Adsorption of Nitrate onto ZnCl 2-Modified Coconut Granular Activated Carbon : Kinetics , Characteristics , and Adsorption Dynamics

Coconut granular activated carbon (CGAC) was modified by impregnating with ZnCl2 solution to remove nitrate from aqueous solutions. Sorption isotherm and kinetic studies were carried out in a series of batch experiments. Nitrate adsorption of both ZnCl2-modified CGAC and CGAC fitted the Langmuir and Freundlich models. Batch adsorption isotherms indicated that the maximum adsorption capacities of ZnCl2-modified CGAC and CGAC were calculated as 14.01mgN·g and 0.28mgN·g, respectively. .e kinetic data obtained from batch experiments were well described by pseudo-second-order model. .e column study was used to analyze the dynamic adsorption process. .e highest bed adsorption capacity of 1.76mgN·g was obtained by 50mgN·L inlet nitrate concentration, 20 g adsorbents, and 10ml·min flow rate. .e dynamic adsorption data were fitted well to the .omas and Yoon–Nelson models with coefficients of correlation R> 0.834 at different conditions. Surface characteristics and pore structures of CGAC and ZnCl2-modified CGAC were performed by SEM and EDAX and BETand indicated that ZnCl2 had adhered to the surface of GAC after modified. Zeta potential, Raman spectra, and FTIR suggested the electrostatic attraction between the nitrate ions and positive charge. .e results revealed that the mechanism of adsorption nitrate mainly depended on electrostatic attraction almost without any chemical interactions.

Among several technologies for nitrate removal, adsorption process was widely used because of its convenience, low consumption of energy, and ease of operation.Adsorption, in general, is the removal process of soluble substances that are in solution on a suitable interface [1].Activated carbon has gained wide attention as an efficient adsorbent which can adsorb various pollutants in aquatic phase especially organic pollutants [12].However, it shows poor adsorption towards anionic pollutants.As reported in Table 1, most of the papers found in the literature review were devoted to the nitrate adsorption with modified and various synthesis materials.
In addition, studies showed the nitrate adsorption due to the different mechanisms [26][27][28][29][30]. e mechanism of nitrate adsorption by magnetic amine-cross-linked biopolymer includes electrostatic attraction, ion exchange, and surface complex formation [27].Soybean was modified with calcium chloride and hydrochloric acid and by calcination, and results confirmed that the mechanism of nitrate adsorption was mainly ion exchange [29].Nitrate was adsorbed by modified biochar (amine-cross-linked reaction that amine group played the main role in nitrate uptake) depending on the electrostatic attraction mechanism [28,30].In addition, nitrate was adsorbed by magnetic multiwalled carbon nanotubes depending on the magnetic reaction to the desired separation [26].
Although studies have examined nitrate adsorption by modified activated carbon using ZnCl 2 , little research examined nitrate adsorption mechanisms by modified activated carbon and the adsorption dynamics.erefore, the main objectives are as follows: (1) nitrate adsorption properties by ZnCl 2 -modified coconut granular activated carbon (ZnCl 2modified CGAC), including the kinetics and isotherms, in batch experiments; (2) physicochemical properties confirmed by SEM, BET, DFT, zeta potentials, and Raman spectra to fully understand the adsorption mechanisms; and (3) column adsorption experiments with different parameters, including adsorbent bed depth, initial concentration and flow rate, and data analysis from column study using omas and Yoon-Nelson models.

Materials and Specimen Preparation.
Coconut granular activated carbon (CGAC) was gained from Huansheng Company, Henan Province, China.It was sieved through a 2 mm sieve, washed with deionized water, and dried at room temperature for 48 h for future use.CGAC was mixed with ZnCl 2 solution.In this study, 200% impregnation ratio was used, which meant that 10 g coconut granular activated carbon was added into the ZnCl 2 solution containing 20 g ZnCl 2 and 100 ml deionized water.e mixture was stirred for 1 h at 80 °C.en, the sample was dried in an oven for 24 h at 106 °C.e resulting sample was placed in the furnace and carbonized at 500 °C for 1 h.e product was washed with 0.5 M HCl sequentially and then washed by deionized water repeatedly until the pH of the solution reached about 5.5.After that, the samples were dried at 106 °C and stored in a desiccator for further use.All the chemicals including ZnCl 2 , KNO 3 , and HCl were of analytical grade.e nitrate stock solutions (100 mgN•L −1 ) were prepared by dissolving 0.7218 g KNO 3 in 1000 ml deionized water.

Batch Experiments. Batch experiments were carried out at different NO 3
− -N concentrations (5,10,15,20,50, 100 mgN•L −1 ).For adsorption equilibrium studies, experiments were conducted in 500 ml conical flasks containing 2.0 g of adsorbent, 200 ml nitrate solution, and 0.01 M NaCl to keep ion strength at 20 °C.e pH was not adjusted during the experiments.e mixtures were shaken at 170 rpm for 12 h and then filtered using a 0.22 μm membrane filter.e concentration of residual NO 3 − -N was measured by ion chromatograph ( ermo Scientific Aquion IC 1100).e amount of NO 3 − -N adsorbed by per mass unit of adsorbent was calculated by the following equation: where q e (mgN•g −1 ) is the adsorbent capacity, C 0 and C e (mgN•L −1 ) are the concentrations of NO 3

−
-N at initial and at equilibrium, respectively, V is the volume of the solution (L), and W is the mass of adsorbent used (g).

−
-N concentration at any time.
2.3.Characterization.Surface properties of CGAC and ZnCl 2 -modified CGAC were observed through a scanning electron microscope (SEM) (HITACHI s4800).e distribution of elements on the surface or in the pores of carbon particles was determined by the same SEM together with energy dispersive X-ray spectroscopy.
Pore structure characteristics of CGAC and ZnCl 2modified CGAC were determined by nitrogen adsorption at 77 K (Quantachrome Autosorb iQ2).Prior to gas adsorption measurements, the carbon was degassed at 170 °C in a vacuum condition for a period of at least 10 h. e BET surface area was determined by application of the Brunauer-Emmett-Teller (BET) equation [16].e pore size distributions of the CGAC and ZnCl 2 -modified CGAC were determined by the density functional theory (DFT) method [16].
Zeta potential measurements of CGAC, ZnCl 2 -modified CGAC, and nitrate-loaded ZnCl 2 -modified CGAC were carried out by the microelectrophoresis (Malvern Zetasizer Nano ZS). e samples were grounded in an agate mortar and sieved through a 74 μm sieve.100 mg of powder samples was mixed with 1 L deionized water by ultrasonic cleaner (KQ-300DE, China) for 10 min at 20 °C.After that, the samples were settled for 30 min.e suspension was collected to determine the zeta potential at several pH values from 2.0 to 12.0 using 0.01 M of NaOH and HCl to adjust pH. e pH PZC was the corresponding pH value when the zeta potential was 0 mV.
To make an intensive study on the interaction mechanisms of nitrate onto ZnCl 2 -modified CGAC, Raman spectroscopic analysis was performed.In the Raman analysis, 0.1 g ZnCl 2 -modified CGAC was placed in 50 ml of nitrate solution with concentration of 0.5 mol•L −1 .e pure solid samples of KNO 3 , ZnCl 2 -modified CGAC, and nitrateloaded ZnCl 2 -modified CGAC were analyzed by Raman spectroscopy (DXR Microscope).e laser wavelength used in Raman measurement was 1050 nm.
FTIR spectroscopy ( ermo Nicolet 6700 Spectrometer, USA) was done to identify the chemical functional groups presented on ZnCl 2 -modified CGAC, nitrate-loaded ZnCl 2modified CGAC, and solid samples of KNO 3 .e saturated samples after adsorption were prepared by mixing the ZnCl 2modified CGAC (2 g) with solution (200 ml) containing 500 mgN•L −1 of nitrate.Samples of particle size <45 μm were first dried for 24 h at a temperature of 383 K.
e dried samples were mixed with finely divided KBr at a ratio of 1 : 100.
e spectrum was scanned from 400 to 4000 cm −1 .

Column Studies.
In column experiment, a glass column (30 cm height and 2.3 cm inner diameter) was filled with ZnCl 2 -modified CGAC on glass wool support.In a typical experiment, a synthetic NO 3 − -N solution was fed into the column from the top at a desired flow rate using a peristaltic pump.A known quantity of the prepared ZnCl 2 -modified CGAC was packed in the column to yield the desired bed height of the adsorbent 39 mm, 78 mm, 117 mm (equivalent to 10 g, 20 g, and 30 g of ZnCl 2 -modified CGAC) at flow rate of 10 ml•min − -N concentration and 20 g adsorbents.Samples were collected from the bottom of the column at regular time intervals and analyzed for residual nitrate concentrations.
e studies were conducted at room temperature (20 ± 2 °C), and natural pH of solutions was about 6.5.e flow of the column was continued until the effluent concentration (C t ) approached the influent concentration (C 0 ), C t /C 0 � 0.95.
e equilibrium NO 3 − -N uptake per unit mass of adsorbent (q 0 ) was calculated by the following equation: e value of the total mass of NO 3 − -N adsorbed, q total (mg), could be calculated from the area under the breakthrough curve: where V total is the effluent volume at equilibrium, Q is the volumetric flow rate (ml•min −1 ), and t total is the total flow time [28,31].Equilibrium metal uptake or maximum capacity of the column, q eq (mgN•g −1 ), in the column was calculated by the following equation: where m is the mass of the adsorbent (g).Total amount of NO 3

−
-N ion entering column (m total ) was calculated by the following equation [32]:

Results and Discussion
3.1.Adsorption Kinetics.e adsorption kinetics described the uptake rate of nitrate ion on the ZnCl 2 -modified CGAC, which controlled the equilibrium time [33].
e kinetic studies were helpful for predicting the adsorption rate with time and explaining the dynamic interactions of nitrate ions with adsorbents, which gave important information for designing and modeling the processes [27,33].e experimental data were fitted by three different kinetic models: the pseudo-first-order model, pseudo-second-order model, and intraparticle diffusion model, and the results were discussed below.
Advances in Materials Science and Engineering e linear equation for pseudo-first-order kinetic model, widely used to predict sorption kinetics, was given by Langergren and Svenska [33], defined as follows: ln q e − q  � ln q e − k 1 t.(7) Pseudo-second-order kinetic model, based on equilibrium adsorption, was expressed as follows: Intraparticle diffusion model described by Weber and Morris [34] was widely used to explain the rate-limiting step: where q e and q t (mg•g −1 ) are the amount of adsorbate adsorbed at equilibrium and at any time, t (min), respectively, and k 1 (min −1 ), k 2 (g•mg −1 •min −1 ), and k dif (mgN•(g•min 1/2 ) −1 ) were the adsorption rate constant of pseudo-first-order model, pseudo-second-order model, and intraparticular diffusion model, respectively.e nitrate adsorption by ZnCl 2 -modified CGAC increased with the increase in initial nitrate concentration as shown in Figure 1(a), and the linear plots of different kinetic models are shown in Figures 1(b)-1(d).Moreover, kinetic constants of different kinetic models for the nitrate adsorption are shown in Table 2.
e correlation coefficients (R 2 ) for the pseudo-first-order kinetic model, pseudo-second-order kinetic model, and intraparticle diffusion model were 0.413-0.995,0.977-0.995,t/q t (min g/mg) 4 Advances in Materials Science and Engineering and 0.593-0.966,respectively.Meanwhile, according to the calculated value (q e,cal ) and experimental uptake value (q e,exp ), the pseudo-second-order model was considered as the best-fit model in describing the nitrate adsorption from aqueous solution [29,35].Pseudo-second-order model has been frequently invoked to describe adsorption of inorganic pollutants on GAC-based materials [29], and a similar study reported the nitrate adsorption onto magnetic amine-cross-linked biopolymer fitted with pseudo-second-order model as well [27].

Adsorption Isotherms.
e adsorption isotherm was conducted to determine the maximum adsorption capacities and expressed the relationship between the amount of sorption and residual nitrate concentration at equilibrium [36].e adsorption isotherm indicated how the adsorption molecules distributed between the liquid phase and the solid phase when the adsorption process reached an equilibrium state [37].In order to optimize the design of an adsorption system, three adsorption isotherms, namely, the Langmuir, Freundlich, and Temkin isotherm models in their linear forms were applied to the equilibrium data to find the suitable model that could be used for design purpose [37].Figure 2 typically shows the nitrate adsorption isotherms on the ZnCl 2 -modified CGAC and CGAC.All the correlation coefficient, R 2 values, and the parameters obtained for the models are summarized in Table 3.
e adsorption data fitted satisfactorily to both Langmuir (R 2 � 0.970) and Freundlich (R 2 � 0.982) models, better than Temkin model (R 2 � 0.828), are shown in Table 3 and Figure 2. Application of the Langmuir model for ZnCl 2 -modified CGAC and CGAC allowed the determination of the maximum equilibrium adsorption capacity [36], which were 14.01 mgN•g −1 and 0.28 mgN•g −1 , respectively.e nitrate adsorption capacity of modified adsorbents was much higher than that of raw adsorbents.e maximum adsorption capacity obtained for ZnCl 2 -modified CGAC was higher than the corresponding values assumed by others for polyethylene glycol/chitosan and polyvinyl alcohol/chitosan (11.44 mgN•g −1 and 7.91 mgN•g −1 ) [21].In addition, the constant 1/n in the range of 0-1 showed the favorable conditions for adsorption [38].

SEM Analysis.
e scanning electron microscopy images of CGAC and ZnCl 2 -modified CGAC are shown in Figures 3(a e elemental composition of ZnCl 2 -impregnated activated carbon determined by EDAX is shown in Table 4.After modified by ZnCl 2 , the carbon content decreased from 74.75% to 37.61% and the zinc contents and the chloride contents increased to 53.61% and 0.49%, respectively.

BET Analysis.
e surface area and pore characteristics are shown in Table 5. e BET specific surface area of CGAC was 876.752 m 2 •g −1 , and the surface area of ZnCl 2modified CGAC decreased to 567.524 m 2 •g −1 .It was obvious that the pore width decreased after impregnation with ZnCl 2 , from 0.518 nm to 0.492 nm. e decrease of surface area was ascribed to blockage of pore openings by the ZnCl 2 that prohibited access of adsorbing gas molecules [29]. is indicated that the ZnCl 2 had adhered to the surface of the CGAC and the result was confirmed by the SEM analysis as shown in Figure 3.It had been proved that activated carbon with large amounts of pore structures was inefficient for adsorption capacity [4,28,39].As a result, micropore adsorption was absent for the potential nitrate adsorption by activated carbon [28].

Zeta Potentials.
Zeta potentials of original activated carbon, ZnCl 2 -modified CGAC, and nitrate-loaded ZnCl 2modified CGAC as a function of pH are shown in Figure 4.All the samples of the zeta potentials became more negative with the increase in pH, probably because of the deposition of more OH − on the adsorbent surface [40].Zeta potentials of CGAC were in the range of +12.55 to −30.50 mV as the initial pH of the suspensions increased from 2.01 to 11.99.After modified by ZnCl 2 , zeta potentials of ZnCl 2 -modified CGAC increased slightly (+14.60 to −29.65 mV) in designed pH range.Point of zero charge pH (pH pzc ) of CGAC was located at 2.39.After the ZnCl 2 modification, pH pzc of modified GAC had a slight increase to 2.56. is suggested that ZnCl 2 loaded on the surface of activated carbon increased positive charge on activated carbon surface.After adsorption, zeta potentials had an apparent decrease at the pH of 5.3∼12, this illustrated that nitrate ions had been adsorbed on the surface of ZnCl 2 -modified CGAC, and the adsorption mechanism of ZnCl 2 -modified CGAC for nitrate was based on electrostatic attraction [28].C e is the equilibrium concentration (mg•L −1 ); Q is the monolayer saturation adsorption capacity; K L is the Langmuir constant (L•mg −1 ); K f is the adsorbentadsorbate relative affinity in the adsorption process ((mgN•g −1 )•(L•mg −1 ) 1/n ); K T is equilibrium blinding constant (L•mg −1 ); and B is the Temkin constant related to the heat of adsorption.

Raman Spectra.
6 Advances in Materials Science and Engineering peak at 1042.3 cm −1 .After the adsorption process, the nitrate on ZnCl 2 -modified CGAC illustrated the Raman peak at 1039.4 cm −1 , which was almost overlapped with the peak of KNO 3 .e results indicated that nitrate ions were adsorbed onto the surface of ZnCl 2 -modified CGAC through electrostatic attraction between the free nitrate ions and the positively charged ions of ZnCl 2 sites [28], which corresponded well to the decrease of zeta potentials after nitrate adsorption onto ZnCl 2 -modified CGAC.
In FTIR spectra of ZnCl 2 -modified CGAC and nitrateloaded ZnCl 2 -modified CGAC, the bands at about 3430 cm −1 were ascribable to ] (O-H) vibrations in hydroxyl groups.e low-frequency values for these bands suggested that the hydroxyl groups were involved in hydrogen bonds and the position of the band due to nonbonded OH groups was usually above 3500 cm −1 for alcohols, phenols, and carboxylic acids [41].
e O-H stretching vibrations occurred within a broad range of frequencies indicating the presence of "free" hydroxyl groups and bonded O-H bands of carboxylic acids.e bands observed at 1800-1000 cm −1 were presumed to be associated with oxygenated C�O and C-O-R structures as the range was reported to reflect the presence of moieties of different C�O (amide, esters, carboxylic acids, quinines, etc.) and C-O-R (aryl and alkyl esters, carboxylic) structures depending on the extent of coalification [42].
Nitrate curve in Figure 6 and the pure KNO 3 crystal had a characteristic peak at 1384.66 cm −1 in FTIR spectra.Compared with the curve of ZnCl 2 -modified CGAC, a new peak at 1384.66 cm −1 (NO 3 − ) was observed in the curve of nitrate-laden ZnCl 2 -modified CGAC which was assigned to the special vibration of nitrate.Meanwhile, few special vibrations could be shown in FTIR spectra, so that the main mechanism of adsorption nitrate based on electrostatic attraction almost without any chemical interactions existed.

Column Studies.
To investigate the adsorption ability to the continuous nitrate removal from solution onto   ZnCl 2 -modi ed CGAC, dynamic column adsorption tests and the dynamic models were conducted to evaluate the performance of a continuous system.

E ect of the Column Depth on the Breakthrough Curves.
As shown in Figure 7(a) and Table 6, when the mass of adsorbent was 10 g, 20 g, and 30 g, the breakthrough points occurred at the time of 200 min, 320 min, and 650 min, respectively.e saturated adsorption capacities of ZnCl 2modi ed CGAC (10 g, 20 g, and 30 g) in column were about 0.25 mgN•g −1 , 0.20 mgN•g −1 , and 0.17 mgN•g −1 , respectively.
e breakthrough time increased with the increase in mass which might be due to the more contact time.e increase in NO 3 − -N uptake capacity with the increasing bed depth in the column may be due to increased adsorbent surface area, which provided more binding sites for the column adsorption [43,44].

E ect of Flow Rates on the Breakthrough Curves.
As shown in Figure 7(b) and Table 6, the breakthrough curve generally occurred faster with higher ow rate at 20 ml•min −1 .With the increases in ow rate, the adsorption capacities were 0.18 mgN•g −1 , 0.20 mgN•g −1 , and 0.27 mgN•g −1 , respectively.As indicated in Figure 7(b), the breakthrough curve became steeper as the ow rate increased.is was due to that, at a high rate of in uent, NO 3 − -N did not have enough time to contact with ZnCl 2 -modi ed CGAC [43].At a low rate of in uent, NO 3 − -N maybe had more time to be in contact with adsorbent [44].Breakthrough time reaching saturation was increased signi cantly with a decrease in the ow rate.

E ect of In uent Concentration on the Breakthrough
Curves.As shown in Figure 7(c) and Table 6, it is illustrated that the adsorption process reached saturation faster and the breakthrough time decreased with increasing in uent NO 3 − -N concentration.e adsorption capacities of ZnCl 2 -modi ed CGAC for nitrate were 0.12 mgN•g −1 , 0.20 mgN•g −1 , and 0.26 mgN•g −1 , respectively.e maximum capacity at 50 mgN•L −1 was higher than those at 10 mgN•L −1 and 20 mgN•L −1 . is might be attributed to high in uent NO 3 − -N concentration providing higher driving force for the transfer process to overcome the mass transfer resistance [45].
is result indicated that the

Application of omas Model and Yoon-Nelson Model.
omas model was widely used to describe the performance of adsorption process in a xed-bed column.is model assumed plug ow behavior in the bed and used Langmuir isotherm for equilibrium and the second-order reversible kinetics [48].
e model was described by the following equation: where K th (L•min −1 •mg −1 ) is the omas rate constant, q 0 (mgN•g −1 ) is the maximum sorption capacity, m (g) is the mass of adsorbent, and Q (ml•min −1 ) is the ow rate.

Advances in Materials Science and Engineering
A simple theoretical model developed by Yoon and Nelson was also tested to investigate the breakthrough behavior of nitrate onto ZnCl 2 -modified CGAC [49,50].e linearized model for a single component system was expressed as follows: where K YN (min −1 ) is the rate constant of model and τ (min) is the time required for 50% adsorbate breakthrough.Among omas and Yoon-Nelson models, the parameters listed in Tables 7 and 8, both of them provided good fit (R 2 > 0.834) to the experimental data at various conditions.In a comparison of values of R 2 and breakthrough curves, both omas and Yoon-Nelson models could be used to describe the behavior of the nitrate adsorption in a fixed-bed column.e results indicated that the external and internal diffusions would not be the limiting step [44].

Conclusions
(1) Nitrate adsorption behavior of CGAC and ZnCl 2modified CGAC was described successfully by both Langmuir and Freundlich models, and the maximum adsorption capacity was predicted to be 14.01 mgN•g −1 and 0.28 mgN•g −1 , respectively.e kinetic data indicated that the adsorption process obeyed the pseudo-second-order model.(2) e characteristics (SEM and EDAX, surface area, pore structure, zeta potential, Raman spectra, and FTIR) indicated that the mechanism of nitrate adsorption mostly depended on the electrostatic attraction between the free nitrate ions and the positively charged ions, almost without any chemical interactions.(3) In column study, the breakthrough curves were strongly dependent on the mass of adsorbents, flow rate, and initial NO 3 − -N concentration.Both omas and Yoon-Nelson models were found to be in good agreement with the experimental data and could be used for prediction of the experimental results as well.
−1 and initial NO 3 − -N concentration of 20 mgN•L −1 .e effect of NO 3 − -N concentration on the adsorption capacity was studied using initial NO 3 − -N concentrations of 10, 20, and 50 mgN•L −1 with column flow rate of 10 ml•min −1 and 20 g ZnCl 2 -modified CGAC.e effect of different flow rates on the adsorption capacity was studied at 5, 10, and 20 ml•min −1 with 20 mgN•L −1 initial NO 3

FIGURE 1 :
FIGURE 1: Linear plots of adsorption kinetics: (a) the variation of adsorption capacity with adsorption time at various initial nitrate concentrations; (b) pseudo-first-order adsorption plot; (c) pseudo-second-order adsorption plot; (d) intraparticle diffusion adsorption plot.

Table 1 :
Literature information on various adsorbents for nitrate removal from water.

Table 2 :
Parameters for different adsorption kinetic models.

Table 7 :
omas parameters at different conditions using linear regression analysis.

Table 8 :
Yoon-Nelson parameters at different conditions using linear regression analysis.

Table 6 :
Parameters in fixed-bed column for nitrate adsorption by ZnCl 2 -modified CGAC.