Sage Journals: Discover world-class research

Abstract

This work provides a simple and convenient method to manufacture the sorbent of Al-eggshell. The influence of AlCl₃ concentration and pH values as well as the dosage of sorbent and their interactions on adsorption of phosphate was investigated. Therefore, Box–Behnken design coupled with response surface method was adopted to explore the empirical model for phosphate species removal. It was observed that there is an optimal point, C(AlCl₃)(0.29 mol/l)–pH(6.12)–dosage(6.72 g/l), for the goal of maximizing phosphate species removal. The second-order polynomial model for phosphate reduction was given as Removal(%) = 96.43 +10.82X1 + 4.29X2 − 0.70X3 + 2.06X1X2 − 1.72X1X3 +8.24X2X3 − 13.10X1²− 17.26X2²− 1.72X3². Contour pictures implied that the interaction between pH values and sorbent dosage was the strongest, followed by C(AlCl₃) versus dosage and C(AlCl₃) versus pH. The adsorption of phosphate data had a good agreement with the Freundlich isotherm equation at 313 and 323 K. Otherwise, Langmuir–Freundlich model described the best fitness at the temperature of 293, 298, and 303 K. The process of adsorption of phosphate on Al-eggshell fitted a pseudo-second-order kinetic equation, which indicates the exothermic reaction. In conclusion, the present work suggests Al-eggshell as an efficient and environmental friendly sorbent for phosphate species adsorption from aqueous solutions.

Keywords

Phosphate Al-eggshell adsorption modeling response surface methodology mechanism

Introduction

Phosphorus is a key limiting nutrient for the growth of aquatic plants and algae in water bodies, which can be confirmed by the following reaction

\begin{matrix} 106 {CO}_{2} + 16 {NO}_{3^{-}} + {HPO}_{4}^{2 -} + 122 H_{2} O + 18 H^{+} + Energy + Traceelement \to \\ C_{106} H_{263} O_{110} N_{16} P + 138 O_{2} \end{matrix}

(Grubb et al., 2000). Algal blooms and eutrophication occurring in lakes, coastal areas, and reservoirs lead to esthetic problems, dissolved oxygen depletion, and destruction of aquatic ecosystems (Biswas et al., 2007; Lurling et al., 2014). Hence, many nations protect their water resources via regulations that limit phosphorus discharge. The usual species of phosphorus observed in aqueous solutions are orthophosphates, polyphosphates, and organic phosphates (Xiong et al., 2017; Zong et al., 2016). Chen et al. (2015) reported that a criterion maximum contaminant level for phosphorus is less than the USEPA’s 0.05 mg/l level. Consequently, exploring efficient methods to remove phosphate species from water is necessary. To our knowledge, the works of phosphate removal from wastewater have been carried out using biological processes (Seviour and McIlroy, 2008), precipitation by chemical methods (Lai et al., 2016; Yang et al., 2016), iron exchange (Seo et al., 2013), crystallization (Bian et al., 2013), and electrochemistry (Shu et al., 2016). Phosphate removal from water and wastewater via adsorption processes offers an alternative treatment, especially if the sorbent is economical and easily available. A considerable attraction has been paid to economic and environmental concerns to the research of applying various kinds of inexpensive sorbents, such as zeolite (Aljbour et al., 2017; Zhang et al., 2014), sludge (Liu et al., 2016), fly ash (Novais et al., 2016), periwinkle shell (Bello and Ahmad, 2011a), and slag (Park et al., 2007).

Previous research suggests that calcite, aluminum hydroxides/oxides, and iron hydroxides/oxides are significant phosphate sorbents (Karageorgiou et al., 2007; Lai et al., 2016). Waste eggshells—generally from residences, restaurants, and bakeries—consist primarily of calcium carbonate (Francis and Abdel Rahman, 2016; Guo et al., 2017) and are considered as a waste disposal and landfill problem. In recent years, different studies have explored useful applications for eggshells (Elabbas et al., 2016; Francis and Abdel Rahman, 2016). Those studies concluded that eggshell may be useful as a fertilizer and a feed additive for livestock. Eggshells have an estimated 7000–17,000 pores (William and Owen, 1995). The porous nature of eggshells makes them an attractive material to employ as an adsorbent. Eggshell appears to effectively adsorb certain heavy metals and organic compounds (Elabbas et al., 2016; Park et al., 2007); however, the application of discarded eggshells in the removal of phosphate by adsorption has received very little focus. Especially the use of AlCl₃ doping waste eggshell was seldom studied.

In this work, we reuse waste eggshell as a sorbent of phosphate and simultaneously reusing a conventional waste. The prepared eggshell was cationized by applying AlCl₃ to activate its phosphate sorption capacity. The eggshell powder before and after modification was characterized using N₂-BET, X-ray diffraction (XRD), and morphological characterization (scanning electron microscope (SEM)) analyzer. The model of Design Expert was used to determine the optimum conditions for phosphate removal from polluted aqueous solutions. The effects of three key parameters in adsorption of phosphate including AlCl₃ concentration, pH values, and sorbent dosage and their interactions on light phosphate removal were investigated using the response surface method. For this purpose, a Box–Behnken design was used to obtain statistical model for phosphate species removal percentage. Adoption equilibrium and kinetics isotherms experiments were performed to assess the behaviors and mechanisms of sorption.

Materials and methods

Materials

Eggshell samples were collected from the canteen of Shandong University of Science and Technology, Qingdao, China. To remove impurity and pollutants, the particles were rinsed with deionized water. Then the samples were dried at 100℃ overnight in the dry oven.

For the preparation of the Al-eggshell, 2.0 g eggshell powder was immersed in 100 ml solution of AlCl₃ (0.29 mol/l) for 24 h with stirring speed 150 r/min. The mixture was agitated on a shake table at 30℃ at equilibrium. Then, the particles were centrifuged and dried, screened through a sieve of 100 mesh, and subsequently used for sorption experiments.

Stock solutions were prepared by dissolving the anhydrous potassium dihydrogen phosphate (KH₂PO₄) in distilled water to give a concentration of 1000 mg PO₄³⁻/l and diluted when necessary.

Design of experiments using Box–Behnken

In order to obtain the largest amount of information from a small number of experiments, the Box–Behnken method was used (Ahmad et al., 2016; Moghddam et al., 2016). Three key parameters including AlCl₃ concentration, pH, and dosage were measured to generate a second-order polynomial model. The objective is selected to be increasing percentage of phosphate removal.

The design levels and parameters are shown in Table 1. A quadratic polynomial equation, which is considered as a response of key factors to analyze quadratic interactions and squared terms, was calculated applying “Design Expert” (Ahmad et al., 2017; Arabi et al., 2016; Shams et al., 2016).

Table 1.

Predictor variables and their coded levels and actual values applied for experimental design.

Variables	Symbol	Real values of coded levels
Variables	Symbol	−1	0	1
C(AlCl₃) (mol/l)	X1	0.15	0.25	0.35
pH	X2	5	6	7
Dosage (g/l)	X3	5	7	9

Characterization

The chemical components of sorbents were estimated by energy dispersive X-ray spectroscopy (EDX, JB-750). The particle morphology of the eggshell and the Al-eggshell was characterized by powder XRD, which was obtained on a Shimadzu XRD-6000 diffractometer, using Cu Kα radiation at 40 kV, 30 mA, a scanning rate of 10°/min, a step size of 0.02°/s, and a 2θ angle ranking from 20° to 70°. The surface and matrix morphologies of the natural eggshell and the Al-eggshell were obtained on a KYKY-2800B SEM. The specific surface area of the eggshell samples was measured by the BET nitrogen gas sorption method. The evaluation of the porosity of natural waste eggshell and Al-eggshell sample was carried out using physical adsorption of N₂ at the temperature of −196℃ (Quantachrome, Autosorb-1 C).

Adsorption procedure

Equilibrium studies

Adsorption equilibrium studies were conducted at initial phosphate solution pH of 6.12. Equilibrium data were obtained by adding 0.672 g of prepared Al-eggshell into a series of Erlenmeyer ﬂasks each ﬁlled with 100 ml of phosphate solution. The initial phosphate concentration ranges from 0.5 to 200 mg/l. The Erlenmeyer ﬂasks were covered with aluminum foil and then placed in a thermostatic shaker for specified time. During the adsorption, the temperature of the system was kept at 293, 298, 303, 313, and 323 K, respectively.

Kinetic studies

Adsorption experiments were carried out by adding a ﬁxed amount of sorbent (0.672 g) to a series of conical ﬂasks ﬁlled with 100 ml diluted solutions (3 and 50 mg/l). The conical ﬂasks were placed in a thermostatic shaker at 298 K and a 6.12 pH.

Results and discussion

Response surface analysis of phosphate removal

Model equation construction

The effect of key factors including AlCl₃ concentration, pH values, dosage, and their interactions is investigated for the percentage of phosphate removal. Phosphate removal is quantified as the difference between the percentage of phosphate removed at the initial time and after 30 min of adsorption. The matrix design with experimental results is given in Table 2.

Table 2.

The design matrix and experimental results adopting Box–Behnken method.

Run	Factors			Removal (%)
Run	X1	X2	X3	Removal (%)
1	0.25	7	5.0	Observed	Predicted
2	0.25	5	9.0	74.16	74.22
3	0.25	6	7.0	64.28	64.22
4	0.25	5	5.0	96.48	96.43
5	0.25	7	9.0	82.05	82.10
6	0.25	6	7.0	89.34	89.29
7	0.35	7	7.0	96.39	96.43
8	0.35	6	9.0	83.22	83.25
9	0.15	6	5.0	89.99	90.02
10	0.25	6	7.0	69.80	69.78
11	0.15	7	7.0	96.41	96.43
12	0.35	6	5.0	57.52	57.49
13	0.15	6	9.0	94.94	94.86
14	0.25	6	7.0	71.72	71.81
15	0.25	6	7.0	96.47	96.43
16	0.25	6	7.0	96.45	96.43
17	0.35	5	7.0	96.37	96.43
18	0.15	5	7	70.51	70.54

The experimental design includes 18 runs by considering six replicates at center point to evaluate the error between each experiment. The experimental data versus the simulated data of phosphate removal are summarized in Table 2. As seen in Table 2, the fitted values of the developed model are close to observed real data obtained from experiments. The good agreement between these data confirms the precision of the model developed to predict the adsorption of phosphate by Al-eggshell. In order to quantify the quadratic interactions and squared terms of phosphate removal, the experimental results were subjected to analysis of variance (ANOVA). Table 3 presents the modeled data.

Table 3.

Analysis of variance for phosphate removal.

Source	Sum of square	df	Mean square	F-value
X1–C(AlCl₃)	937.01	1	937.01	18 × 10⁵
X2–pH	147.58	1	147.58	2.8 × 10⁴
X3–Dosage	3.95	1	3.95	7.6 × 10⁴
X1X2	16.93	1	16.93	3.2 × 10⁴
X1X3	11.80	1	11.80	2.3 × 10³
X2X3	271.43	1	271.43	5.2 × 10⁴
X1²	748.89	1	748.89	1.4 × 10⁵
X2²	1299.27	1	1299.27	2.5 × 10⁵
X3²	12.84	5	12.84	2.4 × 10³
Pure error	0.010	5	0.010	–
Total	3735.13	17	0.002	–

In this case, the related factor or interaction is significant. Therefore, parameters X1, X2, and X3 and their interactions (i.e. X1 × X2, X1 × X3, X2 × X3, X1², X2², and X3²) for the percentage of phosphate removal are statistically significant. Meanwhile, the second-order polynomial model for phosphate reduction given by OPTIMIZATION of “Design Expert” is presented in equation (1)

\begin{matrix} Removal (%) = 96.43 + 10.82 X 1 + 4.29 X 2 - 0.70 X 3 \\ + 2.06 X 1 X 2 - 1.72 X 1 X 3 + 8.24 X 2 X 3 \\ - 13.10 {X 1}^{-} 17 . {26 X 2}^{-} 1 . {72 X 3}^{2} \end{matrix}

(1)

where X1, X2, and X3 are variables of AlCl₃ concentration, pH, and dosage, respectively.

Here equation (1) is adopted to construct the contour plots where the third independent variable is in midpoint level. Equation (1) implies that phosphate removal efficiency improved with the AlCl₃ usage (X1) and pH (X2) yet decreased dosage (X3).

The ANOVA was adopted to assess the significance of the developed model. As a valid statistical model, the adjusted R² should be within R²± 0.2 (Bello and Ahmad, 2011b). In this present work, the difference between R² and R²_adj is 0.001 (R²= 0.999 and R²_adj = 0.9998). Meanwhile, the value of lack of fit is 0.0538, which is not significant relative to the pure error. These results indicate that this developed adsorption behavior could be described well by the predicted model.

3D response surface and contour plots

Interaction of each of the independent actors on the response variables in the whole design space is presented by the contour plots (Figure 1). The effects of AlCl₃ concentration, pH values, and dosage on percentage reduction of phosphate are plotted in Figure 1.

Figure 1.

3D response surface and contour plots of phosphate removal as the function: (a) C(AlCl₃) and pH, (b) pH and dosage, and (c) C(AlCl₃) and dosage.

As per Figure 1(a), phosphate removal increases with the simultaneous behaviors of C(AlCl₃) and pH value to 0.29 mol/l and 3.16, respectively. At this point, phosphate removal reaches 99.04%. The interaction between C(AlCl₃) and dosage gives quite a different trend. As shown in Figure 1(b), the maximum phosphate removed is 98.94% when C(AlCl₃) is 0.29 mol/l with a 6.14 g/l dose. Meanwhile, the effect of C(AlCl₃) is much more obvious than the dose on the phosphate adsorption process. It has been found that removal of phosphate could reach more than 98.80% when C(AlCl₃) ranges from 0.28 to 0.32 mol/l and dosage ranges from 5.54 to 6.78 g/l. Phosphate removal does not change significantly with a varying dose. As for the interaction effect between parameters of pH and dosage, the optimal point for phosphate removal (96.70%) is pH 6.12 with a 7.28 g/l dose.

In a combination of ANOVA and outcomes from the section of Model equation construction, the suggested solution was considered to be 0.29 mol/l C(AlCl₃), pH 6.12, and a 6.72 g/l dose. At this point, phosphate species removal may reach 99.06%.

Moreover, the shape of the relationship of factors implies the interaction intensity between variables, that is the oval contours indicate a stronger interaction intensity than circular ones. Therefore, Figure 1 indicates that the interaction between pH and dose is stronger than that of both C(AlCl₃) versus dose and C(AlCl₃) versus pH.

Characteristics of the eggshell composite

The chemical composition of the natural eggshell and Al-eggshell was estimated by EDX and is summarized in Table 4.

Table 4.

Chemical components of natural eggshell and Al-eggshell.

Component	% (w/w)
Component	Natural eggshell	Al-eggshell
Na₂O	0.492	0.432
MgO	0.842	0.823
Al₂O₃	0.054	0.281
SiO₂	0.011	0.010
P₂O₅	0.183	0.178
SO₃	0.745	0.731
K₂O	0.051	0.048
CaCO₃	97.025	96.883
Fe₂O₃	0.028	0.026
SrO	0.141	0.132
Cl	0.127	0.212
Volatile matter	0.301	0.244

Park et al. (2007) stated that all eggshells have similar chemical components which mainly include CaCO₃ and a few of other elements, i.e. Mg, S, Na, P, Al, K, Sr, and Si. Before modification, CaCO₃ as limestone (97.025%) and MgO (0.842%) were the important inorganic compositions of eggshell. After modification, the dominant inorganic composition was identified as lime (96.883%) as well. The proportion of other elements decreased, excluding Cl. The result taken from Table 4 implies that the composition of the eggshell can be modified.

The surface area, pore volume, and pore size distribution were calculated from nitrogen adsorption data using the Quantachrome software and are given in Table 5.

Table 5.

Physical properties of the natural and Al-eggshells.

	Property
	S_BET (m²/g)	V_micro (cm³/g)	V_meso (cm³/g)	V_T (cm³/g)
Natural eggshell	6.012	0.093 × 10⁻³	1.46 × 10⁻²	1.47 × 10⁻²
Al-eggshell	20.84	2.92 × 10⁻³	5.07 × 10⁻²	1.47 × 10⁻²

BET: Brunauer,Emmett and Teller (BET), a surface area method named by these three famous persons.

Note: S_BET is BET surface area (m²/g), V_micro is micropore volume (cm³/g), V_meso is mesopore volume (cm³/g), and V_T is total pore volume.

Table 5 suggests that the Al-eggshell has improved S_BET, V_micro, V_meso, and V_T compared with natural eggshell sample. Figure 2 provides the pore size distribution of the natural eggshell and Al-eggshell samples. Both the natural eggshell and Al-eggshell exhibit one peak around 2–10 nm. As it is seen from Figure 2, these both samples include mesopores.

Figure 2.

Pore size distribution of natural eggshell and Al-eggshell.

The XRD patterns of the natural eggshell and Al-eggshell were explored in Figure 3. The main peak appeared at 2θ = 29.5°. Simultaneously, Figure 3 also shows several other small peaks at 2θ = 23.2°, 39.5°, 47.3°, 48.7°, and 57.6° in both sorbents. This spectrum confirms the presence of calcium carbonate according to JCPDS Card No.82-1690, where the main characteristic peaks of calcium carbonate were investigated in the XRD graph of the eggshell powder. This outcome also confirms that the calcium carbonate is the main constituent in the eggshell powder (Elabbas et al., 2016; Li et al., 2016). The comparison of the peak information of both the natural eggshell and Al-eggshell implied that the main peak intensity increased; however, no aluminum hydroxide/oxide peak information was found. The reasons for this phenomenon maybe include (1) the mass of Al was too small and (2) aluminum hydroxide/oxide XRD diffraction peak is wide.

Figure 3.

XRD spectra of natural eggshell and Al-eggshell.

SEM was used to examine the surface morphology of both the natural and Al-eggshells. SEM images of the eggshell sample before and after modification are shown in Figure 4(a) and (b), respectively. The natural eggshell has irregular granular crystal with 0.5–18 µm (Figure 4(a)). After modification, the surface of Al-eggshell is a relatively smooth sheet-like shape (Figure 4(b)).

Figure 4.

SEM images of natural eggshell (a) and Al-eggshell (b).

Adsorption isotherms

The adsorption of phosphate onto the Al-eggshell at 293, 298, 303, 313, and 323 K was determined as a function of the equilibrium phosphate concentration and the corresponding adsorption plot, shown in Figure 5 (a) to (e). The analysis of equilibrium data is a requirement for the sign of adsorption systems. The experimental data were fitted to the Langmuir, Freundlich, Langmuir–Freundlich (Sips), and Temkin equations and the constant parameters of the isotherm equations were calculated. These model equations are expressed by equations (2) to (5) (Cheung et al., 2000; Chun et al., 2005; Mezenner and Bensmaili, 2009), respectively

q_{e} = q_{max} \frac{{bC}_{e}}{1 + {bC}_{e}}

(2)

q_{e} = K_{f} \cdot C_{e}^{1 / n}

(3)

q_{e} = q_{max} \frac{C_{e}^{β}}{K + C_{e}^{β}}

(4)

q_{e} = A + Bln C_{e}

(5)

where q_e (mg/g) is the adsorption capacity at equilibrium, q_max is the maximum adsorption capacity, and b (l/mg) is the binding constant which is related to the heat of adsorption. C_e gives the equilibrium phosphate concentration, and both K_f and n are the Freundlich constants related to adsorption capacity and adsorption intensity, respectively. K presents the saturation constant (mg/l) and β shows the cooperative binding constant. A and B are the constants.

Figure 5.

Isotherm for phosphate removal at various solution temperatures on Al-eggshell (pH = 6.12, sorbent dosage=6.72 g/l): (a) 293 K, (b) 298 K, (c) 303 K, (d) 313 K, and (e) 323 K.

Figure 5(a) to (e) suggests that the phosphate loading capacity of the Al-eggshell increased with the increase of temperature up to 323 K. This indicates that the adsorption process of phosphate was favored at higher temperatures and that the sorption course is endothermic.

The Temkin model had the poorest fit to the experimental data in comparison to other isotherm equations, with R² ranging from 0.8565 at 293 K to 0.9154 at 298 K. A comparison of the experimental isotherms with the adsorption models described suggests that the Freundlich isotherm equation provides the best fit at 313 and 323 K (R²= 0.9828 and 0.9945). The Langmuir–Freundlich equation gave the greatest overlap with the experimental data at 293, 298, and 303 K. This can be verified by a high correlation coefficient of 0.9917, 0.9951, and 0.9940.

The n parameter of the Freundlich equation ranging from 2.44 to 3.36 implies that adsorption sites with low energetic heterogeneity of the adsorption system (Chardon and Blaauw, 1998). The Freundlich parameter K_f reached the corresponding maximum value of 3.60 at 323 K and n of 3.36 in this work. This indicated that binding capacity reaches the highest value and that the afﬁnity between the Al-eggshell and phosphate was also higher than other temperature values applied. The constant b represents afﬁnity between the sorbent and sorbate, which increases from 0.038 to 0.284 l/g, respectively, with increasing temperature from 293 to 323 K.

Figure 5(a) to (e) suggests that the Langmuir–Freundlich model describes the adsorption of phosphate data with higher precision than the other fitted models. Generally, three-parameter models represent adsorption isotherm data well, because of high correlation coefficients and low percentage error values in comparison with the two-parameter models. In the course of the experiments, the value of K and β varied as a function of solution temperature. Accordingly, on increasing the temperature from 293 to 323 K, the value of K decreased from 22.33 to 8.96 mg/l and the value of β from 0.53 to 0.35.

Adsorption kinetics and effect of contact time

Kinetic studies can give further information on the adsorption process mechanisms. In this work, the experimental data were fitted to three of the most widely used kinetic models, which are pseudo-second-order, intraparticle diffusion, and Elovich models.

The pseudo-second-order model (Ho and McKay, 1999) is given as

\frac{t}{q_{t}} = \frac{1}{k_{2} q_{e}^{2}} + \frac{t}{q_{e}}

(6)

where q_e (mg/g) and q_t (mg/g) are the amount of phosphate adsorbed at equilibrium and any time t (min), and k₂ (g/mg min) is the pseudo-second-order rate constant.

The intraparticle diffusion model (Weber and Morris, 1963) is given as

q_{t} = k_{t} t^{0.5} + C

(7)

where k_t is the intraparticle diffusion rate constant (mg/gmin^0.5). Values of C suggest the thickness of the boundary layer (mg/g).

The Elovich model (Inyang et al., 2016) has been accepted that the chemisorption process can be described below

q_{t} = \frac{ln a_{e} b_{e}}{b_{e}} + \frac{1}{b_{e}} lnt

(8)

where a_e presents the initial adsorption rate (mg/g min), and b_e is related to the extent of surface coverage and activation energy for chemisorption (g/mg).

The contact time effect on the process of phosphate adsorption and the kinetics fitting lines is assessed in Figure 6. The parameters for each equation are listed in Table 6.

Figure 6.

Contact time effect on phosphate adsorption by Al-eggshell and kinetics fitting lines: (a) 3 mg/l and (b) 50 mg/l.

Table 6.

Parameters for kinetic equations.

Models	Parameters	3 mg/l	50 mg/l
Pseudo-second-order equation (6)	K₂ (g/mg min)	7.76 × 10⁻²	2.24 × 10⁻²
	q_e (mg/g)	0.7801	3.3389
	h (mg/g min)	0.047	0.249
	R²	0.9902	0.9992
Intraparticle diffusion equation (7)	k_t (mg/g min^0.5)	6.71 × 10⁻²	1.97 × 10⁻¹
	C (mg/g)	0.0942	0.9484
	R²	0.8631	0.8310
Elovich equation (8)	a_e (mg/g min)	0.0944	0.6052
	b_e	5.54939	1.5184
	R²	0.9633	0.9597

The phosphate uptake process onto the Al-eggshell appears to occur over three stages: (1) a first sharper reaction stage at 0–60 min for section of 3 mg/l and 0–90 min for 50 mg/l; (2) a low reaction stage at 60–120 min for phosphate concentration of 3 mg/l and 90–120 min for 50 mg/l; and, (3) an equilibrium stage after 120 min for both these sections (Figure 6). The first stage is similar to a report by Yang et al. (2013) who performed experiments applying Kanuma mud to adsorb phosphate. Phosphate was rapidly uptaken for 60 and 90 min for two phosphate solutions, possibly due to surface adsorption. Meanwhile, higher availability of vacant adsorption sites supplied advantageous effects in stage 1. Moreover, a higher driving force (Peleka and Deliyanni, 2009) provided by high concentration improved the contact chance between the sorbent and sorbate, thus, the higher phosphate adsorption efficiency by Al-eggshell can be achieved.

As seen from Figure 6 and Table 6, the pseudo-second-order kinetic equation with higher R² values (0.9902 for 3 mg/l and 0.9992 for 50 mg/l) revealed better compliance with the experimental outcomes, implying that the adsorption is controlled by chemosorption. It can also be seen that equilibrium adsorption capacity (q_e) increases from 0.7801 to 3.3389 mg/g as the initial phosphate concentration increases from 3 mg/l to 50 mg/l. Furthermore, the variations of the rate constant (k₂) have a declining trend with improving initial phosphate concentration. The initial adsorption rate, h (mg/g min), is expressed as $h = k_{2} q_{e}^{2}$

The fraction h was calculated from the intercept of the line obtained by plotting t/q_t versus t. There is a big increase in terms of h values of 0.202 mg/g min between 3 and 50 mg/l phosphate solutions.

Intraparticle diffusion model gave the poorest fitting efficiency, with correlation coefficients of 0.8631 and 0.8310 for these sections. The plot of q_t versus t^0.5 does not pass through the origin. This indicates some degree of boundary layer control and this also further implies that the intraparticle diffusion is not the only rate-limiting step, all of which could be happening simultaneously (Arami et al., 2008). It was obtained that intraparticle rate parameter values (k_t) augmented with initial phosphate concentration. This can be described by the growing effect of driving force leading to a decrease in the diffusion of phosphate in the sorbent.

The Elovich equation coefficients might be calculated from the plot q_t versus ln t. It can be seen that the values of a_e and b_e varied as a function of the initial phosphate concentration. Hence, on increasing the initial phosphate concentration from 3 to 50 mg/l, the value of a_e increases from 0.0944 to 0.6052 mg/gmin. Regarding the constant b_e, an increase of the initial phosphate concentration from 3 to 50 mg/l results in a decrease in the values of a_e from 5.54939 to 1.5184 g/mg due to the decreased surface availability for phosphate.

Further discussion

Phosphate dissociation equilibrium in aqueous solution is pH dependent, presenting as (Yang et al., 2013)

H_{2} {PO}_{4}^{-} + H_{2} O = H_{3} {PO}_{4} + {OH}^{-}

(9)

{HPO}_{4}^{2 -} + H_{2} O \to H_{2} {PO}_{4}^{-} + {OH}^{-}

(10)

{PO}_{4}^{3 -} + H_{2} O \to {HPO}_{4}^{2 -} + {OH}^{-}

(11)

H_{2} O = H^{+} + {OH}^{-}

(12)

As per the research from Karageorgiou et al. (2007), the dominant species are H₂PO₄⁻ between a pH of 5–7; HPO₄²⁻ from 7 to 12; and, PO₄³⁻ when pH is higher than 12. The preferred pH of this study is 6.12. Thus, the dominant species is H₂PO₄⁻ and is followed by HPO₄²⁻.

It is well known that metal oxides and hydroxides primarily existed bound to oxygen atoms and hydroxyl groups formed on the surface, which could provide an absorbing site for adsorption of phosphate (Yang et al., 2013).

The binding sites are protonated (MOH₂⁺) when phosphate is at various pH solution systems. Therefore, the reaction of electrostatic adsorption, $\equiv Al - {OH}_{2}^{+} + H_{2} {PO}_{4}^{-} \to (\equiv Al - {OH}_{2}^{+}) H_{2} {PO}_{4}^{-}$ , can occur due to the coulombic attraction between negative anions and positive protonated AlOH₂⁺. The following reactions potentially occurred in phosphate removal by Al-eggshell in the present experimental conditions. The presence of Al–OH group may contribute to the adsorption of phosphate as given by Borah and Senapati (2006)

\equiv Al - OH + H_{2} {PO}_{4}^{-} \to \equiv Al - H_{2} {PO}_{4} + {OH}^{-}

(13)

2 \equiv Al - OH + H {PO}_{4}^{2 -} \to (\equiv Al)_{2} - {HPO}_{4} + 2 {OH}^{-}

(14)

The proportion of Al increased sharply after it was modified by AlCl₃ (Table 2), which enhanced the adsorption of phosphate efficiency effectively. Thus, this phenomenon seems to imply that for adsorption of phosphate onto the Al-eggshell, outer sphere complexes between the sorbent component surface and P, including electrostatic bonding mechanisms, potentially coexist with inner sphere surface complexes between the Al-eggshell and phosphate, which necessarily containing quantities of covalent bonding. Collins et al. (1999) reports that phosphate species are usually considered to be removed through an inner sphere ligand exchange mechanism (Collins et al., 1999). Meanwhile, a higher coulomb force between binding sites and phosphate in addition to chemical interaction results in higher phosphate loading, as presented in equations (13) and (14). The schematic diagram of phosphate adsorbed on Al-eggshell is explained in Figure 7.

Figure 7.

Schematic diagram of phosphate adsorbed on Al-eggshell.

In order to illustrate the potential application of Al-eggshell, the comparison of maximum adsorption of phosphate capacity of Al-eggshell from this study and that of different materials from the previous studies is presented in Table 7. From Table 7, it has been seen that this comparison obviously indicated that Al-eggshell is an effective sorbent for phosphate removal.

Table 7.

Reported sorption capacity of phosphate for various sorbents.

Sorbent	q (mg/g)	References
Sediment mineral matrix	0.63	Xiao et al. (2013)
Natural iron oxide coated sand	0.88	Boujelben et al. (2008)
Iron oxide tailings	8.21	Zeng et al. (2004)
Ferrihydrite–humic acid	5.43	Yan et al. (2016)
Goethite	4.67
Goethite–humic acid	3.27
Diatomite	3.12	Wang et al. (2016)
Fly ash	10.70
Zeolite	0.12
NaOH–LaCl₃-modified zeolite	6.63 (293 K)	He et al. (2016)
	7.45 (303 K)
	9.10 (313 K)
This study	11.52 (323 K)

Conclusions and further research

The experimental and modeling results indicated that Al-eggshell could potentially be adopted as a sorbent for phosphate species.

As per Box–Behnken design, a quadratic model was developed for prediction of phosphate species adsorption by Al-eggshell. The prediction model exhibited that phosphate species removal efficiency increased by AlCl₃ concentration and pH values and decreased with dosage. The optimum solution suggested by OPTIMIZATION of “Design Expert” was 0.29 mol/l C(AlCl₃), pH 6.12, and a 6.72 g/l dose with 99.06% removal. Three-dimensional response surface and contour pictures reported that the interaction between pH and dosage was the strongest, followed by C(AlCl₃) and dosage, and finally C(AlCl₃) and pH.

Present work described that this adsorption process the Freundlich model at temperature 313 and 323 K provided good fit. On the other hand, the Langmuir–Freundlich equation agreement better fit with the data at other temperatures. The pseudo-second-order kinetic equation described the experimental results well, which implies that the adsorption is controlled by chemosorption.

In addition, Al-eggshell adsorbing phosphate species is friendly to the environment, requires no further treatment, and the absorbed sorbent can be used for soil fertilization.

Based on previous and present work, further research should focus on expanding the application of the Al-eggshell to heavy metals and organic contaminants, for example. The point of the selectivity of phosphorous over other ions also will be paid more attention in the following studies.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: publication of this article: Funding for this work by National Natural Science Foundation of China (51674161), Innovation Team Project of Shandong University of Science and Technology (2012KYTD102), Specialized Research Fund for the Doctoral Program of Higher Education (20133718110005), and Shandong Provincial Education Association for International Exchanges is gratefully acknowledged.

References

Ahmad

Afandi

Adegoke

et al. (2016) Optimization and batch studies on adsorption of malachite green dye using rambutan seed activated carbon. Desalination and Water Treatment 57: 21487–21511.

Ahmad

Bello

(2017) Statistical optimization of adsorption process for the removal of synthetic dye using watermelon rinds. Modelling Earth System Environment 3: 1–9.

Aljbour

Al-Harahsheh

Aliedeh

et al. (2017) Phosphate removal from aqueous solutions by using natural Jordanian zeolitic tuff. Adsorption Science & Technology 35: 284–299.

Arabi

Ghaedi

Ostovan

et al. (2016) Synthesis and application of molecularly imprinted nanoparticles combined ultrasonic assisted for highly selective solid phase extraction trace amount of celecoxib from human plasma samples using design expert (DXB) software. Ultrasonics Sonochemistry 33: 67–76.

Arami

Limaee

Mahmoodi

(2008) Evaluation of the adsorption kinetics and equilibrium for the potential removal of acid dyes using a biosorbent. Chemical Engineering Journal 139: 2–10.

Bello

Ahmad

(2011a) Response surface modeling and optimization of Remazol Brilliant Blue reactive dye removal using Periwinkle shell-based activated carbon. Separation Science and Technology 46: 2367–2379.

Bello

Ahmad

(2011b) Removal of Remazol brilliant violet-5 R dye using periwinkle shells. Chemistry and Ecology 27: 481–492.

Bian

Liu

et al. (2013) Treatment of phosphorus wastewater using crystallization method. Journal of Environmental Sciences 23: S106–S109.

Biswas

Inoue

Ghimire

et al. (2007) The adsorption of phosphate from an aquatic environment using metal-loaded orange waste. Journal of Colloid and Interface Science 312: 214–223.

10.

Borah

Senapati

(2006) Adsorption of Cd(II) from aqueous solution onto pyrite. Fuel 85: 1929–1934.

11.

Boujelben

Bouzid

Elouear

et al. (2008) Phosphorus removal from aqueous solution using iron coated natural and engineered sorbents. Journal of Hazardous Materials 151: 103–110.

12.

Chardon

Blaauw

(1998) Kinetic Freundlich equation applied to soils with a high residual phosphorus content. Soil Science 163: 30–35.

13.

Chen

Zhao

Pan

et al. (2015) Preferable removal of phosphate from water using hydrous zirconium oxide-based nanocomposite of high stability. Journal of Hazardous Materials 284: 35–42.

14.

Cheung

Porter

McKay

(2000) Sorption kinetics for the removal of copper and zinc from effluents using bone char. Separation and Purification Technology 19: 55–64.

15.

Chun

Jeon

Kim

et al. (2005) The phase-shift method for determining Langmuir and Temkin adsorption isotherms of over-potentially deposited hydrogen for the cathodic H₂ evolution reaction at the poly-Pt/H₂SO₄ aqueous electrolyte interface. International Journal of Hydrogen Energy 30: 1423–1436.

16.

Collins

Ragnarsdottir

Sherman

(1999) Effect of inorganic and organic ligands on the mechanism of cadmium sorption to goethite. Geochim Cosmochim Acta 63: 2989–3002.

17.

Elabbas

Mandi

Berrekhis

et al. (2016) Removal of Cr(III) from chrome tanning wastewater by adsorption using two natural carbonaceous materials: Eggshell and powdered marble. Journal of Environmental Management 166: 589–595.

18.

Francis

Abdel Rahman

(2016) The environmental sustainability of calcined calcium phosphates production from the milling of eggshell wastes and phosphoric acid. Journal of Cleaner Production 137: 1432–1438.

19.

Grubb

Guimaraes

Valencia

(2000) Phosphate immobilization using an acidic type F fly ash. Journal of Hazardous Materials 76: 217–236.

20.

Guo

et al. (2017) Phosphorus removal from aqueous solution in parent and aluminum-modified eggshells: thermodynamics and kinetics, adsorption mechanism, and diffusion process. Environmental Science and Pollution Research 24: 14525–14536.

21.

Lin

Dong

et al. (2016) Simultaneous removal of ammonium and phosphate by alkaline-activated and lanthanum-impregnated zeolite. Chemosphere 164: 387–395.

22.

McKay

(1999) Pseudo-second order model for sorption processes. Process Biochemistry 34: 451–465.

23.

Inyang

Onwawoma

Bae

(2016) The Elovich equation as a predictor of lead and cadmium sorption rates on contaminant barrier minerals. Soil and Tillage Research 155: 124–132.

24.

Karageorgiou

Paschalis

Anastassakis

(2007) Removal of phosphate species from solution by adsorption onto calcite used as natural adsorbent. Journal of Hazardous Materials 139: 447–452.

25.

Lai

Xie

Chi

et al. (2016) Adsorption of phosphate from water by easily separable Fe₃O₄@SiO₂ core/shell magnetic nanoparticles functionalized with hydrous lanthanum oxide. Journal of Colloid and Interface Science 465: 76–82.

26.

Xin

Bian

et al. (2016) The physical properties of poly(L-lactide) and functionalized eggshell powder composites. International Journal of Biological Macromolecules 85: 63–73.

27.

Liu

Zhao

Sibille

et al. (2016) Evaluation of natural organic matter release from alum sludge reuse in wastewater treatment and its role in P adsorption. Chemical Engineering Journal 302: 120–127.

28.

Lurling

Waajen

van Oosterhout

(2014) Humic substances interfere with phosphate removal by lanthanum modified clay in controlling eutrophication. Water Research 54: 78–88.

29.

Mezenner

Bensmaili

(2009) Kinetics and thermodynamic study of phosphate adsorption on iron hydroxide-eggshell waste. Chemical Engineering Journal 147: 87–96.

30.

Moghddam

SRM

Ahad

Aqil

et al. (2016) Formulation and optimization of niosomes for topical diacerein delivery using 3-factor, 3-level Box–Behnken design for the management of psoriasis. Materials Science & Engineering C – Materials for Biological Applications 69: 789–797.

31.

Novais

Buruberri

Seabra

et al. (2016) Novel porous fly-ash containing geopolymer monoliths for lead adsorption from wastewaters. Journal of Hazardous Materials 318: 631–640.

32.

Park

Jeong

Yang

et al. (2007) Removal of heavy metals using waste eggshell. Journal of Environmental Sciences 19: 1436–1441.

33.

Peleka

Deliyanni

(2009) Adsorptive removal of phosphates from aqueous solutions. Desalination 245: 357–371.

34.

Seo

Hong

Kim

et al. (2013) Phosphorus removal from wastewater by ionic exchange using a surface-modified Al alloy filter. Journal of Industrial and Engineering Chemistry 19: 744–747.

35.

Seviour

McIlroy

(2008) The microbiology of phosphorus removal in activated sludge processes – The current state of play. Journal of Microbiology 46: 115–124.

36.

Shams

Dehghan

Nabizadeh

et al. (2016) Adsorption of phosphorus from aqueous solution by cubic zeolitic imidazolate framework-8: Modelling, mechanical agitation versus sonication. Journal of Molecular Liquids 224: 151–157.

37.

Shu

Liu

et al. (2016) Solidification/stabilization of electrolytic manganese residue using phosphate resource and low-grade MgO/CaO. Journal of Hazardous Materials 317: 267–274.

38.

Wang

et al. (2016) Comparison study of phosphorus adsorption on different waste solids: Fly ash, red mud and ferric–alum water treatment residues. Journal of Environmental Sciences 50: 79–86.

39.

Weber

Morris

(1963) Kinetics of adsorption on carbon from solution. Journal of the Sanitary Engineering Division 89: 31–60.

40.

William

Owen

(1995) Egg Science and Technology, New York: Food Product Press.

41.

Xiao

Zhu

Cheng

et al. (2013) Characteristics of phosphorus adsorption by sediment mineral matrices with different particle sizes. Water Science and Engineering 6: 262–271.

42.

Xiong

Tong

Yang

et al. (2017) Adsorption of phosphate from aqueous solution using iron-zirconium modified activated carbon nanofiber: Performance and mechanism. Journal of Colloid and Interface Science 493: 17–23.

43.

Yan

Jiang

Yao

et al. (2016) Preliminary investigation of phosphorus adsorption onto two types of iron oxide-organic matter complexes. Journal of Environmental Sciences 42: 152–162.

44.

Yang

Ding

Zhao

et al. (2013) Investigation of phosphate adsorption from aqueous solution using Kanuma mud: Behaviors and mechanisms. Journal of Environmental Chemical Engineering 1: 355–362.

45.

Yang

Jin

Wang

et al. (2016) Phosphate recovery through adsorption assisted precipitation using novel precipitation material developed from building waste: behavior and mechanism. Chemical Engineering Journal 292: 246–254.

46.

Yang

Zhao

Ding

et al. (2013) An electrochemically modified novel tablet porous material developed as adsorbent for phosphate removal from aqueous solution. Chemical Engineering Journal 220: 367–374.

47.

Zeng

Liu

(2004) Adsorptive removal of phosphate from aqueous solutions using iron oxide tailings. Water Research 38: 1318–1326.

48.

Zhang

Kou

et al. (2014) The feasibility of adopting zeolite in phosphorus removal from aqueous solutions. Desalination and Water Treatment 52: 4298–4304.

49.

Zong

Liu

Jiang

et al. (2016) Preparation and characterization of zirconia-loaded lignocellulosic butanol residue as a biosorbent for phosphate removal from aqueous solution. Applied Surface Science 387: 419–430.

Statistical optimization and batch studies on adsorption of phosphate using Al-eggshell

Abstract

Keywords

Introduction

Materials and methods

Materials

Design of experiments using Box–Behnken

Characterization

Adsorption procedure

Equilibrium studies

Kinetic studies

Results and discussion

Response surface analysis of phosphate removal

Model equation construction

3D response surface and contour plots

Characteristics of the eggshell composite

Adsorption isotherms

Adsorption kinetics and effect of contact time

Further discussion

Conclusions and further research

Footnotes

Declaration of Conflicting Interests

Funding

References