Adsorption characteristics of alumina nanoparticles for the removal of hazardous dye, Orange G from aqueous solutions

2.1 Chemicals and reagents

All the reagents used in the synthesis were of analytical (a. r.) grade with the mass fraction purity of 0.99 and were used as received without any further purification. AlCl₃·6H₂O was purchased from Alfa Aesar, England and Orange G from E. Merck, Mumbai, India. Highest purity analytical reagent grades of liquid ammonia and ethanol (95%), HCl and NaOH were supplied by E. Merck, Mumbai, India. The stock solutions of Orange G were prepared by dissolving 1.0 g of dye in 1000 mL of double distilled water and working solutions of 50, 75, 100 and 125 mg/L were prepared daily with the required dilution.

2.2 Synthesis of alumina nanoparticles

γ-phase alumina nanoparticles were synthesized using the sol-gel precipitation in ethanol process as reported by Wang et al. (2008) with little modification. A precursor solution of 0.6 M was prepared using AlCl₃·6H₂O dissolved in ethanol (95%). The precipitating agent, liquid ammonia (1.0 M) was added dropwise to the ethanol solution with continuous stirring and this resulted in formation of a gelatinous white precipitate of Al(OH)₃. The resultant precipitate was filtered and washed with ethanol followed by oven drying at 90 °C for 6 h. The dried white gel was calcined in a muffle furnace at 600 °C for 3 h in the presence of air at a heating rate of 10 °C/min, thus transforming Al(OH)₃ into Al₂O₃ powder. After calcination, the calcined powder was milled and sieved. A yield of 91% was obtained. The advantage of ethanol based precipitation over aqueous precipitation is that in the former case, particles were formed having high degree of dispersivity due to formation of soft agglomerates, while in the latter case, hard heavy agglomerates were formed due to chemical interaction of absorbed water molecules with surface hydroxyl groups between adjacent particles.

2.3 Characterization of the adsorbent

The FTIR spectra of as-synthesized material were recorded on a Fourier Transform Infrared Spectrophotometer (Varian 1600 FT-IR Scimitar Series) to identify the functional groups present on the adsorbent. For recording IR spectra, the sample was encapsulated in KBr using a hydraulic pellet press and the spectra were taken in the wavelength range from 4000 cm⁻¹ to 400 cm⁻¹ with a resolution of 4 cm⁻¹. The TG/DTA was performed in a Simultaneous Thermal Analysis System (NETZSCH STA 409 PC/PG). The sample, mass 11 mg, was heated at a scanning rate of 10 °C/min under dynamic N₂ flow of 75 mL/min at temperature range 31 °C to 1000 °C. The X-ray diffraction pattern of the adsorbent was recorded by an X-ray diffractometer (XPERT-PRO) operated at 45 kV/40 mA using CuKα radiation with wavelength of 1.54 Å and a scanning rate of 1° per min. The crystallite size was evaluated from the diffraction peaks by using the Scherrer method. The adsorbent size was further assessed by analyzing the images recorded by a Transmission Electron Microscope (Philips CM200) at an accelerating electron voltage of 200 kV. Samples for TEM analysis were prepared by introducing a drop of suspension over a copper grid. The surface feature and elemental composition of the as-synthesized material were determined by Scanning Electron Microscope (Zeiss, Germany) and Energy Dispersive X-ray Spectroscopy at an accelerating voltage of 15.0 kV. The textural properties of the adsorbent were determined by N₂ adsorption-desorption isotherms measured at −196 °C using specialized analyzer (ASAP 2020, Micromeritics, USA). Prior to analysis, the sample was degassed at 250 °C for 2 h in the degas port of the analyzer. The adsorption data in the P/P₀ range of 0.01–0.18 were used to calculate the specific surface area of the product by the Brunauer−Emmett−Teller (BET) method. The pore size distribution was analyzed with the help of desorption curve of the isotherm through Barrett−Joyner−Hallenda (BJH) method.

2.4 Determination of pH_ZPC (point of zero Charge)

pH_ZPC of an adsorbent is an important characteristic as it helps in determining the pH value at which the surface of the adsorbent becomes electrically neutral and facilitates in predicting the surface behavior of adsorbent materials. For determination of pH_ZPC, the solid addition method was adopted using 0.01 M NaCl. The solution was placed in six different flasks: each containing 200 mg of alumina nanoparticles at different pH values. The pH of the solutions was adjusted in 2.0–12.0 range by using NaOH and HCl solutions. The flasks were then agitated at 120 rpm for 24 h and the pH of the solutions was measured by a pH meter (Systronics 335). A graph of pH_final vs. pH_initial was plotted and the point of intersection of the two graphs represents the pH_ZPC of the alumina nanoparticles.

2.5 Batch adsorption experiments

A series of batch adsorption experiments was carried out to investigate the optimum operating conditions for adsorption of OG on alumina nanoparticles. The effect of important process variables including initial dye concentration, contact time, solution pH and temperature was studied using 100 mL of dye solution in 250 mL Erlenmeyer conical flasks. In all adsorption experiments, 0.1 g of adsorbent was added. The mixture was agitated at 200 rpm for the desired time in a shaking thermostatically controlled water bath (Macro Scientific Works Pvt. Ltd., Delhi) at 303 K. The samples were collected from the flasks at regular intervals followed by centrifugation. The residual dye concentration in the supernatant was determined by UV-visible spectrophotometer (Model-2203, Systronics, Ahmadabad, India) at 480 nm. The percentage removal of OG and adsorption capacity (q_e) of alumina nanoparticles were calculated using the following equation:

Blank experiments were carried out using 100 mg/L dye solution in a flask without adsorbent to ensure that no dye was adsorbed onto the walls of flasks, while another experiment was conducted using adsorbent and water only to check the sorbent leaching issue which may interfere with the measurement of dye concentrations during spectroscopic analysis.

In order to maintain accuracy, batch experiments were replicated to ensure reproducibility of obtained data. Each batch adsorption experiment was performed in triplicate producing results with deviations <5%. If the error was found to be >5%, another set of experiments was performed. The data of the batch experiments used for the experimental analysis were the average values of three tests.

2.6 Error analysis

An error function assessment is required in order to avoid the biased result that usually occurs due to use of linear expressions resulting from the transformation of nonlinear expressions. Therefore, in the present study, the quality of fit was verified and quantitatively compared with the applicability of other models through the coefficient of determination (R²) and hybrid fractional error function (HYBRID) (Porter et al. (1999):

The error function represents the ameliorated version of the sum of squares of the errors (ERRSQ) fit at low concentration. Accordingly, each ERRSQ value is divided by experimental q_e,_exp, values. This function also includes the number of degrees of freedom of the system which can be evaluated as the number of data points, n, minus the number of parameters, p, of the isotherm equation—as a divisor.

2.7 Desorption and regeneration experiments

To evaluate the reusability of the adsorbent, the adsorption of Orange G and the regeneration of dye loaded alumina nanoparticles were performed by repeating adsorption–desorption cycles five times. In each cycle, 50 mg/L of 100 mL dye solution was mixed with 0.1 g of sorbent for equilibrium period. The loaded sorbents were separated through centrifugation and the supernatant dye solutions were analyzed. The resultant loaded sorbent was tested with different desorbing solutions such as 0.1 M HCl, 0.1 M HClO₄, 0.1 M HNO₃, 0.1 M H₂SO₄, and 0.1 M H₃PO₄ in order to examine the efficiency of each solution toward desorption. Prior to the next adsorption–desorption cycle, the regenerated sorbent was washed thoroughly with distilled water to pH–7.0.

2.8 Stability test

In order to evaluate the stability of the alumina nanoparticles, the leaching of aluminum ions from the adsorbent at variable pH conditions was investigated. Experiments were conducted by dispersing 0.1 g of alumina nanoparticles in 100 mL of aqueous solution with pH values ranging from 2.5 to 9.5. The resultant solutions were kept for shaking in a temperature-controlled shaker at 303 K for 120 min. The concentration of leached aluminum ions in the supernatant was quantified by atomic absorption spectroscopy (AA7000, Shimadzu, Japan).

3.1 Characterization of alumina nanoparticles

The assessment of the surface functional groups of the adsorbent material was carried out by FTIR spectrum analysis (Fig. 1). In the fresh sample (Fig. 1aA), the presence of the prominent peak at 3457 cm⁻¹ is attributed to the stretching vibration of the —OH group from Al—OH framework and another peak at 1629 cm⁻¹ is ascribed to physically adsorbed water molecules (Yang et al., 2010). The presence of a wide pattern extending in 400–1000 cm⁻¹ range substantiates the formation of γ-phase alumina. The absorption bands at 911, 804 and 637 cm⁻¹ correspond to asymmetric stretching, symmetric stretching and bending vibration of the Al—O—Al bond (Afkhami et al., 2010), respectively. The presence of peaks below 700 cm⁻¹ is related to an octahedral arrangement of Al³⁺ ions in the hcp lattice of the oxide ions while peaks between 700 and 950 cm⁻¹ indicate the occupancy of Al³⁺ ions in the tetrahedral sites in the ccp lattice of the oxide ions (Busca et al., 1993). Therefore, the as-synthesized γ-phase alumina comprises both octahedral and tetrahedral coordination units. Fig. 1aB represents adsorbent material after the sorption of dye ions. It can be clearly noticed that after dye adsorption, there has been slight change in peak positions. The peak intensity band at 3457 cm⁻¹ was found to be reduced slightly and a small shift in peak position was observed in the absorption band 1000–400 cm⁻¹. This indicates that dye ions most probably interact with Al—OH and Al—O groups present on the adsorbent surface.

Close

Figure 1 a: FTIR of fresh and dye loaded alumina nanoparticles; b: TG/DTA of alumina nanoparticles; c: XRD diffraction patterns; d: TEM image of alumina nanoparticles; e: particle size distribution analysis of alumina nanoparticles.

Export to PPT

The thermal behavior of the sample has been studied by TG-DTA analysis (Fig. 1b). The TG curve of the sample exhibited three weight losses. The first weight loss amounts to ca 11% of the initial sample weight and terminates at ∼148 °C, and this weight loss corresponds to a well distinguished exothermic peak in the DTA profile. This peak represents desorption of physically adsorbed water (Shen et al. 2012). The second significant weight loss of about 34% is recorded between 150 and 300 °C, which may be attributed to elimination of ethanol and impurities. The third weight loss can be observed at 375 °C and accounts for nearly 65% weight loss with a prominent exothermic peak observed in the DTA pattern ascribed to the dehydration of amorphous aluminum hydroxide (Isfahani et al., 2012). Beyond 600 °C, no significant weight loss is observed and the material becomes thermally stable. Hence, the sorbent was calcined at this temperature in order to obtain γ-phase alumina particles.

The phase confirmation of the as-synthesized adsorbent was accomplished through investigating the X-ray diffractogram as shown in Fig. 1c. The diffractrogram clearly indicates the amorphous nature of the sample. The four characteristic reflections at 2θ = 32°, 37.4°, 45.2° and 67.5° with corresponding reflection planes of 220, 311, 400 and 440 reveal the formation of the γ-species. The obtained peaks were also found to be in good agreement with the standards' database (JCPDS card 00-029-0063) (Huang et al., 2013). Besides phase confirmation, the diffraction peaks were also inspected for the determination of the particle size of the as-synthesized material by using Scherrer formula (Patterson, 1939):

The morphological aspects and the diameter of the nanoparticles were ascertained through TEM analysis. The TEM image of the as-synthesized material is depicted in Fig. 1d. It is obvious from the figure that alumina particles were found to be in somewhat agglomerated state; further, the image also depicted that the particle sizes were in nanorange. The particle size distribution analysis was also investigated through TEM technique and the obtained results (Fig. 1e) suggested that most of the particles size lie within the range of 30–40 nm.

The surface morphology of the alumina nanoparticles was investigated through SEM as shown in Fig. 2a. The SEM micrograph also verified that the particle size existed in the nanoscale range. The particles were found to be clumped together as a result of the formation of a weak van der Waal bond between the particles. The particles appear to be smooth and spindle shaped. The EDX spectrum in Fig. 2b clearly shows the peak of Al and O as major constituents which confirms the formation of alumina nanoparticles. The presence of the C peaks may be due to use of carbon coating during the SEM analysis. The EDX of the sample after dye adsorption (Fig. 2c) has also been investigated. The spectra reveal the presence of relevant ions such as sulfur, nitrogen and sodium ions of the dye molecules. This provides direct evidence for the sorption of dye ions onto alumina nanoparticles.

Close

Figure 2 a: SEM micrograph of alumina nanoparticles at magnification of 50 KX; b: EDX spectra of fresh nano alumina; c: EDX spectra of dye loaded adsorbent; d: N₂ adsorption-desorption isotherm for alumina nanoparticles (Inset: Pore size distribution curves).

Export to PPT

The N₂ sorption experiments were employed for the assessment of textural properties of the sample. Fig. 2d exhibits typical type IV isotherm with an apparent hysteresis loop of type H1, which suggested that adsorbent comprises of mesoporous structures (He et al., 2015). The specific surface area was estimated by the Brunauer-Emmett-Teller (BET) equation (Barret et al., 1951) and was found to be 128 m²/g. The estimated specific surface area was found to be larger than 125 m²/g, which reaffirms the formation of thermodynamically stable γ-phase alumina (Zhou et al., 2007). The pore sizes were also determined by analysis of the BJH pore size distribution curves. It is clearly visible from Fig. 2d (inset) that most of the pores (approximately 90%) were lying in mesoporous range (2–50 nm diameters) and rest of the 10% were found to be in the microporous range (<2 nm). The adsorbent average pore size and pore volume were estimated to be 13.6 nm and 0.32 cm³/g respectively. Thus, the presence of mesoporous structures and high pore volume offers a favorable condition for the liquid phase adsorptive removal of dye ions.

The pH of zero point charge (pH_ZPC) of the alumina nanoparticles was determined from the point of intersection of the pH_initial vs pH_final curve. pH_ZPC indicates the electrical neutrality of the adsorbent at certain pH values when the concentration of positively charged surface groups is the same as that of negatively charged ones. The pH_ZPC of alumina nanoparticles was found to be 7.4 (fig not shown) which is close to that reported by Li et al. (2011).

3.2 Effect of pH

The pH of solution is key parameter as it controls the sorption process by influencing the surface charge of adsorbent as well as adsorbate (Banerjee et al., 2014). Therefore for dye adsorption, the surface charge of the adsorbent and the dissociated forms of dyes in the solution pH is considered as a vital factor. The effect of solution pH on removal of OG by alumina nanoparticles was investigated by varying the solution pH from 2.5 to 9.5 and the results are given in Fig. 3. However, prior to analyzing the effect of pH on the absorbance of aqueous solutions of OG, pH absorbance tests were performed to make sure that changing the pH condition would have no effect on the dye absorbance characteristics. It was found that over pH range 2.5–9.5, the λ_max (480 nm) of the dye remains unaltered. However, between pH 10.0 and 11.5 there was a marked decrease in absorbance due to dissociation of the hydroxyl on the β-naphthol ring. Likewise studies at pH > 2.5 were not selected for experimental purposes as the pK_a value of the OG lies between 1 and 2 (Bird, 2012) which means the dye is available only in the completely ionized form beyond this value and exists as an anionic species for possible interaction with oppositely charged sites on the adsorbent. Thus, the experiments were carried out in this specified pH range and the removal of OG observed was solely due to adsorption not due to hydrolysis or chemical degradation of the dye structure. According to Fig. 3, it can be clearly seen at all initial dye concentrations that maximum removal takes place at pH 2.5, with the highest removal of 96.8% being achieved with an initial dye concentration of 50 mg/L. However, the removal percentage declined considerably with increasing pH, and minimum removal of 40.2% was recorded at pH 9.5 for a dye concentration of 125 mg/L. The removal mechanism may be elucidated on the basis of the surface chemistry of the alumina nanoparticles as it has been reported that the presence of a metal oxide in the aqueous phase results in the hydroxylation of the metal oxide surface (Wawrzkiewicz et al., 2015). Thus, metal oxide surfaces behave amphoterically and can undergo acid-base reactions in response to variable pH condition. Accordingly the probable mechanism that may be taking place is as follows:

Close

Export to PPT

Close

Figure 3 Effect of pH on the adsorption of Orange G on alumina nanoparticles (Temperature = 303 K; Adsorbent dose = 1.0 g/L).

Export to PPT

The above reactions suggest that the surface properties of alumina nanoparticles are strongly governed by the aqueous pH condition. Reaction (1) suggests that at a pH 7.4, the net surface charge of the adsorbent is zero and it becomes electrically neutral. Conceptually at pH > pH_ZPC in basic medium (reaction (2)), the adsorbent surface becomes negatively charged due to deprotonation; thus, the adsorption of anionic dye ions gets severely hindered as a result of columbic repulsion. In acidic pH condition (reaction (3)), when the pH < pH_ZPC the surface is charged positively and this electrostatically attracted the anionic OG species and leads to a higher adsorption of dye anions (reaction (4)). Hence, a pH of 2.5 was taken as the optimum pH for which the adsorbent efficiency for the removal of dye is a maximum. A similar observation has been reported by Zheng et al. (2012) and Monash and Pugazhenthi (2014) for the adsorption of the anionic dye, Methyl orange.

3.3 Effect of co-existing ions

It is well known that wastewater discharged from the textile industries is often contaminated with inorganic anions which impede the adsorption process. For this reason, the effect of inorganic anions on adsorption capacity of OG ions was investigated. Fig. 4 shows the effect of the coexisting ions for different inorganic anions, including NO₃⁻, Cl⁻, CO₃²⁻, SO₄²⁻, PO₄³⁻ and C₂O₄²⁻, have been selected as the competitive ions in the adsorption experiments. It has been observed that the adsorption of OG in the presence of NO₃⁻, Cl⁻ and CO₃²⁻ ions was not affected significantly. However SO₄²⁻, PO₄³⁻ and C₂O₄²⁻ anions exhibit a remarkable effect on the adsorption performance of OG. With the slight increase in concentration of PO₄³⁻ and C₂O₄²⁻ ions, the adsorption efficiency of OG plunges down below 30 mg/g. For the case of SO₄²⁻ the adsorption efficiency seems to be affected appreciably at higher SO₄²⁻, concentrations. The high affinity of alumina nanoparticles for SO₄²⁻, PO₄³⁻ and C₂O₄²⁻ ions may be attributed to formation of outer-sphere surface complexes on the surface of metal hydroxides (Hordern, 2004). Thus, the study of anions envisaged that the presence of SO₄²⁻, PO₄³⁻ and C₂O₄²⁻ ions in aqueous solution was identified as a critical limiting factor in the treatment process when alumina nanoparticles were employed as adsorbent.

Close

Figure 4 Effect of dye uptake in the presence of coexisting anions (dye concentration 50 mg/L; temperature 303 K; dose 1.0 g/L; pH 2.5).

Export to PPT

3.4 Effect of contact time and sorption kinetics

The effect of contact time on OG sorption by alumina particles was studied at 303 K and results are shown in Fig. 5. It was observed that the extent of sorption increased with the increase in contact time up to 30 min and then the sorption attained equilibrium. This suggested that the time required for OG-alumina nanoparticles interactions to attain steady state was just 30 min for all initial dye concentrations which further implies that the interaction is independent of initial OG concentrations. Therefore, all experiments were conducted for 30 min. The rapid removal in the initial stages (>50% removal accounted for all dye concentrations within 5 min) may be due to the fact that initially all the sorption sites were vacant and could adsorb more dye anions from the solution phase (Ghaedi et al., 2012). As the sorption process advanced, the adsorption rate decreased steadily owing to incursion of OG ions deep into the pores of the adsorbent and finally the adsorption curves maintained a flattening profile indicating almost complete saturation of the sorption sites.

Close

Figure 5 Effect of contact time for adsorption of Orange G on alumina nanoparticles for different initial dye concentrations (pH 2.5, adsorbent dose 1.0 g/L, temperature 303 K).

Export to PPT

Investigation of sorption kinetics helps in elucidating the mechanism that controls the sorption process. As it is quite certain, process variables play an imperative role in adsorption processes by influencing the physicochemical nature of the adsorbate-adsorbent and consequently control the removal rate. Therefore, interpretation of sorption kinetics data in terms of different process parameters including pH, temperature and initial dye concentration will provide valuable insights about the sorption mechanism. In the present study, kinetic data were analyzed by fitting them into two important kinetic models, namely the pseudo-first order (Lagergren, 1898) and the pseudo-second order (Ho and McKay, 1999).

The linearized form of the pseudo-first order equation is (Ho and McKay, 1999):

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

Close

Figure 6 Pseudo-second order kinetic plots obtained for various process parameters.

Export to PPT

Meanwhile, on inspection of the first and second order rate constants (Table 1), a decreasing trend has been noticed with an increase of pH and temperature while the rate constant follows a rising trend with increasing initial dye concentration. The trend observed for pH values may be due to a reluctance of the dye anions to diffuse across the barrier layer when negatively charged hydroxyl groups dominate, and similarly with the increase in solution temperature, rate constant values decline which may be attributed to the poor interaction between dye ions and active sites on the adsorbent due to the weakening of physical forces. This further interrupted effective mass transfer of dye ions from bulk to solid surface. In case of initial dye concentration, the rate constant values increased with dye concentrations which might have occurred due to the enhanced concentration gradient that easily surmounts the mass transfer resistance experienced by dye ions (Allen et al., 2005).

3.5 Adsorption mechanism

The actual mechanistic step that controls the overall rate of the sorption process can be investigated by employing the kinetic data into well-known mathematical models i.e. intraparticle diffusion model recommended by Weber and Morris (1963). Theoretically, plots with a linear relationship and a zero intercept are the requisite aspect that determines this model’s suitability as well as indicative of pore diffusion as a rate controlling step. The intraparticle diffusion model is given as follows (Weber and Morris, 1963):

Close

Figure 7 Intra-particle diffusion plots obtained for various process parameters.

Export to PPT

It has already been confirmed from the intra-particle diffusion study that more than one mechanisms seem to control the sorption process, but the model fails to express the precise rate determining step. This setback can be overcome by treating experimental kinetic data further with the external diffusion model and internal diffusion model (Piccin et al., 2011). The external mass transfer model assumes that the concentration at the solid surface approaches to zero and the intraparticle diffusion is negligible during initial period of contact time. External mass transfer has been represented as follows (Piccin et al., 2011):

A reliable estimate of the diffusivities for film diffusion and pore diffusion model can be evaluated using models proposed based on Fickian diffusion principle given as (Crank 1975):

Thus, the model film diffusivity and pore diffusivity have been fitted with the experimental data acquired from the corresponding first linear plot and second linear plot, respectively from the Weber-Morris model. The parameters D_film, and D_pore were evaluated from the plot based on Eqs. (9) and (10) and values are presented in Table 2.

Now, the parameters including k_f and D_pore, that have already been evaluated from their respective models have been further employed for the determination of a dimensionless parameter named as Biot number (Bi_N). This parameter facilitates determination of the rate limiting step of the sorption process. The parameter basically relates external mass transfer coefficient to the diffusion coefficient (Dizge et al., 2008). Thus, the influence of various parameters on the dye adsorption resistance can be examined through the Biot number given by the following expression (Dizge et al., 2008):

It can be seen from Table 2 that with the increase of pH and temperature, the resistance to the mass transfer increased and likewise, film diffusion and pore diffusion values were found to follow a retarding trend. However, opposite trend can be noticed for the Biot number which indicates an increase of resistance in the diffusion mechanism causing delay in the sorption process. In the case of initial dye concentration, the decrease in the resistance to mass transfer with the increase in dye concentration has been identified by the increase of the k_f values. This behavior is attributed to the effective driving force assisted to overcome the mass transfer resistance of the dye ions between liquid and solid phases. Meanwhile, the diffusion coefficients also exhibited a rising trend but the Bi_N values noticeably decline with dye concentration. Thus, high k_f and D_film values indicate that the external diffusion mechanism is significant on control of the sorption process which is further substantiated by a decrease in the Biot number with an increase in the dye concentrations. Moreover, as suggested by Dizge et al. (2008), the Biot number is a decisive factor that helps in deciding the predominance of surface diffusion against external diffusion, in case where the Biot number >100 adsorption processes are considered to be controlled by internal diffusion mechanisms. Cooney (1993) and Dotto et al. (2014), recommend that for Bi_N < 0.5, a complete dominance of the external mass transfer resistance can be anticipated while for Bi_N > 10, a considerable dominance of the intraparticle resistance is possibly expected. Thus, on the basis of above hypothesis, in the present sorption system, the as-evaluated Bi_N values indicate that external mass transfer predominantly controls the sorption of OG anions on alumina nanoparticles.

3.6 Adsorption isotherm

The isotherm study helps in interpreting the interaction between the dye concentration in aqueous phase and the amount of dye adsorbed per unit weight of adsorbent at constant temperature. The study was performed by employing the well-known isotherm models viz. Langmuir (1918), Freundlich (1906), and Dubinin-Radushkevich (1947), to the present system at different temperatures. It has been observed that for all temperatures, the sorption capacity for OG anions by alumina nanoparticles enhanced steadily with increase in initial dye concentration and ultimately approached saturation phase at higher dye concentration (Fig. 8). The adsorption capacity at equilibrium increased from 48.4 to 84.4 mg/g with the increase in initial OG concentrations from 50 to 125 mg/L. The increase in uptake capacity with concentration was anticipated as a result of high driving force for mass transfer (Banerjee et al., 2015).

Close

Figure 8 Isotherm plots for adsorption of Orange G on alumina nanoparticles at different temperatures.

Export to PPT

For the isotherm investigations, the experimental data are correlated by the isotherm expressions given as follows:

Langmuir model in linearized form:

Freundlich model in linearized form:

D-R model in linearized form:

The calculated values of the adsorption isotherm parameters and error functions for OG sorption onto alumina nanoparticles are listed in Table 3. As the error function and R² values determine the applicability of the isotherm model, according to the results, high R² values and low HYBRID values obtained for the Langmuir isotherm model indicated suitability of this model in expressing equilibrium sorption data. Furthermore, the suitability of the model indicates that dye is adsorbed as a monolayer onto a sorbent surface having a finite number of identical sorption sites. Meanwhile, the parameters calculated for D-R model help in predicting the nature of sorption process and possibility of adsorbent regeneration. The value of E can be evaluated from Eq. (15). If the values lie in between 8 and 16 kJ/mol, the adsorption process is supposed to be controlled by a chemisorption mechanism, while the values of E less than 8 kJ/mol indicate involvement of physical adsorption (Banerjee et al., 2015). Thus the estimated E values were found to be <8 kJ/mol which suggests that physical forces might be responsible for the interaction between OG anions and alumina nanoparticles. As the physisorption process is known to be reversible in nature (Dabrowski, 2001), this further recommends that the exhausted adsorbent can be easily regenerated via desorption.

3.7 Thermodynamic investigation

The thermodynamic behavior of the sorption process was investigated by estimating the thermodynamic parameters viz. change in free energy, enthalpy and entropy using the following well-known equations (Uma et al., 2013; Gautam et al., 2015):

3.8 Regeneration of adsorbent

Regeneration of adsorbent for repeated reuse is highly desirable and this not only makes the treatment process economically viable but also offers a prospective solution for the management of contaminant loaded adsorbent. In the present work, in order to determine the effectiveness of different solutions toward desorption, desorbing solutions such as 0.1 M HCl, 0.1 M HClO₄, 0.1 M HNO₃, 0.1 M H₂SO₄, and 0.1 M H₃PO₄ were tested. The study was conducted in batch mode using 1 g/L of dye loaded sorbent in 100 mL of desorbing solution and the contents agitated in a water bath shaker at 303 K until equilibrium was achieved. The solution was analyzed spectrophotometrically for the determination of desorbed dye concentration. Desorption results have been illustrated in increasing manner as H₃PO₄ 83% < H₂SO₄ 87% = HNO₃ 87% < HCl 91% < HClO₄ 93% (Fig. 9). The desorption trend further indicated that ion exchange mechanism governed the desorption process. Thus, with significant desorption potential, HClO₄ was selected for the regeneration and reusability study. It has been investigated that after four cycles of the adsorption-desorption process, the dye removal percentage was still more than 67% after which the removal of OG declined drastically. Thus, the study clearly demonstrated that alumina nanoparticles can be efficiently reused up to four cycles for the removal of OG from aqueous solutions without much loss of removal capacity.

Close

Figure 9 Regeneration of alumina nanoparticles using different desorbing reagents.

Export to PPT

3.9 Stability of adsorbent

As it is a well-known fact that solution pH influences the mobility of metal ions, possibility of leaching of aluminum ions from alumina nanoparticles into treated water can be anticipated when the same was employed for water treatment. Therefore, it is desirable to investigate the stability of the synthesized alumina nanoparticles by performing a leaching test in aqueous solution at variable pH condition. The experimental results suggested that the concentration of leached aluminum ions for all pH conditions (2.5–9.5) was estimated to be not more than 0.082 ppm which further infers that alumina nanoparticles exhibited reasonable stability in various aqueous media and can be safely used for treatment of the wastewater.

3.10 Assessment of sorption behavior of alumina nanoparticles using real wastewater

For this study, dye contaminated real wastewater was procured from the carpet industry located near Varanasi (Bhadohi, Uttar Pradesh, India). As the study comprised of model dye Orange G, therefore, wastewater sample was collected particularly from that industry only where OG was used. Moreover, during dyeing operations, other dyes were also used. In present investigations, it has been examined that along with Orange G two other dyes namely Indigo carmine and Reactive red 180 were also present in the sample and all these dyes were chemically anionic in nature. The effluent was then analyzed for various physicochemical parameters as depicted in Table 4. The BOD₅ measurement was carried out by BOD incubator (Adair Dutt & Company Pvt. Ltd., New Delhi) and for COD analysis was carried out by digestion apparatus (Mac Scientific Equipment, New Delhi). The results of the sampled wastewater analysis revealed that as expected, the wastewater contained high BOD₅ and COD. The concentration of the dyes present in the water sample was measured spectrophotometrically, the measurement for ternary dye system (OG + IC + RR180) has been carried out at their maximum absorbance wavelength λ_1,max (480 nm), λ_2,max (608 nm) and λ_3,max (543 nm) respectively, and absorbance was recorded as A₁, A₂ and A₃. The concentrations of specific dye forms were estimated using equa∗∗∗tion given as follows:

Now, the most important part of the investigation was the assessment of the performance of alumina nanoparticles in a multi-solute system. The study is found to be very important from the industrial viewpoint as the dyeing effluents comprised of mixed solute system; therefore, the results of the adsorption experiments performed using single solute system may vary from those of multi-solute system. Therefore, it is desirable to compare the removal efficiency of alumina nanoparticles for simulated and real wastewater. Batch adsorption experiments were carried out using 100 mL of sample (real wastewater) with adsorbent dose of 1.0 g/L at constant temperature of 303 K and at pH 7.6 and 2.5, respectively. The results (Table 5) imply that treatment of effluents without making any pH adjustment resulted in insignificant adsorption of dye effluents; however, when the pH of the effluents is adjusted to 2.5, an enhanced sorption performance can be observed for all the dyes. Nevertheless, a pH dependent removal process follows a similar trend as that for batch studies but the amount of dye uptake reduces significantly for mixed dye systems in comparison with single solute systems. However, the results obtained for the uptake capacity appear to be significant as the removal percentages for all the dyes remain more than 50% which indicate that the adsorbent is equally effective for a multi-solute system. Furthermore it has also been investigated that the adsorbent is highly proficient in lowering the organic load such as BOD and COD from the real industrial waste effluents.