Scientific paper Sorption Kinetic Studies of Ammonium from Aqueous Solution on Different Inorganic and Organic Media Dajana Ku~i},* Ivana ]osi}, Marija Vukovi} and Felicita Bri{ki University of Zagreb, Faculty of Chemical Engineering and Technology, Department of Industrial Ecology, Maruli}ev trg 19, 10 000 Zagreb, Croatia * Corresponding author: E-mail: dkucic@fkit.hr Received: 12-07-2012 Abstract In this study, the sorption of ammonium from aqueous solution onto activated carbon, natural zeolite, peat and potting soil was studied by performing batch kinetic sorption experiments. The activated carbon wasn't efficiently removing ammonium at concentrations higher than 50 mg L-1. Sorption isotherms of ammonium on zeolite, peat and potting soil were determined at 25 °C and 200 rpm with the initial concentration of 50-7000 mg L-1. Equilibrium data were fitted by Freundlich, Langmuir and Temkin isotherm and parameters were evaluated according these models. Langmuir model gives better fit to experimental data than Freundlich and Temkin models. Maximum adsorption capacities were for activated carbon 0.631 mg g-1, zeolite 58 mg g-1, peat 595 mg g-1 and for potting soil 575 mg g-1. The equilibrium kinetic data were analyzed using adsorption kinetic models: the pseudo-first and second-order equations and were found to follow the pseudo-second-order kinetic model. A comparison between linear and non-linear regression method for estimating the adsorption and kinetics parameters was examined. The obtained results showed that non-linear method may be a better way to determine the kinetic parameters. Thermodynamic studies showed exothermic and endothermic nature of the adsorption of NH4+ on inorganic and organic adsorbents, respectively. From present results it can be seen that zeolite, peat and potting soil are good adsorbents for removal ammonium from aqueous solution. Keywords: Adsorption of ammonium ions; Inorganic and organic adsorbents; Kinetics; Linear method; Non-linear method, Thermodynamic studies. 1. Introduction Ammonia is a colorless, toxic, odorous and corrosive gas produced from several sources, wastewater treatment plants, in outlet gases from composting plants and fertilizer plants, chemical and manufacturing industries.1 Ammonia is also found as ammonium ion (NH4+) in wastewaters,2 leachate3 and condensate evolved during composting process.4 Excess ammonia-nitrogen can cause a health problem involving many living species including humans and animals. It also contributes to the eutrophica-tion of lakes and rivers, depletion of dissolved oxygen and toxicity to aquatic organisms. Therefore, it is important to control emission of this compound for the protection of public health and environment.5 The traditional methods for ammonium removal from waste streams are based on physicochemical and biological treatments. Nowadays, ion exchange and adsorption become more interesting as possible treatment meth-ods.6 Adsorption is often referred to as passive uptake and physical-chemical binding of chemical species or ions to a solid surface.7 Inorganic materials, natural zeolites (clinoptilolite) are the most important cation exchangers that exhibit high ion exchange capacity, selectivity and compatibility with the natural environment. It is found that zeolites are very promising for adsorption of ammonium ions from aqueous solutions. The advantages of zeolites are: wide spreading in nature, low cost, high selectivity towards the ammonium ions in the presence of other concurrent cations in the wastewaters.8 Also, inorganic materials are inert to biological degradation. The use of organic materials for sorption of ammonium is adequate because of their high availability, low cost and the presence of complex microbial communities capable of degrading ammonium ions as pollutant.9,10 Ammonium which is adsorbed on different organic materials can be used for agricultural purpose. The aim of the study was to investigate the adsorption potential of selected inorganic and organic materials for the removal of ammonium from aqueous solution. Adsorption isotherms and kinetics were investigated and different adsorption isotherm models were used to evaluate the experimental data by linear and non-linear method. Also, the influence of pH value and temperature was investigated. 2. Materials and Methods 2. 1. Materials The standard solution used in this study was ammonium sulfate p.a. (Kemika, Croatia). The inorganic materials used in the experiments were zeolite (clinoptilolite) from the Krapina region, Croatia and activated carbon -silcarbon K835, Germany. Farkas et al. (2007) used the same natural zeolite for their experiments. Their results showed that the chemical composition of clinoptilolite is: SiO2 (64.93%), Al2O3 (13.39%), Fe2O3 (2.07%), Na2O (2.40%), K2O (1.30%), CaO (2.00%), MgO (1.08%) and loss on ignition at 1000 °C was 9.63%. These adsorbents were washed with demineralized water to remove the particles adhered to surface, and dried at 105 °C for 2 hours. The organic materials which were used were peat and potting soil from Olaines Kudra, Latvia and Zrinjevac, Zagreb. They were only dried to calculate the moisture of samples. The dried zeolite was crushed with laboratory jaw crusher (Matest, A092 TE, Italy) and granulometric analysis was conducted on rotary sieve shaker (Sieve-Shaker Model RA-86-1, WS Tyler, SAD). Physical characteristics of selected adsorbents are presented in Table 1. Table l. Physical characteristics of sorbents. Zeolite Activated Peat Potting carbon soil pH value 6.7 9.3 4.7 6.5 Moisture, % 2.89 2.25 26.99 37.68 Particle size, mm 1.0 - 2.0 0.6 - 2.4 0.15-0.78 0.08-0.47 2. 2. Characterization of the Adsorbents Fourier Transform InfraRed Spectroscopy - FTIR (Perkin Elmer, USA) analysis was performed in order to establish the structural changes during the modification of activated carbon, zeolite, potting soil and peat with ammonium ions. Pressed pellets were prepared by grinding the powder specimens with spectroscopic grade KBr for FTIR spectra test. The spectra were recorded from 4000-450 cm-1. 2. 3. Adsorption Studies Adsorption experiments were carried out by adding 0.5 g of different sorbents, zeolite, activated carbon, peat and potting soil into 100 mL Erlenmeyer flasks containing 50 mL of different initial concentrations (50-7000 mg L-1) of ammonium solution. The sorption experiments were conducted 24 hours and the samples were taken every one hour. All flasks were shaken at 200 rpm. After filtration (membrane pore size 0.45 pm, Sartorius, Germany) the concentrations of ammonium ions in aqueous solution were determined spectrophotometricaly (Spectrophotometer DR/2400, Hach at USA) at X = 680 nm, using sodium citrate, sodium nitroprussiate dihydrate and sodium-salicylate.11 The linearity range was from 0 to 0.5 mg N L-1 and the detection range was to 0.5 mg N L-1. The solutions were acidified with 1:3 H2SO4 to preserve them for further experiments. Concentrations of NH4+ were expressed as mg N L-1. The amount of ammonium ion adsorbed qe (mg g-1) was determined by using following equation: HÜ = (C„-Ce)V (1) where Co and Ce represent the initial and equilibrium concentration of ammonium ion (mg L-1); V is the volume of the ammonium ion in solution (L) and m is the mass of dry adsorbent (g). The ammonium removal percentage can be calculated as follows: removal percentage C -C 100 (2) 2. 4. Batch Kinetics Studies Kinetic experiments were identical to those of equilibrium tests. The aqueous samples were taken at preset time intervals and the concentrations of ammonium ions were determined spectrophotometricaly.11 The amount of sorption at time t, qt (mg g-1), was calculated by expression: h = 0c0-c,yv (3) where Ct (mg L-1) is the liquid phase concentrations of ammonium ions at different time. 2. 5. Effect of pH The effect of pH solution (pH = 2-9) on sorption of ammonium ions on activated carbon, zeolite, potting soil and peat was determined by agitating 0.5 g of different adsorbents and 50 mL of ammonium sulphate solution (Co = 100 mg L-1 ). Agitation was provided for 4 h contact time which was sufficient to reach equilibrium with a constant agitation speed of 200 rpm. The pH value was adjusted by adding a few drops of diluted NaOH or HCl. 3. Results and Discussion 3. 1. Characterization of the Adsorbents The particle size of inorganic and organic materials which were used in this study were in the range with other proposed in literature.5'8'12'13 To establish if the structure of activated carbon, zeolite' peat and potting soil is modified during adsorption of ammonium ions, FTIR analysis was performed (Figure 1a-d). The number of adsorption peaks in the spectra indicated the complex nature of the analyzed samples. Figure 1a shows that the FTIR spectra of activated carbon were not significantly changed after adsorption of ammonium ions. The peak localized at 1047.13 cm1 in FTIR spectra of zeolite corresponds to the vibration of the bands connected with the internal Si-O(Si) and Si-O(Al) vibrations in tetrahedral or alumino- and silico-oxygen bridges.14 The changes in the FTIR spectra of zeolite indicated a little shift of these band positions and changes in their intensity. The broad and strong band ranging from 3700 to 3000 cm1 might attribute to -OH stretching vibrations which are presented in FTIR spectra of all investigated adsorbents (Figure 1a-d). The peaks at 2919.94 and 1384.41 (Figure1c-d) are attributed to the symmetric and asymmetric C-H stretching vibration, and the band around 1615 cm-1 (potting soil, peat) is due to asymmetric and symmetric stretching vibrations of C=O.15,16 The peaks ranging from 1200 to 950 cm-1 are due to C-O stretching. The wave numbers of potting soil (similar for peat) shifted from 3409.02, 1641.08, 1384.41 and 1035.55 cm-1 to 3382.34, 1634.80, 0 (peak disappeared) and 1046.79, respectively, after ammonium adsorption suggesting that hydroxyl, phenol and carboxyl groups might be the main active groups for ammonium adsorption on potting soil and peat.1516 3. 2. Effect of Solution pH on Aammonium Adsorption The removal of NH4+ from aqueous solution using different sorbents was studied at pH values 2-9 and the data obtained were given in Figure 2. The adsorption effi- a) c) b) d) JtT fi j V 3WW -'W lftl.8 \ \ \ h l ma A IM I v s J »5« I / \ \ ^ \ [ | v v t 1I74JC ~\KIiH j 1 «s.» ISM« \ Imz \l M M m M 1600 UDO 1200 1050 300 ) 430 Figure 1: FTIR spectra of a) activated carbon, b) zeolite, c) peat and d) potting soil before (-) and after ammonium adsorption (---). ciency of zeolite, potting soil and peat, increased by raising the pH from 4 to 8, and reached a maximum value at pH 7, and then it decreased at pH 9. The decrease in the NH4+ removal efficiency after pH 8 may be due to the conversion of ammonium ions (NH4+) to ammonia gas (NH3).6,17,18 Also, at low pH (below 4) the adsorption efficiency decreased due to the competition of H ions for the active sites on the adsorbent surface.6,19 and NH4+ 80 70 ■ £ 60 ■ T z c o 50 ■ CL O 1ft 4(1 - T7 < 30 - 20 - 10 - 0 - □ □ A □ A □ D A ¿A O O O A 6 O o Zeolite o Activated carbon □ Polling soil A Peat « O o 0 O 4 6 PH 10 Figure 2: The effect of pH on adsorption of NH4+ using inorganic and organic adsorbents (activated carbon, zeolite, peat and potting soil) (Co(NH4+) = 100 mg L-1; m^^ = 0.5 g; V = 50 mL; t = 240 min; T = 25 °C). 3. 3. Adsorption Isotherm Adsorption isotherms describe how pollutants interact with adsorbent materials and so they are critical to optimize the use of adsorbents. In order to optimize the design of an adsorption system for removal of ammonium ions from aqueous solutions, it is important to establish the most appropriate correlation for the equilibrium curve (Figure 3). From Figure 3 it can be seen that the ammonium adsorption capacity of zeolite, peat and potting soil increased with the increase of initial ammonium concentra- tion in aqueous solutions, indicating that plenty of readily accessible sites are available at the start of adsorption. After equilibration, when the adsorbent becomes saturated, a plateau is reached indicating that no more sites are available for further adsorption. The obtained results show that zeolite has lower adsorption capacity than potting soil and peat. Also, it can be seen from Figure 3 that the adsorption isotherm for activated carbon in concentration range from 50 to 7000 mg L-1 lies almost on the axis Ce. The adsorption capacity of activated carbon was calculated for concentration range from 5 to 100 mg L-1 (results are not shown). Adsorption capacities of adsorbents from present study were compared with other adsorbents reported in previous researches for ammonium removal and is compiled in Table 2. From Figure 4 it can be seen that removal percentage of ammonia is the highest for peat and potting soil. Zeolite has a high removal percentage at lower concentrations < 500 mg L-1 but as concentration of ammonium increases the removal percentage decreases. Activated carbon was not efficiently removing ammonium ions from aqueous solution.5 500 400 £. 300 « 200 100 O O B O □ Patting soil 8 i Peat o Zeolite o o Activated carbon □ ° ¡TiV ft O o o o 60 - 50 m 40 f ■o 30 I e c s—1 - 20 * ■ 10 2000 4000 C, (mg L-') 6000 0 8000 Figure 3: Adsorption isotherms of ammonium adsorption on inorganic (activated carbon and zeolite) and organic (potting soil and peat) materials (V = 50 mL, madsolbent = 0.5 g, T = 25 °C and pH = 2.1). Table 2. Comparison of adsorption capacity of different adsorbents for ammonium ions removal from water. Adsorbent Contact Concentration Particle size of Adsorption time (h) range (mg L-1) adsorbent, mm capacity (mg g-1) Reference Activated 2 35-280 0.70 - 1.25 10.25 [5] carbon (AC) Natural zeolite 5 60-800 1.25 9.22 [8] Natural zeolite 4 1000-7000 1.0-2.0 58.00 Present study Strawberry 14 25-1000 0.30 6.05 [15] leaf powder Peat 10 0-0.8 - 1.59 [9] Peat 5 1000-7000 0.15-0.78 595.00 Present study Potting soil 5 1000-7000 0.08-0.47 575.00 Present study Figure 4: Percentage of removed NH4+ from aqueous solutions by sorption on different adsorbents. The adsorption equilibrium data were analyzed using Freundlich, Langmuir and Temkin isotherm models in the concentration range of 1000 to 7000 mg L-1. This concentration range was selected because the wider concentration range between 50 and 7000 mg L-1 did not give a good estimation of adsorption parameters. The concentrations of ammonium ions between 50 to 1000 mg L-1 in solutions were less diluted than concentrations between 1000-7000 mg L-1 for spectrophotomet-rical measurements of ammonium ions and it resulted in analytical error. The error increases as the concentration is higher. The Freundlich isotherm is an empirical equation used to describe multilayer adsorption with interaction between adsorbed ions.20 The Freundlich equation is ex- pressed as: (l+i-C.) and linearized equation: Ce = 1 1 if b ■ <7,™* 4max (6) (7) where qe is the amount of pollutant adsorbed on the sor-bent at equilibrium (mg g-1), b is the Langmuir adsorption constant related to the energy of adsorption (L/mg) and qmax is the mono-layer adsorption capacity of adsorbent (mg g-1). Therefore, a plot of CJq versus Ce gives a straight line of slope 1/qmax and intercept 1/(b qmax). Adsorption of NH4+ by zeolite, peat and potting soil can also be evaluated to see whether it is favorable by using a separation factor RL that is given by:15 _J_ (1+b-Cg) (8) where Co (mg L-1), is the highest initial concentration of adsorbate and b is Langmuir constant. If the value of RL is higher than 1, adsorption is considered to be unfavorable but if it is between 0 and 1 it is favorable.19 The values of RL for zeolite, peat and potting soil in this study have been found to be 0.0820, 0.1369 and 0.1369, respectively, indicating that adsorption of ammonium ions on zeolite, peat and potting soil are favorable. Temkin and Pyzhec24 considered the effect of the adsorbate interaction on adsorption. They suggested that these interactions would influence the heat of adsorption of all the molecules in the layer and it would decrease linearly with coverage. The Temkin isotherm has been used in the following form, Eq. (9): (4) RT qe =—(lnATr + lnCJ^+B-lnQ b and linearized equation: RT, . RT, _ qe —In A + ■—\nCe (9) (10) = ItiÄV + —-InC, (5) values of KF and n are calculated from the intercept and slope of the plot. The Langmuir isotherm theory assumes monolayer coverage of adsorbate over a homogenous adsorbent surface. A basic assumption is that all adsorption sites are equivalent and adsorption on active sites is independent of whether the adjacent is occupied.16'22 sorption model can be expressed as:23 The Langmuir ad- where KF (an indicator of adsorption capacity) and 1/n (is where the adsorption intensity) are Freundlich constants. The dimension 1/n gives an indication of the favorability of adsorption and values n > 1 represent favorable adsorption conditions.16 Equation (4) may also be written in the logarithmic form as RT (11) The adsorption data can be analyzed according to Eq. (10). A plot of qe versus ln Ce enables the determination of the constants A (K) and B. The constant B is related to the heat of adsorption and A (K) is Temkin equilibrium constant (L mg1). The values of KF, b, n, qmax, A, B and R2 for zeolite, peat and potting soil are given in Table 3. Figure 5 a-c, shows the experimental equilibrium data for ammonium concentrations from 1000 to 7000 mg L-1 and the predicted theoretical Freundlich, Langmuir and Temkin isotherm. The isotherm parameters were estimated by linear and non-linear regression analysis. The non-linear regression analysis was estimated using Least Square method Table 3. Langmuir, Freundlich and Temkin isotherm constants and correlation coefficients for adsorption of ammonium on zeolite, peat and potting soil. Method Isotherm Parameters Zeolite Adsorbent Peat Potting soil Linear regression Langmuir b (L mg-1) 0.0014 0.0006 0.0006 qmax (mg g-1) 63.69 714.29 714.29 R2 0.956 0.983 0.994 Non-linear regression b (L mg-1) 0.0016 0.0009 0.0009 qmax (mg g-1) 58.36 595.02 575.21 Rm2ax 1.000 1.000 1.000 Linear regression Freundlich Kf (mg g-1) 0.0122 0.2402 0.2719 n 3.17 1.66 1.63 R2 0.949 0.931 0.924 Non-linear regression Kf (mg g-1) 0.0416 0.2311 0.2424 n 3.86 1.47 1.63 R2 1.000 1.000 1.000 Linear regression Temkin A (L mg-1) 0.0020 0.0095 0.0077 B 23.41 133.88 132.65 R2 0.943 0.941 0.931 Non-linear regression A (L mg-1) 0.0011 0.0062 0.0051 B 10.10 97.86 95.61 R2 0.999 0.998 0.998 implemented in Scientist software. The correlation coefficients obtained from non-linear method for Langmuir, Freundlich and Temkin were identical. Thus, to estimate which model describes better the experimental data the regression coefficients of linear method were used. Based on correlation coefficients the highest R2 was obtained by Langmuir model. Finally, it can be concluded that Langmuir model was suitable for describing the sorption equilibrium of ammonium ions by zeolite, potting soil and peat. The Langmuir parameter qmax shows that peat and potting soil have a much higher adsorption capacity than zeolite. 3. 4. Sorption Kinetics Data The study of adsorption kinetics is very useful for understanding the mechanisms that are involved, dynamics of adsorption of ammonium ions onto zeolite, peat and potting soil and also for design of future facilities for adsorption on a large scale. Kinetics data for peat and potting soil were similar and because of that only results of peat were presented. Figure 6 shows the sorption kinetics of ammonium ions on the zeolite and peat at the lowest and the highest concentrations, 500 and 7000 mg L-1. As can be seen, the sorption rate was quite fast at the initial stage and with time it gradually slowed down. The faster initial rate may be due to the availability of the uncovered sorption sites of the adsorbent at the start of experiment. More than 90% of the amount of ammonium adsorbed at equilibrium was reached in 4 hours.18 In order to predict the mechanism involved during the present sorption process and the potential rate controlling such as chemical reaction processes, two kinetic models were used to fit the experimental data, pseudofirst order and pseudo-second order. The parameters were evaluated according linear and non-linear method. Results of present study suggest that non-linear method is better than linear method in determining kinetic parameters of a particular kinetic model (Table 4a and b). This is mainly because transforming a non-linear kinetic model to a linearized form tends to alter the error distribution, and thus distort the parameters. Non-linear analysis conducted on the same abscissa and ordinate results in the same error distribution and is therefore a better way to obtain the kinetic parameters than linear method.7,25 Also, it has been reported that it is not suitable to use the correlation coefficient of a linear regression analysis for comparing the best appropriate model.26 The pseudo-first-order kinetics model known as the Lagergern equation is one of the most widely used rate equation to describe the adsorption of an adsorbate from given liquid phase. A kinetic model, pseudo-first order is given in the form:27 at Integrating this for the boundary conditions t = 0 to t = t and qt = 0 to qt = qt, Eq. (12) may be rearranged for linearized data plotting as shown by Eq. (13): kf = (13) where qe and qt (mg g-1) are the amount of ammonium adsorbed at equilibrium and at time t and kf is the pseudofirst-order rate constant (min-1). The slope and intercept of plots of log(qe-qt) versus t were used to determine the Figure 5: Isotherm plots for ammonia adsorption on a) zeolite, b) peat and c) potting soil. first-order constant kf and equilibrium adsorption capacity qe by linear regression (Table 4a and b). From Figure 7 it can be seen that pseudo-first order does not fit well for the Figure 6: Effect of contact time on adsorption of ammonium on zeolite and peat (V = 50 mL, m adsolbents = 0.5 g, T = 25 °C and pH = 2.1). whole range of contact time and is generally applicable over the initial 20 to 30 minutes of the sorption process.28 The pseudo-second-order model is based on the assumption that the rate-limiting step may be a chemical sorption involving forces trough the sharing or exchange of electrons between adsorbent and adsorbate.20 The pseudo-second-order kinetic model is given as:28' 29' 30 dq, , , .2 —=ks(qe-qt) d, (14) For the boundary conditions t = 0 to t = t and qt = 0 to qt = qt, the integrated form of equation (15) becomes: l q, =■ I ks <7, (15) Figure 7: Pseudo-first-order kinetic model plots of ammonium adsorption on natural zeolite. Table 4. Characteristic parameters of the different kinetic models (pseudo-first and second order) estimated by linear and non-linear regression for ammonia sorption on a) zeolite and b) peat. a) Zeolite Pseudo-first order Pseudo-second order Co qe,exp_ kf qe, calc ks qe, calc (mg L-1) (mg g-1) (min 1) (mg g-1) R2 (g mg-1 min-1) (mg g-1) R2 Linear 500 17.87 0.0062 12.64 0.878 0.0014 18.59 0.969 regression 1000 31.89 0.0053 26.07 0.928 0.0006 31.65 0.948 2000 39.86 0.0064 31.46 0.909 0.0005 42.19 0.959 3000 54.42 0.0081 52.48 0.966 0.0003 52.63 0.927 5000 54.92 0.0074 71.24 0.893 0.0001 55.87 0.913 7000 56.02 0.0051 55.69 0.937 0.0001 55.25 0.952 Non-linear 500 17.87 0.0149 16.60 0.913 0.0010 19.16 0.969 regression 1000 31.89 0.0099 28.56 0.964 0.0005 35.36 0.975 2000 39.86 0.0119 37.35 0.938 0.0003 44.15 0.960 3000 54.42 0.0076 56.87 0.984 0.0001 55.80 0.989 5000 54.92 0.0020 67.90 1.000 0.0001 57.34 1.000 7000 56.02 0.0031 78.10 0.999 0.0001 58.56 1.000 b) Peat Pseudo-first order Pseudo-second order C o q e, exp. kf qe, calc k s qe, calc (mg L-1) (mg g-1) (min 1) (mg g-1) R2 (g mg-1 min-1) (mg g-1) R2 Linear 500 38.54 0.0191 21.39 0.741 0.0012 29.67 0.980 regression 1000 85.54 0.0233 51.61 0.850 0.0045 86.21 0.970 2000 174.84 0.0281 130.32 0.961 0.0001 175.44 0.950 3000 220.85 0.0318 220.85 0.992 0.0006 227.27 0.999 5000 394.79 0.0329 395.46 0.986 0.0006 400.00 0.999 7000 421.12 0.0175 533.70 0.876 0.0001 434.78 0.974 Non-linear 500 38.54 0.0633 29.67 0.841 0.0106 29.90 0.968 regression 1000 85.54 0.0454 85.37 0.894 0.0018 88.05 0.982 2000 174.84 0.0409 175.26 0.975 0.0007 182.24 0.997 3000 225.45 0.0304 222.22 0.968 0.0003 225.56 0.977 5000 404.79 0.0305 380.26 0.989 0.0002 405.89 0.995 7000 421.12 0.0141 440.76 0.979 0.0001 425.78 0.981 which has a linear form: t 1 1 — = 1-T +--' 9, ktgr q. (16) where qe and qt (mg g-1) are the amount of ammonium adsorbed at equilibrium and at time t and ks is the pseudo-second-order rate constant (g mg-1 min-1). Figure 8a and b shows the plots of qt against t for the pseudo-second order model for sorption of ammonium onto natural zeolite and peat. The best-fit model was selected based on both regression correlation coefficient (R2) and the calculated qe values. It is seen from Table 4a and b and Figure 9 that the theoretical qe(calc) values for zeolite and peat calculated from the pseudo-first order did not give reasonable values with regard to the experimental uptake ones, qe(exp) This suggests that this sorption system is not a first-order. The qe(calc) values agree very well with the experimental values for the case of the pseudo-second-order kinetics, and a re- gression coefficient of above 0.95 shows that the model can be applied for the entire concentration range of adsorption process, suggesting that chemisorption might be the rate-limiting step that controlled the adsorption process of ammonium on the zeolite and peat.25 Also, from Table 4a and b it can be seen that the pseudo-second-order rate constant (ks) decreased with increased initial concentration. 3. 5. Thermodynamic Studies Thermodynamic parameters were calculated to confirm the adsorption nature of the present study. The thermodynamic constants, free energy change (AG0), enthalpy change (AH°) and entropy change (AS°) were calculated to evaluate the thermodynamic feasibility of the process and to confirm the nature of the adsorption process.29,31 The Gibb's free energy change of the process is related to the Kc by the following equation, Eq. (17): 60 SO i- 40 rr> cA £ ff 30 20 • 10 a) Zeolite x O O / .Vigo o^// J^— _O / —S /äc o 0.5 g L-1 □ 3 g l-1 ¡v i i g L-1 * 5 g L-1 [f 0 2 g L-1 x 7 g L-1 r -2 nd order model AC" =-RT hi Kc (17) 100 200 300 Ume (min) 400 500 Figure 8: Pseudo-second-order kinetic model plots of ammonium adsorption on a) zeolite and b) peat. Figure 9: Estimated qe values for pseudo-first and second order model. where T is temperature in K, R the ideal gas constant (8.314 J mol-1K-1) and Kc is the thermodynamic equilibrium constant, which is defined as, Eq. (18): K = C -C (18) where Co is initial concentration of ammonium solution, mg L-1, and Ce is the equilibrium concentration of solution, mg L-1. According to thermodynamics, the Gibb's free energy is also related to the enthalpy change (AH0) and entropy change (AS°) at constant temperature by the van't Hoff equation, Eq. (19): In K„ = - A G" RT AH" AS0 RT R (19) In order to determine the thermodynamic parameters, experiments were carried out at different temperature in the range of 25-55 °C (Figure 10). The values of ental-phy (AH°) and entropy (AS°) were calculated from the slope and intercept of the plot ln Kc versus T-1. The calculated values are listed in Table 5. £ J X 90 30 70 GO 50 40 30 20 10 0 O D □ □ □ A A □ AS A O O O o Zeolite o Activated ca re on □ Potting soil i Peat O O O 10 20 30 40 Temperature (°C) 50 60 Figure 10: The effect of temperature on NH4+ adsorption on inorganic and organic adsorbents (activated carbon, zeolite, peat and potting soil) (Co(NH4+) = 100 mg L-1; m^^ = 0.5 g; V = 50 mL; t = 240 min; T = 25 °C). Figure 10 indicates the effect of the temperature (25, 28, 35, 45 and 55 °C) on the adsorption of NH4+ ions by activated carbon, zeolite, potting soil and peat. The amount of NH4+ removed from aqueous solution by adsorption on activated carbon and zeolite decreased by rising temperature from 25 to 55 °C. On the contrary, the amount of NH4+ removed from aqueous solution by adsorption on peat and potting soil increased by rising tem- Table 5. Thermodynamic parameters for the adsorption of NH4+ by different adsorbents Adsorbent T, K K AG°, kj mol1 AH°, kj mol1 AS°, J mol1 K1 Zeolite 298 4.76 -3.87 -2G.39 -58.25 3G1 2.63 -2.42 3G8 2.GG -1.77 318 1.92 -1.73 328 1.85 -1.68 Activated carbon 298 1.G6 -G.15 -1.G6 -3.G8 3G1 1.G5 -G.13 3G8 1.G4 -G.1G 318 1.G3 -G.G8 328 1.G2 -G.G5 Potting soil 298 2.86 -2.6G 11.68 48.4G 3G1 3.33 -3.G1 3G8 3.57 -3.26 318 4.17 -3.77 328 4.55 -4.13 Peat 298 2.5G -2.27 9.83 4G.41 3G1 2.56 -2.36 3G8 2.63 -2.48 318 3.13 -3.G1 328 3.57 -3.47 perature from 25 to 55 °C. For both, activated carbon and zeolite, the values of AH° were negative, indicating the exothermic nature of the process, which further explain the fact that the adsorption efficiency decreased with the increase of temperature. Negative values of AS° indicate a decrease in randomness at the solid/solution interface during the adsorption process while low value of AS° indicates that no remarkable change on entropy occurs.29 The positive values of AH° for peat and potting soil indicated the endothermic nature of the process. Change in the entropy, AS° was found to be for peat and potting soil 40.41 and 48.40 J mol1 K1, respectively, for NH4+ removal from aqueous solution. This result implies that NH4+ ions on solid phase (surface of adsorbent) were in a much more chaotic distribution compared to the relatively ordered state of bulk phase (aqueous solution).17 The negative values of AG° for investigated adsorbents confirmed the feasibility and spontaneous nature of the adsorption process. It is clear that the adsorption process using organic materials can occur in a wide temperature range as compared to inorganic (mineral) materials. For the most mineral materials investigated, the adsorption capacity increases with increasing the temperature up to 25-35 °C and above this range of temperature, the adsorption capacity of adsorbents decreases.32 4. Conclusions From obtained experimental results and their evaluation it can be concluded that zeolite, peat and potting soil are good adsorbents for removal of ammonium ions from aqueous solution. Activated carbon showed low potential for removal of ammonium at concentration higher than 50 mg L-1. The removal percentage was about 75% for peat and potting soil at concentration < 5000 mg L-1, 90% for zeolite at concentration < 500 mg L-1. The equilibrium data have been analyzed using Freundich, Langmuir and Temkin isotherms. The Langmuir isotherm was demonstrated to provide the best fit for the sorption of ammonium to zeolite, peat and potting soil. The adsorption capacity for inorganic materials, activated carbon and zeolite, were 0.631 mg g-1 and 58 mg g-1, respectively, for organic materials, peat and potting soil were 595 mg g-1 and 575 mg g-1 at 25 °C, respectively. Kinetic studies suggest that ammonium adsorption on zeolite, peat and potting soil as adsorbents could be described more favorably by the pseudo-second order kinetic model and the parameters are better estimated by non-linear method. 5. Acknowledgment This work was supported by the Ministry of Science, Education and Sports of the Republic of Croatia under Research Project 125-1251963-1968. 6. References 1. T. Kaosol, N. Pongpat, Proc. Wrld. Acad. Sci. E. 2011, 77, 642-646. 2. A. Farkaš, M. Rožic, Z. Barbaric-Mikocevic, J. Hazard. Mater. 2005, B117, 25-33. 3. F. Briški, Z. Gomzi, N. Horgas, M. Vukovic, Acta Chim. Slov. 2003, 50, 715-729. 4. D. Kučic, N. Kopčic, I. Cosic, M. Vukovic, F. Briški, Proc., 3rd International Symposium in Environmental Management Towards Sustainable Technologies,Faculty of Chemical Engineering and Tecnology, Zagreb, Croatia 20ll 280-285. 5. P. Vassileva P. Tzvetkova, R. Nickolov, Fuel 2008, 88, 387-390. 6. A. Hedstrom, J. Environ. Eng. 2001, 127, 434-437. 7. S. Chowdhury, P. Saha The IIOAB Journal 2010, 3, 3-7. 8. E. Ivanova M. Karsheva, B. Koumanova, J. UCTM2010, 45, 3, 295-302. 9. E. Pagans X. Font, A. Sanchez, Biotechnol. Bioeng., 2006, 3, 515-525. 10. N.-J. Kim, M. Hirai, M. Shoda, J. Hazard. Mater. 2000, B72, 77-90. 11. APHA, Standard Methods for the examination of Water and Wastewater, 16th ed., American Public Health Association, Washington, DC, 1985. 12. S. J. Allen, G. Mckay, J. F. Porter, J. Colloid. Interf. Sci. 2004, 280, 323-333. 13. M. L. Cornelius, W. L. A. Osbrink, J. Econ. Entomol. 2010, 103, 799-807. 14. A. Peter, L. Mihaly-Cozmuta, A. Mihaly-Cozmuta, C. Nicu-la, Fascicula Biologie, 2011, Tom. XVIII, 128-135. 15. H. Liu, Y. Dong, H. Wang, Y. Liu, Desalination 2010, 263, 70-75. 16. S. Liang X. Guo, N. Feng, Q. Tian, J. Hazard. Mater. 2010, 174, 756-762. 17. K. Saltali, A. Sari, M. Aydin, J. Hazard. Mater. 2007, 141, 258-263. 18. H. Huang, X. Xiao, B. Yan, L. Yang, J. Hazard. Mater. 2010, 175, 247-252. 19. B. H. Hameed, D. K. Mahmoud, A. L. Ahmad, J. Hazard. Mater. 2008, 158, 499-506. 20. A. B. Perez Marin, M. I. Aguilar, V. F. Meseguer, J. F. Ortuno, J. Saez, M. Llorens, Chem. Eng. J. 2009, 155, 199206. 21. H. Freundlich, J. Phys. Chem. 1906, 57, 385-470. 22. Y.-F. Wang, F. Lin, W.-Q. Pang, J. Hazard. Mater. 2007, 142, 160-164. 23. I. Langmuir, J. Amer. Chem. Soc. 1916, 38, 2221-2295. 24. M. I. Temkin, Zh Fiz Chim. 1941, 15, 296-332. 25. D. Wen, Y.-S. Ho, X. Tang, J. Hazard. Mater. 2006, B133, 252-256. 26. Y. Khambhaty, K. Mody, S. Basha, B. Jha, Colloids Surf. 2008, A328, 40-43. 27. S. Lagergren, K Sven Vetenskapsakad Handl. 1898, 24, 1-39. 28. Y. S. Ho, G. McKay, Trans IChemE 1998, 76, 332-340. 29. A. Bhatnagar, A. K. Minocha, M. Sillanpaa, Biochem. Eng. J. 2010, 48, 181-186. 30. Y. S. Ho, G. McKay, Process Biochem. 1999, 34, 451-465. 31. M. M. El Jamal, M. C. Ncibi, Acta Chim. Slov. 2012, 59, 2431. 32. M. A. Wahab, S. Jellali, N. Jedidi, Bioresource Technol. 2010, 101, 8606-8615. Povzetek Raziskovali smo adsorpcijo amoniaka iz vodnih raztopin na aktivnem oglju naravnem zeolitu, šoti in glini pri 25 °C pri začetni koncentraciji med 50-7000 mg NH4+-N L-1. Eksperimentalne podatke smo analizirali z Freundlichovo and Langmuirjevo izotermo, kjer se Langmuirjev model izkazal bolje. Ugotovili smo, da adsorpcija na aktivno oglje pri koncentracijah nad 50 mg NH4+-N L-1 ni več učinkovita, medtem ko so največje sorpcisjke kapacitete na zeolitu 58 mg NH4+-N g-1, na šoti 595 mg NH4+-N g-1 in za glino 575 mg NH4+-N g-1 suhe snovi. Z uporabo dveh kinetičnih modelov, psevdo-prvega in drugega reda, smo proučevali tudi kinetiko adsorpcije, pri čemer drugi model bolje opiše hitrost procesa. Zeolit, šota in glina so torej dobri adsorbenti za odstranjevanje amoniaka iz vodnih raztopin.