Removal of Cu²⁺, Fe²⁺ and SO₄²⁻ ions from industrial wastewater by ion exchange resins contained in a rotating perforated cylindrical basket of different heights

Abstract

The present study is concerned with the development of a new cylindrical basket filled with ion exchange resin. The performance of the reactor was examined by removing Cu²⁺, Fe²⁺ and SO₄²⁻ ions from synthetic wastewater. Variables studied were the initial ion concentration in the solution, contact time, resin height inside the cylindrical basket and cylindrical basket rotational speed. Dimensionless analysis was used to obtain a mass transfer correlation for each of the mentioned ions suitable for scale up and design of the present reactor. The experimental results revealed that both the percentage and the rate of removal of (Cu²⁺, Fe²⁺ and SO₄²⁻) ions decrease as the initial ion concentration in the solution increases, while they increase as the contact time, rotational speed and (L/d) ratio increase. Both Langmuir’s and Freundlich’s adsorption isotherms were examined and it was found that Langmuir’s adsorption isotherm gives a better fitting for the obtained data than Freundlich’s. Regeneration ability was tested, which revealed the high resin efficiency upon operating several consequence cycles that could reach 4 cycles with a slight decrease in the removal efficiency.

Introduction

Many industrial activities generate a significant amount of heavy metal ions in water via their wastewater effluents, which negatively impact the environment, aquatic life, and human health¹. Social pressure for a safe environment has forced governments to legislate more stringent measures for wastewater treatment, especially for non-biodegradable pollutants such as heavy metals, to avoid contamination of the water bodies (rivers, oceans, and seas). According to this policy, the present work is concerned with (Cu²⁺, Fe²⁺ and SO₄²⁻) removal from wastewater. Copper and iron ions are found with variable concentrations in several industries, such as the metal finishing industry, the printed circuit industry, and the effluent of the mining industry.

An example of the simultaneous presence of (Cu²⁺, Fe²⁺ and SO₄²⁻) in the wastewater of copper plating plants where steel is coated with copper from a CuSO₄ bath. The rinsing solution contains Cu²⁺and SO₄²⁻ ions which should be removed to recycle water. Fe²⁺, SO₄²⁻ comes also from the acid pickling step of the steel work piece in H₂SO₄ to remove oxides prior to plating.

Copper and iron ions in wastewater, even with low concentrations, could be harmful and cause serious diseases like liver damage in the worst case. Copper discharge to surface water could be of great concern as it will have the form of a suspended solid (sludge) or even as a free ion starts with small traces and builds up till it exceeds a relatively high concentration of which soil and most attached plants won’t be able to survive, moreover animals on lands polluted with copper would be poisoned as the worst case². The presence of iron is essential for the biological system due to its vital role in the synthesis of hemoglobin, however higher dose of iron is considered to be a serious poisonous to newborn babies and even humans in general as the digestive system absorbs the iron content from water which will cause the failure of its cells, iron also cause kidney and liver damage as well as a malfunction in the cardiovascular system³. Knowing that the maximum allowable limit of Cu²⁺ and Fe²⁺ in water bodies is 1 and 0.3 mg/L respectively according to EPA^4,5. The presence of sulfate ions in water, either surface or groundwater, is of high concern to many environmental, industrial, and health institutions because the high content of sulfate ions in water could favor the formation reaction of hydrogen sulfide, which is harmful to both aquatic life and also industrial equipment. Sulfate ions in water could be sourced from acid mine drainage and also from other several industries, high content of sulfate ions moving through metal pipes such as Mn, Ni, Cu...etc. could cause the release of these metals into the water causing a direct threat to aquatic life and human health⁶. High SO₄²⁻ concentrations in water bodies above the allowable limit (250 mg/L) can cause several bad effects on humans and animals according to EPA⁷. Besides, SO₄²⁻ can be reduced by anaerobic sulphate reducing bacteria to sulphide ions which is highly lethal and corrosive to metallic structures⁸.Sulfates are discharged from several industries such as textiles, dyes, fertilizers, metal, and plating industries. Several different methods for the removal of heavy metal ions and other pollutants from wastewater have been studied over the past decades, such as chemical precipitation, electrodeposition, membrane filtration, liquid–liquid extraction, ion exchange process, electrocoagulation etc. The ion exchange process was found to be the most suitable method for wastewater containing low to moderate amounts of heavy metal ions due to its high removal efficiency in addition to its reasonable cost⁹. Several types of adsorbents have been tested for the removal of heavy metal ions over the past decades like activated carbon, zeolites (organic resin), and synthetic resin, there are several types of synthetic resin used for different removal purposes like strong acidic cation exchange resin with sulfonic acid as a functional group, strong base anion exchange resin with the quaternary amino group as a functional group, weak acidic cation exchange resin with carboxylic acid as a functional group, and weak base anion exchange resin with primary, secondary and/or ternary amino group as a functional group^10,14 Ion removal by ion exchange resin involves two steps, namely (1) mass transfer from the bulk solution to the resin material. (2) ion exchange between the removable ion and the exchanger resin¹⁵. The aim of the present work is to develop a new ion exchange reactor made of a rotating cylindrical perforated basket, filled with ion exchange resins. The rate of removal of Cu²⁺, Fe²⁺ and SO₄²⁻ from synthetic wastewater was studied by the suggested reactor. Variables studied were ion concentration, the effect of resin height inside the basket to the basket diameter ratio (L/d), and the rotation speed of the basket. Previous studies proved that using either fluidized bed (Agitation of resin freely in solution) reactor or fixed bed reactor has drawbacks, where mass transfer of a fluidized bed is low because of low slip (relative) velocity between the resin particles and the solution while fixed bed suffers from high-pressure drop and hence high pumping power and high operating costs of the process^9,16,17. The present study seeks to reach a dimensionless mass transfer correlation that can be used in practice for the design and scale up of the present reactor.

Experimental part

Resin and pretreatment

A strong-acid cation (SAC) exchange resin TRILITE MC-10H was used for the removal of (Cu²⁺, Fe²⁺), while a strong-base anion (SBA) exchange resin TRILITE MA-12 was used for the removal of (SO₄²⁻). Table 1 shows the physical, and chemical properties and pretreatment of the resin¹⁸.

Table 1 Physical, chemical properties and pretreatment of the used resin.

Apparatus

The experimental setup shown in Fig. 1 describes the present work which consists of a plexiglass vessel of 2.5 L with 4 baffles fixed on the inside wall of the vessel, a stainless-steel screen basket of mesh no.40 with a top removable cover for ease of resin packing and unpacking, where the basket is connected to a rotating shaft attached to an electric motor with a variable speed controller.

Figure 1

schematic diagram for the experimental setup.

Procedures

• Prior to each run, both the plexiglass vessel and the stainless-steel basket were washed with distilled water and dried.

• 2 L of freshly prepared required solution with the required concentration of either copper, iron or sulfate was added into the vessel.

• The required amount of previously treated resin (either cation or anion) was added into the stainless-steel basket, and then the top of the basket was closed tightly with a stainless-steel cover of the same basket material to avoid resin escape during basket rotation.

• The stainless-steel basket shaft was connected to the motor and fully immersed in the solution inside the vessel.

• Adjustment of the motor speed at the required value was carried out by a variac; an optical tachometer was used to measure the rotation speed.

• Withdrawal of a 10 ml sample from the vessel (from the same point) every 10 min for Cu²⁺ ions removal and every 5 min for Fe²⁺, SO₄²⁻ ions removal.

• Copper ion samples were analyzed using iodometric titration while iron ion samples were analyzed using redox titration, while sulfate ion samples were analyzed using “Hach DR3900” spectrophotometer¹⁹.

• All runs were carried out under a constant temperature of 25 ± 1 °C and pH 5. Solution viscosity and density were obtained using a viscometer and a density bottle, respectively; then the diffusion coefficient was obtained from literature at 25 °C^20,21.

Studied parameters

• Initial (Cu²⁺, Fe²⁺, SO₄²⁻) ion concentration (200,300,400 mg/L)

• Cylindrical basket rotational speed (200,300,400,500,600,700 rpm)

• Ratio between resin height inside the cylindrical basket and the diameter of the cylindrical basket (0.59, 0.47, 0.35) for strong cation resin and (0.69, 0.55, 0.41) for strong anion resin.

• Regeneration ability of the strong cation resin

Results and discussion

Mass transfer analysis & data correlation

According to previous studies, the rate-determining step of the ion exchange process is either the film diffusion or the intra-particle diffusion step¹⁶. The former is due to the resistance resulting from the boundary layer around the ion exchange resin, while the latter is due to the resistance resulting from the porosity of the resin particle. Agitation affects the film diffusion step; conversely, the particle diffusion step is not sensitive to agitation. To determine the rate of the (Cu²⁺, Fe²⁺, SO₄²⁻) ion removal via the ion exchange process using a batch reactor, the following equation was used, assuming a pseudo-first order reaction²².

$$- V\frac{dc}{{dt}} = KAC$$

(1)

By integration at t = 0, C = C₀ and t = t, C=C the equation becomes

$$Ln \frac{{C_{0} }}{C} = \frac{KA}{V} t$$

(2)

where C₀: initial ion concentration (mg/L), C: ion concentration at a given time (mg/L), V solution volume (cm³), t: time of reaction (s), K: rate constant (cm/s), A: resin particle surface area (cm²).

$$A = \frac{{6W}}{{d_{p} \uprho _{r} }}$$

(3)

where W: resin mass (g), d_p: resin particle average diameter (cm), ρ_r: resin density (g/cm³).

The data presented in Fig. 2a–c show a typical plot of Ln (C₀/C) versus time at different rotational speeds, the data fit Eq. (2) which emphasizes that the reaction is a first-order diffusion reaction. From Fig. 2a–c which represents Eq. (2), the slope of each straight line is (KA/V), from which the (Cu²⁺, Fe²⁺, SO₄²⁻) removal mass transfer coefficients were calculated under different operating parameters.

Figure 2

Relation between C0/C and time at different basket rotational speed for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

The data presented in Fig. 3a–c show the relation between rotational speed and the mass transfer coefficient, as the cylinder rotation speed increases, the mass transfer coefficient (K) increases. According to the equations:

$${\text{For}}\;{\text{Cu}}^{2 + } \quad \quad K = \alpha N^{0.528}$$

(4)

$${\text{For}}\;{\text{Fe}}^{2 + } \quad \quad K = \alpha N^{0.6388}$$

(5)

$${\text{For}}\;SO_{4}^{2 - } \quad \quad K = \alpha N^{0.553}$$

(6)

Figure 3

Relation between K and rpm at different initial ion concentration for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

The exponent 0.528 from Eq. (4) is in agreement with the value of 0.5 obtained by El Shazly et al. (for his rate of mass transfer study at the wall of a rectangular agitated vessel with a 90° pitched blade turbine)²³.

Previous studies explained the increase in mass transfer coefficient with cylinder rotational speed by the fact that cylinder rotation induces an upward axial flow through the porous bed^17,24,25,26. The axial flow then moves radially from inside the rotating bed to the outer solution. During this journey of the solution inside the basket, Fig. 4, it enriches the bed with fresh solution and decreases the diffusion layer thickness around each particle, with a consequent increase in the mass transfer coefficient according to Eq. (7)¹⁶. In addition, it is well known that rotating cylinders are characterized by the formation of a high degree of turbulence at their outer rotating basket surface, with a consequent increase in the rate of (Cu²⁺, Fe²⁺ and  SO₄²⁻) removal²⁷ (Fig. 4).

Figure 4

Solution flow pattern during cylinder rotation.

$$K= \frac{D}{\updelta }$$

(7)

Previous mass transfer studies showed that film diffusion is the rate-determining step, which is consistent with the present finding¹⁶.

The obtained data were treated mathematically using dimensionless groups Re, Sh, Sc, and an additional dimensionless factor (L/d) was added to account for the effect of the aspect ratio of the reactor on the rate of mass transfer, where l is the resin height inside the cylindrical basket and d is the diameter of the cylinder.

Re, Sh, and Sc are defined as:

$$Re = \frac{{\uprho {\text{Nd}}^{2} }}{\upmu }$$

(8)

$$Sh= \frac{Kd}{D}$$

(9)

$$Sc = \frac{\upmu }{{\uprho {\text{D}}}}$$

(10)

where ρ is the solution density, N is the cylinder rotation speed, d is the cylinder diameter, µ is the solution dynamic viscosity and D is the mass transfer diffusion coefficient.

The data presented in Fig. 5a–c fit the following equations:

$${\text{For}}\;{\text{Cu}}^{2 + } \quad \quad Sh\,\alpha {\text{Re}}^{0.528}$$

(11)

$${\text{For}}\;{\text{Fe}}^{2 + } \quad \quad Sh\,\alpha {\text{Re}}^{0.622}$$

(12)

$${\text{For}}\;SO_{4}^{2 - } \quad \quad Sh\,\alpha {\text{Re}}^{0.553}$$

(13)

Figure 5

Relation between Sh and Re at different Sc number for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

The exponent 0.528 from Eq. (11) is in agreement with the previously obtained values of 0.523 which was obtained by Hagag et al.¹⁶.

The data presented in Fig. 6a–c fit the following equations:

$${\text{For}}\;{\text{Cu}}^{2 + } \quad \quad Sh\,\alpha \left( \frac{L}{d} \right)^{0.259}$$

(14)

$${\text{For}}\;{\text{Fe}}^{2 + } \quad \quad Sh\,\alpha \left( \frac{L}{d} \right)^{0.32}$$

(15)

$${\text{For}}\;{\text{SO}}_{4}^{2 - } \quad \quad Sh\,\alpha \left( \frac{L}{d} \right)^{0.454}$$

(16)

Figure 6

Relation between Sh and L/d at different rpm for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

Equations (11, 14) were used in Fig. 7a to obtain an overall mass transfer correlation for Cu²⁺ ion removal as shown in Eq. (17) with an average deviation of ± 9.1% and standard deviation of ± 13.2% under the conditions of 1261.5 < Sc < 1352.9, 3233.94 < Re < 11,697.4, 0.59 < L/d <0.35

Figure 7

Overall mass transfer correlation for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

$$Sh = 0.6114Sc^{{0.33}} \text{Re} ^{{0.528}} \left( {\frac{L}{d}} \right)^{{0.259}}$$

(17)

Equations (12, 15) were used in Fig. 7b to obtain an overall mass transfer correlation for Fe²⁺ ion removal as shown in Eq. (18) with an average deviation of ±8.58% and standard deviation of ±11.8% under the conditions of 1348.9 < Sc < 1510.4, 3234.15 < Re < 11,698.53, 0.59 < L/d < 0.35

$$Sh = 0.566Sc^{{0.33}} \text{Re} ^{{0.622}} \left( {\frac{L}{d}} \right)^{{0.32}}$$

(18)

Equations (13, 16) were used in Fig. 7c to obtain an overall mass transfer correlation for SO₄²⁻ ion removal as shown in Eq. (19) with an average deviation of ± 3.72% and standard deviation of ± 4.5% under the conditions of 1261.5 < Sc < 1352.92, 3233.9 < Re < 11,697.4, 0.69 < L/d < 0.41

$$Sh = 0.5188Sc^{{0.33}} \text{Re} ^{{0.553}} \left( {\frac{L}{d}} \right)^{{0.454}}$$

(19)

Factors affecting the % removal

Initial ion concentration

The data presented in Fig. 8a–c show that, as the initial ion concentration increases from 200 to 400 mg/L, the percentage removal decreases. This could be explained by the fact that the amount of resin is constant, thus the same amount of active sites are available for the exchange process, which will be fully occupied and any further ions will remain in the solution. Also, it could be explained by the decrease in diffusion coefficient with increasing the initial ion concentration, which is attributed to the increase in interionic attraction between ions in the solution and the increase in the viscosity of solution¹⁷.

Figure 8

%Removal versus rotational speed at different initial ion concentration for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

Contact time

The data presented in Fig. 9a–c show the relation between the percentage removal and the contact time under different initial ion concentrations. It was found that % removal increases as the contact time increases, reaching approximately the equilibrium concentration of (Cu²⁺) after 60 min and of (Fe²⁺, SO₄²⁻) after 30 min. Figure 10a–c also show that the % removal increases as the contact time increases under different (L/d) ratio.

Figure 9

% Removal VS contact time at different initial ion concentration for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

Figure 10

% Removal VS contact time at different L/d ratio for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

L/d ratio

The increase in K and the % pollutant (Cu²⁺, Fe²⁺, SO₄²⁻) removal with increasing L/d as shown in Fig. 10a–c may be explained in the light of the following two opposing effects:

• 1. As L increases, the solution flowing through the bed Fig. 4 becomes depleted in the pollutant. As a result of the decrease in pollutant concentration with bed height, the driving force for the reaction decreases with a consequent decrease in the mass transfer coefficient according to Eq. (2)

• 2. On the other hand, as the depleted solution arises in the rotating bed, it meets fresh (unexhausted) resin particles with many active centers, which react with the residual ions with a higher degree of removal than the lower partially or fully exhausted resin particles. The increase in K and % pollutant removal with increasing L under the present range of conditions suggests that the second enhancing effect (2) is predominating over the first retarding effect (1). The relatively weak retarding effect arising from pollutant depletion in the upper zones may be ascribed to the partial replenishment of the upper zone by pollutant ions which diffuse from the turbulent outside solution to the upper zone of the rotating basket by virtue of the concentration gradient between the inner solution and the outer turbulent solution of the rotating cylinder basket²⁷.

Ion exchange equilibrium

Equilibrium data is an essential requirement for designing an adsorption system. Adsorption equilibrium is related to temperature, thus at a given temperature, this equilibrium is called an adsorption isotherm. It is used to obtain the maximum possible capacity of the sorbent, where the model which gives the best fit for the obtained experimental results is the best adsorption isotherm to represent those results. Adsorption capacity (q) at equilibrium was obtained from Eq. (20)^16,17,23.

$${q}_{e}= \frac{\left({C}_{o}-{C}_{e}\right)V}{W}$$

(20)

where q_e is the adsorption capacity after reaching equilibrium (mg/g), C₀ is the initial concentration in solution (mg/L), C_e is the equilibrium concentration in solution (mg/L), V is the used solution volume (L) and W is the dry resin mass (g), knowing that equilibrium time was set to be 60 min for Cu²⁺ ion removal, and 30 min for both Fe²⁺ and SO₄²⁻ after plotting saturation curves during experimental work.

Langmuir isotherm

Langmuir assumed the following points to build up his isotherm^17,23,24:

• Finite number of active sites arrayed onto the adsorbent surface.

• All active sites show equal adsorption affinity (Homogeneously).

• Only monolayer (single) adsorption takes place.

• Almost negligible interaction occurs between adsorbed molecules.

Langmuir isotherm linear form equation is as follows:

$$\frac{{C}_{e}}{{q}_{e}}= \frac{1}{{K}_{L}{Q}_{m}}+ \frac{{C}_{e}}{{Q}_{m}}$$

(21)

where K_L is the adsorption constant for Langmuir (L/mg), Q_m is the maximum theoretical adsorption capacity (mg/g).

The data presented in Fig. 11a–c show the Langmuir adsorption isotherm for different initial ion concentrations, plotting C_e/q_e versus C_e, from the graphs K_L, Q_m R² of each line were obtained as shown in Table 2.

Figure 11

Langmuir adsorption isotherm at different initial ion concentration for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

Table 2 Langmuir adsorption isotherm.

Freundlich isotherm

This adsorption isotherm is based on the assumption of, a heterogonous surface of an exponential distribution of the adsorption heat over the surface^9,15.

Freundlich linear form equation is as follows

$$Log \,{q}_{e}=Log \,{K}_{F}+ \frac{1}{{\mathrm{n}}_{F}} Log \,{C}_{e}$$

(22)

where both K_F (L/g) and 1/n_F are the adsorption constants for Freundlich.

The data presented in Fig. 12a–c show the Freundlich adsorption isotherm at different initial ion concentrations, plotting Log q_e versus Log C_e, where from the graph K_F, 1/n_F and R² were obtained as shown in Table 3.

Figure 12

Freundlich adsorption isotherm at different initial ion concentration for (a) Cu²⁺ ion, (b) Fe²⁺ ion, (c) SO₄²⁻ ion.

Table 3 Freundlich adsorption isotherm.

The obtained R² values in Tables 2, and 3 show that Langmuir adsorption isotherm gives the best fitting for the data obtained from experimental work. The negative values of K_L and 1/n_f indicate that the adsorption process complies with previous results and confirms that the process won’t be efficient with high adsorbate concentrations¹⁶.

Regeneration

Since the cost of ion exchange is a major cost item in the process of wastewater remediation by ion exchange, we decided to examine the efficiency of regenerated resin to make sure that the resin can be used many times without significant loss of efficiency, thus saving on the operating costs of the process.

The used strong acid cation (TRILITE MC-10H) reliability was tested by reusing the same amount of resin several cycles, each cycle is meant by the adsorption–desorption reaction. After each adsorption process, the used amount of resin was regenerated using 0.1 M HCl. The data presented in Fig. 13 show the relation between % removal of (Cu²⁺) ion and reaction time at constant 400 mg/L initial (Cu²⁺) concentration and 500 rpm for several cycles for the resin, the graph revealed that the resin showed almost a constant % removal of (Cu²⁺) ion for the first 4 cycles and a slight decrease for the 5th cycle, while a significant decrease for the 6th cycle. This proves that (TRILITE MC-10H) showed great reliability for the (Cu²⁺) ion removal even after using the resin several times.

Figure 13

% Removal of (Cu²⁺) versus contact time for adsorption–desorption cycles.

Conclusion

The present study has revealed that the rotating ion exchange basket reactor is a highly promising reactor in ion removal from wastewater. Examination of the reactor in the separate removal of (Cu²⁺, Fe²⁺ and SO₄²⁻) at a rotating fixed bed of ion exchange resin has produced the following results for the conditions: initial (Cu²⁺, Fe²⁺ and SO₄²⁻) concentration range 200–400 mg/L, pH 5, constant temperature of 25 ± 1 °C, (L/d) ratio range 0.35–0.69 and basket rotating speed range 200–700 rpm.

(1) The system ion exchange process is a first-order diffusion-controlled reaction.

(2) Dimensionless treatment for mass transfer data has yielded the following correlations for (Cu²⁺, Fe²⁺ and SO₄²⁻) respectively: Eqs. (17)–(19).

These equations can be used in practice to scale up (design) and operate the present reactor.

(3) (TRILITE MC-10H) resin showed great performance to adsorb (Cu²⁺) ion, even after several adsorption–desorption cycles using HCl for regeneration which reflects the magnificent efficiency of the resin particles against heavy duty.

(4) The system showed a higher (Cu²⁺, Fe²⁺ and SO₄²⁻) ion removal efficiency with increasing basket rotating speed thanks to the flow pattern of axial (inlet)-radial (outlet) which allowed a better circulation for the solution on the resin particles, thus an increased rate of mass transfer.

(5) Equilibrium data studies showed that Langmuir adsorption isotherm gives a better fitting for the obtained experimental data than Freundlich isotherm.