Bengal gram seed husk as an adsorbent for the removal of dyes from aqueous solutions – Column studies

1.1 Theory of bed depth service time (BDST) model

The problem in designing adsorption columns is to predict how much effluent the bed will treat, or how long the bed will last before regeneration is necessary. All existing models are based on determining breakthrough curves for specific systems. Therefore, column operations are essential for industrial scale designing of technical systems. A number of models for the design of fixed bed adsorbers have been developed based on mathematical analysis and prediction of the shape of breakthrough curves.

A simplified approach for fixed bed adsorbers is available to correlate the service time, t, with the operation variables. Such a model is the Bed Depth Service Time (BDST) model (Hutchins, 1973; Dole and Klotz ,1946). The BDST (bed depth service time) model proposed by Hutchins (1973) is adopted for the design of fixed-bed adsorbers. It deals with the movement of an adsorption wave front through the adsorbent bed. The BDST has a linear relationship as given below

S_t = Bed service time (min)

A = N₀/(C₀ u) and $\gamma =[u/(K_0{N}_0)][\ln (C_0/C_t)-1]$

B_d = Bed depth (cm)

N₀ = Dynamic capacity by using column

K₀ = Rate constant of adsorption by using column (L/mg min)

u = Dye flow rate (ml/min)

By plotting service time S_t against bed depth from the experimental data, N₀ can be evaluated from the slope of the graph, and K₀ is obtained from the intercept at S_t = 0.

The service time of a column can be related to a number of process variables such as concentration of adsorbate, column height, flow of solution of adsorbate, pH of adsorbate solution and presence of foreign ions, size of the adsorbent and nature of adsorbate. However, at 50% breakthrough, the BDST model reduces to a form which predicts that a plot of BDST at 50% breakthrough against bed height should be a straight line passing through the origin. The critical bed depth is the point where the best-fit line through the data intersects the abscissa.

2.1 Preparation of adsorbent, SHBG

The preparation of adsorbent is already explained in previous paper (Somasekhara Reddy et al., 2017a,b). However, the procedure is again given here. The seed husk of Bengal gram (SHBG) is discarded as a waste in a small-scale industry where dal of Bengal gram (which is used in the preparation of certain food items) is separated from seed of Bengal gram. This waste is used in certain areas as foodstuff to the animals in addition to use as fire wood in hotels and restaurants. The SHBG is collected from a local industry, which is in a nearby town, Nandyal and washed thoroughly with de-ionized water for removing dirt. The dried husk material is ground and sieved to desired mesh size like 53<–<75 μm. It is abbreviated as SHBG. It is used as an adsorbent for the removal of dyes like CR, DTB and DB.

2.2 Adsorbate

CR, DTB and DB are obtained from M/S Sipka Sales Corporation, New Delhi free of cost and are used without further purification. The wave length of the maximum absorbency for CR, DTB and DB is 497, 615 and 500 nm, respectively. The physical properties of selected dyes are shown in Table 1. The chemical structures of selected dyes are shown in Fig. 1.

Close

Figure 1

Structures of Congo red (CR), Direct turquoise blue 86 (DTB) and Direct black 38 (DB).

Export to PPT

Stock solution of 1000 mg L⁻¹ was prepared by dissolving accurate quantity of the dye in double distilled water. The experimental solution was obtained by diluting the stock solution to the designed initial dye concentration.

2.3 Experimental

3.1 Characterization of the adsorbent

The characterization of the adsorbent, SHBG was done in the previous papers where the SHBG was used for removal of CR through batch kinetics experiments and equilibrium studies (Somasekhara Reddy et al., 2017a,b).

3.2 Effect of initial concentration of CR

The effect of initial concentration of CR (25, 50, 75 and 100 mg L⁻¹) on the adsorption of CR in column containing SHBG was studied at a constant flow rate (0.67 ml min⁻¹), a constant particle size of SHBG (53<–<75), a constant cross sectional area of column (same column is used for all CR concentrations), constant pH of the CR solution (7.02) and constant height (3.5 cm). The experimental details like time taken for emerging first colourless drop and colour drop from the column are shown in Table 2. The breakthrough curves for all initial concentrations were drawn between the volume treated and C_t/C₀ as shown in Fig. 2. Higher t_1/2 is observed for the lowest concentration of 25 mg L⁻¹ of CR and the order of decreasing t_1/2 was as follows 25 mg L⁻¹ of CR (831.69 min) > 50 mg L⁻¹ of CR (343.55 min) > 75 mg L⁻¹ of CR (217.5 min) > 100 mg L⁻¹ of CR (97.42 min). It is observed that the breakthrough time decreased with increasing inlet CR concentration. It was observed from Fig. 2 that the sharper breakthrough curves were obtained as the inlet concentration of CR increased. The larger the inlet concentration, the steeper is the slope of breakthrough curve and the smaller is the breakthrough time. The low initial CR concentration resulted in better adsorption into SHBG than the high initial CR concentration. This was because the former could be adsorbed more slowly through the SHBG. In other words the CR molecules were not adsorbed onto SHBG in a short time period at high initial CR concentration. It was confirmed further by the column capacity and the values are arranged in Table 2. A similar observation was made in case of adsorption of methylene blue on chemically modified wheat straw (Zhang et al., 2011a,b), adsorption of reactive black on granular activated carbon prepared from waste (Ahmad and Hameed, 2010), adsorption of methylene blue on phoenix tree leaf powder (Han et al., 2009) adsorption of DTB onto polyamide-epichlorohydrin-cellulose polymer (Hwang and Chen, 1993) and adsorption of brilliant yellow on cross-linked chitosan fibre (Yoshida and Takemori, 1997).

Close

Figure 2

Breakthrough curve of the effect of influent concentration on CR adsorption on SHBG. Conditions: flow rate = 0.667 ml min⁻¹; size of SHBG ⩾ 53 < 75 μm; pH of CR solution = 7.01; column height = 3.5 cm.

Export to PPT

3.3 Effect of particle size of the adsorbent

The effect of SHBG particle size (53<–<75, 75<–<90, 90<–<150 and >150 μm) on the adsorption of CR in column containing SHBG was studied at constant flow rate (0.667 ml min⁻¹), constant cross sectional area of column (same column is used for all particle sizes), constant initial concentration of CR solution (50 mg L⁻¹), constant pH of CR solution (pH 7.02) and constant height (3.5 cm). The experimental details like time taken for the first colourless drop and the colour drop emergence from the column are shown in Table 3. The breakthrough curves for all particle sizes were drawn between the volume treated and C_t/C₀ as shown in Fig. 3. Higher t_1/2 is observed for the smaller sized particles of SHBG and the order of decreasing t_1/2 was as follows (53<–<75) size of SHBG (343.55 min) > (75<–<90) size of SHBG (186.00 min) > (90<–<150) size of SHBG (126.39 min) > (>150) size of SHBG (96.65 min), respectively. This reveals that smaller size was favourable for high adsorption of CR. It was confirmed further by the column capacity and the values are arranged in Table 3. It was an expected trend only because a smaller particle provides more surface area. Fig. 3 shows that the breakthrough point curves shifted to the left when particle size increased. However, this shift was greater when the particle size increased from 53<–<75 μm to 75<–<90 μm than when it passed from 75<–<90 μm to 90<–<150 μm to >150 μm. This similar trend was observed in equilibrium studies also (Somasekhara Reddy et al., 2017a,b). A close observation of the results reveals that t_1/2 was becoming closer at higher sizes. Probably, at more higher sizes, t_1/2 may be the constant due to less surface area available. A similar observation was made in case of biosorption of reactive black 5 on chitosan (Barron-Zambrano et al., 2010), adsorption of astrazone blue on wood (Poots et al., 1978) and the adsorption of DTB on polyamide-epichlorohydrin-cellulose polymer (Hwang and Chen, 1993).

Close

Figure 3

Breakthrough curve of the effect of size of SHBG on CR adsorption on SHBG. Conditions: flow rate = 0.667 ml min⁻¹; influent concentration of CR = 50 mg L⁻¹; pH of CR solution = 7.01; column height = 3.5 cm.

Export to PPT

3.4 Effect of pH

The effect of pH (5.6, 7.02 and 9.1) of CR solution on the adsorption of CR in column containing SHBG was studied at a constant flow rate (0.667 ml min⁻¹), a constant particle size of SHBG (53<–<75 μm), a constant cross sectional area of column (same column is used for all pHs), a constant initial concentration CR solution (50 mg L⁻¹) and a constant height (3.5 cm). The experimental details like time taken for the first colourless drop and the colour drop emergence from the column are shown in Table 4. The breakthrough curves for all pHs of solutions of CR were drawn between the volume treated and C_t/C₀ as shown in Fig. 4.

Close

Figure 4

Breakthrough curve of the effect of pH of CR solution on CR adsorption on SHBG. Conditions: flow rate = 0.667ml min⁻¹; influent concentration of CR = 50 mg L⁻¹; influent concentration of CR = 50 mg L⁻¹; size of SHBG ⩾ 53 < 75 μm; column height = 3.5 cm.

Export to PPT

Higher t_1/2 is observed for the lower pH of CR solution and the order of decreasing t_1/2 was as follows pH 5.6 of CR solution (627.21 min) > pH 7.02 of CR solution (343.55 min) > pH 9.1 of CR solution (92.31 min). It was confirmed further by the column capacity and the values are arranged in Table 4. This reveals that lower pH is favourable for the removal of CR by using SHBG. This may be due to the interaction between more positive CR ion and the more negative SHBG due to acquiring a positive charge by the cellulose present in SHBG as it was in contact with water. This similar trend was observed in kinetics and equilibrium studies also (Somasekhara Reddy et al., 2017a,b). As shown in Fig. 4 with an increase of pH in the influent, the breakthrough curves shifted from right to left, which indicated that less CR was removed. The adsorption capacity of SHBG would decrease with an increase in pH of GR solution. It requires less time to reach the saturation, and the efficiency of biosorption was much lower. A similar observation was made in case of adsorption of brilliant yellow on cross-linked chitosan fibre (Yoshida and Takemori, 1997). It was observed contrary to this that the adsorption of methylene blue (MB) on rice husk was increased with an increase in pH of MB solution (Han et al., 2007)

3.5 Effect of dyes

The effect of similar type of dyes like DTB and DB (both are direct dyes because CR is also direct dye) on the adsorption of CR in column containing SHBG was studied at constant flow rate (0.67 ml min⁻¹), constant particle size of SHBG (53<–<75 μm), constant cross sectional area of column (same column is used for all dyes), constant pH of the CR solution (7.02), constant height (3.5 cm) and at constant influent dye, DTB/DB/CR concentration (50 mg L⁻¹). The experimental details like time taken for the first colourless drop and the colour drop emergence from the column are shown in Table 5. The breakthrough curves for all the combinations were drawn between the volume treated and C_t/C₀ as shown in Fig. 5. The more service time at which 50% breakthrough was achieved (t_1/2) was observed for the CR and the order of decreasing t_1/2 is as follows CR (343.55 min) > DB (53.41 min) > DTB (21.07 min). Therefore, SHBG was capable to remove CR effectively compared to DB and DTB even though the CR, DB and DTB belong to the same type of dyes (direct dyes). It is confirmed further by the column capacity and the values are arranged in Table 5. The low uptake is possibly due to the fact that the DB and DTB have much larger dye molecules and a long chain than the CR, so that it could not penetrate the internal pore structure of the SHBG particles. As shown in Fig. 5, breakthrough curves shifted from right to left, for CR, DB and DTB, respectively. The adsorption capacity of SHBG would decrease for DB and DTB. DB and DTB require less time to reach the saturation when compared with CR, and the efficiency of biosorption was much lower for DB and DBT. The absorption of direct red 83 in the column contains polyamide-epichlorohydrin-cellulose polymer was less compare to the adsorption of DTB and direct red 83 on the same material because direct red 83 is larger than that of DTB and direct red 83 (Hwang and Chen, 1993).

Close

Figure 5

Breakthrough curve of the effect of different direct dyes on dye adsorption on SHBG. Conditions: flow rate = 0.667 ml min⁻¹; influent concentration of dyes = 50 mg L⁻¹; size of SHBG ⩾ 53 < 75 μm; column height = 3.5 cm.

Export to PPT

3.6 Effect of foreign ions

For the determination of any interference caused by the presence of other inorganic cations, monovalent sodium (Na) and potassium (K), and divalent calcium (Ca) and magnesium (Mg) were added to the CR solution passed through the column for adsorption studies. These elements are major constituents of saline and hard waters and are likely to be encountered in most industrial effluents from which dyes are intended to be removed by these column studies. Not only that in the dye manufacturing process, salts are usually discharged into the wastewater for their salting-out effect. Many ions are present in the dye-house wastewater. Therefore, the influence of foreign ions was studied in this paper.

The influence of foreign ions like Na, K, Ca and Mg on the adsorption of CR in column containing SHBG was studied at a constant flow rate (0.667 ml min⁻¹), a constant particle size of SHBG (53<–<75 μm), a constant cross sectional area of column (same column is used for all foreign ions), a constant initial concentration CR solution (50 mg L⁻¹) and a constant height (3.5 cm). The concentrations of NaCl, KCl, CaCl₂ and MgSO₄ salts in the present experiments were 1, 1, 0.5 and 0.5 M, respectively. The experimental details like time taken for the first colourless drop and the colour drop emergence from the column are shown in Table 6. The breakthrough curves for CR in the presence of foreign ions were drawn between the volume treated and C_t/C₀ as shown in Fig. 6. Higher t_1/2 is observed for the presence of monovalent ion, potassium along with CR (CR-K) and the order of decreasing t_1/2 is as follows CR-K (540.82 min) > CR-Mg (473.83 min) > CR-Ca (403.28 min) > CR-Na (384.67 min) > CR alone (343.55 min). It was confirmed further by the column capacity and the values are arranged in Table 6. This reveals that the presence of either monovalent ion or divalent ion increases adsorption of CR. The enhancement of adsorption caused by salts was also documented (Han et al., 2008; Jeng 1992; Randtke and Jepsen, 1982; Verwey and Overbeek, 1948) in the literature. A similar trend was observed in case of adsorption of CR on rice husk (Han et al., 2008), adsorption of basic red 46, basic red 18 and basic red 28 on activated clay (Hsu et al., 1997), adsorption of direct orange 39 (McKay, 1982a,b), adsorption of phenol (Cooney 1998), adsorption of p-nitrophenol (Snoelyink et al., 1969), adsorption of sodium benzene sulphonate (Coughlin and Tan, 1968) and various organic compounds (benzoic acid, aniline, m-phenylenediamine and anthranilic acid) (Cooney and Wijaya, 1987) on activated carbon in the presence of NaCl, Na₂ SO₄, NaCl, NaCl, CaCl₂ and NaCl respectively when organic compounds are fully ionized. It is thought that the presence of salts, electrolytes, the charge density in the diffusion layer is significantly increased, which in turn minimizes the volume of diffusion layer required to neutralize the surface charge. These cations have a thinning effect on the diffusion layer formed on the surface of the particles. Although the total net charge did not change, the thickness of the diffusion layer had diminished significantly (Benefield et al., 1982). Further the more complex structure of CR would indicate a more complex charge distribution throughout its structure making it more prone to adsorption.

Close

Figure 6

Breakthrough curve of the effect of different foreign ions on CR adsorption on SHBG. Conditions: flow rate = 0.667 ml min⁻¹; influent concentration of dyes = 50 mg L⁻¹; size of SHBG ⩾ 53 < 75 μm; column height = 3.5 cm.

Export to PPT

The adsorption of CR was more in the presence of KCl compared to other salts. Probably the diffusion layer thickness may be further reduced in the presence of KCl. The variation of concentrations of salts may further increase the adsorption capacity.

The adsorption of methylene blue in the presence of salts (NaCl and CaCl₂) onto rice husk was decreasing (Han et al., 2007) instead of an increase in case of adsorption of CR on the same rice husk in the presence of NaCl (Han et al., 2008).

3.7 Effect of height

The effect of bed depth of the column (3.5, 7, 10 and 12.5 cm) on the adsorption of CR in column containing SHBG was studied at a constant flow rate (0.67 ml min⁻¹), a constant particle size of SHBG (53<–<75 μm) a constant cross sectional area of column (same column is used for all bed depths), a constant initial concentration CR solution (50 mg L⁻¹) and a constant pH of CR solution (7.08). The experimental details like time taken for the first colourless drop and the colour drop emergence from the column are shown in Table 7. The breakthrough curves for all bed depths were drawn between the volume treated and C_t/C₀ as shown in Fig. 7.

Close

Figure 7

Breakthrough curve of the effect of different column heights on CR adsorption on SHBG. Conditions: flow rate = 0.667 ml min⁻¹; influent concentration of dyes = 50 mg L⁻¹; size of SHBG = > 53 < 75 μm; column height = 3.5 cm.

Export to PPT

Higher t_1/2 was observed for the higher bed depth and the order of decreasing t_1/2 was as follows, 12.5 cm (957.45 min) > 10 cm (739.52 min) > 7.0 cm (547.86 min) > 3.5 cm bed depth of column (343.54 min). It was confirmed further by the column capacity and the values are arranged in Table 7. This reveals that higher bed depth was favourable for the removal of CR by using SHBG. So the higher bed depth of column resulted in a decrease in the solution concentration in the effluent at the same time. The slope of breakthrough curve decreased with increasing bed depth, which resulted in a broadened mass transfer zone. Higher uptake was observed at the highest bed depth due to an increase in the surface area of the biosorbent, SHBG, which provided more binding sites for the sorption (Han et al., 2007, 2008, 2009; Sivakumar and Palanisamy, 2009a,b; Singh et al., 2009; Tan et al., 2008).

3.7.1

3.7.1 Bed depth service time (BDST) model

BDST model was used to describe the fixed-bed column behaviour and to extend the laboratory experiments for industrial applications. The BDST model proposed by Hutchins (1973) states that the service time of a column can be related to a number of process variables. However, in the present studies the service time at which 50% breakthrough is achieved (t_1/2) is related to the height. The experimental data in column were fitted through BDST model (Eq. (1)). The plot is made between bed depth and the service time, t_1/2 and the plot is shown in Fig. 8. The straight line is observed in Fig. 8. The dynamic capacity, N₀ of SBP for CR was calculated from the slope of Fig. 8 and it was arranged in Table 8. The dynamic capacity of SHBG for removal of CR is 2244.33 mg ml⁻¹.

Close

Figure 8

Analysis of BDST model.

Export to PPT

Table 8 Analysis of results of BDST model.

Column height (cm)	Time taken for the first colour drop to emerge	N0 (mg L−1)	R2
3.5	3 h 25 min	2244.3	0.9917
7.0	6 h 5 min
10.0	9 h
12.5	12 h

3.8 Effect of regeneration

Recovery of adsorbate material (CR) as well as the regeneration of adsorbent is an important process in wastewater treatment.

The SHBG gets exhausted on continuous usage of the same for column studies. At this point, the concentration of influent (dye solution pouring into the column) is equal to the concentration of effluent (dye solution coming out from the column). Therefore, the recovery of the adsorbed material (dyes) as well as regeneration of the SHBG becomes quite necessary. Thermal regeneration is not possible in our case as we are using SHBG without any modification. Elution of dye with simultaneous chemical regeneration by a suitable solvent is a definite alternative to thermal regeneration and is tried with the help of acetone in these investigations.

The exhausted column of SHBG which was loaded with CR was regenerated with acetone. Acetone was added to the exhausted column with 10 ml increments for all the experiments till the regeneration was completed. In the first experiment, acetone was added with 0.6667 ml/min to the 3.5 cm height exhausted column of SHBG in which 50 mg L⁻¹ concentrated CR is passed. 120 ml of acetone is required for the completion of regeneration of this column and the total amount of CR was recovered in this case. In the second experiment, acetone is added with 0.6667 ml min⁻¹ to the 3.5 cm height exhausted column of SHBG in which 100 mg L⁻¹ concentrated CR was passed. 80 ml of acetone is required for the completion of regeneration of this column and the total amount of CR was recovered in this case also. In the third experiment, acetone was added with 0.3333 ml min⁻¹ to the 5.0 cm height exhausted column of SHBG in which 50 mg L⁻¹ concentrated CR was passed. 120 ml of acetone was required for the completion of regeneration of this column and the total amount of CR was recovered in this case. The regeneration curves are shown in Fig. 9. The results reveal that the regeneration of CR was almost the same for different flow rates. At the beginning more amount of CR was eluted due to more accumulation of CR.

Close

Figure 9

Desorption curve of CR from SHBG.

Export to PPT

After de-sorption, the columns of SHBG were washed with 100 ml of hot water in 10 ml fractions at the same flow rate which is maintained for the flow of dye. These columns are again loaded with the two different concentrations (50 mg L⁻¹ and 100 mg L⁻¹) of CR solution at the flow rate of 0.6667 ml min⁻¹ to check the adsorption efficiency of the SHBG. After regeneration the column is exhausted early. It is known from the t_1/2 values and the values are shown in Table 9. The data may be helpful in designing a fixed bed column for the treatment of dye of known concentrations.

3.9 Cost estimation

Basically, India is an agricultural country. The disposal of the solid agricultural wastes/by-product, SHBG is a big problem and it creates a lot of pollution problems. Therefore, using this waste material, SHBG for the useful wastewater treatment purpose is appreciable and these materials are/will be supplied free of cost by the agriculturists because they want to get rid of them. The cost involved is only for transportation and process of the collected material.

The selected agricultural waste/by-product, SHBG was procured free of cost or very less cost and the average and approximate cost of the materials is Rs. 25/- per ton. The cost which includes transport, process charges etc. of the final material which was used for the adsorption studies was Rs. 150/- per ton. The cost of final material was nothing when compared to the cost of activated carbon. Of course, the final cost estimation was also based on the removal capacity of different dyes and on regeneration. Even the regeneration of exhausted SHBG was not that much efficient, still the process was economical as compared to carbon because of low cost and abundant availability of SBP. The cost estimation of some agricultural wastes/by-products was available in the literature (Somasekhara Reddy et al., 2012; Asfour et al., 1985; Nassar et al., 1991; Hameed et al., 2008).