Nitrate adsorption from aqueous solution using granular chitosan-Fe3+ complex

Nitrate adsorption from aqueous solution using granular chitosan-Fe3+ complex

Accepted Manuscript Title: Nitrate adsorption from aqueous solution using granular chitosan-Fe3+ complex Author: Qili Hu Nan Chen Chuanping Feng WeiWu...

1MB Sizes 0 Downloads 32 Views

Recommend Documents

No documents
Accepted Manuscript Title: Nitrate adsorption from aqueous solution using granular chitosan-Fe3+ complex Author: Qili Hu Nan Chen Chuanping Feng WeiWu Hu PII: DOI: Reference:

S0169-4332(15)00889-2 http://dx.doi.org/doi:10.1016/j.apsusc.2015.04.049 APSUSC 30142

To appear in:

APSUSC

Received date: Revised date: Accepted date:

9-11-2014 28-3-2015 7-4-2015

Please cite this article as: Q. Hu, N. Chen, C. Feng, W.W. Hu, Nitrate adsorption from aqueous solution using granular chitosan-Fe3+ complex, Applied Surface Science (2015), http://dx.doi.org/10.1016/j.apsusc.2015.04.049 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Highlights: Granular chitosan-Fe3+ complex had high performance for nitrate adsorption Granular chitosan-Fe3+ complex had shorter equilibrium time (1.5 h)

ip t

Nitrate adsorption was ascribed to ion exchange and electrostatic attraction

Ac ce p

te

d

M

an

us

cr

Granular chitosan-Fe3+ complex could be regenerated using NaCl solution

Page 1 of 38

Nitrate adsorption from aqueous solution using granular chitosan-Fe3+ complex Qili Hua,b, Nan Chena,b,*, Chuanping Fenga,b, WeiWu Huc School of Water Resources and Environment, China University of Geosciences

ip t

a

b

cr

(Beijing), Beijing, 100083, China

Key Laboratory of Groundwater Cycle and Environment Evolution (China

us

University of Geosciences (Beijing)), Ministry of Education, Beijing, 100083, China

The Journal Center, China University of Geosciences (Beijing), Beijing, 100083,

an

c

M

China

* Corresponding author: Nan Chen, School of Water Resources and Environment,

Ac ce p

Fax: +86 10 82321081

te

Tel: +86 10 82322281

d

China University of Geosciences (Beijing), Beijing, 100083, China.

E-mail: [email protected] (N. Chen)

Page 2 of 38

Abstract In the present study, In order to efficiently remove nitrate, granular chitosan-Fe3+ complex with high chemical stability and good environmental adaptation was

ip t

synthesized through precipitation method and characterized using SEM, XRD, BET and FTIR. The nitrate adsorption performance was evaluated by batch experiments.

cr

The results indicated that granular chitosan-Fe3+ complex was an amorphous and

mesoporous material. The BET specific surface area and average pore size were 8.98

us

m2 g-1 and 56.94 Å, respectively. The point of zero charge was obtained at pH 5. The

an

maximum adsorption capacity reached 8.35 mg NO3--N g-1 based on LangmuirFreundlich model. Moreover, no significant change in the nitrate removal efficiency

M

was observed in the pH range of 3.0-10.0. The adverse influence of sulphate on nitrate removal was the most significant, followed by bicarbonate and fluoride, whereas

d

chloride had slightly adverse effect. Adsorption process followed the pseudo-second-

te

order kinetic model, and the experimental equilibrium data were fitted well with the Langmuir-Freundlich and D-R isotherm models. Thermodynamic parameters revealed

Ac ce p

that nitrate adsorption was a spontaneous and exothermic process. Granular chitosanFe3+ complex could be effectively regenerated by NaCl solution. Keywords:

Nitrate, Granular chitosan-Fe3+ complex, Adsorption, isotherm,

Regeneration

Page 3 of 38

1

Introduction The current problem of excess nitrate had gained significant attention in surface

and ground water around the world. The extensive use of nitrogen fertilizers and

ip t

improper treatment of wastewater from industrial sites were presumably responsible for nitrate contamination [1]. High concentration of nitrate in drinking water would

cr

pose a serious threat to human health such as liver damage, cancers and

methemoglobinemia [2]. Therefore, it was essential to lower the nitrate concentration

us

within a maximum contaminant level (MCL) of 50 mg L-1 (as NO3-) in drinking water

an

[3]. Elevated nitrate concentration surpassing self-purification capacity of water body had a negative impact on ecosystem sustainability including eutrophication caused by

M

heavy algal growth [4].

Nitrate could chronically exist in surface and ground water due to its high solubility

d

and poor adsorption property, which increased difficulty of nitrate removal [5].

te

Various methods had been developed for the nitrate removal from water and wastewater to reduce the risks to environment, including biological denitrification [6],

Ac ce p

ion exchange [7], electrochemical reduction [8], reverse osmosis [9], catalytic reduction [10] and adsorption [11]. However, biological method was in need of addition of carbon source and disposal of large amounts of biomass waste [12]. Ion exchange required disposal of waste brine and renewal of costly ion exchange resin. The electrodes were easily inactive in electrochemical reduction. Reverse osmosis had high operational cost. Catalytic reduction could produce the hazardous by-products such as nitrite and ammonia. Adsorption was found to be superior to other techniques in terms of its low cost, ease of operation and simplicity of design [13]. Recently, many adsorbents such as carbon-based adsorbent [14], mineral-based adsorbent [15] and agricultural waste [16]

Page 4 of 38

were employed to remove nitrate from aqueous solution. However, these materials had some inherent limitation including small particle size or low desorption efficiency, which might hinder their practical application. Some modified techniques such as

ip t

surface protonation [11], metal ions immersion [17] and surfactant modified mineral [18] were also applied to enhance their adsorption of nitrate. Nevertheless, some

cr

modifiers for adsorbent synthesis had a potential adverse effect on environment. As a

consequence, developing an easily reusable, technically feasible and environmentally

us

friendly adsorbent was of prime importance for the nitrate removal.

an

Chitosan, a natural product derived from deacetylation of the polysaccharide chitin, had received vast interests due to its excellent properties, such as abundant source,

M

non-toxicity, biodegradability and adsorption properties [19]. Chitosan containing enormous free amino and hydroxyl groups had the ability to coordinate with many

d

transition metal ions through a chelation mechanism [20]. Granular chitosan-Fe3+

te

complexes prepared by different approaches had successfully employed to adsorb textile dye [21], Cr(VI) [22], phosphate [23] and As(III) [24]. However, there were

Ac ce p

few reports for using granular chitosan-Fe3+ complex as an adsorbent to remove

nitrate from aqueous solution.

To verify the feasibility of granular chitosan-Fe3+ complex for the nitrate removal,

nitrate adsorption performance was evaluated by investigating the effect of adsorbent dosage, initial NO3--N concentration, contact time, pH and co-existing ions on the nitrate removal in this study. Adsorption isotherm and adsorption kinetics were studied to reveal the nitrate adsorption mechanism of granular chitosan-Fe3+ complex.

Furthermore, desorption studies were also conducted to evaluate the reusability of adsorbent.

2

Materials and methods

Page 5 of 38

2.1 Materials Chitosan (deacetylation degree = 80.0%–95.0%) was purchased from Sinopharm Chemical Reagent Co., Ltd, Shanghai, China. Ferric chloride was obtained from

ip t

Tianjin Fuchen Chemical Regeats factory, Tianjin, China. Ammonia solution was provided by Beijing Chemical Works, Beijing, China. A series of various

cr

concentrations of nitrate solution required were prepared by dissolving NaNO3 in deionized water. All chemicals used in this study were of analytical grade without

us

further purification.

an

2.2 Adsorbent synthesis

Chitosan powder (9.0 g) was dispersedly dissolved in 0.2 M FeCl3 aqueous solution (300 mL), and then stirred at room temperature for 1.5 h. Subsequently, the resulting

M

chitosan-Fe3+ solution was drop wise added into ammonia solution (12.5% v/v) using

d

a disposable syringe. After stabilized for 1.0 h, the spherical hydrogel beads formed

te

were separated and sufficiently washed with deionized water, then dried at 50°C for 7.5 h. The dried beads were immersed in deionized water (liquid-solid ratio of 10 mL

Ac ce p

g-1) at 40°C for 4 h in a horizontal shaker (HZC-280, Bing, China). The swelling effect occurred during the immersion, which would contribute to improving adsorbent surface property and promoting adsorption capacity. After separation, washing and drying, granular chitosan-Fe3+ complexes were obtained, and then stored in sealed

plastic bag at room temperature for further study. Additionally, the chitosan beads were prepared to compare the nitrate adsorption capacity with granular chitosan-Fe3+ complex. In the present study, granular chitosan-Fe3+ complex with a particle size range of 1.04-1.16 mm was a spherical and epigranular adsorbent, which was superior to hydrotalcite [25] and modified beet residue [26] whose particle sizes were less than 150

and 4.1

, respectively.

Page 6 of 38

2.3 Batch experiments 100 mL of nitrate solution was poured in 250 mL conical flasks for batch experiments, which were sealed and agitated at 140 rpm in a thermostatic shaker

ip t

(HZC-280, Bing, China) at 25°C for 1.5 h. 2.3.1 Evaluation of adsorption performance

cr

The experiments for evaluation were performed at initial nitrate concentration of 50 mg NO3--N L-1 and 20 g L-1 of adsorbent dosage.

us

The effect of adsorbent dosage was investigated at the addition of 5, 10, 15, 25, 30

an

and 40 g L-1. The initial pH of nitrate solution was adjusted in the pH range from 3.0 to 12.0 using 0.1 M HCl or 0.1 M NaOH solution. In order to investigate the effect of co-existing ions, anions (Cl-, HCO3-, F- and SO42-) were added at every anion

M

concentration of 4 mmol L-1, and cations (K+, Na+, NH4+, Ca2+, Mg2+) were added by

d

fixing a chloride ion concentration at 4 mmol L-1.

te

2.3.2 Kinetic, isotherm and thermodynamic studies These studies were performed at adsorbent dosage of 20 g L-1.

Ac ce p

Kinetic studies were carried out at different nitrate concentrations (20, 50 and 100

mg NO3--N L-1) at certain intervals (5, 10, 15, 20, 25, 35, 45, 60, 90 and 120 min). Isotherm studies were performed at different initial concentrations (20, 30, 40, 50, 75, 100, 120, 150, 180 and 200 mg NO3--N L-1). Thermodynamic studies were conducted

at 15, 25, 35, 45 and 55°C.

Adsorption capacity of granular chitosan-Fe3+ complex was calculated at

equilibrium using the following equation [13]: (1) where qe (mg g-1) is the amount of nitrate adsorbed at equilibrium, C0 (mg L-1) and Ce (mg L-1) are the initial and equilibrium NO3--N concentration in aqueous solution,

Page 7 of 38

respectively. V (L) is volume of nitrate solution and M (g) is the mass of adsorbent used. 2.4 Analysis

ip t

The concentration of NO3--N was determined by standard colorimetric method using a UV/vis spectrophotometer (DR 6000, HACH, USA). The surface morphology

cr

and particle size were observed with a scanning electron Microscope (SEM) (SSX–

550, Shimadzu, Japan). The crystal structure of granular chitosan-Fe3+ complex was

us

carried out by quantitative X-ray diffraction (XRD) (D8 Focus, Bruker, Germany).

an

The point of zero charge of granular chitosan-Fe3+ complex was estimated by batch equilibrium techniques described by Chutia et al. [27]. The specific surface area and

M

BJH (Barrett-Joyner-Halenda) pore size distribution of granular chitosan-Fe3+ complex were determined using Brunauer-Emmett-Teller (BET) specific surface

d

analysis device (JW-BK132F, JWGB, China). FTIR spectra of chitosan beads and

te

granular chitosan-Fe3+ complex before and after nitrate adsorption were recorded on a Fourier transform infrared spectrometer using KBr pellets over the wavenumber range

Ac ce p

of 4000-400 cm-1 with a resolution of 4 cm-1 (Nexus 470, Thermo Nicolet, USA).

3

Results and discussion

3.1 Characterization of granular chitosan-Fe3+ complex

3.1.1 SEM

SEM images of granular chitosan-Fe3+ complex before and after nitrate adsorption

were shown in Fig. 1a and Fig. 2b. It was evident that granular chitosan-Fe3+ complex

possessed poorly developed pore and extremely heterogeneous surface morphology, which were partially attributed to the uneven volatilization of ammonia and water molecule on the chitosan-Fe3+ hydrogel beads surface during desiccation. The irregular surface with fingerprint-like textures increased specific surface area and

Page 8 of 38

thereby provided more active sites for nitrate adsorption. Adsorbent surface before nitrate adsorption was observed at a compact gel state, while adsorbent surface after nitrate adsorption had higher surface roughness. This phenomenon was due to the

ip t

occurrence of swelling effect of granular chitosan-Fe3+ complex in the process of

cr

nitrate adsorption.

Fig. 1. SEM images before nitrate adsorption (a) and after nitrate adsorption (b); XRD

an

us

pattern (c); Adsorption BJH (Barrett–Joyner–Halenda) pore size distribution (d).

3.1.2 XRD

M

The XRD pattern of granular chitosan-Fe3+ complex was illustrated in Fig. 1c. The crystalline peak of this adsorbent was not observed, suggesting that granular chitosan-

d

Fe3+ complex exhibited amorphous nature. This behavior was in agreement with other

Thakre et al. [28].

te

study where Ti-Al binary metal oxide supported chitosan beads was prepared by

Ac ce p

3.1.3 Specific surface area

The BET specific surface area, total pore volume and average pore size of granular

chitosan-Fe3+ complex were found to be 8.98 m2 g-1, 0.019 cm3 g-1 and 56.94 Å

respectively. Granular chitosan-Fe3+ complex with the small specific surface area had high adsorption capacity, indicating this adsorbent had a good affinity for nitrate. As was shown in Fig. 1d, the observed pore sizes mostly varied between 2 and 50 nm, accounting for 89.6%. According to the IUPAC classification, the BJH adsorption pore size distribution revealed the mesoporous nature of granular chitosan-Fe3+ complex. Furthermore, this pore size range was conductive to the penetration of nitrate into the inner layer of the adsorbent due to very small ionic radius of nitrate.

Page 9 of 38

3.1.4 FTIR The FTIR spectra were analyzed to identify the vital functional groups accounting for the nature of nitrate adsorption. The broad and intense adsorption band at

ip t

approximately 3419 cm-1 for the spectrum of pristine chitosan (curve a in Fig. 2) was ascribed to intermolecular hydrogen bonds and the overlap between O–H and N–H

cr

stretching vibration. The characteristic peaks at 2923 and 2881 cm-1 were assigned to

C–H stretching vibration of –CH2 and –CH3 aliphatic groups, respectively [29]. Two

us

new peaks appeared at 2337 and 2363 cm-1 were attributed to the amino protonation

an

of –NH2 and –NHCOCH3 group. The peak at 1648 and 1559 cm-1 referred to C=O stretching vibration of acetyl group and the stretching vibration of amine group,

M

respectively. The peaks at 1155, 1032, 1419 and 1384 cm-1 were related to C–O–C stretching vibration and C–H symmetric bending vibration, separately. It had been

d

reported that the characteristic peak of NO3- existed at around 1384 cm-1 [30]. The

te

peak at 1261 and 1073 cm-1 corresponded to the stretching vibration of C–O in chitosan molecule on C-3 and C-6 position, respectively. It was observed that C–H

Ac ce p

out-of-plane bending vibration and N–H wagging vibration occurred at ca. 898 and 660 cm-1 [31].

It was clearly seen from the curve b in Fig. 2 that the peaks at 1559 and 660 cm-1

disappeared for chitosan after reaction, and the transmittances at 2374 and 2341 cm-1 sharply increased due to the occurrence of dipole moment change compared with original chitosan, which implied that the interaction did occur between nitrate ions and protonated amine groups. Moreover, it was observed from the curve c and d in Fig. 2 that there were few changes as to the FTIR spectra of granular chitosan-Fe3+ complex before and after nitrate adsorption, indicating no new chemical bond appeared between nitrate ions and granular chitosan-Fe3+ complex.

Page 10 of 38

Fig. 2. FTIR spectra: chitosan before nitrate adsorption (a); chitosan after nitrate adsorption (b); chitosan-Fe3+ complex before nitrate adsorption (c); chitosan-Fe3+

ip t

complex after nitrate adsorption (d).

cr

3.2 Effect of adsorbent dosage

Adsorbent dosage was a vital parameter influencing adsorption capacity and

us

effluent concentration [32]. The effect of adsorbent dosage on nitrate removal was depicted in Fig. 3. It was evident that the nitrate removal efficiency increased from

an

27.6% to 90.4% with the increase in adsorbent dosage from 2 to 40 g L-1. The

M

increment was ascribed to availability of more active sites and larger surface area at higher dosage [33]. However, the equilibrium adsorption capacity of adsorbent

d

decreased from 6.99 to 1.15 mg NO3--N g-1, which was due to the fact that the

te

increase in adsorbent dosage for a given amount of nitrate resulted in unsaturation of adsorption sites [34]. Moreover, it was found that further increase in dosage beyond

Ac ce p

20 g L-1 rarely affected adsorption capacity of adsorbent. Adsorption capacity of granular chitosan-Fe3+ complex reached 2.04 mg NO3--N g-1 compared with granular

chitosan-Fe3+ complex (1.38 mg NO3--N g-1) that was not immersed in deionized water.

Fig. 3. Nitrate removal efficiency and adsorption capacity under different dosage

conditions (initial concentration = 50 mg NO3--N L-1, contact time = 1.5 h, temperature = 25°C, pH = 5.9 ± 0.2).

3.3 Effect of pH

Page 11 of 38

The pH of solution affected not only surface charges and dissociation of functional groups, but also chemical speciation and diffusion rate of solute [35,36]. In the present study, the point of zero charge was obtained at pH 5.4. The influence of initial

ip t

pH on nitrate removal was illustrated in Fig. 4. It was observed that no significant change occurred in terms of nitrate removal efficiency exceeding 80% in the pH range

cr

of 3.0-10.0, indicating adsorption process was almost independent of initial pH.

Moreover, adsorption capacity of granular chitosan-Fe3+ complex (2.04 mg NO3--N g) was superior to chitosan (1.95 mg NO3--N g-1) at an initial concentration of 50 mg

us

1

an

NO3--N L-1 and adsorption dosage of 20 g L-1, implying electrostatic attraction between negatively charged nitrate ions and positively charged anime groups of

M

chitosan was not a unique mechanism of nitrate removal [37]. However, the free amine groups of chitosan could not be protonated at alkaline medium, which did not

d

facilitate electrostatic attraction. Anion was proposed to be adsorbed on adsorbent

te

through nonspecific adsorption including electrostatic attraction or specific adsorption involving ion exchange [38]. Besides, when the pH of solution was higher than the

Ac ce p

point of zero charge, negatively charged adsorbent surface will fail to adsorb nitrate ions due to electrostatic repulsion. For these reasons, ion exchange mainly occurred between exchangeable chloride ions of adsorbent and nitrate ions in alkaline solution, which was consistent with the results of D-R isotherm model. These results could be elucidated by the following reactions. (2) (3) (4) (5)

Page 12 of 38

The sharp decrease in nitrate removal efficiency appeared when initial pH exceeded 11.0, which was probably ascribed to the competition between hydroxide and nitrate ions for adsorption sites and the increase in diffusion resistance of nitrate caused by

ip t

abundant hydroxide ions [28]. In addition, the final pH of solution was located between 5.0 and 6.0, indicating granular chitosan-Fe3+ complex had buffer capacity in

cr

the pH range of 3.0-10.0.

us

Fig. 4. Effect of pH on nitrate removal efficiency (adsorbent dosage = 20 g L-1, initial

an

concentration = 50 mg NO3--N L-1, contact time = 1.5 h, temperature = 25°C, pH =

M

5.9 ± 0.2).

3.4 Effect of co-existing ions

d

The nitrate-contaminated water generally contained a series of chemical

te

compositions, which could exert a negative influence on adsorption process. Consequently, it was extremely important to explore the effect of co-existing ions on

Ac ce p

nitrate removal in binary component solution. As was shown in Fig. 5, anions presenting in nitrate solution had a detrimental effect on adsorption process, resulting in the decrease in nitrate removal efficiency. The adverse influence of sulphate on nitrate removal was the most significant, followed by bicarbonate and fluoride, whereas chloride had slightly adverse effect. These anions occupied adsorption sites on the adsorbent surface and increased electrostatic repulsion between granular chitosan-Fe3+ complex and nitrate ions, leading to the decrease in nitrate removal efficiency. Wan et al. [25] pointed out that multivalent anion with higher charge density was adsorbed more readily than monovalent anion. Thus, granular chitosanFe3+ complex had a greater affinity for sulphate ions compared with other anions.

Page 13 of 38

On the other hand, the nitrate removal efficiency had no distinct difference among five cations, implying these cations did not disturb nitrate adsorption. The decrease in nitrate removal efficiency was caused by the competition between nitrate and chloride

ip t

ions for adsorption sites.

cr

Fig. 5. Effect of co-existing ions on nitrate adsorption (adsorbent dosage = 20 g L-1,

initial concentration = 50 mg NO3--N L-1, contact time = 1.5 h, temperature = 25°C,

an

us

pH = 5.9 ± 0.2).

3.5 Effect of initial nitrate concentration and adsorption isotherm In the present study, adsorption capacity increased dramatically from 0.90 to 4.62

M

mg NO3--N g-1 with the increase in initial NO3--N concentration, because higher

d

concentration gradient acting as a driving force overcame mass transfer resistance

te

between bulk solution and adsorbent surface [39]. Few sufficient adsorption sites accommodated nitrate ions for a certain amount of adsorbent when nitrate content

Ac ce p

exceeded a certain concentration, resulting in the decrease in removal efficiency from 93.1% to 47.1% with the increase in initial concentration from 20 to 200 mg NO3--N

L-1.

Adsorption isotherm models were commonly used to predict maximum adsorption

capacity, which helped to realize how optimized an adsorption system. The mutual correlation of adsorption capacity and equilibrium concentration was investigated by fitting experimental data using the Langmuir [24], Freundlich [21], LangmuirFreundlich [37] and Dubinin-Radushkevich [40] isotherm models, respectively. The nonlinear forms of these models were given as:

Page 14 of 38

(6) (7)

ip t

(8)

cr

(9)

where qe (mg g-1) is the calculated adsorption capacity at equilibrium, Ce (mg L-1) is

1

us

the equilibrium nitrate concentration, KL (L mg-1) is the Langmuir constant, qm (mg g) is the fitting maximum adsorption capacity, KF ((mg g-1) (L mg-1)1/n) is the

an

Freundlich constant related to adsorption capacity, n is an empirical parameter related to adsorption intensity, KF-L (L mg-1) is the Langmuir-Freundlich adsorption constant,

M

m is the heterogeneity factor, β (mol2 kJ-2) is a constant related to the adsorption

d

energy, ε (kJ mol-1) is Polanyi potential, R (8.314J mol-1 K-1) is the universal gas constant, T (K) is the temperature in Kelvin. In Dubinin-Radushkevich isotherm

te

model, the unit of the equilibrium nitrate concentration (Ce) should be mol L-1.

Ac ce p

In addition to correlation coefficient (R2), chi-square analysis (χ2) and normalized

standard deviation (NSD) were employed to reasonably evaluate the validity of various models, which were defined as [41,42]: (10)

(11) where

(mg g-1) is the experimental equilibrium adsorption capacity,

(mg g-1)

is the adsorption capacity calculated from models, N is the number of measurements.

Page 15 of 38

Fig. 6. Adsorption isotherm for nitrate adsorption on granular chitosan-Fe3+ complex: Langmuir isotherm (the dash-dotted line), Freundlich isotherm (the dash line), Langmuir-Freundlich isotherm (the solid line) (a); Dubinin-Radushkevich isotherm

ip t

(b).

cr

Table 1

Error analysis and constants of the Langmuir, Freundlich, Langmuir-Freundlich and

an

us

D-R isotherm models.

The corresponding information was shown in Fig. 6 and Table 1. The essential

M

feature of Langmuir isotherm could be conveyed by a dimensionless constant separation factor (RL), which was defined as [37]:

d

(12)

te

It was generally stated that the RL value with the range 0 < RL < 1 indicated

Ac ce p

favorable, while the n value in the range from 2 to 10 represented good [43]. It was obvious that the calculated RL and n values in this study were 0.21 and 2.77 respectively, suggesting nitrate adsorption on the granular chitosan-Fe3+ complex was

favorable. It was difficult to decide which isotherm was better fitted with the experimental data because the R2 values exceeded 0.97 for the first three isotherm models. The χ2 and NSD values of Freundlich and Langmuir-Freundlich isotherm outdistanced those of Langmuir isotherm. Additionally, the m value of LangmuirFreundlich isotherm was not more than 1. These results indicated nitrate adsorption on granular chitosan-Fe3+ complex was a multilayer adsorption. The relatively high correlation coefficient (R2 = 0.998) reflected that the

experimental data was also well described by the D-R isotherm, and that the fitting β

Page 16 of 38

value could logically predict the mean free energy E (kJ mol-1) that could be calculated as [2]: (13)

ip t

It was known that adsorption type was physical adsorption (E < 8 kJ mol-1) or ion

cr

exchange (8 kJ mol-1 < E < 16 kJ mol-1). The E value obtained in this study was found to be 10.46 kJ mol-1, revealing nitrate adsorption on granular chitosan-Fe3+ complex

bromide (CPB) modified zeolite adsorbed nitrate [2].

an

3.6 Adsorption kinetics

us

was ion exchange, which was in agreement with other study where cetylpyridinium

In order to accurately reflect the variation trend of nitrate concentration with time

M

and reveal the reaction pathway of adsorption process, several kinetic models were employed to examine the experimental data. The nonlinear pseudo-first-order and

Ac ce p

te

d

pseudo-second-order kinetic models were expressed as follows [44]: (14) (15)

where qe (mg g-1) and qt (mg g-1) represent the amount of nitrate adsorbed at

equilibrium and at time t, respectively, k1 (min-1) and k2 (g mg-1 min-1) denote the

pseudo-first-order and pseudo-second-order rate constants, separately, t (min) is the contact time.

In addition, the intraparticle diffusion model [45] was utilized to predict whether intraparticle diffusion was rate-determining step, which was given as: (16) where kp (mg g-1 min-0.5) is the intraparticle diffusion rate constant, C (mg g-1) is the intercept related to the thickness of the boundary layer.

Page 17 of 38

Fig. 7. Kinetic studies of nitrate removal with an initial concentration of 50 mg NO3-N L-1: pseudo-first-order kinetic (the dash line), pseudo-second-order kinetic (the

ip t

solid line) (a); intraparticle diffusion (b).

Table 2

cr

Kinetic constants for adsorption of nitrate on the chitosan-Fe3+ complex.

us

It was observed from Fig. 7a that adsorption reaction was rapid in the first 15 min, followed by a sluggish stage until equilibrium. The equilibrium time of granular

an

chitosan-Fe3+ complex (1.5 h) was inferior to that of chitosan hydrogel beads (10 h)

M

[37] and organoclay (4 h) [1]. Compared with the pseudo-first-order kinetic, it was evident from Table 2 that the pseudo-second-order kinetic was fitted well with the

d

experimental data in terms of correlation coefficient. For the intraparticle diffusion

te

model (Fig. 7b), three segments of straight line could perfectly match a fitting curve of qt versus t0.5, implying intraparticle diffusion was not an exclusive rate-determining

Ac ce p

step. Nitrate adsorption involved three steps [44]: a rapid external surface diffusion process (step 1), a slow intraparticle diffusion stage (step 2) and a dynamic adsorption-desorption equilibrium (step 3). These three processes often occurred simultaneously.

3.7 Thermodynamic parameters Temperature was practically regarded as a key factor adjusting reaction rate and

affecting adsorption capacity. The results with respect to the effect of temperature on nitrate removal were shown in Table 3. It was evident that the adsorption capacity slightly declined with the rise of temperature. The increase in desorption trend of

Page 18 of 38

nitrate ions from adsorption sites to bulk solution might be applied to account for this phenomenon. The results as mentioned above were further verified by thermodynamic parameters,

ip t

such as Gibbs free energy change (∆Go), enthalpy change (∆Ho) and entropy change (∆So). The ∆Go (kJ mol-1) value was calculated using the following equation [32]:

cr

(17)

where R (8.314 J mol-1 K-1) is the universal gas constant, T (K) is the temperature in

an

ions. The Kd value was calculated as follows [43]:

us

Kelvin, Kd (mL g-1) is the distribution coefficient reflecting the selectivity for nitrate

(18)

M

where qe (mg g-1) and Ce (mg mL-1) is the adsorption capacity on granular chitosanFe3+ complex and equilibrium concentration in bulk solution, respectively. The ∆So (J

d

mol-1 K-1) and ∆Ho (kJ mol-1) values could be estimated from the intercept and slope

(19)

Ac ce p

te

of linear plot of lnKd versus 1/T, respectively through the following equation [44]:

Fig. 8. The plot of lnKd versus 1/T for nitrate adsorption on granular chitosan-Fe3+ complex.

Table 3

Thermodynamic parameters for nitrate adsorption on the chitosan-Fe3+ complex.

Fig. 8 showed the exothermic nature of nitrate adsorption was confirmed by the negative ∆Ho value. The negative value of ∆Go at various temperatures indicated that

Page 19 of 38

adsorption reaction was a thermodynamically feasible and spontaneous process. These results were in agreement with other studies where crosslinked chitosan-Fe(III) complex adsorbed textile dye [21] and Cr(IV) [22]. The higher negative of ∆Go with

ip t

the increase in temperature revealed nitrate adsorption on granular chitosan-Fe3+ complex was more likely to occur at higher temperature. The positive value of ∆So

cr

clearly showed the affinity of granular chitosan-Fe3+ complex for nitrate ions and the increase in the randomness at the solid/solution interface during nitrate adsorption.

us

3.8 Regeneration

an

High adsorption capacity and good reusability were of utmost importance for any adsorbent, which would significantly promote economic value of adsorption method.

M

In the present study, desorption efficiencies of 0.1 M NaCl, 0.1 M NaOH and 0.1 M Na2SO4 solution were found to be 84.3%, 95.3% and 93.8%, respectively.

d

Consequently, granular chitosan-Fe3+ complex was better than layered double

te

hydroxide whose desorption efficiency was less than 2% [46]. It was evident that these desorption reagents could efficiently regenerate granular chitosan-Fe3+ complex

Ac ce p

based on ion exchange reaction, which conduced to recovering nitrate from nitratecontaminated water. However, when regenerated granular chitosan-Fe3+ complex was firstly applied to adsorb nitrate, nitrate removal efficiencies were 78.1%, 0.8% and 22.5% for NaCl, NaOH and Na2SO4 solution, respectively. These results were explained by the fact that hydroxide and sulphate ions irreversibly occupied adsorption sites during regeneration, and thereby available adsorption sites for the first reuse significantly decreased. Moreover, nitrate removal efficiency slightly decreased from 78.1% to 76% undergoing four circles using 0.1 M NaCl solution as the eluent, indicating there were few irreversible adsorption sites on adsorbent surface [13]. Therefore, NaCl was a promising desorption reagent.

Page 20 of 38

4

Conclusion Granular chitosan-Fe3+ complex had high performance for nitrate adsorption. The

maximum adsorption capacity was 8.35 mg NO3--N g-1 based on Langmuir-

ip t

Freundlich model. Granular chitosan-Fe3+ complex had relatively large particle size (1.04-1.16 mm) and short equilibrium time (1.5 h). XRD analysis indicated that

cr

granular chitosan-Fe3+ complex was an amorphous material. The BET specific surface

us

area, total pore volume and average pore size were 8.98 m2 g-1, 0.019 cm3 g-1 and 56.94 Å, respectively. The point of zero charge was obtained at pH 5.4. Adsorption

an

process infrequently suffered from initial pH and temperature, reflecting high chemical stability and good environmental adaptation. The adverse influence of

M

sulphate ions on nitrate adsorption was mostly significant compared with other anions (Cl-, HCO3-, F-), while commonly cations (K+, Na+, NH4+, Ca2+, Mg2+) rarely

d

interfered with nitrate adsorption. Nitrate adsorption was mainly attributed to ion

te

exchange and electrostatic attraction. Thermodynamic study revealed that nitrate adsorption on granular chitosan-Fe3+ complex was a spontaneous and exothermic

Ac ce p

process. Granular chitosan-Fe3+ complex could be easily regenerated using NaCl

solution.

Acknowledgements

The authors acknowledge financial supports from the Foundation for the Advisor of

Beijing Excellent Doctoral Dissertation (No. 20121141501; No. 20131141502), the National Natural Science Foundation of China (NSFC) (No. 21407129) and the Fundamental Research Funds for the Central Universities (No. 2652014101).

Page 21 of 38

References [1] S. Bagherifam, S. Komarneni, A. Lakzian, A. Fotovat, R. Khorasani, W.Y. Huang, J.F. Ma, S.Q. Hong, F.S. Cannon, Y.J. Wang, Highly selective removal of

ip t

nitrate and perchlorate by organoclay, Appl. Clay Sci. 95 (2014) 126–132. [2] Y.H. Zhan, J.W. Lin, Z.L. Zhu, Removal of nitrate from aqueous solution using

cr

cetylpyridinium bromide (CPB) modified zeolite as adsorbent, J. Hazard. Mater. 186 (2011) 1972–1978.

us

[3] World Health Organization (WHO), Guidelines for Drinking Water Quality, 4th

an

ed, 2011.

[4] J. Schick, P. Caullet, J.L. Paillaud, J. Patarin, C. Mangold-Callarec, Batch-wise

M

nitrate removal from water on a surfactant-modified zeolite, Microporous Mesoporous Mat. 132 (2010) 395–400.

d

[5] P. Loganathan, S. Vigneswaran, J. Kandasamy. Enhanced removal of nitrate from

te

water using surface modification of adsorbents – A review, J. Environ Manage. 131 (2013) 363–374.

Ac ce p

[6] J. Chung, K. Amin, S. Kim, S. Yoon, K. Kwon, W. Bae, Autotrophic denitrification of nitrate and nitrite using thiosulfate as an electron donor, Water

Res. 58 (2014) 169–178.

[7] M. Alikhani, M.R. Moghbeli, Ion-exchange polyHIPE type membrane for removing nitrate ions: Preparation, characterization, kinetics and adsorption studies, Chem. Eng. J. 239 (2014) 93–104.

[8] T. Öznülüer, B. Özdurak, H.Ö. Doğan, Electrochemical reduction of nitrate on graphene modified copper electrodes in alkaline media, J. Electroanal. Chem. 699 (2013) 1–5.

Page 22 of 38

[9] L.A. Richards, M. Vuachère, A.I. Schäfer, Impact of pH on the removal of fluoride, nitrate and boron by nanofiltration/reverse osmosis, Desalination 261 (2010) 331–337.

ip t

[10]K. Wada, T. Hirata, S. Hosokawa, S. Iwamoto, M. Inoue, Effect of supports on Pd-Cu bimetallic catalysts for nitrate and nitrite reduction in water, Catal. Today

cr

185 (2012) 81–87.

[11]S. Chatterjee, D.S. Lee, M.W. Lee, S.H. Woo, Nitrate removal from aqueous

us

solutions by cross-linked chitosan beads conditioned with sodium bisulfate, J.

an

Hazard. Mater.166 (2009) 508–513.

[12]A. Bhatnagar, M. Sillanpää, A review of emerging adsorbents for nitrate removal

M

from water, Chem. Eng. J. 168 (2011) 493–504.

[13]C.S. Shen, Y. Shen, Y.Z. Wen, H.Y. Wang, W.P. Liu, Fast and highly efficient

d

removal of dyes under alkaline conditions using magnetic chitosan-Fe(III)

te

hydrogel, Water Res. 45 (2011) 5200–5210. [14]A. Afkhami, T. Madrakian, Z. Karimi, The effect of acid treatment of carbon

Ac ce p

cloth on the adsorption of nitrite and nitrate ions, J. Hazard. Mater. 144 (2007) 427–431.

[15]D. Bhardwaj, M. Sharma, P. Sharma, R. Tomar, Synthesis

and

surfactant

modification of clinoptilolite and montmorillonite for the removal of nitrate and preparation of slow release nitrogen fertilizer, J. Hazard. Mater. 227–228 (2012) 292–300.

[16]X. Xing, B.Y. Gao, Q.Q. Zhong, Q.Y. Yue, Q. Li, Sorption of nitrate onto aminecrosslinked wheat straw: Characteristics, column sorption and desorption properties, J. Hazard. Mater. 186 (2011) 206–211.

Page 23 of 38

[17]M. Zhang, B. Gao, Y. Yao, Y.W. Xue, M. Inyang, Synthesis of porous MgObiochar nanocomposites for removal of phosphate and nitrate from aqueous solutions, Chem. Eng. J. 210 (2012) 26–32.

ip t

[18]H. Guana, E. Bestland, C.Y. Zhu, H.L. Zhu, D. Albertsdottir, J. Hutson, C.T. Simmons, M. Ginic-Markovic, X. Tao, A.V. Ellis, Variation in performance of

cr

surfactant loading and resulting nitrate removal among four selected natural zeolites, J. Hazard. Mater. 183 (2010) 616–621.

us

[19]R. Laus, V.T.D. Fávere, Competitive adsorption of Cu(II) and Cd(II) ions by

an

chitosan crosslinked with epichlorohydrin–triphosphate, Bioresour. Technol. 102 (2011) 8769–8776.

[20]J. Qu, Q.L. Hu, K. Shen, K. Zhang, Y.Y. Li, H. Li, Q.R. Zhang, J.Q. Wang, W.Q.

M

Quan, The preparation and characterization of chitosan rods modified with Fe3+

d

by a chelation mechanism, Carbohydr. Res. 346 (2011) 822–827.

te

[21]C.A. Demarchi, M. Campos, C.A. Rodrigues, Adsorption of textile dye Reactive Red 120 by the chitosan–Fe(III)-crosslinked: Batch and fixed-bed studies, J.

Ac ce p

Environ. Chem. Eng. 1 (2013) 1350–1358.

[22]A.C. Zimmermann, A. Mecabô, T. Fagundes, C.A. Rodrigues, Adsorption of Cr(VI) using Fe-crosslinked chitosan complex (Ch-Fe), J. Hazard. Mater. 179 (2010) 192–196.

[23]T. Fagundes, E.L. Bernardi, C.A. Rodrigues, Phosphate adsorption on chitosanFeIII-crosslinking: batch and column studies, J. Liq. Chromatogr. Relat. Technol. 24 (2001) 1189–1198. [24]H.H.D Santo, C.A. Demarchi, C.A. Rodrigues, J.M Greneche, N. Nedelko, A. Slawska-Waniewska, Adsorption of As(III) on chitosan-Fe-crosslinked complex (Ch-Fe), Chemosphere 82 (2011) 278–283.

Page 24 of 38

[25]D.J. Wan, H.J. Liu, R.P. Liu, J.H. Qu, S.S. Li, J. Zhang, Adsorption of nitrate and nitrite from aqueous solution onto calcined (Mg–Al) hydrotalcite of different Mg/Al ratio, Chem. Eng. J. 195–196 (2012) 241–247.

ip t

[26]M. Karimi, M. Entezari, M. Chamsaz, Sorption studies of nitrate ion by a modified beet residue in the presence and absence of ultrasound, Ultrason.

cr

Sonochem. 17 (2010) 711–717.

[27]P. Chutia, S. Kato, T. Kojima, S. Satokawa, Arsenic adsorption from aqueous

us

solution on synthetic zeolites, J. Hazard. Mater. 162 (2009) 440–447.

an

[28]D. Thakre, S. Jagtap, N. Sakhare, N. Labhsetwar, S. Meshram, S. Rayalu, Chitosan based mesoporous Ti–Al binary metal oxide supported beads for

M

defluoridation of water, Chem. Eng. J. 158 (2010) 315–324. [29]T.B. Mostafa, A.S. Darwish, An approach toward construction of tuned

d

chitosan/polyaniline/metal hybrid nanocomposites for treatment of meat industry

te

wastewater, Chem. Eng. J. 243 (2014) 326–339. [30]P. Yuan, F. Annabi-Bergaya, Q. Tao, M.D. Fan, Z.W. Liu, J.X. Zhu, H.P. He,

Ac ce p

T.H. Chen, A combined study by XRD, FTIR, TG and HRTEM on the structure of delaminated Fe-intercalated/pillared clay, J. Colloid Interface Sci. 324 (2008) 142–149.

[31]A. Kamari, I.D. Pulford, J.S.J. Hargreaves, Chitosan as a potential amendment to remediate metal contaminated soil – A characterisation study, Colloids Surf B Biointerfaces 82 (2011) 71–80.

[32]A. Sarı, M. Tuzen, M. Soylak, Adsorption of Pb(II) and Cr(III) from aqueous solution on Celtek clay, J. Hazard. Mater. 144 (2007) 41–46.

Page 25 of 38

[33]R.A.K. Rao, F. Rehman, Adsorption studies on fruits of Gular (Ficus glomerata): Removal of Cr(VI) from synthetic wastewater, J. Hazard. Mater. 181 (2010) 405– 412.

ip t

[34]H.B. Senturk, D. Ozdes, A. Gundogdu, C. Duran, M. Soylak, Removal of phenol from aqueous solutions by adsorption onto organomodified Tirebolu bentonite:

cr

Equilibrium, kinetic and thermodynamic study, J. Hazard. Mater. 172 (2009) 353–362.

us

[35]D. Kołodyńska, Chitosan as an effective low-cost sorbent of heavy metal

an

complexes with the polyaspartic acid, Chem. Eng. J. 173 (2011) 520–529. [36]Y.H. Li, P. Zhang, Q.J. Du, X.J. Peng, T.H. Liu, Z.H. Wang, Y.Z. Xia, W. Zhang,

M

K.L. Wang, H.W. Zhu, D.H. Wu, Adsorption of fluoride from aqueous solution by graphene, J. Colloid Interface Sci. 363 (2011) 348–354.

d

[37]S. Chatterjee, S.H. Woo, The removal of nitrate from aqueous solutions by

te

chitosan hydrogel beads, J. Hazard. Mater. 164 (2009) 1012–1018. [38]P. Ganesan, R. Kamaraj, S. Vasudevan, Application of isotherm, kinetic and

Ac ce p

thermodynamic models for the adsorption of nitrate ions on graphene from aqueous solution, J. Taiwan Inst. Chem. Eng. 44 (2013) 808–814.

[39]V.C. Srivastava, M.M. Swamy, I.D. Mall, B. Prasad, I.M. Mishra, Adsorptive removal of phenol by bagasse fly ash and activated carbon: Equilibrium, kinetics and thermodynamics, Colloids and Surfaces A: Physicochem. Eng. Aspects 272 (2006) 89–104.

[40]A. Olgun, N. Atar, S.B. Wang, Batch and column studies of phosphate and nitrate adsorption on waste solids containing boron impurity, Chem. Eng. J. 222 (2013) 108–119.

Page 26 of 38

[41]A. Mirmohseni, M.S.S. Dorraji, A. Figoli, F. Tasselli, Chitosan hollow fibers as effective biosorbent toward dye: Preparation and modeling, Bioresour. Technol. 121 (2012) 212–220. Karamanis,

P.A.

Assimakopoulos,

Efficiency

of

aluminum-pillared

ip t

[42]D.

montmorillonite on the removal of cesium and copper from aqueous solutions,

cr

Water Res. 41 (2007) 1897–1906.

[43]M. Jaina, V.K. Garg, K. Kadirvelu, Adsorption of hexavalent chromium from

us

aqueous medium onto carbonaceous adsorbents prepared from waste biomass, J.

an

Environ. Manage. 91 (2010) 949–957.

[44]D.H. Ding, Y.X. Zhao, S.J. Yang, W.S. Shi, Z.Y. Zhang, Z.F. Lei, Y.N. Yang,

M

Adsorption of cesium from aqueous solution using agricultural residue Walnut shell: Equilibrium, kinetic and thermodynamic modeling studies, Water Res. 47

d

(2013) 2563–2571.

te

[45]E. Kumar, A. Bhatnagar, J.A. Choi, U. Kumar, B. Min, Y. Kim, H. Song, K.J. Paeng, Y.M. Jung, R.A.I. Abou-Shanab, B.H. Jeon, Perchlorate removal from

Ac ce p

aqueous solutions by granular ferric hydroxide (GFH), Chem. Eng. J. 159 (2010) 84–90.

[46]R. Katal, M.S. Baei, H.T. Rahmati, H. Esfandian, Kinetic, isotherm and thermodynamic study of nitrate adsorption from aqueous solution using modified rice husk, J. Ind. Eng. Chem. 18 (2012) 295–302.

Page 27 of 38

Table 1 Langmuir qm 5.00 (mg g-1) KL 0.074 (L mg-1)

Freundlich KF ((mg g-1) (L mg-1)1/n)

R2

0.975

R2

0.991

R2

0.999

R2

χ2

0.535

χ2

0.060

χ2

0.017

χ2

NSD (%)

0.179

NSD (%)

0.066

NSD (%)

0.041

NSD (%)

8.35 0.091

Dubinin-Radushkebbich qm 9.30 (mg g-1) β 4.6 × 10-5 (mol kJ-2)

0.567

ip t

0.998 0.019

0.041

Ac ce p

te

d

M

an

us

cr

n

Langmuir-Freundlich qm 0.90 (mg g-1) KL-F 2.77 (L mg-1) m

Page 28 of 38

Table 2

2

Pseudo-second-order qe 2.20 (mg g-1) k2 0.081 (g mg-1 min-1) 2 R 0.998 2

kp1 (mg g-1 min-0.5) C1 (mg g-1) R2 2

Intraparticle diffusion kp2 0.418 (mg g-1 min-0.5) C2 0.032 (mg g-1) 0.992 R2 2

0.040

χ

0.005

χ

0.004

χ

NSD (%)

0.054

NSD (%)

0.017

NSD (%)

0.022

NSD (%)

1.27

0.961

4.5 × 10-4

0.013

Ac ce p

te

d

M

an

us

cr

χ

0.108

ip t

Pseudo-first-order qe 2.01 (mg g-1) k1 0.116 (min-1) 2 R 0.988

Page 29 of 38

Table 3 qe (mg g-1)

288

2.07

-12.75

298

2.03

-12.99

308

2.01

318

1.98

-13.53

328

1.96

-13.83

-13.30

-5.06

So (J mol-1 K-1)

27.65

Ac ce p

te

d

M

an

us

0.996

Ho (kJ mol-1)

cr

R2

ip t

Go (kJ mol-1)

Temperature (K)

Page 30 of 38

Fig. 1 (b)

(c)

(d)

Ac ce p

te

d

M

an

us

cr

ip t

(a)

Page 31 of 38

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 2

Page 32 of 38

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 3

Page 33 of 38

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 4

Page 34 of 38

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 5

Page 35 of 38

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 6

Page 36 of 38

Ac ce p

te

d

M

an

us

cr

ip t

Fig. 7

Page 37 of 38

an

us

cr

ip t

Fig. 8

Ac ce p

te

d

M

1 2

Page 38 of 38