THE BIOSORPTION POTENTIAL OF MODIFIED BOTTLE GOURD …

Chemical Industry & Chemical Engineering Quarterly

Available on line at Association of the Chemical Engineers of Serbia AChE www.ache.org.rs/CICEQ

Chem. Ind. Chem. Eng. Q. 24 (4) 319−332 (2018) CI&CEQ

319

DRAGANA Z.

MARKOVIĆ-NIKOLIĆ1 ALEKSANDAR LJ. BOJIĆ2

DANIJELA V. BOJIĆ2

MILORAD D. CAKIĆ3

DRAGAN J. CVETKOVIĆ3

GORAN S. NIKOLIĆ3

1High Technologically Artistic Professional School, Leskovac,

Serbia 2University of Niš, Faculty of

Science and Mathematics, Department of Chemistry, Niš,

Serbia 3University of Niš, Faculty of

Technology, Leskovac, Serbia

SCIENTIFIC PAPER

UDC 635.627:544.3/.4

THE BIOSORPTION POTENTIAL OF MODIFIED BOTTLE GOURD SHELL FOR PHOSPHATE: EQUILIBRIUM, KINETIC AND THERMODYNAMIC STUDIES

Article Highlights • A new low-cost and efficient cationic biosorbent for phosphate is presented • Phosphate biosorption process is essentially exothermic and spontaneous • The anion exchange mechanism is the dominant biosorption process • Freundlich isotherm describes favorable multilayer phosphate biosorption • Biosorbent can be easily separated, recycled, and reused multiple times Abstract

In order to preserve the environment and prevent the occurrence of eutrophi-cation, a new quaternary ammonium biosorbent based on the bottle gourd (Lagenaria vulgaris) shell (QALVS) was prepared and characterized in the pre-sent study. The results of FTIR and elemental analyses confirmed the suc-cessful modification of the lignocellulosic biomass and showed that 15.68 mg N g-1 was incorporated as cationic –N+R3 group. The biosorbent was tested to remove phosphate from aqueous solution in batch mode. Kinetic studies indi-cated that the biosorption of phosphate onto QALVS can be fitted by a non-lin-ear pseudo nth-order model (n ≈ 1) very well. The Freundlich model provided the best description of the biosorption process. The maximum phosphate bio-sorption capacity was 16.69 mg P g-1 for QALVS at 20 °C and pH 6. Thermo-dynamic parameters indicate that the biosorption of phosphate by QALVS is exothermic. The anion exchange mechanism was the dominant process, in addition to physisorption. Coexisting ions exhibited a distinct effect on phos-phate biosorption with the order of NO3

- > SO42- > Cl-. The QALVS biosorbent

could be repeatedly used in phosphate biosorption with slight losses in initial biosorption capacities.

Keywords: Lagenaria vulgaris, phosphate, biosorption, kinetic, equilib-rium, thermodynamic.

Agro-industrial and municipal wastewaters are hazardous to human health and the environment because they release large amount of phosphorus into natural waters (from 4 to 15 mg P dm-3 as phos-phate). Excessive phosphate in water is one of the main factors leading to the eutrophication and det-erioration of water quality [1]. Leaching of phosphate

Correspondence: G.S. Nikolić, Faculty of Technology, Bulevar oslobodjenja 124, Leskovac 16000, Serbia. E-mail: [email protected] Paper received: 19 October, 2017 Paper revised: 1 January, 2018 Paper accepted: 15 February, 2018

https://doi.org/10.2298/CICEQ171019006M

into the ground water affects the drinking water qua-lity leading to potential risk to human health and ani-mals [2]. That is why the discharge of industrial and municipal wastewater rich in phosphate into natural watercourses is regulated by numerous legislation. In order to solve water quality problems, state environ-mental agencies have demanded from a wastewater treatment plant to achieve a very low level of total phosphorus in effluents (< 0.05 mg dm-3) in order to prevent eutrophication [1]. This has led to the deve-lopment of many treatment methods that can improve the removal and recovery of phosphate from various effluents prior to discharge into natural waters.

D.Z. MARKOVIĆ-NIKOLIĆ et al.: THE BIOSORPTION POTENTIAL OF… Chem. Ind. Chem. Eng. Q. 24 (4) 319−332 (2018)

320

A number of studies have been reported using physical, chemical, biological, or combined processes for the removal of phosphate from wastewater [1-4]. According to current practice, various advanced tech-niques have been developed and successfully applied for reducing phosphorus to low concentrations, such as: tertiary membrane filtration (to 0.04 mg dm-3), chemical precipitation (0.005-0.04 mg dm-3), biologi-cal treatment (0.02-0.1 mg dm-3), reverse osmosis (0.008 mg dm-3), electrodialysis (< 0.005 mg dm-3), etc. [5]. However, these techniques are characterized by many weaknesses, such as: high capital invest-ment, specific conditions and their very strict control, potentially new and poisonous water pollutants, inab-ility to recycle phosphorous, disposal of the generated phosphate sludge, and waste management costs [4]. Therefore, these methods are not sustainable [5].

Over the past decade, considerable attention has been paid to the economic and ecological issues of adsorptive removal of phosphate, using conventi-onal (activated carbon) and low-cost adsorbents (fly ash, sludge, oxide tailings) [5]. However, the adsorp-tion method has inherent application limitations, which are described in Text S-1, Supporting Inform-ation (available from the author upon request). On the other hand, biosorption of phosphate onto low-cost and easily available plant materials in wastewater treatment has attracted great interest [5]. Biosorption has become a promising technique due to high effi-ciency, simplicity, and safety during application, as well as the availability of various natural materials without harmful environmental impact. In addition, nutrients loaded products can be further used as fer-tilizers for agricultural production [6].

Further development of biosorption method has been focused on the research of biosorbent prepared from agricultural residue or waste. Therefore, most of the researchers look for locally available agricultural by-products as they offer efficient, viable and cost-effective solutions. Agricultural by-products benefits are presented in Text S-2, Supporting Information. However, available plant waste has very low biosorp-tion capacity for anions and shows an extremely poor affinity to phosphorous. The natural by-products can efficiently remove metal ions (or other cationic spe-cies) from wastewater due to the presence of various functional groups in the biomass [7]. The biosorption capacity of lignocellulosic materials for anions can be increased by some surface modifications [5]. Chem-ical modification is a newer technique that implies the functionalization of surface groups on the solid with or without carbonization. The newly developed functi-onal groups may provide new sites for biosorption of

phosphate from aqueous solutions. For removal of anions, lignocellulosic materials have been cationized through hybridizing with inorganic chemicals as well as grafting with ammonium-type reagents [5].

In the past, numerous attempts have been made in finding inexpensive and effective anion exchangers produced from agricultural by-products. Studies showed that many materials, such as rice husk [2], sugarcane bagasse [8,9], wheat straw [10], etc., could be modified into anion exchangers and utilized to remove phosphate, as well as other anionic species from the aqueous solution [11,12]. The agricultural by-product represents a potential alternative as an anion exchanger because of its particular properties (high surface area, porosity, thermal resistance, chemical stability) and high reactivity, resulting from the presence of reactive hydroxyl groups in polymer chains. One of these materials can be the bottle gourd (Lagenaria vulgaris) shell (LVS), described in Text S-3, Supporting Information. The dried fruit shell, as a biocompatible natural material of hard and por-ous structure, is an exceptionally valuable agricultural residue for making potentially attractive biosorbents of different functionalities. This is due to the large amount of easily available hydroxyl groups existing in the cellulose, hemicelluloses, and lignin, which can easily make a series of chemical reactions, such as conden-sation, etherification and copolymerization [13,14]. The use of LVS as an efficient and inexpensive alter-native biosorbent in natural and activated forms (acti-vated carbon) for the removal of some toxic metal ions from wastewater (Pb2+, Cd2+, Zn2+, Cu2+ and Ni2+) [7,15,16], and other cationic pollutants, such as dyes [17] or pharmaceuticals [18], was examined. Accord-ing to the literature, LVS has not been studied as bio-sorbent for removal of anionic pollutants from aque-ous solutions.

The present study illustrates the preparation of a cationic biosorbent from LVS. The biosorbent system for anions, containing trimethylammonium-hydroxy-propyl groups (QALVS), was obtained through the chemical reactions of the lignocellulosic LVS biomass with epichlorohydrin and trimethylamine. The import-ance of using these agents is described in Text S-4, Supporting Information. The synthesis success of the quaternary ammonium product with anion exchange property was confirmed using physicochemical and spectroscopic methods. Our main objective is to study the biosorption potential and efficiency of QALVS for phosphate removal from aqueous solution in batch conditions. Bearing in mind that successful phosphate biosorption requires a better understanding of the mechanism, the main emphasis in this paper is dev-


321

oted to the analysis of the nature of biosorption, thermodynamic parameters, equilibrium and biosorp-tion kinetics. The analyzed data, together with the known effect of coexisting ions on phosphate biosorp-tion and the possibility of regeneration and reuse of biosorbent, can provide valuable information for the design and operation of wastewater treatment plants. This represents a great practical value for the tech-nological applications of phosphate removal from aqueous solutions by QALVS as a potential anion exchanger.

MATERIALS AND METHODS

Reagents and materials

All reagents used in this study (epichlorohydrin, trimethylamine, dimethylformamide and pyridine) were of analytical grade (Sigma-Aldrich Chemie GmbH). The solutions for modification and synthesis processes, as well as to rinse the obtained products, were prepared with deionized water (18 MΩ). Stock solution of phosphate (1000 mg P dm-3) was obtained by dissolving 4.392 g of pure KH2PO4 in 1 dm3 of deionized water. Working standard solutions (5, 25, 50 and 100 mg P dm-3) were prepared by suitable dilution of the stock solution.

Mature fruit L. vulgaris from the southern part of Serbia (near Leskovac), grown at an altitude of around 250 m, was used as the initial lignocellulosic material in this experiment. The naturally dried shell of L. vul-garis was manually cleaned and crushed into small pieces of 1 to 2 cm. The raw material was soaked and washed with deionized water to remove dust and sol-uble impurities. The washed material was dried in an oven for 24 h at 60 °C, milled in a crusher mill (Waring 8010 ES, Germany) and sieved (Oct-Digital 4527) to separate the sizes between 400 and 800 μm.

Characterization methods

The constituents and physical properties of LVS were determined according to the standard AOAC procedures [19]. Elemental analysis (CHNS/O) of unmodified LVS and QALVS biosorbent was per-formed using an Elemental Analyzer (Vario EL III CHNS/O systeme GmbH). The infrared spectra of investigated samples were recorded by a FTIR spec-troscope (BOMEM MB-100, Hartmann & Braun, Canada), in the range of 4000-400 cm-1, with 8 scans at resolution of 2 cm-1. Win Bomem Easy software was used to analyze the obtained FTIR spectra.

Biosorbent synthesis

The dried and milled LVS (10 g) was suspended in dimethylformamide (60 cm3) at room temperature

(23 °C). After the adjustment of pH value at 10 using NaOH solution (5 M), the mixture was stirred for 30 min at 80 °C in a three-neck round bottom flask (250 cm3). Afterwards, in the flask was added epichloro-hydrin (50 cm3) and the reaction was carried out at 80 °C for 2 h. The reaction product was washed with an aqua diluted ethanol (1:1) at 40 °C. Quaternary ammonium groups were then introduced into the epo-xypropyl by-product after reaction with trimethylamine (50 cm3) in the presence of pyridine (20 cm3) as a catalyst. This synthesis was performed in a three-necked flask at 80 °C for 4 h. The product was washed with ethanol solution (50%), followed by NaOH (0.1 M) and HCl (0.1 M) to remove the residual chemicals. Finally, the product was washed with deionized water at 40 °C and vacuum dried at 40 °C for 2 h. The resulting QALVS product was then used in all the biosorption experiments.

Biosorption experiments

Phosphate biosorption experiments were con-ducted in batch mode (pH 6, 60 min, 150 rpm), using 50 cm3 of the phosphate working solution (5, 25, 50, or 100 mg P dm-3), which were optimized by preli-minary testing. The solution was stirred at 150 rpm (using a magnetic stirrer) in order to ensure a good mixing without vortex effect. Heating was done using a water bath set to the desired temperature (20, 30, 40 and 50, ±1 °C). The defined dose of QALVS bio-sorbent (2 g dm-3) was added to the working solution. Aliquots of the working solutions (3 cm3) were with-drawn at appropriate time intervals (up to 60 min), fol-lowed by filtering through a 0.45 μm microfiltration membrane (Agilent Technologies, Germany), diluted in 4% HNO3 and preserved with CCl4 to measure-ment. Filtrates were analyzed for residual phosphate concentration (expressed over phosphorus), axially at 213.618 nm using an inductively coupled plasma-opti-cal emission spectrometer (ICP-OES, Spectro ARCOS FHE12, Germany), according to the manufacturer’s instructions. The equilibrium phosphate concentration was calculated as follows:

( )−= 0 e

e

V C CQ

m (1)

where Qe (mg P g-1) is the amount of phosphorus biosorption per gram QALVS at equilibrium; C0 and Ce (mg P dm-3) are the concentrations of phosphorus at original and equilibrium, respectively; V (dm-3) is the volume of solution, and m (g) is the dry mass of QALVS biosorbent.


322

Data presentation and statistical analysis

Biosorption phosphate tests were replicated three times and only mean values are presented. The kinetics and equilibrium model parameters were eva-luated by linear and non-linear regression (Leven-berg-Marquardt method) using OriginPro 8.0 software (OriginLab Corporation, USA). Significant levels were set at α = 0.05. Statistical analysis was processed using the following parameters: determination coef-ficient (R2), standard deviation (SD), probability for a given statistical model (p-value), sum of the squares of the errors (SSR), average relative error (ARE in %), and chi-squared distribution (χ2). The best matching of the applied models was verified using the lowest values of the statistical parameters (except R2).

RESULTS AND DISCUSSION

Compositional and elemental analysis

The initial LVS material was subjected to exam-ination of physical (density and bulk weight), proxi-mate (moisture, volatile matter and ash), and bio-chemical properties (cellulose, hemicelluloses, lignin, protein). All values are means of triplicate determin-ations and expressed in percentage (%) based on dry matter (Table 1). Lignin is most represented compo-nent in the LVS biomass. A high proportion of holo-cellulose in raw material suggests the more hydroxyl groups available in the epichlorohydrin-ammonium reaction. The protein content is extremely low, while the volatile matters were not detected considering that the LVS fruit is naturally dried. Also, the plant material was generally low in ash. The biomass is characterized by lower moisture content and bulk density. The obtained results are in accordance with the properties of other agricultural residues from the genus Lagenaria [7].

Elemental analysis (CHNS/O) was performed to assess functional groups in the biosorbent during the modification process (Table 1). Based on that, the

amount of nitrogen added (Nadd, mmol N g-1) the reaction efficiency (RE, %) and the degree of substi-tution (DS) of QALVS were determined by the follow-ing equations [18]:

( )= −add QALVS LVS0.714N N N (2)

( )=× −

tot

tot w

16214 100

NDSN M

(3)

= add100NREF

(4)

where NQALVS and NLVS are contents of nitrogen in the quaternized and raw biomass (%) estimated from ele-mental analysis; the factor 0.714 converts % N to mmol N g-1 of biomass; the factor F is amount (mmol) of trimethylamine added to the reaction mixture per gram of LVS biomass; Ntot (%) is the total nitrogen of QALVS; Mw is the molecular weight (g mol-1) of tri-methylamine; 162 is the molar weight of anhydro-glucose units, and 14 is the relative atomic mass of nitrogen.

The H/C ratio was 0.13% in LVS, as well as in QALVS. The presence of sulfur has not been iden-tified. Attachment of trimethylamine was evaluated by increasing the nitrogen content. It was calculated (Eq. (2)) that 15.68 mg N g-1 was incorporated onto LVS surface as cationic –N+R3 group, indicating a theo-retical exchange capacity of 1.12 mEq g-1. It is clear that the reaction between the LVS and the quater-nizing reagent was not particularly efficient. This is supported by a small value of the substitution degree, indicating that the substitution takes place on each fifth glucopyranose unit in the cellulose chain. Hence it can be assumed that the amount of quaternizing reagent was not the limiting factor of the reaction effi-ciency. In this regard, barriers to reactivity with the reactant could include bulk density, which can be a function of biomass composition [11].

Table 1. Compositional and elemental analysis of lignocellulosic LVS biomass and QALVS biosorbent; RE - reaction efficiency; DS – degree of substitution; Nadd - amount of nitrogen added to LVS biomass

Compositional analysis Elemental analysis

Parameter LVS Parameter LVS QALVS

Cellulose, % 39.58 C, % 45.46 46.81

Hemicellulose, % 18.22 H, % 5.98 6.53

Lignin, % 41.90 O, % 48.55 45.08

Proteins, % < 0.1 N, % 0.01 1.58

Ash, % 0.28 Nadd / mmol g-1 - 1.12

Density, g cm-3 0.46 RE / % - 4.39

Moisture, % 3.80 DS - 0.20


323

Spectroscopic analysis

The FTIR structural characterization of both raw LVS and QALVS biosorbent aims to confirm the successful modification of the biomass to a functional biosorbent by the quaternization process (Figure 1). As shown in Figure 1a, the spectrum displays a num-ber of absorption bands, indicating the complex nat-ure of LVS. It is clear that LVS contains functional groups that are characteristic for cellulose (900-1300 cm-1) and lignin (1200-1750 cm-1) [20]. Bands iden-tified at 1036, 1106 and 1161 cm-1 are typical for O-H, C-OH, C-O and CH2 glycoside groups in cellulose. Main peaks at higher wavenumbers (1260, 1324, 1376, 1424, 1462, 1510, 1601 and 1738 cm-1) come from the absorption of C-OH, C-H, CH2, CH3, C-O, C=C and C=O groups from lignin and hemicelluloses, respectively. These peaks are used for the detection of structural changes during chemical treatment of raw LVS.

It can be noticed from Figure 1b that chemical modification of LVS by using both epichlorohydrin and

trimethylamine induces some changes of several FTIR bands. Moreover, the number of bands in the finger print region of QALVS has increased with res-pect to those in LVS. Likewise, in some bands, dec-reasing intensity and movement have been expres-sed. For example, a lowering of intensity (at 3411, 1260, 1036, 608 and 443 cm-1 originates from O-H, at 1738 cm-1 originates from C=O, and at 897 cm-1 ori-ginates from β-glucosidic linkage), increase in inten-sity (at 2890 cm-1 originates from -CH2), shifting of positions for cellulose (at 1266, 1066, and 911 cm-1) and hemicellulose (at 1718 cm-1) groups in QALVS were observed. Obviously, the breakdown of the gly-cosidic structure of hemicellulose occurs during the LVS modification to QALVS, and the -OH groups of cellulose are involved in the reaction with epichloro-hydrin. The decrease in band intensity typical for –OH groups may be due to the reaction of glucopyranose units with epichlorohydrin [21]. In addition, it has been observed that there is even a diminishing of bands at 1601 cm-1 (originates from C=C, which is simultane-

Figure 1. FTIR spectra of raw LVS biomass (a) and QALVS biosorbent (b).


324

ously moving to 1594 cm-1) and shifting of bands at 1371 cm-1 (originates from the O-H phenolic group) and at 1475 cm-1 (originates from -OCH3) in the case of lignin groups as compared with LVS. These obs-ervations indicate a partial destruction of the lignin during the modification [22]. From the other side, it is characteristic that the IR band at about 1260 cm-1 (originates from the C-OH groups) doubled to two bands (at 1266 and 1227 cm-1), indicating the appear-ance of -OH groups of different origins, probably from cellulose and attached epichlorohydrin, with respect to those in LVS. A similar phenomenon was also obs-erved with the band at 1036 cm-1 (doubling to the two bands at 1066 and 1029 cm-1) which confirms the previous assumption. Most importantly, the FTIR spectrum of QALVS (Figure 1b) shows the emer-gence of a new weak band at 1490 cm-1, correspond-ing to the C-H bending vibration from quaternary –N+-(CH3)3 groups [20]. These observations clearly indicate the incorporation of a cationic –N+R4 group onto the biosorbent surface [10].

Biosorption kinetics

Kinetics of a biosorption process is important in water treatment because it offers valuable information on the reaction pathways, the rate constant and the control mechanism of the biosorption process [23,24]. In order to demonstrate the phosphate biosorption process, a kinetic investigation was conducted using the linear (pseudo first-order, pseudo second-order and intra-particle diffusion) and non-linear (pseudo nth-order) kinetic models [25-28]. The experimental values of the phosphate biosorption onto QALVS obtained under optimum conditions (20 °C, pH 6, agit-ation speed of 150 rpm for 60 min, and initial phos-phate concentration in the range of 5-100 mg P dm-3) were analyzed by selected kinetic models. The results, illustrated in Figure 2, indicate that pseudo first-order and intra-particle diffusion models can only describe the biosorption process before reaching equilibrium (0–20 min), while pseudo nth-order and pseudo second-order models can be applied for the entire biosorption process.

The corresponding kinetic parameters for all applied models are shown in Table 2. The deter-mination coefficient values of pseudo nth-order model for phosphate biosorption are satisfactory (R2 > 0.998) and the best compared to those of other applied models, as shown in Table 2. In addition, theoretic-ally, Qcal values by nth-order model are very close to Qexp values (ARE = 0.03-0.36%). In another case, although the R2 > 0.992 by pseudo first-order model, the values of Qexp are different from Qcal (ARE = 5.03-

-8.19%). A very similar situation was observed with the kinetic model of the pseudo-second order. These parameters indicate that the phosphate biosorption process do not follow both the pseudo first-order and pseudo second-order model. Along with statistical parameters, these facts suggest that the phosphate biosorption by QALVS can be well represented by the pseudo nth-order kinetic model (n ≈ 1) for the entire biosorption process. Besides, the non-linear form of kinetic model (nth-order) is best suited to explain the present biosorption study as compared to linear forms (pseudo first-order and pseudo second-order).

Figure 2. Kinetic models for biosorption of phosphate onto QALVS.

For clarifying the phosphate biosorption mech-anism and determining factors that affect the rate of the process, the experimental values are fitted with


325

the Weber model of intra-particle diffusion (Figure 2d). According to the model [28], there should be linear dependence between Qt and t 1/2 and that the plot starts from the origin. In this case, the rate of the process should be limited only by intra-particle dif-fusion. Any deviation from this model suggests that another mechanism is involved. Besides, the Cid value in the intra-particle diffusion equation (Table 2), which is estimated from the linear plot (based on the intercept), represents the thickness of the boundary layer. Accordingly, the larger intercept indicates a greater effect of the boundary layer [28]. As shown in Figure 2d, it is obvious that linear dependence does not start from the origin (Cid ≠ 0), which is probably the result of a different rate of mass transfer in the initial and final period of the biosorption process [29].

This fact indicates that the intra-particle diffusion is not the only rate limiting mechanism in the biosorption process. Likewise, the small values of Cid parameters, which are very close to zero and which are reduced with an increase in the initial phosphate concentration (Table 2), suggest that the boundary layer effect at phosphate biosorption is smaller. Therefore, other kinetic processes probably occur simultaneously and contribute to the biosorption mechanism, such as ion exchange and physical sorption on the biosorbent surface area [10].

Thermodynamic studies

The effect of temperature on the biosorption pro-cess was analyzed as a function of the contact time of the phosphate with QALVS biosorbent. Experiments

Table 2. Expressions of kinetic models and values of corresponding linear and non-linear regression parameters for phosphate biosorption onto QALVS; The subscripts “exp” and “cal” denote the experimental and calculated values; Qe (mg P g-1) is the biosorption capacity at the equilibrium; Qt (mg P g-1) is the individual capacity in a given time t (min); k1 (min-1) is the pseudo-first order rate cons-tant; k2 (g mg-1 min-1) is the pseudo-second order rate constant; kn is the rate constant and its unit depends on the order of the reaction ((min-1) (mg g-1)1–n); n is the reaction order; kid (mg g-1 min-0.5) is the intra-particle diffusion rate constant; Cid is the thickness of the boundary layer; Rp is the particle radius and D is intra-particle diffusivity

Models Parameters

Initial concentrations of phosphate, mg P dm-3

5 25 50 100

Qexp. 2.48 9.89 12.28 16.13

Pseudo nth-order [25]

( )− −= − − −1

1 1eQ 1n n

t e nQ Q n k t Levenberg-Marquardt method: a = Qe, b = kn, c = n

Qcal. 2.48 9.92 12.28 16.18

ARE 0.16 0.31 0.03 0.36

kn 0.246 0.151 0.154 0.143

n 1.09 0.89 0.89 0.91

R2 0.999 0.998 0.998 0.999

χ2 4.66·10-4 0.023 0.039 0.027

SSR 0.004 0.208 0.356 0.242 Pseudo first-order [23]

− = − 1ln( ) lne t eQ Q Q k t

= intercepteQ e

=1 slopek

Qcal. 2.27 10.51 12.90 17.34

ARE 8.19 6.18 5.03 7.50

k1 -0.223 -0.126 -0.130 -0.130

R2 0.995 0.995 0.999 0.992

SD 0.093 0.069 0.036 0.112

p <0.0001 <0.0001 <0.0001 <0.0001 Pseudo second-order [24]

= +2

e2 e

1 1

t

t tQ Qk Q

= 1/ slopeeQ

= 22 slope / interceptk

Qcal. 2.56 11.24 13.89 18.18

ARE 3.31 13.55 13.10 12.72

k2 0.279 0.015 0.013 0.010

R2 0.998 0.978 0.981 0.985

SD 0.333 0.275 0.206 0.138

p <0.0001 <0.0001 <0.0001 <0.0001 Intra-particle diffusion [26]

= +0.5 t id idQ k t C = interceptidC

π

= =

12e

idp

6slope

Q DkR

Qcal. 2.38 9.93 12.30 16.12

ARE 3.58 0.32 0.16 0.06

kid 0.737 2.196 2.769 3.518

Cid 0.015 -0.499 -0.604 -0.567

R2 0.984 0.977 0.972 0.982

SD 0.130 0.552 0.772 0.779

p <0.0001 <0.0001 <0.0001 <0.0001


326

were conducted at four temperatures (20, 30, 40 and 50 °C), with the biosorbent dose of 2 g dm-3 in 50 cm3 of phosphate solution (initial concentration of 25 mg P dm-3), under optimal conditions (pH 6, agitation speed of 150 rpm for 60 min). Maximum phosphate removal of 79.16% was achieved at 20 °C for 30 min of con-tact time. The biosorption of phosphate onto QALVS shows a decrease in amount of biosorption (from 9.89 to 4.88 mg P g-1) when increasing the temperature (from 20 to 50 °C). The observed decrease in the bio-sorption capacity with an increase of the temperature indicates that lower temperatures favor the removal of phosphate by QALVS. On the other hand, phosphate desorption from the solid surface occurs due to the increase in temperature of the solution [28]. These observations point to the exothermic nature of the biosorption, as described further. A similar effect was reported for the phosphate removal from aqueous sol-ution by chemically modified biomass of sugarcane bagasse [9], modified wheat residue [10], agro-waste rice husk ash [2], etc.

The typical thermodynamic parameters (ΔG0, ΔH0 and ΔS0) were determined from the variations of the thermodynamic distribution coefficient (Kd) with change in temperature to clarify the process of phos-phate biosorption [18]. The Kd coefficient for the bio-sorption reaction was determined as quotient (Csorb/Ce), where Csorb is the equilibrium concentration of sorbed phosphate (mg P dm-3) and Ce is the equi-librium phosphate concentration in the solution (Table 3). The standard free enthalpy and entropy changes were determined by plotting lnKd versus 1/T (Figure 3), from the slope and intercept of the plot, where ΔH0 = −R·slope and ΔS0 = R·intercept. The thermo-dynamic parameter ΔG0 was calculated according to the following thermodynamic equation:

Δ = −0d G RT lnK (5)

The values of thermodynamic parameters for phosphate biosorption onto QALVS are presented in Table 3. The determinate values of ΔH0 and ΔS0 for the biosorption were calculated to be -46.38 kJ mol-1 and -0.147 kJ mol-1 K-1, respectively. The negative value of ΔS0 shows the stability process with no struc-tural change at solid-liquid interface, and suggests

that the system exhibits random behavior [30]. The negative values of ΔH0 and ΔG0 confirm that the bio-sorption is essentially an exothermic and spontane-ous one [9]. However, the positive value of ΔG0 at 50 °C implies that the biosorption is not spontaneous. Thus, the increase in the value of ΔG0 with increasing temperature (from 20 to 50 °C) shows that the bio-sorption is less favorable at higher temperature [30]. Besides, it is established that the ΔH0 value is in the range from -4 to -40 kJ mol-1 for multilayer physisorp-tion, in the range from -40 to -80 kJ mol-1 for ion exchange (electrostatic interactions between the solid charged surface and the ions from the solution), and in the range from -80 to -800 kJ mol-1 (the same order of magnitude as for the average chemical reaction) for monolayer chemisorption [31]. In this case, based on the calculated ΔH0 value, the ion exchange mech-anism for the biosorption is implied to be the main one.

Figure 3. Thermodynamic plot of lnKd versus 1/T for phosphate

biosorption onto QALVS (initial phosphate concentration of 25 mg P dm-3).

Isotherm studies

The two-parameter (Langmuir, Freundlich, Tem-kin, Dubinin-Radushkevich) and three-parameter (Sips) isotherm models were used to describe the equilibrium between the phosphate biosorbed onto QALVS and in the aqueous solution [32-36]. The lin-ear and non-linear forms of these isotherm models were used to analyze the equilibrium isotherm data by OriginPro 8.0 software. The fitness of these models was evaluated by the isotherm parameters, as shown in Table 4.

50 323.16 9.77 15.23 0.641 +1.194

Table 3. Thermodynamic parameters for phosphate biosorption onto QALVS

T Csorb Ce Ke

ΔG0 ΔH0 ΔS0 R2

°C K mg dm-3 mg dm-3 kJ mol-1 kJ mol-1 kJ mol-1 K-1

20 293.16 19.79 5.21 3.798 -3.253 -46.376 -0.147 0.997

30 303.16 16.94 8.06 2.102 -1.873

40 313.16 13.66 11.34 1.204 -0.483


327

Table 4. Expressions of isotherm models and values of corresponding linear and non-linear regression parameters for phosphate biosorption onto QALVS; Qe is equilibrium solid phase concentration (mg g-1); Ce is equilibrium liquid phase concentration (mg dm-3); Qexp, Qmax and Qcal are relevant with experimental, maximum and calculated biosorption capacity, respectively; KL is Langmuir constant related to the energy of biosorption (dm3 mg-1); RL is Langmuir separation factor in the function of initial concentration C0 of phosphate; KF is Freundlich constant characterizing biosorption capacity ((mg g-1) (mg dm-3)1/n); nF is Freundlich parameter related to energy or intensity (1/nF) of biosorption; bT is the Temkin constant related to heat of biosorption (J mol-1 K-1); KT is the Temkin equilibrium binding constant (dm3 mg-1), R is the gas constant (8.314 J mol-1 K-1) and T is the absolute temperature (K); QDR is is the Dubinin-Radushkevich theoretical monolayer saturation biosorption capacity(mg g-1); KDR is Dubinin–Radushkevich isotherm constant (mol2 kJ-2); E is free energy of biosorption (J mol-1) and ε is Polanyi's potential; KS is Sips biosorption constant (dm3 g-1); αS is Sips constant related to energy of biosorption (dm3 mg-1) and βS is exponent

Linear regression

Model Expressions and isotherm parameters Statistical parameters

R2 SD p ARE

Langmuir [32]

= +1e e

e max L max

C CQ Q K Q

Qexp 16.13 0.986 0.277 0.007 3.35

Qmax 16.67

=+ 0

1(1

)L

LR

K C KL 0.283

Freundlich [33] = + 1

ln ln lne F eF

Q K Cn

Qcal 16.69 3.45

KF 5.624 0.988 0.112 0.006

constraint: nF > 1 1/nF 0.258

nF 3.876

Temkin [34] = +R Rln lne T e

T T

T TQ K Cb b

Qcal 14.96 7.24

KT 69.219 0.973 1.149 0.013

bT 1377.024

1/bT 7.262×10-4

Non-linear regression

Model Expressions and isotherm parameters Statistical parameters

R2 χ2 SSR ARE

Langmuir [32] =+1

max L ee

L e

Q K CQ

K C Qcal 14.90 7.56

Qmax 15.63 0.909 4.539 9.078

KL 0.305

Freundlich [33] =

1

Fne F eQ K C

Qcal 16.07 0.37

KF 6.160 0.986 0.797 1.594

1/nF 0.227

nF 4.405

Temkin [34] = Rln( )e T e

T

TQ K Cb

Qcal 14.91 7.55

KT 69.09 0.973 1.320 2.641

bT 1375.516

1/bT 7.27×10-4

Dubinin-Radushkev [35]

ε−=2

DRDR K

eQ Q e

= DR1/ 2E K , E > 0

ε = + 1R ln(1 )

eT

C

Qcal 14.37 10.91

QDR 14.411 0.872 6.354 12.708

KDR 2.12×10-6

E 485.64

Sips [36] β

βα=

+

S

S

S

S1e

ee

K CQ

C

0 < βS < 1 =S max SK Q

Qcal 15.77 2.23

KS 7.432 0.988 1.222 1.222

βS 0.311

αS 0.201

Qmax 36.98


328

Linear regression analysis

Figure 4 illustrates the isotherm models in the linear forms (Langmuir, Freundlich, Temkin) for the phosphate biosorption onto QALVS under optimal conditions (20 °C, pH 6, agitation speed of 150 rpm for 60 min, and initial phosphate concentration in the range of 5-100 mg dm-3). The analysis of the isotherm data is significant to define equations that correctly represent the experimental results and they can be used for design purposes of biosorption. Table 4 shows the fitting parameters for the measured iso-therm data for phosphate biosorption onto QALVS on the linear forms. The values of isotherm parameters (R2, SD and p-value) for the Langmuir and Freundlich models indicate a better fit to the experimental data than the Temkin isotherm in this study. Also, there is similarity between Qmax and Qcal values obtained from the Langmuir and Freundlich models, respectively. The magnitude of the Freundlich constant (nF >1)

represents a favorable nature of the phosphate bio-sorption by QALVS. However, even though the R2 values are above 0.98 for the Langmuir and Freund-lich isotherm models, the ARE parameter is more pro-nounced in relation to R2 as a criterion for evaluating the experimental data. The actual deviation between the predicted and experimental values indicates that ARE values are significantly larger (Table 4). This fact suggests that Qcal parameters obtained by the Lang-muir and Freundlich isotherm equations in the linear forms are not in accordance with the experimental values.

Non-linear regression analysis

Bearing in mind that the non-linear regression is a better way to get the equilibrium parameters [37], isotherms for the phosphate biosorption by QALVS were analyzed according to the non-linear form (Langmuir, Freundlich, Temkin, Dubinin-Radush-kevich and Sips), using OriginPro 8.0 software. Figure 5 shows the fit of the isotherm models to the expe-rimental data for the biosorption. It can be seen that the isotherms are characterized by a large increase in the Qe values at low phosphate concentrations. The results of non-linear regression obtained by the application of the two-parameter (Langmuir, Freund-lich, Temkin and Dubinin-Radushkevich) and three- -parameter (Sips) isotherm equations are presented in Table 4. Among these models, the Freundlich and Sips models were in good agreement with the experi-mental data (R2 ≈ 0.990). However, equilibrium Qcal value estimated by the Freundlich model (16.07 mg P g-1) corresponded to the experimentally obtained value (16.13 mg P g-1) slightly better than one esti-mated by the Sips model (15.77 mg P g-1). Moreover, the average relative error (ARE) and chi-squared dis-tribution (χ2) for the Sips model were slightly higher. The relatively poor determination coefficients (R2 < < 0.980) and very high levels of error function values (χ2, ARE and SSR) observed for the Temkin, Lang-muir and Dubinin-Radushkevich models may be attri-buted to the unequal distribution of the biosorption surface area and active sites in QALVS biosorbent.

The Freundlich model is the most appropriate to describe the biosorption of phosphate onto QALVS as a heterogeneous system. According to this model, phosphate saturation is not predicted [33]. Therefore, the infinite coverage of the solid surface can be pre-dicted only mathematically. In this study, the esti-mated Freundlich parameters (Table 4) reflect hetero-geneous surface and a multilayer biosorption on the surface. The values of constants related to biosorp-tion capacity (KF) and intensity of biosorption (1/nF) were found to be 6.160 and 0.227 (i.e., < 1). These

Figure 4. Isotherm plots for phosphate biosorption onto QALVS

by the linear regression.


329

results suggest that the biosorption of phosphate by the cationic QALVS biosorbent was favorable.

Furthermore, the Langmuir isotherm model, which assumes the phenomenon of monolayer bio-sorption on a homogeneous surface [32], does not show satisfactory determination coefficient (R2 = = 0.909) for describing the phosphate biosorption by QALVS. The values of error function are higher than those of the other applied models. The maximum biosorption capacity (Qmax) for phosphate from the Langmuir model was 15.63 mg P g-1. Yet, the low cal-culated values of Langmuir separation factor RL (between 0.396 and 0.032) in the function of initial phosphate concentration (between 5 and 100 mg P g-1), which denotes the biosorption nature, indicated that the biosorption process onto QALVS was favorable.

According to the Sips model, the capacity of bio-sorption is limited by high sorbate concentrations [36]. Unlike the Langmuir model, to which it is very similar, this model also includes a parameter that relates to a heterogeneous system. Thus, the Sips model reduces to the Langmuir model when the dis-sociation parameter βS = 1. The higher R2 value for the three-parameter isotherm suggests the applic-ability of this model to represent the equilibrium bio-sorption of phosphate by QALVS (Table 4). The Sips isotherm model provided the best fit to the experi-mental biosorption data for phosphate ions, revealing a theoretical maximum biosorption capacity of 36.98 mg P g-1 (from Qmax = KS /αS, Table 4). It is believed that biosorption capacity obtained from the Sips equation could be more realistic than that from the Langmuir equation.

Desorption, reuse and regeneration tests

Recovery of the biosorbed anions and repeated usability of the biosorbent (as exchanger) is important

in terms of the practical application of industrial efflux-ent treatment [9,21]. Removal of the biosorbed phos-phate ions from the QALVS biosorbent was studied in a batch system (20 °C, 150 rpm and 60 min). The phosphate loaded QALVS (from the phosphate sol-ution of 25 mg P dm-3 at 20 °C) were eluted with 50 cm3 of 0.1M NaOH solution as optimal [38]. The reaction mixture was then centrifuged at 5000 rpm for 5 min. The concentration of phosphate released into the solution was determined using the ICP-OES method. On the basis of the obtained phosphate con-tent, the ratio of phosphate desorption (DRP, %) was calculated using the following equation [39]:

= dRP

b

100D

QQ

(6)

where Qd is amount of phosphate desorbed (mg P g-1) and Qb is amount of phosphate biosorbed (mg P g-1). In order to show the reusability of the biosorbent (as an anion exchanger), the already exploited QALVS biosorbent was washed with deionized water 4–5 times and treated at pH 2 with HCl (0.1 M) for 60 min. The biosorbent was thoroughly washed with deion-ized water to eliminate the excess chloride ions. The regenerated biosorbent again suspended in phos-phate containing solutions (initial concentration of 25 mg P dm-3) for the next biosorption cycle.

In order to demonstrate the reusability of the QALVS biosorbent, the biosorption/desorption cycle was repeated five times for QALVS biosorbent. The possibility of phosphate desorption in 0.1 M NaOH solution, as well as the regeneration efficiency of the QALVS biosorbent are shown in Figure 6. The applied tests showed that about 60% of the biosorbed phosphate ions were desorbed from QALVS after five consecutive biosorption/desorption cycle. The des-

Figure 5. Biosorption isotherms on the non-linear forms for the phosphate biosorption onto QALVS.


330

orption efficiency level of QALVS was significant to 98.9% for the phosphates tested after the first cycle of ions solution removal. This indicates an almost rev-ersible process, as well as the fact that there is no strong interaction between the biosorbent surface and the biosorbed phosphate ions. However, there was a certain decrease in the biosorption capacity for phosphate after the following cycles of reuse (95.6%), and the second cycle of desorption efficiency level was about 89%. In general, regeneration procedure resulted in 20% reduction of biosorption capacity and about 40% of desorption efficiency after the fifth cycle. It is probably due to a certain physical degrad-ation of the recovered biosorbent [29], i.e., ascribed to the destruction of cellulose and hemicellulose in biosorbents by using very alkaline (pH 12) and acidic (pH 2) solutions [30], as well as reducing the number of active sites and accumulating phosphate ions in the biosorbent bulk. Although the biosorption capacity decreased after the fifth cycle, the results showed that QALVS biosorbent could be repeatedly used in phos-phate biosorption studies, similar to other biosorbents [21,38]. On the other hand, this effect was more pro-nounced than in the case of recycled date-palm wastes as biosorbent, which showed very low desorbability (≈ 10%), indicating that the phosphate species were strongly attached to the sorbent material (a kind of chemisorption) and the sorption process cannot be considered as a completely reversible process for this case [6]. Besides, compared to some anionic resins, the phosphate loaded QALVS biosorbent is less expensive (Table S-1, Supporting Information) and can still be used as fertilizer in agriculture [6,29].

Figure 6. Reuse and regeneration tests for the phosphate

loaded QALVS.

Effect of coexisting anions

Wastewater, as well as natural water, normally contains many other anions, which may compete with

phosphate for the active biosorption sites. Because of that, it is necessary to examine the potential interfer-ences of coexisting anions on the biosorption of QALVS towards phosphate. In this investigation, the influences of nitrate, sulfate, and chloride on phos-phate biosorption were examined. These results were summarized in Figure S-3, Supporting Information. It can be seen that in presence of nitrate or sulfate anions, the phosphate biosorption capacity decreased significantly when the concentration of competitive ions was similar. Chloride anions did not affect the phosphate biosorption under test conditions (pH 6, 20 °C, 150 rpm, 60 min). The phosphate biosorption interference follows the following order: NO3

- > SO42-

> Cl-. It is certain that nitrate among these anions caused the greatest decrease of phosphate biosorp-tion, whereby the biosorption capacity of QALVS dec-reased from 9.88 to 7.86 mg P g-1 (about 20%). It is presumed that nitrate replaces phosphate due to its monovalent nature and potential selectivity onto the QALVS biosorbent. It is clear that QALVS interacts with phosphates (or nitrate and sulfate) mainly through the exchange with the chloride ions at the –NR4

+Cl- active sites. Therefore, the effect of these anions on phosphate biosorption is due to their affinity towards the biosorbent and it competes effectively with the phosphate biosorption.

Comparison of sorption capacity

By comparing the results obtained in this study, it was found that raw LVS biomass does not show biosorption properties for tested anions (Table S-2, Supporting Information). Regarding the performance of QALVS as biosorbent, it was found that the modi-fication significantly increases the removal efficiency of phosphate. To illustrate the potential in the use of QALVS in actual applications, a comparative evalu-ation of the sorption capacities of various types of low-cost sorbents for the phosphate is provided in Text S-6, Supporting Information. This comparison clearly indicated that quaternary ammonium biosor-bent based on the bottle gourd (Lagenaria vulgaris) shell is an effective biosorbent for phosphate removal from the aqueous solution.

CONCLUSIONS

This study confirmed that a cationic biosorbent can be prepared from a lignocellulosic agricultural by-product such as LVS biomass, and used to rem-ove phosphate from an aqueous solution in order to prevent eutrophication. Successful modification of LVS biomass into functional biosorbent was verified by elemental and FTIR analyses. Thermodynamic


331

parameters confirm that the phosphate biosorption is exothermic and favored process in the range of 20-40 °C. The anion exchange mechanism is the dominant biosorption process. The non-linear pseudo nth-order model (n ≈ 1) showed the best agreement with the experimental data in the concentration range between 5 and 100 mg P dm-3. A good fit of the Freundlich non-linear equation reflects a heterogeneous biosor-bent surface and favorable phosphate biosorption, indicating a multilayer biosorption. The maximum phosphate biosorption capacity was 0.54 mmole P g-1 for QALVS at 20 °C and pH 6. QALVS present com-parable or higher biosorption capacity of phosphate when compared with other modified biomasses and conventional sorbents. This biosorbent could be rep-eatedly used in phosphate biosorption with slight losses in initial biosorption capacities. The exhausted biosorbent contains abundance of valuable nutrients (N, P, K) and can be used in agriculture as a fertilizer.

Acknowledgements

This work was financed by the Serbian Ministry of Education, Science and Technological Develop-ment through Grants TR34008 and 34012. The authors are thankful to Dr Olga Cvetković, scientific counselor, Center of Chemistry, ICTM, Belgrade, Serbia, for her expert assistance in running elemental analyses for our study.

REFERENCES

[1] A.Z. Gu, L. Liu, J.B. Neethling, H.D. Stensel, S. Murthy, Water Sci. Technol. 63 (2011) 804-810

[2] S. Mor, K. Chhoden, K. Ravindra, J. Clean. Prod. 129 (2016) 673-680

[3] S. Smith, I. Takacs, S. Murthy, G.T. Daigger, A. Szabo, Water Environ. Res. 80 (2008) 428-438

[4] Y. Yao, B. Gao, M. Inyang, A.R. Zimmerman, X. Cao, P. Pullammanappallila, L. Yang, J. Hazard. Mater. 190 (2011) 501-507

[5] T.A.H. Nguyen, H.H. Ngo, W. Guo, T.V. Nguyen, J. Water Sustain. 2 (2012) 193-207

[6] Z. Ismail, Int. J. Environ. Stud. 69 (2012) 135-149

[7] D.L. Mitic-Stojanovic, A. Zarubica, M. Purenovic, D. Bojic, T. Andjelkovic, A.Lj. Bojic, Water SA 37 (2011) 303-312

[8] U.S. Orlando, A.U. Baes, W. Nishijima, Bioresour. Technol. 83 (2002) 195-198

[9] S. Hena, S. Atikah, H. Ahmad, Int. J. Eng. Sci. 4 (2015) 51-62

[10] X. Xu, B. Gao, W. Wang, Q. Yue, Y. Wang, S. Ni, Colloids Surfaces, B 70 (2009) 46-52

[11] L.H. Wartelle, W.E. Marshall, J. Environ. Manage. 78 (2006) 157-162

[12] P. Loganathan, S. Vigneswaran, J. Kandasamy, J. Environ. Manage. 131 (2013) 363-374

[13] B.L. Peng, N. Dhar, H.L. Liu, K.C. Tam, Can. J. Chem. Eng. 89 (2011) 1191-1206

[14] D.Z. Marković-Nikolić, A.Lj. Bojić, G. Petković, N. Ristić, M.D. Cakić, G.S. Nikolić, Adv. Technol. 6 (2017) 56-68

[15] D.L Mitić-Stojanović, D. Bojić, J. Mitrović, T. Andjelković, M. Radović, A.Lj. Bojić, Chem. Ind. Chem. Eng. Q. 18 (2012) 563-576

[16] D.V. Bojić, G.S. Nikolić, J.Z. Mitrović, M.D. Radović, M.M. Petrović, D.Z. Marković, A.Lj. Bojić, Chem. Ind. Chem. Eng. Q. 22 (2016) 235-247

[17] D.V. Bojić, M.S. Ranđelović, A.R. Zarubica, J.Z. Mitrović, M.D. Radović, M.M. Purenović, A.Lj. Bojić, Desalin. Water Treat. 51 (2013) 6871-6881

[18] D. Bojić, M. Momčilović, D. Milenković, J. Mitrović, P. Banković, N. Velinov, G.S. Nikolić, Arab. J. Chem. 10 (2017) 956-964

[19] Association of Official Analytical Chemists (AOAC), Offi-cial methods of analysis, 20th ed., AOAC International, 2016

[20] R.M. Silverstein, G.C. Bressler, T.C. Morril, Spectrometric Identification of Organic Compounds, 5th ed., John Wiley & Sons, New Jersey, 2005

[21] T.S. Anirudhan, F. Noeline, D.M. Manohar, Environ. Sci. Technol. 40 (2006) 2740-2745

[22] M.A. Khan, S.M. Ashraf, V.P. Malhotra, Int. J. Adhes. Adhes. 24 (2004) 485-493

[23] N.Y. Mezenner, A. Bansmaili, Chem. Eng. J. 147 (2009) 87-96

[24] M. Pan, X. Lin, J. Xie, X. Huang, RSC Adv. 7 (2017) 4492-4500

[25] S. Lagergren, K. Sven. Vetenskapsakad. Handl. 24 (1898) 1-39

[26] Y.S. Ho, G. McKay, Can. J. Chem. Eng. 76 (1998) 822-827

[27] A. Özer, J. Hazard. Mater. 141 (2007) 753-761

[28] W.J. Weber, J.C. Morris, J. Sanit. Eng. Div. Proceed. Am. Soc. Civil Eng. 89 (1963) 31

[29] A.C.A. de Lima, R.F. Nascimento, F.F. de Sousa, J.M. Filho, A.C. Oliveira, Chem. Eng. J. 185 (2012) 274-284

[30] X. Xu, B.Y. Gao, Q.Y. Yue, X. Zhan, Carbohyd. Polym. 82 (2010) 1212-1218

[31] J.S. Mattsom, H.B. Mark, Activated Carbon: Surface Chemistry and Adsorption from Aqueous Solution, Marcel Dekker, New York, 1971

[32] I. Langmuir, J. Amer. Chem. Soc. 40 (1918) 1361-1403

[33] H.M.F. Freundlich, J. Phys. Chem. 57 (1906) 385-471

[34] M.I. Temkin, V. Pyzhev, Acta Physic. Chim. USSR 12 (1940) 327-356

[35] M.M. Dubinin, L.V. Radushkevich, in Proceedings of the USSR Academy of Sciences, Vol. 55, 1947, pp. 331-337

[36] R. Sips, J. Chem. Phys. 16 (1948) 490-494

[37] S. Chowdhury, P. Saha, Clean 39 (2011) 274-282

[38] W. Xiong, J. Tong, Z. Yang, G. Zeng, Y. Zhou, D. Wang, P. Song, R. Xu, C. Zhang, M. Cheng, J. Colloid Interface Sci. 493 (2017) 17-23

[39] K. Riahi, B.B. Thayer, A.B. Mammou, A.B. Ammar, M.H. Jaafoura, J. Hazard. Mater. 170 (2009) 511-519.


332

DRAGANA Z.

MARKOVIĆ-NIKOLIĆ1 ALEKSANDAR LJ. BOJIĆ2

DANIJELA V. BOJIĆ2

MILORAD D. CAKIĆ3

DRAGAN J. CVETKOVIĆ3

GORAN S. NIKOLIĆ

1Visoka tehnološko umetnička strukovna škola, Vilema Pušmana 17,

Leskovac 16000, Srbija 2Univerzitet u Nišu, Prirodno-

matematički fakultet, Departman za hemiju, Višegradska 33, Niš 18000,

Srbija 3Univerzitet u Nišu, Tehnološki fakultet,

Bulevar oslobođenja 124, Leskovac 16000, Srbija

NAUČNI RAD

BIOSORPCIONI POTENCIJAL MODIFIKOVANE KORE TIKVE SUDOVNJAČE ZA FOSFATE: RAVNOTEŽNE, KINETIČKE I TERMODINAMIČKE STUDIJE

U cilju očuvanja životne sredine pripremljen je i okarakterisan novi kvaternerni amonijum biosorbent (QALVS) na bazi kore tikve sudovnjače (Lagenaria vulgaris). Rezultati FTIR i elementalne analize potvrdili su uspešnost hemijske modifikacije lignocelulozne bio-mase kore tikve i pokazali da je 15.68 mg N g-1 ugrađeno u obliku katjonske -N+R3 grupe. Modifikovana biomasa kore tikve je testirana kao biosorbent za uklanjanje fosfata iz vodenog rastvora, u šaržnom režimu. Od matematičkih modela koji su korišćeni za opisivanje kinetike biosorpcije, nelinearni model pseudo n-tog reda (n ≈ 1) pokazao je najbolje slaganje. Različiti termodinamički parametri (∆G0, ∆H0, ∆S0) potvrdili su da je biosorpcija egzoterman i spontan proces. Biosorpcija fosfata prati Freundlich-ov model izoterme, otkrivajući maksimalni kapacitet biosorpcije od 16,69 mg P g-1 na 20 °C i pH 6. Testovi biosorbcije su pokazali da se uklanjanje fosfata iz rastvora vrši prema meha-nizmu jonske izmene, koji je dominantniji u odnosu na fizisorpciju. Koegzistirajući joni u rastvoru su pokazali izrazito dejstvo na biosorpciju fosfata u nizu: NO3

- > SO42- > Cl-.

Biosorbent se može više puta koristiti nakon regeneracije, uz izvesni gubitak polaznog biosorpcionog kapaciteta. Ova istraživanja sugerišu da se dobijeni katjonski biosorbent može uspešno iskoristiti za prečišćavanje voda u cilju sprečavanja fenomena eutrofifi-kacije.

Ključne reči: Lagenaria vulgaris, fosfati, biosorpcija, kinetika, ravnoteža, termo-dinamika.

Documents

THE BIOSORPTION POTENTIAL OF MODIFIED BOTTLE GOURD …