Removal of nickel(II) and lead(II) ions from aqueous solution using peat as a low-cost adsorbent: A kinetic and equilibrium study

2.1 Preparation of peat as a natural adsorbent

The studied peat was mechanically excavated from a peatland located in the village of Leszno, near the Ner river (Poland, GPS coordinates 52°5′N, 18°56′E). It was dried at 105 °C for 24 h, then refined and classified. Peat prepared in this way was tested as a potential adsorbent of Ni(II) and Pb(II) ions.

The peat used as an adsorbent was subjected to thorough physicochemical analysis. The dispersive–morphological character of the material was determined on the basis of SEM images and particle size distribution using a Mastersizer 2000 analyser (Malvern Instruments Ltd., UK). This instrument uses the laser diffraction method, and its measuring range is from 0.2 to 2000 μm. The SEM images were recorded with a scanning electron microscope (Zeiss EVO40, Germany). The electrokinetic properties of the peat were also established. First, electrophoretic mobility was measured at a constant ionic force of 0.001 M NaCl, and then the zeta potential was calculated using the Henry equation (Zetasizer Nano ZS equipped with an autotitrator – Malvern Instruments Ltd., UK). Its elemental composition was determined using the Vario El Cube system (Elementar Analysensysteme GmbH, Germany), and its FT-IR spectrum was evaluated using an EQUINOX 55 spectrophotometer (Bruker, Germany), so that the functional groups present on the peat surface could also be identified. The materials were analysed in the form of tablets, made by pressing a mixture of anhydrous KBr (ca. 0.1 g) and 1 mg of the tested substance in a special steel ring under a pressure of approximately 10 MPa. The investigation was performed over a wavenumber range of 4000–400 cm⁻¹ (at a resolution of 0.5 cm⁻¹). The aforementioned methods enabled the full characterization of peat as an adsorbent of natural origin. The sorption properties were determined by plotting the nitrogen adsorption/desorption (at 77 K) isotherms, and indirectly by finding the surface area and total volume as well as mean size of pores. Such parameters as specific surface area (A_BET), size distribution (S_p) and volume (V_p) of pores were obtained by making a precise analysis of the adsorption/desorption isotherms in stationary conditions and using the BET and BJH algorithms. These measurements were performed using an ASAP 2020 apparatus (Micromeritics Instruments Co., USA).

2.2 Batch adsorption studies

The adsorption process was carried out using inorganic salts as metal ion precursors. The precursor of nickel(II) ions was nickel(II) nitrate hexahydrate (Ni(NO₃)₂⋅6H₂O; POCh SA), and the precursor of lead(II) was lead(II) nitrate (Pb(NO₃)₂⋅H₂O; POCh SA). Model solutions were prepared with concentrations in the range 15–100 mg/L. An important part of the study was the determination of optimum conditions for carrying out the process of removing metal ions (Table 1). The effect of the following factors was investigated: time of contact between adsorbate and adsorbent (1–240 min), pH of the system (1–9), and mass of sorbent (0.5–5 g).

Prepared solutions of salts in the appropriate concentrations were placed in conical flasks, and the appropriate quantity of adsorbent was added. The system was then mixed using a magnetic stirrer (Ika Werke GmbH, Germany) over suitable time intervals. The resulting mixture was filtered under reduced pressure (Sartorius, Germany). The product was dried for 24 h at 105 °C. To determine the effectiveness of the adsorption process, further analysis was carried out on the filtrate obtained at each stage of adsorption. The effectiveness of adsorption was estimated with the use of atomic absorption spectrometry (AAS) applying a Z-8200 spectrometer equipped with premix fishtail type burner – air/acetylene and N₂O/acetylene, graphite furnace – flame and furnace on the same beam made by Hitachi, Japan, this being one of the most accurate methods of quantitative analysis for measuring nickel(II) and lead(II) ion (the wavelength for the determination of Ni is 232.0 nm and 283.3 nm for Pb) concentrations in the filtrate following adsorption. The adsorption process was also carried out over a wide pH range.

The effectiveness of adsorption of metal ions on natural peat was determined using the following formula (1):

The kinetics of the process of adsorption of nickel(II) and lead(II) ions were also determined. For this purpose, adsorption was carried out using 0.5 g of peat and 100 mL of solutions of nickel(II) and lead(II) in different concentrations (15–100 mg/L), over different time intervals (1–240 min). The residual concentrations of nickel(II) and lead(II) ions in the solution were determined using AAS analysis. The quantity of metal ions adsorbed in unit time q_t was calculated from the following formula (2):

To determine the adsorption isotherms it was necessary to find the equilibrium concentration (q_e) of adsorbed metal ions on natural peat. The equilibrium concentration (q_e) was determined using the following formula (3):

2.3 Desorption studies

Desorption of nickel(II) or lead(II) ions from the spent peat was carried out with varying concentrations of HCl solution (0.05–0.1 M). The spent peat (0.5 g) was placed in a beaker with 100 mL of hydrochloric acid in the appropriate concentration, and the system was then mixed using a magnetic stirrer (Ika Werke GmbH, Germany) for 60 min. The resulting mixture was filtered under reduced pressure using specialized equipment and filters (Sartorius, Germany). The filtrate was analysed using AAS to determine the degree of desorption of metal ions from the peat.

All measurements of adsorption as well as desorption test were repeated three times. The presented values are averaged.

3.1 Physicochemical characteristics of peat

Elemental analysis provided information about the chemical composition of the peat sample, giving the content of carbon as 47.1%, hydrogen 6.4%, nitrogen 2.8%, and sulphur 0.9%. The peat also contains a considerable amount of oxygen, whose content was indirectly determined as approximately 27.0%.

Analysis of the FT-IR spectrum of natural peat (Fig. 1A) confirmed the presence of functional group characteristic of peat. Hydroxyl groups were identified by the broad band attributed to stretching vibrations of O—H bonds in the range 3600–3100 cm⁻¹, while additionally ether groups were manifested by the band attributed to stretching vibrations of C—O bonds at 1100–1000 cm⁻¹, carbonyl groups by the band attributed to stretching vibrations of C⚌O bonds at 1750–1500 cm⁻¹ and signal at 1413 cm⁻¹ is associated with the stretching vibrations of C—OH groups in the structure of peat. The presence of a developed aliphatic structure is confirmed by the absorption band attributed to the stretching vibrations of C—H bonds in the range 2980–2840 cm⁻¹. The bands originating from the bonds of the aromatic structure may overlap with those attributed to the bonds of other groups. In general, the FT-IR spectrum confirms the diversity of the peat’s structural composition. The detailed analysis of the peat is consistent with the data given by León-Torres et al. (2012). The presence of such a variety of functional groups on the peat surface makes it an attractive adsorbing material.

Close

Figure 1

FT-IR spectrum (a) and zeta potential vs. pH (b) of peat sample.

Export to PPT

Fig. 1B shows the zeta potential as a function of pH, recorded for a peat sample. The measurements were made in a 0.001 M solution of NaCl. The character of the obtained electrokinetic curve suggests that the zeta potential of peat strongly depends on the pH. For the analysed peat sample the zeta potential values were negative over the whole investigated pH range, namely for pH from 1.7 to 11, where the electrokinetic potential varied from −3 to −25 mV. These zeta potential values are in agreement with the data presented by Asadi et al. (2011a,b), who reported that the zeta potential of peat varied from 2.4 to −23 mV. The initial pH of peat, at which the zeta potential is −23 mV, is 6.7. With increasing pH the zeta potential becomes increasingly negative. Peat usually has a negative surface charge, which is attributed to the negative charge of organic functional groups resulting from the dissociation of H⁺ ions according to Eq. (4):

The zeta potential of peat particles in colloidal systems has also been found to depend significantly on the concentration and type of the electrolyte (Asadi et al., 2011a,b; Moayedi et al., 2011). The electrokinetic curve of the peat sample used in this study did not reach the isoelectric point (i.e.p.); however, its shape suggests that such a point would be reached at a pH of about 1.3. It can be concluded that peat demonstrates rather good electrokinetic stability in the alkaline pH range.

The parameters of porous structure of the peat sample were also determined, using the BET method. It was found that, in spite of its good sorption properties, the specific surface area of peat as a natural sorbent is small, amounting to 1.40 m²/g. The total pore volume of the peat is 0.01 cm³/g, while the average pore size is 14.6 nm.

Information about the peat’s dispersion, grain morphology and character of agglomeration was obtained from the particle size distribution diagram according to the volume contribution (Fig. 2A) and SEM image (Fig. 2B).

Close

Figure 2

Particle size distribution according to volume contribution (a) and SEM image (b) of the peat.

Export to PPT

The studied peat sample was found to contain 10% particles with diameters smaller than 4.4 μm, 50% particles with diameters smaller than 31.5 μm, and 90% particles with diameters smaller than 118.1 μm. The mean particle diameter in the whole volume of the peat sample D[4.3] is 49.1 μm. The SEM image confirms the dominance of particles with large diameter and irregular shape.

Based on the physicochemical analysis it can be concluded that the peat offers good possibilities for application as a potential adsorbent of metal ions.

3.2 Adsorption process

3.2.1

3.2.1 Effect of contact time

In the first stage of the study, it was determined how the time of contact between sorbent and adsorbate (with concentrations of nickel(II) and lead(II) ions in the range 15–100 mg/L) affects the quantity of metal ions adsorbed in a given time interval (q_t). The experimental data obtained are shown in Fig. 3a and b.

Close

Figure 3

Effect of contact time on the removal of nickel(II) (a) and lead(II) (b) ions by peat (metal ion concentration 15–100 mg/L, peat dose 0.5 g, volume of sample 100 mL, time 1–240 min).

Export to PPT

The results show that a state of equilibrium of the adsorption of nickel(II) ions on natural peat is reached after just 15 min for the lowest concentration of model solution (15 mg/L), and after 30 min for the other concentrations (30, 50 and 100 mg/L). The highest effectiveness of the process was attained after 60 min. In the case of adsorption of lead(II) ions, adsorption equilibrium is reached after 3 min for ion concentrations of 15 and 30 mg/L, after 5 min for 50 mg/L, and after 15 min for 100 mg/L.

Verification of the results obtained clearly indicates the effectiveness of the process of adsorption of nickel(II) or lead(II) ions from the model aqueous solutions using a material of natural origin. In the case of adsorption of Pb(II) ions from model solutions with concentrations of 15–100 mg/L, a process effectiveness in the range 98.9–100% was obtained. Irrespective of the adsorption time, a higher effectiveness was obtained for the removal of lead(II) ions. This may indicate that the surface of peat has a greater affinity for ions of that metal. The dependences obtained are related to different degrees of hydration of metal ions precursors (Pb(NO₃)₂ and Ni(NO₃)₂⋅6H₂O). The higher degree of metal ion hydration, the more difficult adsorption due to difficult access to the peat surface, and consequently the lower degree of its removal. Moreover, lead atom is bigger than nickel one, which also influences its uptake from aqueous solutions. The electrons are easily polarized and to a lower degree retained by the nucleus (better match of the radius of metal ion, and also the symmetry of its electron orbital). It was also confirmed by other scientists (Kalmykowa et al., 2008). The computed values for the quantity of nickel(II) or lead(II) ions adsorbed vs. time (t), q_t (mg/g), were also used when determining the kinetics of the adsorption process.

3.2.2

3.2.2 Effect of pH

A very important parameter in the process of adsorption of nickel(II) and lead(II) ions is the pH of the reaction system. This applies particularly to peat, a sorbent of natural origin whose surface contains numerous acid functional groups. Manipulation of the pH of the system may cause dissociation of these groups and bring about a significant improvement or worsening of the effectiveness of removal of nickel(II) and lead(II) ions from model solutions.

It was determined how pH (in the range 1–9) influences the effectiveness of removal of nickel(II) and lead(II) ions from model aqueous solutions using peat as a natural adsorbent. The effect of pH values in the acidic (pH = 1, 3, 5), neutral (pH = 7) and alkaline (pH = 9) ranges was determined for a metal ion concentration equal to 50 mg/L. The results, shown in Fig. 4a, imply that the pH of the aqueous system has a significant impact on the effectiveness of the process. The effectiveness of adsorption of nickel(II) ions from model solutions with a concentration of 50 mg/L increases as the system becomes more alkaline. For both nickel(II) and lead(II) ions, the lowest effectiveness (42.7% and 63.1% respectively) was obtained when the process took place in the most acidic environment (pH = 1). The greatest effectiveness for model solutions containing Ni(II) ions (92.5%) was obtained in the most alkaline environment (pH = 9). In the case of adsorption of lead(II) ions the greatest effectiveness was recorded at pH = 5, and attained 100%. Irrespective of the pH of the reaction system, the effectiveness of the process was found to be significantly greater in the case of adsorption of lead(II) ions. This may indicate that the surface of peat has a greater affinity for lead(II) ions than for nickel(II) ions, as has also been confirmed by other researchers in studies in which Sphagnum peat was used (Kalmykowa et al., 2008).

Close

Figure 4

Effect of pH (a) and adsorbent dose (b) on the removal of nickel(II) and lead(II) ions by peat.

Export to PPT

The effectiveness of removal of nickel(II) and lead(II) ions using peat as a natural adsorbent increased as the system became more alkaline. This result is explained by the presence of acid functional groups on the peat surface. At low pH values (1–3) there are no changes in their properties; their binding properties are modified only when the pH of the system is increased, due to the dissociation of hydrogen from the acid functional groups. This may explain to a large extent the mechanism of the adsorption process as proposed in a later chapter. Besides this, the pH = 5 is optimal value for adsorption of both metal ions (nickel(II) and lead(II)) – there is no precipitation of metal hydroxides. Comparison of pH influence on the adsorption of different types of metal ions onto peat is presented in Table 2. Collected information confirms that the optimum pH of the process is not higher than 6.0 to avoid the precipitation of metal hydroxides. According to the Pourbaix diagram above pH 7.5 may precipitate forms hydroxyl metals.

Table 2 pH optimum adsorption of selected metal ions with the use of peat.

Metal adsorbed	Optimal pH	Reference
Ni2+, Pb2+	5.0–7.0	This study
Pb2+, Cd2+, Cu2+	6.0	Qin et al. (2006)
Cu2+, Ni2+, Cd2+	6.0–7.0	Liu et al. (2008)
Pb2+	5.5	Zehra et al. (2015)
Cu2+, Pb2+, Cd2+	5.0–6.5	Villaverde et al. (2009)

3.3 Adsorption kinetics

When adsorption is used for the removal of selected metal ions, it is particularly important to determine the process kinetics, which may facilitate the design of the process in real-life conditions. Kinetics models based on the quantity of substance adsorbed as a function of time have been proposed in other work (Vasanth Kumar and Sivanesan, 2006; Areco et al., 2013). In this study, to determine the rate and mechanism of the process of adsorption, the pseudo-first-order kinetics model of Lagergren (1898) and the pseudo-second-order model of Ho and McKay (1999) were used.

3.3.1

3.3.1 Lagergren pseudo-first-order kinetics model

In a kinetics model of Lagergren’s pseudo-first-order type, the rate of reaction is directly proportional to the difference between the equilibrium concentration of adsorbate in the solid phase of the adsorbent and the momentary concentration in the solid phase. The Lagergren pseudo-first-order model can be expressed by Eq. (5) (Lagergren, 1898):

(5)

$$\frac{{\mathit{dq}}_t}{\mathit{dt}}=k_1(q_e-q_t)$$ where q_e and q_t (mg/g) are the quantities of metal ions adsorbed at equilibrium and at time t (min) respectively, and k₁ (1/min) is the rate constant for the pseudo-first-order rate equation.

After integration and application of the boundary conditions (q_t = 0 at t = 0 and q_t = q_t at t = t), Eq. (5) can be rewritten for linearized data plotting in the form (6):

(6)

$$\text{log}(q_e-q_t)=\text{log}(q_e)-\frac{k_1}{2.303}t$$

The equilibrium adsorption capacity (q_e) and adsorption rate constant (k₁) (Table 3) were computed experimentally from the graph of log(q_e − q_t) versus t, which is shown in Fig. 5.

Table 3 Pseudo-first-order and pseudo-second-order types 1–4 kinetics parameters and coefficient of determination for the adsorption of nickel(II) and lead(II) ions onto peat.

Type of kinetics model		Parameter	Unit	Initial concentration of Ni2+ (mg/L)				Initial concentration of Pb2+ (mg/L)
				15	30	50	100	15	30	50	100
		qe,exp	mg/g	2.610	4.910	7.370	14.440	3.000	6.000	9.940	19.780
Pseudo-first-order		qe,cal	mg/g	0.827	1.780	5.036	3.860	0.041	1.593	0.878	4.488
		k1	1/min	0.164	0.126	0.190	0.108	0.314	−1.054	−1.313	−3.145
		r2	–	0.869	0.689	0.874	0.812	1.000	0.964	0.669	0.973
Pseudo-second order	Type 1	qe,cal	mg/g	2.580	4.813	7.327	14.460	3.000	6.003	9.913	19.781
		k2	g/mg min	0.807	0.366	0.122	0.069	79.491	2.282	3.677	0.051
		h	mg/g min	5.374	8.488	6.529	14.460	715.461	82.239	361.337	19.783
		r2	–	0.999	0.999	0.999	0.999	0.999	0.999	0.999	0.999
	Type 2	qe,cal	mg/g	2.778	4.888	8.083	14.808	3.001	5.986	9.905	19.659
		k2	g/mg min	0.205	0.131	0.039	0.062	31.207	2.727	1.611	0.649
		h	mg/g min	1.582	3.137	2.526	13.629	281.045	97.706	158.078	250.938
		r2	–	0.972	0.925	0.982	0.992	0.919	0.860	0.971	0.871
	Type 3	qe,cal	mg/g	2.800	6.525	−7.530	22.355	1.418	5.996	9.908	19.696
		k2	g/mg min	0.196	0.139	0.014	0.011	−62.714	2.337	1.564	0.563
		h	mg/g min	1.538	1.809	0.799	5.416	−126.167	84.011	153.505	218.546
		r2	–	0.972	0.874	0.927	0.997	0.213	0.860	0.971	0.871
	Type 4	qe,cal	mg/g	2.761	5.067	8.030	14.784	3.002	5.995	9.910	19.688
		k2	g/mg min	0.215	0.105	0.040	0.063	28.462	2.325	1.551	0.563
		h	mg/g min	1.639	2.706	2.600	13.809	256.433	83.580	152.328	218.170
		r2	–	0.870	0.854	0.841	0.974	0.917	0.852	0.968	0.862

The bold values are related with the parameters of the pseudo-second order kinetic model of type 1 which preferably describe realized process of adsorption as compared to bold experimental data.

Close

Figure 5

Pseudo-first-order kinetics fit for the adsorption of nickel(II) (a) and lead(II) (b) ions onto peat.

Export to PPT

Table 3 shows the computed values of the adsorption rate constant (k₁) for particular concentrations of metal ions, calculated from the gradients in Fig. 5. The correlation coefficient (r²) computed for the pseudo-first-order model of adsorption of nickel(II) ions lay within the range 0.689–0.874, while for the adsorption of lead(II) ions the range was 0.669–1.000 for the concentrations used. This indicates that the pseudo-first-order kinetics model fits fairly well to the experimental data obtained in the case of adsorption of lead(II) ions (model solution concentrations 15, 30, 100 mg/L). This is because in the case of the model solution concentrations of lead(II) ions, a state of adsorption equilibrium was attained quickly. Nonetheless the results for the quantity of adsorbed metal ions over time (q_e,cal) computed from the pseudo-first-order kinetics model of adsorption deviated markedly from the experimental values (q_e,exp). It should also be noted that for lead(II) ions in concentrations of 30, 50 and 100 mg/L a negative value of k₁ (the adsorption rate constant) was obtained. A much better fit was obtained using the pseudo-second-order model, as will be described in the following section.

3.3.2

3.3.2 Ho pseudo-second-order kinetics model

In Ho’s pseudo-second-order kinetics model, the rate of the adsorption process is assumed proportional to the square of the difference from the equilibrium concentration of adsorbate. This model is represented by Eq. (7) (Ho and McKay, 1999):

(7)

$$\frac{{\mathit{dq}}_t}{\mathit{dt}}=k_2{(q_e-q_t)}^2$$ where k₂ (g/(mg min)) is the rate constant for the pseudo-second-order rate equation, and q_e and q_t are the quantities (mg/g) of ions adsorbed at equilibrium and at time t (min).

When the boundary conditions are introduced, Eq. (7) takes the form:

(8)

$$\frac{t}{q_t}=\frac{1}{k_2{q}_e^2}+\frac{1}{q_e}t$$

The initial adsorption rate h (mg/(g min)) is defined as follows:

(9)

$$h=k_2{q}_e^2$$

The values of k₂ and q_e can be determined from the gradient and intercept, respectively, of the plot of t/qt vs. t. This relationship can be classified as a pseudo-second-order kinetics model of type 1.

It should also be born in mind that k₂ and q_e can be calculated from the plots for the linear forms of equations given in Table 4 as pseudo-second-order kinetics models of types 1–4 (Vasanth Kumar and Sivanesan, 2006). The values of k₂, h and q_e,cal obtained from the four linear forms of pseudo-second-order equations were found to be different.

Table 4 Linear forms of pseudo-second-order kinetics model.

Type of kinetics pseudo-second order	Linear form	Plot
Type 1	$\frac{t}{q_t}=\frac{1}{k_2{q}_e^2}+\frac{1}{q_e}t$	$\frac{t}{q_t}$ vs. t
Type 2	$\frac{1}{q_t}=\left(\frac{1}{k_2{q}_e^2}\right)\frac{1}{t}+\frac{1}{q_e}$	$\frac{1}{q_t}$ vs. $\frac{1}{t}$
Type 3	$\frac{1}{t}=\frac{k_2{q}_e^2}{q_t}-\frac{k_2{q}_e^2}{q_e}$	$\frac{1}{t}$ vs. $\frac{1}{q_t}$
Type 4	$\frac{q_t}{t}=k_2{q}_e^2-\frac{k_2{q}_t^2}{q_e}$	$\frac{q_t}{t}$ vs. $q_t$

The kinetics of the adsorption of nickel(II) and lead(II) ions from model solutions of concentrations 15, 30, 50 and 100 mg/L onto natural peat was relatively well-described using the pseudo-second-order kinetics model of type 1. Pseudo-second-order models of types 2–4, in spite of the high values of their correlation coefficients, cannot be taken into consideration because of the large differences between the experimental and calculated sorption capacities.

In the case of the pseudo-second-order kinetics model of type 1, the coefficient of correlation (r²), calculated from the slopes of the curves shown in Fig. 6a, was 0.999, for adsorption of both nickel(II) and lead(II) ions, irrespective of their concentration in the model solution. It should also be noted that the quantities of adsorbed metal ions (q_e,cal) computed from the pseudo-second-order kinetics model of type 1 corresponded very well with the experimental data (q_e,exp). The initial adsorption rate (h) was not found to have any tendency to increase with increasing concentration; this may be related to the heterogeneity of the natural material.

Close

Figure 6

Types 1 (a) and 2 (b) pseudo-second-order kinetics model, obtained using a linear method, for the adsorption of Ni²⁺ and Pb²⁺ on peat.

Export to PPT

For pseudo-second-order kinetics models of types 2 and 3 (Figs. 6b and 7a) the correlation coefficient lies in the range 0.874–0.997 for adsorption of nickel(II) ions, and in the range 0.213–0.971 for adsorption of lead(II) ions. The adsorption capacities (q_e,cal) computed from the pseudo-second-order kinetics models of types 2 and 3 did not fit with the experimental capacities (q_e,exp) to the same degree as in the case of the model of type 1. Also the values of the adsorption rate constant (k₂) and initial adsorption rate (h) obtained using the type 3 model for a lead ion concentration of 15 mg/L were found to be negative. The parameters computed using the pseudo-second-order models of types 2 and 3 indicate that these models cannot be used to describe properly the kinetics of the process of adsorption of nickel(II) and lead(II) ions on natural peat.

Close

Figure 7

Types 3 (a) and 4 (b) pseudo-second-order kinetics model, obtained using a linear method, for the adsorption of Ni²⁺ and Pb²⁺ on peat.

Export to PPT

The pseudo-second-order kinetics model of type 4 (Fig. 7b, Table 3) produced unfavourable results. This is confirmed by the low value of the coefficient of correlation (r²), which lies in the range 0.841–0.974 for various initial concentrations of nickel(II) ions, and 0.852–0.968 for lead(II) ions. The coefficients of correlation (r²) obtained using the pseudo-second-order kinetics model of type 4 are markedly lower than those obtained using the type 1 model.

To sum up the results concerning the adsorption kinetics, it can be concluded that the pseudo-second-order kinetics model provides a significantly better description of the kinetics of the adsorption of nickel(II) and lead(II) ions on natural peat.

3.4 Adsorption isotherms

Adsorption isotherms describe the relationship between the quantity of substance adsorbed and the concentration in the equilibrium solution. An important element of the implementation of a metal ion adsorption process is the fitting of experimental data to an appropriate adsorption isotherm model. Analysis of the adsorption isotherms obtained provides much valuable information about the mechanism and nature of the adsorption process. Fig. 8 shows the plot of q_e versus C_e for the adsorption isotherms of nickel(II) and lead(II) on peat at 25 °C. The experimental data were applied to the three-parameter isotherm models of Freundlich (1906) and Langmuir (1918). Table 4 gives the coefficient of correlation, the parameters of the adsorption equation and the quantity of metal ions adsorbed at equilibrium concentration (q_e).

Close

Figure 8

Fitting of the Langmuir and Freundlich isotherm models to equilibrium results for nickel(II) (a) and lead(II) (b) ions adsorbed onto peat.

Export to PPT

3.5 Adsorption mechanism

In order to propose a mechanism for the adsorption of nickel(II) and lead(II) ions onto natural peat, FT-IR analysis was performed (Fig. 9A) in addition to the measurement of the electrokinetic stability (Fig. 9B) of the sorbent before and after the adsorption process. Previously, other researchers (Li et al., 2007; Dahiya et al., 2008) using low-cost adsorbents for the removal of toxic metal ions have confirmed a mechanism based on ion exchange or on the formation of complexes with functional groups present on the adsorbent surface.

Close

Figure 9

FT-IR spectra (a) and zeta potential curves (b) for nickel(II) and lead(II) ions adsorbed by peat (pH = 5).

Export to PPT

In the present work a mechanism is suggested (Fig. 10) based on ion exchange by the following functional groups: carboxyl (groups derived from a carbonyl group), phenolic, hydroxyl and ether, with positively charged ions of nickel(II) or lead(II). These groups are capable of bonding ions of environmentally harmful metals through the dissociation of a hydrogen cation.

Close

Figure 10

Example mechanism for the adsorption of nickel(II) and lead(II) ions on peat.

Export to PPT

Analysis of the FT-IR spectra of the peat before and after adsorption (Fig. 9A) confirms the important role played by functional groups on the peat surface in the metal ion adsorption process. Based on the spectra obtained, it can be concluded that following the adsorption of either nickel(II) or lead(II) onto natural peat, the intensity of the relevant bands is reduced. This is particularly visible in the case of stretching vibrations of —OH groups in the range 3550–3300 cm⁻¹ and stretching vibrations of C⚌O groups in the range 1680–1600 cm⁻¹. After adsorption process lower intensity has also signal at 1413 cm⁻¹, associated with the stretching vibrations of C—OH groups in the structure of peat. The results show clearly that ions of the environmentally harmful metals form bonds with oxygen groups, at the same time causing separation of a hydrogen cation. Confirmation of this thesis are slightly shifts in the maximal wavenumbers of the signals characteristic for groups that are involved in creation of the interactions between hazardous metal ions and peat surface. These shifts are observed especially for the signal attributed to the —OH groups. Maximal wavenumber for this peak, in the spectrum of peat, is around 3420 cm⁻¹, and after adsorption of Ni²⁺ and Pb²⁺ ions is shifted to the 3385 cm⁻¹ and 3394 cm⁻¹ respectively. Similar observations are related to the signal attributed to the C⚌O groups. Before adsorption that signal has maximum at 1635 cm⁻¹ and after the process maxima were shifted to the 1643 cm⁻¹ for Ni²⁺ and to the 1641 cm⁻¹ for Pb²⁺ ions. The spectra in Fig. 9A also show that the adsorption process is more effective for lead(II) than for nickel(II) ions, as is indicated by the smaller intensity of particular bands obtained for peat with adsorbed Pb²⁺ ions. These results show excellent correlation with those obtained using atomic absorption spectrometry, which confirm that a greater effectiveness of adsorption on peat is obtained for lead(II) ions.

Fig. 9B shows the zeta potential curves for the peat samples with and without adsorbed Ni(II) and Pb(II) ions. A pH-dependent curve was plotted for the samples. For an aqueous suspension of peat the zeta potential value ranged from −5 mV for a pH of approximately 2, to −25 mV for a pH of 8. The zeta potential values suggest that the functional groups in the peat were highly deprotonated and negatively charged, indicating that the major binding sites are acidic groups.

At constant pH, the zeta potential value became more positive with an increase in either Pb²⁺ or Ni²⁺ concentration. This indicates that the ions’ surface coordination is caused by the following:

• electrostatic attraction between the negative charge on the biomass and positive metal cations;

• and/or ionic exchange adsorption where Ni²⁺ and Pb²⁺ ions replace a proton from the surface adsorbent (substitution of protons in surface groups on metal ions), as outlined below:

As a consequence of this, a positive remaining charge could be located on the surface, making the zeta potential value more positive. The results of electrokinetic analysis and the FT-IR spectra confirm the correctness of the proposed mechanism of interaction between the analysed metal ions and peat.

The next confirmation of the mechanism based on the ion exchange principle is comparison of pH values of the solution before and after adsorption process (see Table 7). Experimental data collected confirm that increasing amount of metal ions adsorbed onto the adsorbent surface, the more hydrogen ions (H⁺) get released out of peat into the water solution. Consequently, the pH value of the reaction mixture decreases. This effect is consistent with the ion exchange principles – the more ions were adsorbed on the peat, the more hydrogen ions were released into the solution, making the pH lower.

3.6 Desorption test

The possibility of reusing natural adsorbents multiple times increases the attractiveness of the adsorption process. An important element of the process is the development of an effective method for the regeneration of the materials used. For this purpose a desorption test was carried out, using water and hydrochloric acid in various concentrations (0.05 and 0.1 M) as the regenerating agent.

The results presented in Table 8 indicate that water is a weak desorbing agent. This is evidence of a strong bond between the adsorbate and adsorbent. Only when hydrochloric acid was used was a significant increase in the degree of desorption achieved. It was found that as the concentration of the hydrochloric acid used as a desorbing agent increased, there was an improvement in the effectiveness of the biosorbent regeneration process. The mechanism of desorption is based on the exchange of hydrogen ions (H⁺) with the adsorbed metal ions. After regeneration, the sorbent could be used again in further adsorption tests.