Adsorption characteristics of reactive dyes in columns of activated carbon

Adsorption characteristics of reactive dyes in columns of activated carbon

Journal of Hazardous Materials 165 (2009) 944–949 Contents lists available at ScienceDirect Journal of Hazardous Materials journal homepage: www.els...

350KB Sizes 2 Downloads 19 Views

Journal of Hazardous Materials 165 (2009) 944–949

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Adsorption characteristics of reactive dyes in columns of activated carbon Y.S. Al-Degs a,∗ , M.A.M. Khraisheh b , S.J. Allen c , M.N. Ahmad c a

Chemistry Department, The Hashemite University, P.O. Box 150459, Zarqa, Jordan Department of Civil and Environmental Engineering, University College of London, Gower Street, London WCIE 6BT, UK c The Queen’s University of Belfast, School of Chemical Engineering, David Keir Building, Stranmillis Road, BT 9 5AG Northern Ireland, UK b

a r t i c l e

i n f o

Article history: Received 25 June 2008 Received in revised form 19 October 2008 Accepted 22 October 2008 Available online 28 October 2008 Keywords: Activated carbon Reactive dyes Fixed-bed adsorber Mass transfer zone Bed-depth service time model

a b s t r a c t Adsorption behaviour of reactive dyes in fixed-bed adsorber was evaluated in this work. The characteristics of mass transfer zone (MTZ), where adsorption in column occurs, were affected by carbon bed depth and influent dye concentration. The working lifetime (tx ) of MTZ, the height of mass transfer zone (HMTZ), the rate of mass transfer zone (RMTZ), and the column capacity at exhaustion (qcolumn ) were estimated for the removal of remazol reactive yellow and remazol reactive black by carbon adsorber. The results showed that column capacity calculated at 90% of column exhaustion was lower than carbon capacity obtained from equilibrium studies. This indicated that the capacity of activated carbon was not fully utilized in the fixed-bed adsorber. The bed-depth service time model (BDST) was applied for analysis of reactive yellow adsorption in the column. The adsorption capacity of reactive yellow calculated at 50% breakthrough point (N0 ) was found to be 0.1 kg kg−1 and this value is equivalent to about 14% of the available carbon capacity. The results of this study indicated the applicability of fixed-bed adsorber for removing remazol reactive yellow from solution. © 2008 Elsevier B.V. All rights reserved.

1. Introduction Adsorption in column (or in fixed-bed adsorber) is the most common and efficient way for purification of wastewater. Before testing the performance of an adsorbent in fixed-bed adsorber, equilibrium isotherm studies should be conducted to measure the maximum capacity of that adsorbent. In previous studies the authors have reported that FS-400 (a commercially activated carbon) has a high adsorption capacity for removing remazol reactive dyes from solution [1,2]. Adsorption values for reactive dyes were in the range 200–1000 mg g−1 [1]. The high affinity of FS-400 toward reactive dyes was attributed to the unique chemical nature of FS400 [2,3]. Furthermore, part of the high carbon capacity for reactive dyes was attributed to the high porosity of the adsorbent [4]. Adsorption kinetics of remazol reactive dyes was found to be high at the early stages of adsorption, while a gradual uptake was noticed after 30 min from the start of the adsorption process [5]. Many investigations were reported for removing cationic dyes by column adsorption [6]. However, few reports on adsorption of remazol reactive dyes in column were reported [7,8]. More investigations are necessary because remazol reactive dyes have a wide indus-

∗ Corresponding author. E-mail address: [email protected] (Y.S. Al-Degs). 0304-3894/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2008.10.081

trial application than cationic dyes and consequently will have a high impact when discharged to the environment. Due to the high cost of commercially activated carbons and technical difficulties associated with pilot scale experiments, it is advisable to run small-scale column experiments before carrying out the high-cost pilot scale column experiments. It is worth to mention that smallscale column studies, generally, gave accurate prediction for dye removal from real wastewater systems [6]. In addition, it was found that small-scale experiments could predict (within an acceptable size of error) the adsorption in the expensive pilot scale column experiments [9]. In the design of adsorption columns for reactive dyes, the effect of many process parameters should be evaluated. Most column investigations usually considered the effect of solution flow rate, adsorbate concentration and adsorbent mass (or depth) [6,10,11]. In spite of the high adsorption of remazol reactive dyes by activated carbon, industrial application of this technology has apparently not been yet reported [12]. Particularly speaking, the high production and regeneration costs are the main reasons that retard the application of activated carbons for reactive dyes treatment on a wide scale. On the other hand, full-scale biologically activated carbon filters are under investigation to be applied for wastewater treatment [12]. In this research, the adsorption characteristics of remazol reactive dyes in the fixed-bed adsorber are investigated. MTZ characteristics are studied for remazol reactive black and yellow. BDST

Y.S. Al-Degs et al. / Journal of Hazardous Materials 165 (2009) 944–949

945

Nomenclature BDST bed-depth service time C0 inlet dye concentration (kg m−3 ) C0.02 (or Cb ) concentration of dye at 2% of exhaustion (kg m−3 ) C0.90 (or Cx ) concentration of dye at 90% of exhaustion (kg m−3 ) C0.95 concentration of dye at 95% of exhaustion (kg m−3 ) F volumetric flow rate (m3 min−1 ) H carbon bed-depth (m) HMTZ height of mass transfer zone (m) ka BDST adsorption rate constant (m3 kg−1 min−1 ) MTZ mass transfer zone column adsorption capacity (kg m−3 or kg kg−1 ) N0 qcolumn column adsorption capacity (kg kg−1 ) RMTZ rate of mass transfer zone (m min−1 ) time at 50% breakthrough (min) t0.5 tx total time involved for the establishment of MTZ (min) tf time of initial formation of MTZ (min) time required for MTZ movement downward the tı column (min) u linear flow rate (m min−1 ) V0.95 influent volume corresponding to C0.95 (m3 ) Vb (or V0.02 ) influent volume corresponding to C0.02 (m3 ) Vx (or V0.90 ) influent volume corresponding to C0.90 (m3 )

model, which offers a simple approach and rapid prediction of adsorber performance, is applied for modelling adsorption of reactive yellow in activated carbon column.

Fig. 1. Ideal breakthrough curve [9].

and is calculated as [13]: HMTZ tı = H tx − tf

The more rapid the kinetics of adsorption, the shallower is the HMTZ. tı , tf , and H are the times required to move MTZ down the adsorber, the time required for the initial formation of MTZ, and the carbon bed-depth, respectively. The rate of movement of the mass transfer zone (RMTZ) down the column is highly dependent on the adsorption extent of the adsorbate on the adsorbent. This term, RMTZ, is an important measure because it indicates the rate at which the adsorbent will be exhausted and is calculated from Eq. (3) [14]: RMTZ =

2. Theoretical background 2.1. Mass transfer zone and breakthrough curve MTZ is formed at the front of the column where adsorption takes place. The depth of MTZ is controlled by many factors like the nature of adsorbate, characteristics of adsorbent, mass (or bed-depth) of adsorber, particle size of adsorbent, adsorbate inlet concentration, solution pH, and solution flow rate [7,8]. Among these variables, bed-depth, solute concentration and flow rate are considerably affecting the lifetime of the column. Once formed, MTZ moves down through the adsorbent bed until it reaches the adsorber end, where the effluent concentration of solute begins to rise in the aqueous phase [13]. The typical breakthrough curve is usually expressed by plotting Ceffluent or Ceffluent /Cinfluent versus treated volume V or service time t. Fig. 1 depicts a typical breakthrough curve where the column capacity is fully utilized. The concentration at breakthrough point is chosen arbitrarily at some low value, Cb . When the effluent concentration Cx is approaching to 90% of C0 (inlet adsorbate concentration) then the adsorbent is considered to be essentially exhausted [9,13]. MTZ in the fixed-bed adsorber is the portion of the curve between Cx and Cb which is assumed to have a constant depth. The time tx needed for MTZ to establish itself and moves down to the end of carbon bed-depth is calculated from Eq. (1) [11]: tx =

Vx F

(1)

where F is the volumetric flow rate (m3 min−1 ). Vx is the influent volume corresponding to Cx . The height of mass transfer zone HMTZ is a measure of the rate of the removal of adsorbate by the column

(2)

HMTZ × F V0.9 − V0.02

(3)

where F, V0.9 , and V0.02 are the volumetric flow rates of feeding dye solution, the treated volume corresponding to C0.9 and the treated volume corresponding to C0.02 , respectively. The capacity at exhaustion is determined by calculating the total area below the breakthrough curve. The column capacity is estimated as follows [13]:

 V0.9

−1

qcolumn (kg kg

)=

V0.02

(C0.9 − C0.02 ) dV mass

(4)

where C0.9 , C0.02 , V0.9 , and V0.02 have the same meaning as described earlier. 2.2. Bed-depth service time model (BDST) It is generally accepted that BDST offers the simplest approach and rapid prediction of adsorber design and performance [6,7,15]. The basic relations that relating C0 /Cb and column service time (t) for purification in a flowing system were originally proposed by Bohart and Adams [16]. The original BDST theory was developed for the removal of chlorine gas by charcoal column [16]. Although the original work by Adams–Bohart was carried out for gas–charcoal adsorption system, its overall approach was successfully applied in quantitative description of many adsorption systems including dye-activated carbon systems [6,7,13,15]. This model assumes that the adsorption rate is proportional to both the residual capacity of activated carbon and the concentration of the adsorbing solute. Adams–Bohart model is presented as [16]: ln

C

0

Cb



−1

= ln(eka

N0 H/u

− 1) − ka C0 t

(5)

946

Y.S. Al-Degs et al. / Journal of Hazardous Materials 165 (2009) 944–949

where, C0 (kg m−3 ) is the inlet dye concentration. Cb is the maximum acceptable limit concentration (kg m−3 ). t is the fix-bed adsorber service time (min). H is the bed-depth (m). N0 is the column adsorption capacity (kg m−3 ). u is the linear volumetric flow rate (m min−1 ) of dye solution. ka is the BDST rate constant (m3 kg−1 min−1 ). Because the exponential term in Eq. (5) is usually much larger than unity, then Eq. (5) is reduced to: t=

N0 1 ln H− C0 u ka C0

C

0

Cb



−1

(6)

A plot of service time t against H, should generate a straight line with slope equal to (N0 /C0 u) and intercept of (−(1/kaC0 )ln((C0 /Cb )−1)). From the slope and intercept, both N0 and ka are calculated. At 50% breakthrough (C0 /Cb = 1/2), the second term on the right-hand side of Eq. (6) is reduced to zero and generating the following simplified equation [17]:

Fig. 2. Remazol reactive black (anionic dye).

3. Materials and experimental methods

placed in a thermostated shaker (L.H. Engineering Co. Ltd., England) for 3 weeks. The equilibrium time (3 weeks) was determined from earlier kinetic studies [1]. After equilibrium, the solutions were filtered and the remaining free dyes were estimated using predetermined calibration graphs for each dye. Absorbances of dye solution were measured using double-beam spectrophotometer (PerkinElmer UV/Vis, Lambda 12 Model, Germany). The surface concentrations of dyes were estimated from the mass balance equation. Blank solutions, containing no adsorbent, were also involved in the study. Adsorption isotherms were repeated in duplicate and the average values were reported.

3.1. Adsorbent

3.4. Column studies

The commercially activated carbon used in this research was Filtrasorb 400 (FS-400) and it was purchased from Chemviron Carbon, UK. FS-400 was selected due to its high adsorption for remazol reactive dyes from solution [1–5]. The adsorbent was prepared from bituminous coal and it has the following properties: specific surface area 1100 m2 g−1 , bulk density 407 kg m−3 , methylene blue value 500 mg g−1 , and iodine number 1050 mg g−1 . The equilibrium and column studies were conducted using activated carbon of particle diameter 600–710 ␮m.

Reactive dyes are representing 30% of the dyes used worldwide [1,2,4]. Their consumption rate was increased by about 15% per year since 1980. They are characterized by nitrogen-to-nitrogen double bounds and aromatic sulphonated groups. Two reactive dyes of high industrial application were studied: Remazol reactive yellow and remazol reactive black. The working solutions of dyes were prepared from their standard concentrated solutions as received from the manufacturer (Bayer, Frankfurt, Germany). The employed dyes were completely soluble in water and showed a moderate pH values (5–7) in distilled water at 0.1 kg dm−3 concentration. The wavelengths of maximum absorption of dyes were: 413, and 595 nm for reactive yellow and reactive black, respectively. Due to commercial reasons, the suppliers provided the chemical structure of reactive black only (Fig. 2) and did not provide the structure of the remazol reactive yellow.

Traditionally, adsorption isotherms have been used for testing the overall performance of the adsorbent. But in practice the final technical systems normally use column-type operation. Moreover, simple isotherms cannot give accurate scaleup data in a fixed-bed system. Small-scale column tests were carried out to evaluate the capacity of FS-400 for removing reactive dyes from water under continuous flow conditions. The employed column apparatus was based on the scaling approach developed by Crittenden et al. [18] to simulate the performance of the adsorber in the large-scale adsorption units. A glass column (30 cm × 2.5 cm) was filled with activated carbon (particle size: 600–710 ␮m) on a glass support covered with glass wool. To avoid entrapping of air bubbles inside carbon particles, the particles were soaked in appropriate amount of water and agitated until no air bubbles were detected in the solution. The column was loaded with appropriate solution which percolated downward, at a linear flow rate of 1.0 m min−1 (equivalent to 5.0 × 10−6 m3 min−1 volumetric flow rate) using a pressureadjustable peristaltic pump (Smith and Nephew Watson-Marlow, England). Dye concentration in the effluent solution was periodically measured as outlined in Section 3.3. The column operation was terminated when 90% or 95% of the capacity was used up. Effect of carbon mass (or bed-depth) on yellow dye adsorption was studied at inlet dye concentration 0.19 kg m−3 and linear flow rate 1.0 m min−1 . Effect of inlet dye concentration was studied for black dye at carbon mass 5.0 g (equivalent to 7.0 cm bed-depth) and linear flow rate of 1.0 m min−1 .

3.3. Determination of adsorption isotherms of dyes

4. Results and discussion

t0.5 =

N0 H C0 u

(7)

A plot of t0.5 (service time at 50% breakthrough) versus H should yield a straight line with a slope of (N0 /C0 u) and this slope represents the time required to exhaust a unit length of the adsorber under the applied experimental variables.

3.2. Remazol reactive dyes

Adsorption isotherms (at 20 ± 2 ◦ C) for each dye were determined individually using concentration–variation method. Samples of 0.050 g (±0.001 g) of activated carbon (particle range 600–710 ␮m) were added to 100 cm3 solutions of pH 5 containing different concentrations of dyes. For both systems, the initial dye concentration was between 0 and 0.1 kg m−3 . The adsorption solutions (40 solutions for both dyes) were tightly sealed and directly

4.1. Adsorption isotherms Adsorption data were modelled using popular Langmuir isotherm [13]: qe =

bQmax Ce 1 + bCe

(8)

Y.S. Al-Degs et al. / Journal of Hazardous Materials 165 (2009) 944–949

Fig. 3. Adsorption isotherms of reactive yellow and reactive black. Mass of adsorbent: 0.050 g; volume of solution: 50.0 cm3 ; pH: 5; particle diameter: 600–710 ␮m; temperature: 20 ◦ C.

where Ce , qe , Qmax , and b are the equilibrium concentrations of remaining dye in solution (kg m−3 ), the amount of dye adsorbed per mass unit of adsorbent at equilibrium (kg kg−1 ), the amount of adsorbate at complete monolayer coverage (kg kg−1 ) and the Langmuir constant (m3 kg−1 ), respectively. Adsorption isotherms showed a Langmurian shape, i.e. initial slope at the beginning and a plateau covering most experimental points (Fig. 3). The values of Qmax were 0.73 and 0.294 kg kg−1 for reactive yellow and reactive black, respectively. Langmuir constants were 160 and 50 m3 kg−1 for reactive yellow and reactive black, respectively. The isotherm results indicated that yellow dye has a higher adsorption extent (by 2.5-fold) compared to black dye. 4.2. Effect of carbon mass (or bed-depth) on MTZ characteristics Effect of carbon mass on MTZ characteristics was studied for reactive yellow dye. The breakthrough curves (at different masses or bed-depths) are shown in Fig. 4. Breakthrough at 50% was indicated by a horizontal line in the figure. It is obvious in Fig. 4 that the adsorber capacity was increased at higher carbon masses. The treated volume at 50% breakthrough has increased by 6.8-fold when the adsorber mass increased from 2.0 to 8.0 g which corresponds to 0.03–0.12 m bed-depth. This behaviour was expected since at higher carbon masses more adsorption sites are available for dyes. In fact, the breakthrough curves shown in Fig. 4 were not followed the typical “S-shape” curves that produced in ideal adsorption systems (Fig. 1). The deformed breakthrough curves were obtained

Fig. 4. Breakthrough curves of reactive yellow adsorption on different carbon beddepths (or masses). C0 : 0.19 kg m−3 ; flow rate: 5.0 × 10−6 m3 min−1 ; pH: 5; particle diameter: 600–710 ␮m; temperature: 20.0 ◦ C.

947

due to two factors: (a) the slow adsorption kinetics of reactive dyes on porous FS-400 [5], where the slow kinetics of reactive dyes makes the breakthrough faster and consequently an incomplete “S” breakthrough shape is produced, and (b) the use of small-scale column apparatus usually produces premature breakthrough curves as reported in the literature [18]. Similar breakthrough curves were reported for adsorption of tectilon blue 4R-01, tectilon red 2B, and tectilon orange 3G by FS-400 adsorber [8]. To have a better insight into the effect of carbon mass (or carbon bed-depth) on reactive yellow removal by adsorber, the characteristics of MTZ were determined. The MTZ has its own characteristics which are highly dependent on the nature of the adsorbent–adsorbate interactions and to a lesser extent on the other operating conditions such as, carbon bed-depth, inlet dye concentration and volumetric flow rate [11]. HMTZ, RMTZ, qcolumn values and the other related parameters were obtained from breakthrough curves and compiled in Table 1. As indicated in Table 1, the bed-depth has a high effect on the column performance as inferred from the values of V0.90 and V0.02 . Both V0.90 and V0.02 were significantly increased with bed-depth. At 90% exhaustion, the treated volume (V0.9 ) is increased by 1.7fold when the adsorber bed-depth increased from 0.03 to 0.09 m. However, 15-fold was the increment in the treated volume at 2% exhaustion. The time required to form MTZ, tf , has been greatly increased over the studied bed-depth range. It takes 150 min for MTZ to form at 0.09 m bed-depth, while MTZ was formed just after 10 min at 0.03 m bed-depth. The earlier formation of MTZ reflected the earlier exhaustion of the adsorber. Gupta et al. [11] have reported much higher tf values for adsorption of phenolic pollutants in a small-scale activated carbon adsorber and this was attributed to the high adsorption kinetics of the system. The fast breakthrough reported in the current study is, in fact, expected because remazol reactive dyes exhibited slow adsorption kinetics. Adsorption kinetic studies revealed that only 20% of the carbon equilibrium capacity was utilized after 6 h of mixing between activated carbon and reactive dyes [5]. The values of tx (time to reach 90% exhaustion) were also increased with bed-depth and this was expected because the lifetime of the adsorption zone is supposed to increase with carbon bed-depth (or carbon mass). Table 1 indicated that HMTZ was increased with carbon bed-depth. The HMTZ has been increased from 9.9 × 10−3 to 2.79 × 10−2 m over the studied bed-depth range (0.03–0.09 m). In each case, the magnitude of HMTZ was equivalent to about onethird of the total bed-depth. The increase in HMTZ was expected because reactive dyes have slow adsorption kinetics on FS-400, therefore, their adsorption zone will increase at higher bed-depths. Zogorski and Faust [14] reported an independent relationship between HMTZ and carbon bed-depth for adsorption of small (compare to dyes) organic solutes. In a different study, Gupta et al. [11] reported HMTZ values which were about half of the bed-depth. The RMTZ is reflected the affinity of reactive dyes for activated carbon and predicted the exhaustion rate of adsorber. As noted in Table 1, RMTZ values were decreased by increasing bed-depth of adsorber. RMTZ was decreased from 7.75 × 10−6 to 1.19 × 10−6 m min−1 by increasing the bed-depth from 0.03 to 0.09 m. An independent relationship was observed between RMTZ and carbon bed-depth for adsorption of phenolic compounds by activated carbon adsorber [14]. Adsorption values of reactive yellow in the adsorber, qcolumn , given in Table 1 were relatively high (0.29–0.54 kg kg−1 ) when compared to adsorption values reported for similar compounds [8]. However, these capacities are still lower than the maximum equilibrium capacity (0.73 kg kg−1 ). The lower adsorption in the column was attributed to the lower contact time between adsorber particles and adsorbate molecules. From 40% to 75% of the equilibrium

948

Y.S. Al-Degs et al. / Journal of Hazardous Materials 165 (2009) 944–949

Table 1 Effect of bed-depth on MTZ of reactive yellow adsorption in columna . Mass (×10−3 kg)

H (×10−2 m)

Vx(0.9) (×10−3 m3 )

Vb(0.02) (×10−3 m3 )

Cx(0.9) (kg/m3 )

Cx(0.02) (kg/m3 )

tx (h)

tf (min)

HMTZ (×10−2 m)

RMTZ (×10−6 m min−1 )

qcolumn (kg/kg)

2.0 4.0 6.0

3.0 6.0 9.0

6.45 8.40 10.95

0.05 0.15 0.75

0.170 0.170 0.170

0.003 0.004 0.004

21.5 28.0 36.5

10 30 150

0.99 1.96 2.79

7.75 1.37 1.19

0.541 0.349 0.287

a

Experimental conditions: flow rate: 5 × 10−6 m3 min−1 ; dye concentration: 0.19 kg m−3 ; particle size: 600–710 ␮m; pH: 5.0; temperature: 22 ◦ C.

Fig. 5. Breakthrough curves of reactive black on activated carbon at different inlet concentrations. Flow rate: 5.0 × 10−6 m3 min−1 ; carbon mass: 5.0 g (7.5 cm beddepth); pH: 5; particle diameter: 600–710 ␮m; temperature: 20.0 ◦ C.

capacity was utilized in column operation. In related studies, less than 10% of adsorbent capacity was utilized in the fixed-bed adsorber [19,20]. The modest adsorption of yellow dye in adsorber is due to slow adsorption kinetics as mentioned earlier [5]. In contrast, Gupta et al. [21] have reported a higher adsorption value in column compare to equilibrium value for malachite green dye (a cationic dye) by activated carbon. 4.3. Effect of inlet dye concentration on MTZ characteristics Fig. 5 depicts three breakthrough curves of reactive black removal by activated carbon adsorber at different inlet dye concentrations, 0.075 m bed-depth, and flow rate of 5.0 × 10−6 m3 min−1 . At 50% breakthrough point, the treated volume was increased by decreasing inlet dye concentration. The treated volume has been increased from 1.05 to 3.5 dm3 over the studied concentration range. This indicated that a better column performance was obtained at lower inlet dye concentration and the reason for this behaviour will be discussed soon. Adsorption parameters for reactive black were summarised in Table 2. As indicated in Table 2, treated volumes at 2.0% and 90% exhaustion were increased (with different magnitudes) by decreasing inlet dye concentration. The increment in V0.02 was 300%, while V0.9 has increased by 44%. This behaviour was expected because at lower dye concentration the

competition for adsorption was less, accordingly, higher uptake and larger treated volume were obtained. As Table 2 indicated, a small effect on tf was observed over the studied concentration domain, tf value was increased from 5 to 20 min by decreasing dye concentration from 0.115 to 0.025 kg dm−3 . The total service time of the adsorber, tx , is affected by the inlet dye concentration, more service time was obtained at lower dye concentration. The service time at flow rate of 5.0 × 10−6 m3 min−1 and bed-depth of 0.075 m has been increased from 19 to 27.3 h over the indicated concentration range in Table 2. As a main conclusion from Tables 1 and 2, a better column performance was obtained at higher bed-depth and lower inlet dye concentration. At these conditions, longer service times and larger treated volumes were obtained. The values of HTMZ also indicated that a better column performance was obtained at lower dye inlet concentration, the value of HMTZ was reduced from 2.22 × 10−2 to 1.83 × 10−2 m when the dye concentration lowered from 0.115 to 0.025 kg m−3 . In all cases, the HMTZ was presented about 30% of the bed-depth. As was the case in HMTZ, RMTZ was also decreased (but with higher extent) by decreasing inlet concentration which also proved the high column performance at lower dye concentration. qcolumn values for black dye were increased with inlet dye concentration which was expected as more adsorption is accomplished at higher inlet dye concentration. Furthermore, at higher surface loadings the column performance was decreased and this is concluded by comparing the service times and V0.9 values of the adsorber at different inlet concentrations. The lowest service time and V0.9 values were obtained at the highest inlet dye concentration (0.115 kg m−3 ). As reported for reactive yellow, the equilibrium capacity of the carbon was not fully utilized in the column study, from 12% to 36% of the maximum capacity was achieved under the studied experimental conditions. Again, the incomplete utilization of the carbon capacity (0.294 kg kg−1 ) is mainly attributed to the short contact time between dye solution and the adsorber. 4.4. Analysis of reactive yellow adsorption in column by BDST model In this section, the adsorption data of reactive yellow were analysed using BDST model represented by Eq. (7). The previous analyses (Sections 4.2 and 4.3) were essential for characterization of reactive dye removal by the carbon adsorber, however, BDST analysis was necessary to give a clear picture for the practical application of FS-400 in fixed-bed adsorber. As mentioned earlier, BDST model was tested in many column studies [5,13,17]. In the current BDST analysis, the target parameter is the slope of Eq. (7), which estimates

Table 2 Effect of inlet dye concentration on MTZ of reactive black adsorption in columna . C0 (kg/m3 )

H (×10−2 m)

Vx(0.9) (×10−3 m3 )

Vb(0.02) (×10−3 m3 )

Cx(0.9) (kg/m3 )

Cx(0.02) (kg/m3 )

tx (h)

tf (min)

HMTZ (×10−2 m)

RMTZ (×10−5 m min−1 )

qcolumn (kg/kg)

0.115 0.055 0.025

7.5 7.5 7.5

5.7 7.1 8.2

0.025 0.05 0.10

0.103 0.049 0.023

0.007 0.004 0.001

19.0 23.7 27.3

5 10 20

2.22 1.94 1.83

1.95 1.37 1.13

0.107 0.061 0.036

a

Experimental conditions: flow rate: 5 × 10−6 m3 min−1 ; particle size: 600–710 ␮m; pH: 5.0; temperature: 22 ◦ C.

Y.S. Al-Degs et al. / Journal of Hazardous Materials 165 (2009) 944–949

the time required to exhaust a unit length of the fixed-bed at 50% breakthrough under test conditions. Analysis of adsorption data of reactive yellow by BDST model generated the following linear equation: t0.5 = 21,000H − 375 (r2 = 0.982). As shown, a linear relationship was obtained with a fair r2 value, but the line does not pass through the origin. It has been reported that this type of deviation was caused by more than one rate-limiting step in the adsorption process [22,23]. The slope of the straight line is 21,000 min m−1 which indicates that 3.5 h are needed to exhaust 1.0 cm of carbon bed during adsorption of reactive yellow at 50% breakthrough. The BDST adsorption capacity, N0 , is an important constant for evaluating the adsorber efficiency, efficient adsorbers are characterized by high N0 values. The magnitude of N0 is 39.9 kg m−3 , which was simply obtained from the slope of BDST plot (21,000 min m−1 ), C0 (inlet dye concentration, 0.19 kg m−3 ) and u (linear flow rate, 1 × 10−2 m min−1 ) as described in Section 2.2. Using the bulk density of the adsorbent (407 kg m−3 ), the value of N0 in kg kg−1 units was 0.1. This value indicated that only 14% of the available carbon capacity (0.73 kg kg−1 ) was utilized in the adsorber at 50% breakthrough point. Generally speaking, the reported performance of FS-400 for dye removal in column is promising when compared to the other related studies [8,19,20]. 5. Conclusions The results of this work indicated a good removal of problematic reactive dyes by fixed-bed activated carbon column. The characteristics of MTZ of reactive yellow and black adsorption indicated that a better column adsorber is achieved at longer bed-depth and diluted inlet dye concentration. A service time of 36.5 h (at 90% exhaustion) was reported for reactive yellow adsorption at carbon bed-depth 0.09 m, inlet dye concentration 0.19 kg m−3 , flow rate 5.0 × 10−6 m3 min−1 , pH 5, and particle diameter 600–710 ␮m. Analysis of reactive yellow adsorption data by BDST model revealed that 3.5 h are needed to exhaust 1.0 cm of carbon bed-depth at 50% breakthrough point. N0 (BDST column capacity) of reactive yellow was found to represent about 14% of the equilibrium capacity of the adsorber, this percentage is high and promising when compared to the other related adsorption studies. References [1] Y.S. Al-Degs, M.A. Khraisheh, S.J. Allen, M. Ahmad, Effect of carbon surface chemistry on the removal of reactive dyes from textile effluents, Water Res. 34 (34) (2000) 927–935.

949

[2] Y.S. Al-Degs, M.A. Khraisheh, S.J. Allen, M. Ahmad, Sorption behaviour of cationic and anionic dyes from aqueous solution on different types of activated carbons, Sep. Sci. Technol. 36 (2001) 91–102. [3] Y.S. Al-Degs, M.A. Khraisheh, S.J. Allen, M. Ahmad, Adsorption of remazol reactive black B on activated carbon: adsorption on H and L carbon, Adv. Environ. Res. 3 (1999) 132–138. [4] Y.S. Al-Degs, M.I. El-Barghouthi, M.A. Khraisheh, M.N. Ahmad, S.J. Allen, Effect of surface area, micropores, secondary micropores and mesopores volumes of activated carbons on reactive dyes adsorption from solution, Sep. Sci. Technol. 39 (2004) 97–111. [5] M.A. Khraisheh, Y.S. Al-Degs, S. Allen, M. Ahmad, Elucidation of the controlling steps of reactive dyes adsorption on activated carbon, Ind. Eng. Chem. Res. 41 (2002) 1651–1657. [6] G. McKay, M.J. Bion, Simplified optimisation for fixed bed adsorption systems, Water Air Soil Poll. 51 (1990) 33–41. [7] G.M. Walker, L.R. Weatherley, Adsorption of acid dyes on to granular activated carbon in fixed beds, Water Res. 31 (1997) 2093–2101. [8] G.M. Walker, L.R. Weatherley, Textile wastewater treatment using granular activated carbon adsorption in fixed beds, Sep. Sci. Technol. 39 (2000) 1329–1341. [9] M.L. Bao, O. Griffini, D. Santianni, K. Barbieri, D. Burrini, F. Pantani, Removal of bromate ion from water using granular activated carbon, Water Res. 33 (1999) 2959–2970. [10] V.K. Gupta, Equilibrium uptake, sorption dynamics, process development, and column operations for the removal of copper and nickel from aqueous solution and wastewater using activated slag, a low-cost adsorbent, Ind. Eng. Chem. Res. 37 (1998) 192–202. [11] V.K. Gupta, S.K. Srivastava, R. Tyagi, Design parameters for the treatment of phenolic wastes by carbon columns (obtained from fertilizer waste material), Water Res. 34 (2000) 1543–1550. [12] T. Robinson, G. McMullan, R. Marchant, P. Nigam, Remediation of dyes in textile effluent: a critical review on current treatment technologies with proposed alternatives, Bioresour. Technol. 77 (2001) 247–255. [13] S.D. Faust, O.M. Aly, Adsorption Processes for Water Treatment, Butterworth Publishers, 1987. [14] J.S. Zogorski, S.D. Faust, Water-1976. I. Physical, Chemical, Wastewater Treatment. Symp. Series 166.73,54, American Institute of Chemical Engineering, New York, 1977. [15] V.K. Lee, J.F. Porter, G. McKay, Development of fixed-bed adsorber correlation models, Ind. Eng. Chem. Res. 39 (2000) 2427–2433. [16] G.S. Bohart, E.Q. Adams, Some aspects of the behaviour of charcoal with respect of to chlorine, J. Am. Chem. Soc. 42 (1920) 523–544. [17] C.K. Lee, K.S. Low, S.L. Chew, Removal of anionic dyes by water hyacinth roots, Adv. Environ. Res. 3 (1999) 343–351. [18] J.C. Crittenden, J.K. Berrigan, W.H. David, Design of rapid small-scale adsorption tests for a constant diffusivity, J. Water Pollut. Control Fed. 58 (1986) 312–320. [19] J.A. Laszlo, Removing acid dyes from textile wastewater using biomass for decolorization, Am. Dyest. Rep. 83 (1994) 17–21. [20] B. Smith, T. Koonce, S. Hudson, Decolorizing dye wastewater using chitosan, Am. Dyest. Rep. 82 (1993) 18–36. [21] V.K. Gupta, S. Srivastava, D. Mohan, Equilibrium uptake, sorption dynamics, process optimization, and column operations for the removal and recovery of malachite green from wastewater using activated carbon and activated slag, Ind. Eng. Chem. Res. 36 (1997) 2207–2218. [22] D.C. Sharma, C.F. Forster, Column studies into the adsorption of chromium(VI) using sphagnum-moss peat, Bioresour. Technol. 52 (1995) 261–267. [23] K. Low, C. Lee, B. Tan, Quaternized wood as sorbent for reactive dyes, Appl. Biochem. Biotechnol. 87 (2000) 233–245.