with a hysteresis loop in the relative pressure range of 0.8-1.0, hinting at the existence of both mesopores and macropores in these resins. Table 1 lists the essential properties of D311 and AEE-3.
The FT-IR spectra of D311 and AEE-3 samples are shown in Figure 2. Notably, two bands with peaks at 1366 cm −1 and 2930 cm −1 are ascribed to asymmetric bands of the alkyl groups (-CH 2 -and-CH 3 ), and the peak at 1478 cm −1 is assigned to C-N group. The appearance of a new band at 1407 cm −1 Appl. Sci. 2019, 9, 3077 4 of 14 is associated with -CH 2 -of-CH 2 -N + R 3 group [7]. On the other side, the fact that the strong base ion-exchange capacity of the resin increased from 0 to 3.42 mmol/g implied that propyl groups were successfully introduced onto the primary amine after alkylation. These observations proved AEE-3 had been synthesized successfully. is associated with -CH2-of-CH2-N + R3 group [7]. On the other side, the fact that the strong base ionexchange capacity of the resin increased from 0 to 3.42 mmol/g implied that propyl groups were successfully introduced onto the primary amine after alkylation. These observations proved AEE-3 had been synthesized successfully.    Figure 3 describes the effect of contact time on NO3 − -N uptake by AEE-3 and A520E at 298 K, and all the data were simulated using two commonly used kinetic models [21,22]:

Adsorption Kinetics
Pseudo-first-order: The initial adsorption rate was calculated by Equation (5): where Qe (mg/g) is the equilibrium adsorption capacity of NO3 − -N, and Qt (mg/g) refers to the amounts of NO3 − -N in resins at time t (min). k1 (1/min) and k2 (g/(mg·min)) represent the rate constants for the pseudo-first-and pseudo-second-order kinetic models, respectively. h0 is the initial adsorption rate (mg/(g·min)). The kinetic parameters, together with the coefficient of determination (R 2 ), are listed in Table 2. It is evident that the adsorption behavior of NO3 − -N onto both resins was preferable to fit the pseudo-second-order based on the R 2 values. Obviously, AEE-3 exhibited a greater capacity for NO3 − -N than A520E, illustrating that the introduction of alkyl chains would form quaternary ammonium groups which can easily interact with NO3 − -N via electrostatic attraction. Finally, a larger h0 value of AEE-3 was obtained, and the difference in the adsorption rate is related to the difference of the resin matrix.   and all the data were simulated using two commonly used kinetic models [21,22]:

Adsorption Kinetics
Pseudo − sec ond − order : The initial adsorption rate was calculated by Equation (5): where Q e (mg/g) is the equilibrium adsorption capacity of NO 3 − -N, and Q t (mg/g) refers to the amounts of NO 3 − -N in resins at time t (min). k 1 (1/min) and k 2 (g/(mg·min)) represent the rate constants for the pseudo-first-and pseudo-second-order kinetic models, respectively. h 0 is the initial adsorption rate (mg/(g·min)). The kinetic parameters, together with the coefficient of determination (R 2 ), are listed in

Effect of Coexisting Anions
In general, some coexisting anions in water would strongly compete with NO3 − -N to occupy active sites via electrostatic interaction. Thus, the selective absorption of resin is of vital importance in evaluating the practical application, as well as in considering the absorption performance. A520E is known for its good selectivity toward NO3 − -N, and its adsorption capacities are still relatively high in the present of Cl − or SO4 2− [20]. As displayed in Figure 4, a slightly higher adsorption amount of AEE-3 was detected compared to A520E with an identical addition of anions. The phenomena may be rationalized by the fact that the long alkyl chain on AEE-3 is conducive to the adsorption selectivity for NO3 − -N with lower hydration energy than Cl − or SO4 2− .

Effect of Coexisting Anions
In general, some coexisting anions in water would strongly compete with NO 3 − -N to occupy active sites via electrostatic interaction. Thus, the selective absorption of resin is of vital importance in evaluating the practical application, as well as in considering the absorption performance. A520E is known for its good selectivity toward NO 3 − -N, and its adsorption capacities are still relatively high in the present of Cl − or SO 4 2− [20]. As displayed in Figure 4, a slightly higher adsorption amount of AEE-3 was detected compared to A520E with an identical addition of anions. The phenomena may be rationalized by the fact that the long alkyl chain on AEE-3 is conducive to the adsorption selectivity for NO 3 − -N with lower hydration energy than Cl − or SO 4 2− .
in evaluating the practical application, as well as in considering the absorption performance. A520E is known for its good selectivity toward NO3 − -N, and its adsorption capacities are still relatively high in the present of Cl − or SO4 2− [20]. As displayed in Figure 4, a slightly higher adsorption amount of AEE-3 was detected compared to A520E with an identical addition of anions. The phenomena may be rationalized by the fact that the long alkyl chain on AEE-3 is conducive to the adsorption selectivity for NO3 − -N with lower hydration energy than Cl − or SO4 2− .

Effect of pH
The primary mechanism for NO 3 − -N uptake onto a strongly basic anion exchanger can be explained by electrostatic interaction or columbic force [23]. Figure

Effect of pH
The primary mechanism for NO3 − -N uptake onto a strongly basic anion exchanger can be explained by electrostatic interaction or columbic force [23].

Adsorption Isotherms
The isotherms of NO3 − -N adsorption by AEE-3 are displayed in Figure 6. The experimental data were fitted by both Langmuir and Freundlich equations which are represented as follows [25,26]: where Ce (mg/L) is the concentration at equilibrium state, Qe (mg/g) is the adsorption capacity at equilibrium state, Qm (mg/g) is the maximum adsorption capacity, KL is the Langmuir constant and KF and n are the Freundlich constants. The R 2 of the isothermal fit are presented in Table 3. In view of

Adsorption Isotherms
The isotherms of NO 3 − -N adsorption by AEE-3 are displayed in Figure 6. The experimental data were fitted by both Langmuir and Freundlich equations which are represented as follows [25,26]: Appl. Sci. 2019, 9, 3077 8 of 14 where C e (mg/L) is the concentration at equilibrium state, Q e (mg/g) is the adsorption capacity at equilibrium state, Q m (mg/g) is the maximum adsorption capacity, K L is the Langmuir constant and K F and n are the Freundlich constants. The R 2 of the isothermal fit are presented in Table 3.
In view of the R 2 values, the equilibrium adsorption of NO 3 − -N onto AEE-3 was better described by the Langmuir isotherm, which coincided with the prevalent ion-exchange mechanism [7,23]. Moreover, the Q m values of AEE-3 dropped with the increase of temperature, revealing that the NO 3 − -N adsorption by the resin was an exothermic process, and therefore, reducing the temperature benefited NO 3 − -N uptake.
The maximum NO 3 − -N adsorption amount of AEE-3, as evaluated according to the Langmuir model, was compared with other materials including chitosan, activated carbon and commercial anion exchange resins (A520E and D201), etc. As shown in Table 4, the Q m of AEE-3 for NO 3 − -N uptake was higher than that of these commercially available materials, signifying that the novel anion exchange resin can act as a potential material for the technological application of NO 3 − -N removal from water.
Appl. Sci. 2019, 9, x FOR PEER REVIEW 8 of 13 Figure 6. Equilibrium isotherm for NO3 − -N adsorbed using AEE-3.  Figure 7 shows the adsorption amounts of NO3 − -N using AEE-3 and A520E in a bi-solute system (HA/NO3 − , SDBS/NO3 − or TA/NO3 − ) for four adsorption-desorption cycles. It was seen that the   Figure 7 shows the adsorption amounts of NO 3 − -N using AEE-3 and A520E in a bi-solute system (HA/NO 3 − , SDBS/NO 3 − or TA/NO 3 − ) for four adsorption-desorption cycles. It was seen that the adsorption capacities of A520E for NO 3 − -N distinctly declined over the four cycles. DOMs can occupy more active sites through hydrophobic and electrostatic interactions on the A520E resin with a polystyrene matrix [12]. By contrast, AEE-3 exhibited greater NO 3 − -N adsorption in bi-solute systems, and the reduction of NO 3 − -N uptake was slight with increasing the number of adsorption-desorption cycles. This result validated the hypothesis that AEE-3 has a strong resistance to interferential organics due to its hydrophilic matrix. Besides, over 94% of the adsorbed NO 3 − -N on AEE-3 was desorbed, while the desorption rates of A520E in the presence of organic matters declined gradually with the number of adsorption-desorption cycle (Figure 8). In addition, Figure 9 presents the desorption rates of DOMs regenerated by a NaCl solution, and the low values for A520E resin can be attributed to the strong affinity between the polystyrene matrix and the hydrophobic organics, causing the DOMs to not be removed fully by regeneration. Hence, these results demonstrate an improved performance on adsorption/regeneration of AEE-3 compared to A520E.

Effect of DOMs on Static Adsorption and Desorption
Appl. Sci. 2019, 9, x FOR PEER REVIEW 9 of 13 causing the DOMs to not be removed fully by regeneration. Hence, these results demonstrate an improved performance on adsorption/regeneration of AEE-3 compared to A520E.

Effect of DOMs on Column Mode Experiments
It is imperative to test the utilization efficiency of adsorbents in a dynamic fixed-bed experiment related to real operating systems [32]. The influence of three organics on NO3 − -N adsorption was studied using an influent with constant NO3 − -N concentration of 20 mg/L in the presence of HA, TA and SDBS. As can be seen in Figure 10, the breakthrough curves (Run 2 and Run 3) for NO3 − -N adsorption onto AEE-3 have a close coincidence with the original breakthrough curve (Run 1). Nevertheless, a large separation between the three breakthrough curves was observed for NO3 − -N adsorption onto A520E, which is indicative of a limited reusability of A520E towards NO3 − -N solution containing DOMs.

Effect of DOMs on Column Mode Experiments
It is imperative to test the utilization efficiency of adsorbents in a dynamic fixed-bed experiment related to real operating systems [32]. The influence of three organics on NO 3 − -N adsorption was studied using an influent with constant NO 3 − -N concentration of 20 mg/L in the presence of HA, TA and SDBS. As can be seen in Figure 10 studied using an influent with constant NO3 -N concentration of 20 mg/L in the presence of HA, TA and SDBS. As can be seen in Figure 10, the breakthrough curves (Run 2 and Run 3) for NO3 − -N adsorption onto AEE-3 have a close coincidence with the original breakthrough curve (Run 1). Nevertheless, a large separation between the three breakthrough curves was observed for NO3 − -N adsorption onto A520E, which is indicative of a limited reusability of A520E towards NO3 − -N solution containing DOMs.

Assessments of Practical Application
Conventional heterotrophic denitrification techniques are severely restricted by the low C/N ratio of wastewater, thereby causing the high NO3 − -N content in STWW [33,34]. To further assess the practical application of AEE-3, the dynamic column test of NO3 − -N uptake onto AEE-3 was performed using a real STWW effluent which was collected from a local sewage treatment plant, containing the following quality parameters: pH, 6.54; NO3 − -N, 20 mg/L; COD, 39 mg/L; TDS, 436 mg/L. The breakthrough curves of AEE-3 are shown in Figure 11. An inconspicuous deviation between Run 1 and Run 3 breakthrough curves was found, further confirming its potential application in the removal of NO3 − -N from real STWW effluent.

Assessments of Practical Application
Conventional heterotrophic denitrification techniques are severely restricted by the low C/N ratio of wastewater, thereby causing the high NO 3 − -N content in STWW [33,34]. To further assess the practical application of AEE-3, the dynamic column test of NO 3 − -N uptake onto AEE-3 was performed using a real STWW effluent which was collected from a local sewage treatment plant, containing the following quality parameters: pH, 6.54; NO 3 − -N, 20 mg/L; COD, 39 mg/L; TDS, 436 mg/L.
The breakthrough curves of AEE-3 are shown in Figure 11. An inconspicuous deviation between Run 1 and Run 3 breakthrough curves was found, further confirming its potential application in the removal of NO 3 − -N from real STWW effluent.
using a real STWW effluent which was collected from a local sewage treatment plant, containing the following quality parameters: pH, 6.54; NO3 − -N, 20 mg/L; COD, 39 mg/L; TDS, 436 mg/L. The breakthrough curves of AEE-3 are shown in Figure 11. An inconspicuous deviation between Run 1 and Run 3 breakthrough curves was found, further confirming its potential application in the removal of NO3 − -N from real STWW effluent.

Conclusions
The strong base polyacrylic AER AEE-3 could be developed and utilized to absorb NO 3 − -N selectively from aqueous solutions. Batch adsorption studies exhibited that a pseudo-second-order model was appropriate for depicting the kinetic process, and that the best fit for the isotherms data was Langmuir model. In addition, AEE-3 displayed a greater adsorption capacity and better regeneration capability for NO 3 − -N uptake than the commonly-used, nitrate-selective AER A520E in the presence of interferential DOMs. Notably, the effective adsorption of NO 3 − -N in real STWW by AEE-3 packed in fixed-bed columns revealed its promising potential in the removal of NO 3 − -N from actual complex wastewater.