Adsorption of organic matter from industrial phosphoric acid (H₃PO₄) onto activated bentonite

1 Introduction

Industrial phosphoric acid (IPA) is a product of great importance in chemical and food industries. In Tunisia, the IPA is produced from phosphate rock by a sulfuric acid attack. The presence of impurities in the raw material leads to relatively acid loaded toxic chemical species that were naturally present (typically organic matter, Cd, As, Ba, Hg, Mg, Al, V, Zn, Ca, etc.) detrimental to the quality of the IPA for use in the food industry and in the production of fertilizers and detergents.

The removal of organic matter contained in industrial phosphoric acid appears to be an important step, both for the production of phosphoric acid and fertilizers decontaminated (Mellah and Benachour, 2007; Boualia et al., 1993). This led us to search cheap and most feasible treating method. Our choice focused on the process of adsorption by Tunisian bentonite of Oued Tfal (Mhamdi et al., 2010). The clays used were purified-activated. These clays were abundant substances and were among the valuable materials whose valorization was very important to the economy. These materials constituted the raw material for manufacturing products as varied as construction materials (cement, bricks, tiles, pottery … ) molds for foundries, drilling muds and bleaching earth. They were also used in the composition of certain industrial fine chemicals (pharmaceuticals, cosmetics, and catalysts). Clays owed their industrial applications to their physicochemical properties such as plasticity, swelling power and adsorption, rheological properties and cation exchange capacity.

The work presented in this paper not only concerned the characterization of Oued Tfal clay from the region of Gafsa (Tunisia), but also the determination of physicochemical properties of the raw clay, treated by the hydrochloric acid and this utilities for the purification of industrial phosphoric acid.

2 Experimental

2.3

2.3 Bentonite

2.3.1

2.3.1 Crude bentonite

We used a bentonite collected from the deposit of “Oued Tfal” from the region of “Gafsa”, located in south Tunisia. The cation exchange capacity (CEC) determined by the methylthioninium chloride method before purification was about 54.23 meq/100 g air-dried bentonite.

2.3.2

2.3.2 Purified bentonite

The content of impurities was decreased or removed completely by applying a purification protocol proposed by van Olphen (1976). This was achieved by dispersing the crude lumps in hydrochloric acid solution (0.1 M) to facilitate elimination of carbonates by controlled acid attack but not to destroy the structure of our clay. This argillaceous suspension was agitated mechanically during 4 h then centrifuged at 3500 rpm during 15 min. Supernatant was removed, the remaining argillaceous fraction is mixed with 400 mL of NaCl solution(1 M). The clay suspension was shaken for 12 h. Then, it was centrifuged 3000 rpm for 5 min. The clay phase was recovered and the NaCl solution was rejected. This cycle of agitation–centrifugation was carried out seven times under the same conditions in order to exchange the interfoliaceous cations against those of sodium and to eliminate impurities by sedimentation.

After these exchanges, clay was washed three times with distilled water. The suspension obtained was put in dialysis membranes to remove the chloride ions adsorbed onto the surface from the layers. The dialysis water was renewed until what the test with silver nitrate indicates the absence of chloride ions. Then, the purified clay suspension was dried in an oven at a temperature not exceeding 60 °C in order to be activated later. The cation exchange capacity (CEC) was determined by the methylthioninium chloride method after purification was about 93.28 meq/100 g air-dried bentonite and the mass yield was 64.39%.

2.3.3

2.3.3 Activated bentonite

Acid-activation (Amari et al., 2010) was carried out with sulfuric acid (1.5 M) in a jacketed glass reactor equipped with a reflux condenser, a thermometer and a stirrer. At the end of each experiment, the solid content was immediately filtered, washed free of sulfate with hot water until the washing water was neutral, and dried at 110 °C for a few hours, in order to obtain dehydrated samples and a constant weight. The cation exchange capacity (CEC) was determined by the methylthioninium chloride method after purification-activation was about 73.33 meq/100 g air-dried bentonite and the mass yield was 75.25%.

2.3.4

2.3.4 Purified-activated bentonite

The purified bentonite must be activated later by the same method described for the purification and activation, the cation exchange capacity (CEC) determined by the methylene blue method after purification activation was about 120.82 meq/100 g air-dried bentonite and the mass yield was 58.97%.

Table 2 presents the CEC of the crude bentonite “JScr”, simple activated bentonite “JSa”, purified bentonite “JSp” and purified-activated bentonite “JSpa”.

Table 2 CEC of the bentonite.

Bentonite	CEC (meq/100 g)
JScr	54.23
JSp	93.28
JSa	73.33
JSpa	120.82

Table 3 presents the chemical composition of different bentonites (crude “JScr”, simple activated “JSa”, purified “JSp” and purified-activated “JSpa”) at 20 °C.

Table 3 Chemical composition (in wt.%) of the bentonite.

MxOy (%)	SiO2	Al2O3	MgO	Na2O	K2O	Fe2O3	CaO	P2O5	MnO	Ignition loss
JScr	50.09	22.52	1 .25	0.66	0.95	6.18	17.04	0.20	0.03	25.24
JSp	61.05	27.98	1.68	0.28	1.05	5.79	1.46	0.05	0.02	12.66
JSa	58.54	19.81	1.02	0.46	0.88	5 .87	12.43	0.12	0.02	22.96
JSpa	75.42	20.62	1.20	0.06	0.02	1.24	1.10	0.05	0.00	08.12

And Table 4 presents the mass yield of each technique (purification, Acid-activation and purification-activation).

Table 4 The mass yield of each technique (purification, acid-activation and purification-activation).

Technique	The mass yield (%)
Purification	64.39
Acid-activation	75.25
Purification-activation	58.97

Finally, an X-ray diffraction analysis (PANalytical X-Pert PRO powder diffractometer, the source of X-rays is a ceramic-class copper anode tube with 2.2 kW maximum power) and an Infrared Spectroscopy (SHIMADZU 8400S) were performed and the results presented successively in Figs. 1 and 2 confirmed that our sample belongs to the family of montmorillonite bentonites and the distinct increase of infrared absorbency at 3610–916 cm⁻¹, confirmed the dominant presence of dioctahedral smectite with Al–Al–OH stretching and bending bands.

Close

Figure 1 X-ray diffraction analysis of bentonites: Crude (a), purified (b), activated (c) and purified-activated (d) bentonite.

Export to PPT

Close

Figure 2 Infrared Spectroscopy of bentonites: Crude (a), purified (b), activated (c) and purified-activated (d) bentonite.

Export to PPT

3 Results and discussion

Equilibrium isotherms were obtained by contacting 100 mL of H₃PO₄ with different weights of bentonite in a temperature controlled battery of 200 mL special locally designed beakers. The latter are provided with an external mantle in which the temperature regulated water flows continuously.

The operating conditions were as follows: agitation speed, 250 rpm; contact time, 30 min; pH, 1.78; temperatures, 30, 40 and 50 °C. At equilibrium, the mixture was left to settle and the supernatant acid was filtered to remove any particles. These clear solutions were analyzed using UV–Vis spectrophotometer (Beckman Coulter DU-800 spectrophotometer) at 350 nm, it is the maximum wavelength of characterizing the OM present in H₃PO₄, and yielded a linear relation between optical densities of solutions and the OM concentrations. An H₃PO₄, free of OM was used as the blank (Mellah et al., 1991). The amount of OM adsorbed per gram of bentonite, X/m, was given experimentally as (C₀ − C_e) v/m where C₀ and C_e (mg/L) are the H₃PO₄ mass concentration initially and at equilibrium respectively, v is the volume of the solution (L) and m is the mass of dry bentonite used (g).

The equilibrium adsorption isotherms were of fundamental importance to determine the adsorption capacity of bentonite for the organic matter and to diagnose the nature of adsorption. The acid used contains 562.224 mg OM/L H₃PO₄. Isotherms are presented in Fig. 3. The full curves were obtained by a least squares fitting of experimental data. The models used are discussed as follows.

Close

Figure 3 General equilibrium of OM adsorbed onto bentonite JSpa.

Export to PPT

3.1

3.1 Langmuir model

The Langmuir theory was first used to describe the adsorption of gas molecules onto the metal surfaces (Langmuir, 1918). However, this model has found successful application in many other sorption processes. Basic assumptions of the Langmuir model are:

• -

  adsorption occurs only on specific sites,

• -

  the maximum possible adsorption is of a complete monomolecular layer,

• -

  the sites are homogeneous energy-wise. The amount of OM adsorbed per unit weight of adsorbent can then be expressed by:

(1)

$$q_e=\frac{X}{m}=\frac{Q_0{\mathit{bC}}_e}{1+{\mathit{bC}}_e}$$ A linear form of this equation is:

(2)

$$\frac{C_e}{q_e}=\frac{1}{Q_{0}b}+\frac{C_e}{Q_0}$$

Fig. 4 illustrates the Langmuir analysis of the isotherm data. Obvious deviation from linearity was exhibited for each of the three investigated temperatures. The Langmuir plots separate the regions of relatively low and high OM concentrations; therefore, in order to correlate experimental data, it is necessary to use different Q₀ and b values over corresponding concentration ranges (Table 5). These values reported in Eq. (1) allow for the plot of q_e versus C_e (Fig. 3). This fact implies that the solute–solute forces are stronger than the solute-adsorbent forces, at the beginning of the adsorption (the first branch).

Close

Figure 4 Langmuir equilibrium isotherms of O.M. adsorbed onto bentonite.

Export to PPT

Table 5 Langmuir, Freundlich and Redlich–Peterson isotherm constants.

T (°C)	Langmuir				Redlich–Peterson			Freundlich
	Ce (mg/L)	|Q0|(mg/g)	|b|.103 (L/mg)	R2	Log a	B	R2	Log kF	1/n	R2
30	<295.28	1.4513	3.276	0.980	23.54	−8.914	0.949	−30.08	12.89	0.934
	>295.28	14.7058	2.796	0.962	66.04	−25.470	0.990
40	<317.54	0.9990	3.060	0.926	28.88	−10.930	0.975	−42.62	17.71	0.960
	>317.54	5.5865	2.736	0.728	61.37	−23.630	0.987
50	<361.06	0.45187	2.760	0.970	83.21	−32.210	0.967	−82.35	32.93	0.944
	>361.06	18.5180	2.474	0.821	53.25	−20.040	0.962

The energy of adsorption can be diminished by the presence of solvent and/or other solutes (Giles et al., 1974). In our case, the industrial H₃PO₄ contained a high solute concentration in the form of minerals (Table 1).

The bentonite being a polar adsorbent, it preferably adsorbs polar molecules and in particular water. The plateau obtained on Langmuir isotherms was attributed to the monomolecular layer formation which contains both the solute and the solvent molecules. The second branch of the Langmuir isotherm was characterized by a rise in the adsorption capacity. This can be explained either by the penetration of the OM inside the macro- and micropores of the adsorbent, forming new adsorption surfaces, or by the formation of multilayers as a result of the interactions between the molecules of the adsorbate. A least squares fit allows the calculation of isotherm constants b and Q₀ which is the maximum amount of OM adsorbed corresponding to a monomolecular formation (Table 5). Generally, it can be observed that the maxima of adsorption (Q₀) decrease with increasing temperatures, a fact which conformed to the physical adsorption hypothesis.

3.2

3.2 Freundlich model

The Freundlich model stipulates that the ratio of solute adsorbed onto the solute concentration is a function of the solution concentration (Freundlich, 1906). This empirical model was shown to be consistent with an exponential distribution of active centers, characteristic of heterogeneous surfaces (Sips, 1948). The amount of solute adsorbed, q_e, is related to the concentration of solute in the solution at equilibrium, C_e, following:

(3)

$$q_e=k_F{C}_e^{\frac{1}{n}}$$ This expression can be linearized to give:

(4)

$$\log q_e=\log k_F+\frac{1}{n}\log C_e$$ where k_F and n are Freundlich constants related to the adsorption capacity and adsorption intensity respectively of the adsorbent.

Figs. 5 and 6 illustrate the Freundlich analysis of the isotherm data for 30, 40 and 50 °C. The data were quite well fitted by Freundlich equation. The values of k_F and 1/n could be calculated either by the least squares method or graphically. Results are presented in Table 5. As the temperature decreases, k_F increases, confirming the adsorption process.

Close

Figure 5 Freundlich equilibrium isotherms of O.M. adsorbed onto bentonite.

Export to PPT

Close

Figure 6 Logarithmic plot of Freundlich equilibrium isotherms of OM absorbed onto JSpa bentonite.

Export to PPT

3.3

3.3 Redlich–Peterson model

Redlich and Peterson (1959) have developed an empirical model that incorporates the features of Langmuir and Freundlich isotherms as given in the following equation:

(5)

$$q_e=\frac{K_r{C}_e}{1+{\mathit{aC}}_e^B}$$ where K_r is the modified Langmuir constant (dm³/g), a (dm³/mg) and B are constants. For simplicity, K_r = bq₀. This formula reduces to Freundlich isotherm at a low surface coverage and to Langmuir isotherm at a high adsorbate concentration. This model can describe the adsorption process over a wide range of concentrations. It has a linear dependence on concentration in the numerator and an exponential function in the denominator.

a and B can be estimated from the slope and the intercept of the linearized form of Redlich–Peterson equation, respectively:

(6)

$$\log \left(\frac{K_r{C}_e}{q_e}-1\right)=B\log C_e+\log a$$ Figs. 7 and 8 illustrate the Redlich–Peterson analysis of adsorption data at 30, 40 and 50 °C.

Close

Figure 7 Redlich–Peterson equilibrium isotherms of O.M. adsorbed onto bentonite.

Export to PPT

Close

Figure 8 Logarithmic plot of Redlich–Peterson equilibrium isotherms of OM absorbed onto JSpa bentonite.

Export to PPT

4 Conclusions

Adsorption of organic matter contained in industrial phosphoric acid onto bentonite JSpa has been investigated. The mathematical analysis of the adsorption isotherms at different temperatures was discussed using Langmuir, Freundlich and Redlich–Peterson models. The equilibrium adsorption isotherms of organic matter onto bentonite are well described by the Langmuir for low concentrations and Freundlich models for high concentrations but the model of Redlich–Peterson is not used to fit the experimental data because the degree of heterogeneity (0 < B < 1) is negative.

The maximum monomolecular layer capacity given by the Langmuir model was 14.7058 mg/g at 30 °C. This value was too small because the other numerous impurities present in H₃PO₄ like Mg²⁺ influence for the absorbability of the organic matter.

The parameters' influence on the adsorption was not often easy to predict. Moreover, it was to be borne in mind that, due to the difficult operating conditions, the experimental data scatter widely over the limited and narrow temperature range that is covered by this work.