Me

NF

a

ARRA

KCEMCF

1

wplaecerIbd

eimt

1d

Separation and Purification Technology 63 (2008) 213–219

Contents lists available at ScienceDirect

Separation and Purification Technology

journa l homepage: www.e lsev ier .com/ locate /seppur

odeling fixed bed column for cadmium removal fromlectroplating wastewater

alini Sankararamakrishnan ∗, Pramod Kumar, Vivek Singh Chauhanacility for Ecological and Analytical Testing, 302 Southern Laboratories, Indian Institute of Technology, Kanpur, U.P. 208016, India

r t i c l e i n f o

rticle history:eceived 1 October 2007eceived in revised form 17 April 2008ccepted 5 May 2008

eywords:admiumlectroplating wastewaterodeling

hitosan

a b s t r a c t

Removal of cadmium by xanthated chitosan was investigated in a packed bed up-flow column. The exper-iments were conducted to study the effect of important design parameters such as bed height and flowrate. At a bed height of 9 cm and flow rate of 3 ml min−1, the metal-uptake capacity of xanthated chitosanand plain chitosan flakes for cadmium was found to be 132.3 ± 1.5 and 40.1 ± 0.5 mg g−1 respectively.The bed depth service time (BDST) model was used to analyze the experimental data. The computedsorption capacity per unit bed volume (N0) was 2.19 and 14.6 g l−1 for plain and xanthated flakes respec-tively. The rate constant (Ka) was recorded as 0.5514 and 0.0418 l mg−1 h−1 for plain and xanthated chitosanrespectively. In flow rate experiments, the results confirmed that the metal-uptake capacity and the metalremoval efficiency of plain and xanthated chitosan decreased with increasing flow rate. The Thomas model

ixed bed was used to fit the column sorption data at different flow rates and model constants were evaluated.The column regeneration studies were carried out for two sorption–desorption cycles. The eluant usedfor the regeneration of the sorbent was 0.01 N H2SO4. A decreased breakthrough time and an increasedexhaustion time were observed as the regeneration cycle progressed, which also resulted in a broadenedmass transfer zone. The column was successfully applied for the removal of cadmium from electroplatingwastewater. Three hundred sixty-seven bed volumes of electroplating wastewater were treated in column

sorbe

ccitthnamttawtd

experiments using this ad

. Introduction

Cadmium, like other heavy metals, is released into naturalaters by industrial and domestic wastewater discharges. In com-arison with other industries, the electrochemical industries uses

ess water, hence, the volume of wastewater produced is smaller,nd the wastewater is highly toxic in nature because of the pres-nce of high concentrations of metals such as copper, nickel, zinc,admium and cyanides. The use of cadmium cyanide baths in thelectroplating industry generates a strong concern related to envi-onmental impacts due to high cadmium and cyanide toxicity [1].n humans, cadmium is accumulated in the kidneys, which willegin to malfunction at overdoses spilling proteins in the urine andisrupting protein metabolism [1].

Methods proposed for Cd removal from wastewaters are those

mployed for most heavy metals, these include precipitation,on exchange resins, vacuum evaporation, solvent extraction and

embrane technologies [2,3]. Among them, precipitation is thereatment most widely applied due to its simple equipment,

∗ Corresponding author. Tel.: +91 512 2597844; fax: +91 512 2597844.E-mail address: nalini@iitk.ac.in (N. Sankararamakrishnan).

fti[a

bw

383-5866/$ – see front matter © 2008 Elsevier B.V. All rights reserved.oi:10.1016/j.seppur.2008.05.002

nt, reducing the concentrations of Cd(II) from 10 to 0.1 mg l−1.© 2008 Elsevier B.V. All rights reserved.

apability to treat large volumes of water containing high metaloncentrations and low cost. Nevertheless, precipitation is oftenncapable of meeting the legislation requirements for wastewaterso be discharged to surface waters and sewage (0.1 mg l−1). Fur-hermore, this treatment produces large amounts of sludge hard toandle. The other existing methods, although usually effective, areot free of drawbacks, high capital cost with recurring expensesnd capability to treat only small volumes, are the most com-on. Economical, practical and efficient alternative techniques are,

herefore, required to purify Cd-containing wastewaters. New prac-ices have been focused on the study of processes based on sorptionpproaches. Sorption appears as a simple and low-cost method,ith a great potential of becoming an actual alternative to conven-

ional ones, overcoming the problems of insufficient efficiency andifficult waste handling derived from the precipitation method. Dif-erent materials have been subject of study in this regard, most ofhem can be classified in two groups, minerals, such as iron hydrox-des and oxyhydroxides [4–6], aluminium hydroxides and oxides

7,8], and bio-materials, such as agricultural by-products [9–11],lgae [12], fungi [13] and chitosan based sorbents [14,15].

Chitosan has undoubtedly been one of the most popular adsor-ents for the removal of metal ions from aqueous solution and isidely used in wastewater treatment applications [15]. Although

214 N. Sankararamakrishnan et al. / Separation and P

Nomenclature

C0 inlet Cd(II) concentration (mg l−1)Cb outlet Cd(II) Concentration (mg l−1)E elution efficiency (%)F volumetric flow rate (l h−1)Ka bed depth service model rate constant (l mg−1 h−1)kTh Thomas model constant (l mg−1 h−1)M Sorbent mass (g)mad amount of Cd(II) adsorbed in the column (mg g−1)md amount of Cd(II) desorbed from the column (g)mtotal total amount of Cd(II) sent to column (g)N0 sorption capacity of bed (mg l−1)t service time (h)tb time at which Cd(II) concentration in the effluent

reached 0.1 mg l−1 (h)te time at which Cd(II) concentration in the effluent

reached 9.9 mg l−1 (h)�t mass transfer zone (h)ttotal total flow time (h)� linear flow rate (cm h−1)V throughput volume (l)Veff volume of the effluent (ml)Q0 maximum solid-phase concentration of solute

(mg g−1)

tomtatumamuifel

mciwdsoa

2

2

alawu

um

2

ce0gsCNftfaamacS

2

cb6bdip(ou4ctattTcrtrwFs

2

twsiTf

Q uptake capacity (mg g−1)Z bed height (cm)

here have been few studies on the use of chitosan for the removalf cadmium [16,17] the studies have been carried out only in batchode, which are usually limited to the treatment of small quanti-

ies of wastewater. The sorption capacity parameter obtained frombatch experiment is useful in providing information about effec-

iveness of metal–biosorbent system. However, the data obtainednder batch conditions are generally not applicable to most treat-ent systems where contact time is not sufficiently long for the

ttainment of equilibrium [11]. The practical application of heavyetal sorption is most effectively carried out in a packed bed col-

mn, as it efficiently utilize the sorbent capacity and results in anmproved quality of the effluent [18]. Studies have been reportedor the removal of cadmium in a packed bed column [11,19]. How-ver, the adsorbent was not applied to real industrial waster waterike cyanide rich electroplating effluent.

In a recent study, the efficiency of xanthated chitosan for cad-ium removal was demonstrated in the batch mode (adsorption

apacity 325 mg g−1) [20]. In the present investigation, the abil-ty of xanthated chitosan to remove cadmium from electroplatingastewater in a packed up-flow column was evaluated. Effects ofesign parameters, such as bed height and flow rate, on cadmiumorption have been examined. In addition, the removal capacityf cadmium in real cyanide rich electroplating effluent for twodsorption–desorption cycles have also been investigated.

. Materials and methods

.1. Materials

Chitosan flakes were acquired from India Sea foods, Cochin, India

nd used without any further purification. The degree of deacety-ation was reported to be 88% by the manufacturer. Glutaraldehydend carbondisulfide were purchased from Sigma–Aldrich and usedithout further purification. Stock solution of Cd(II) was preparedsing Cd(NO3)2·4H2O (BDH chemicals). All the inorganic chemicals

2

c

urification Technology 63 (2008) 213–219

sed were analar grade and all reagents were prepared in Milliporeilli-Q deionised water.

.2. Chemical modification of the chitosan flakes

Chitosan flakes were cross linked with glutaraldehyde andhemically modified (Scheme 1) and characterized as describedarlier [21]. To obtain 20% crosslinking [22] chitosan flakes (ca..5 g) were suspended in methanol (100 ml), and a 25% aqueouslutaraldehyde solution (0.046 ml, 0.12 mmol) was added. Aftertirring at room temperature for 6 h, the product was filtered.rosslinked chitosan flakes (0.5 g) were treated with 25 ml of 14%aOH and 1 ml of CS2. The mixture was stirred at room temperature

or 24 h. The obtained orange product, crosslinked xanthated chi-osan were washed thoroughly with water, air dried and used forurther experiments. The physical characteristics of the prepareddsorbent are given in Table 1. Energy dispersive X-ray analysis isuseful tool to identify within short analysis time the kinds of ele-ent contained in solid specimen. The surface morphology of the

dsorbent was obtained from the SEM studies and the elementalomposition from energy dispersive analysis by X-ray (EDAX) byEM-FEI Quanta 200 instrument.

.3. Sorption–desorption in fixed bed column reactor

Sorption in a continuous-flow system was done in a fixed bedolumn reactor (2.3 cm i.d., 30 cm column length). Each bed of sor-ent of desired height was underlain by 4 cm3 of glass wool andcm3 of 3 mm glass beads. The addition of glass wool and glasseads was made to improve the flow distribution. The schematiciagram of the reactor is shown in Fig. 1. Metal ion solution hav-

ng an initial concentration of 10 mg l−1 was adjusted to pH 8 andumped through column at a desired flow rate by a peristaltic pumpMiclins) in an up-flow mode. Samples were collected from the exitf the column at different time intervals and analyzed for cadmiumsing atomic absorption spectrophotometer (PerkinElmer Aanalyst00). Operation of the column was stopped when the effluent metaloncentration exceeded a value of 9.9 mg l−1. The column bed washen rinsed by passing 100 ml deionised water in upward directiont the same speed as used for biosorption from the Cd metal solu-ion. Desorption was carried out by passing 0.01 N H2SO4 throughhe column bed in upward direction at a flow rate of 10 ml min−1.he effluent metal solution was collected and was analyzed for Cdontent. On the completion of desorption cycle, the column wasinsed with deionised water in the same manner as for biosorptionill the eluting deionised water attained pH 7.0. The desorbed andegenerated column bed was reused for next cycle. All experimentsere carried out in duplicates and the deviations were within 5%.

or all graphical representations, the mean values were used. Alltatistical analyses were made using ORIGIN PRO 6.1 software.

.4. Wastewater samples

The sample was acquired from a local electroplating indus-ry located in Kanpur City, U.P. India during December 2006. Theastewater sample was analyzed promptly after collection using

tandard analytical methods [23]. The characteristics of electroplat-ng wastewater were: color: colorless, pH: 11.1, TDS: 55,292 mg l−1,SS: 5397 mg l−1, cadmium: 1570 mg l−1, cyanide: 3322 mg l−1, sul-ate: 1784 mg l−1 and carbonate: 50312 mg l−1.

.5. Modeling and analysis of column data

The performance of packed bed is described through the con-ept of the breakthrough curve. Both, the time until the sorbed

N. Sankararamakrishnan et al. / Separation and Purification Technology 63 (2008) 213–219 215

ficatio

sapfcopai(tpfh

V

w

anbb

TP

P

GBSPSNNC

t(9b

�

f

m

c

T

t

Scheme 1. Chemical modi

pecies are detected in the column effluent (breakthrough point)t a given concentration, and the shape of the concentration-timerofile or breakthrough curve, are very important characteristicsor operation, dynamic response and process design of a sorptionolumn because they directly affect the feasibility and economicsf the sorption phenomena. Experimental determination of thesearameters is very dependent on column operating conditions suchs feed pollutant concentration and flow rate. A breakthrough curves usually expressed in terms of adsorbed pollutant concentrationCad = net pollutant concentration (C0) – outlet pollutant concentra-ion (C)) or normalized concentration defined as the ratio of effluentollutant concentration to inlet pollutant concentration (C/C0) as aunction of flow time (t) or volume of effluent (Veff) for a given bedeight. Effluent volume (Veff) calculated from Eq. (1):

eff = Fttotal (1)

here ttotal and F are the total flow time and volumetric flow rate.

The quantity of metal retained in the column represented by the

rea above the breakthrough curve (C versus t), is obtained throughumerical integration [24]. Dividing the metal mass adsorbed (mad)y the sorbent mass (M) leads to the uptake capacity (Q) of theiomass.

able 1hysical and chemical characteristics of xanthated chitosan

roperties Quantitative value

eometric mean size(mm) 0.30ulk Density (g cm−3) 0.26urface Area (m2 g−1) 0.49ore Diameter (A) 70.05(%) 0.47(%) 4.07a (%) 0.18l (%) 0.28

f

E

ups

t

wbct

u

n of plain chitosan flakes.

The breakthrough time (tb, the time at which metal concentra-ion in the effluent reached 0.1 mg l−1) and bed exhaustion timete, the time at which metal concentration in the effluent exceeded.9 mg l−1) were used to evaluate the mass transfer zone (�t) giveny Eq. (2):

t = te − tb (2)

Total amount of metal sent to column (mtotal) can be calculatedrom Eq. (3):

total = C0Fte (3)

Total metal removal percent with respect to flow volume can bealculated from Eq. (4):

otal metal removal (%) = mad

mtotal× 100 (4)

The metal mass desorbed (md) can be calculated from the elu-ion curve (C versus t). The elution efficiency (E) can be calculatedrom Eq. (5):

(%) = md

mad× 100 (5)

A number of mathematical models have been developed for these in design of column parameters. Among various models, modelroposed by Bohart and Adams [25] is widely used [11,18]. Theimplified equation of Bohart and Adams model is as follows:

= N0Z

C0�− 1

KaC0ln

[C0

Cb− 1

](6)

here C0 is the initial metal ion concentration (mg l−1); Cb is the

reakthrough metal ion concentration (mg l−1); N0 is the sorptionapacity of bed (mg l−1); � is the linear velocity (cm h−1) and Ka ishe rate constant (l mg−1 h−1).

Eq. (6) can be used to determine the service time (t), of a col-mn of bed height Z, given the values of N0, C0 and Ka which must

216 N. Sankararamakrishnan et al. / Separation and Purification Technology 63 (2008) 213–219

Fig. 1. Break through curves at different flow rates obtained for plain and xanthatedcbi8

bv

ttfipe

l

wmti

3

3

fict

F

oeiuabpwfttgmflpmtpopar

3

aTiccgtcii

3

hitosan. (a) Plain chitosan; bed height 9 cm; (b) adsorbent – xanthated chitosan;ed height 9 cm; (c) adsorbent – xanthated chitosan; bed height 15 cm. Conditions:

nitial Cd(II) conc. = 10 mg l−1; pH 8. Flow rate: (�) 3 ml min−1; (�) 5 ml min−1; (�)ml min−1; (—) predicted Thomas model.

e determined for laboratory columns operated over a range ofelocity values, �.

Successful design of a column sorption process required predic-ion of the concentration-time profile or breakthrough curve forhe effluent. Various mathematical models can be used to describexed bed adsorption. Among these, the Thomas model [26] is sim-le and widely used. The linearized form of Thomas model can bexpressed as follows

n[

C0

Cb− 1

]= kThQ0M

F− kThC0V (7)

here kTh is the Thomas model constant (l mg−1 h−1), Q0 is theaximum solid-phase concentration of solute (mg g−1), V is the

hroughput volume (l), F is the volumetric flow rate (l h−1) and Ms the sorbent mass (g).

. Results and discussion

.1. Effect of flow rate

In the first stage of removal studies in the continuous-flowxed column with plain and xanthated chitosan, the flow rate washanged from 3 to 8 ml min−1 while the inlet cadmium concentra-ion in the feed was held constant at 10 mg l−1 at pH 8.0. The plots

twbt

ig. 2. Schematic diagram of laboratory based small column for fixed bed studies.

f comparative normalized cadmium concentration versus efflu-nt volume at different flow rates are given in Fig. 2. As indicatedn Fig. 2, at the lowest flow rate of 3 ml min−1, relatively higherptake values were observed for cadmium biosorption to both plainnd xanthated chitosan. In general, for both the adsorbent, sharperreakthrough curves were obtained at higher flow rates. The break-oint time and total adsorbed cadmium quantity also decreasedith increasing flow rate. This behavior can be explained by the

act that cadmium biosorption is affected by insufficient residenceime of the solute in the column. This insufficient time decreaseshe bonding capacity of the cadmium ions on to xanthate and aminoroup present in the biosorbent [27]. Even though more shortenedass transfer zone (usually preferable) was observed at the highest

ow rate, the total metal removal percentage (a reflective of systemerformance) and the metal uptake were actually observed maxi-um at lowest flow rates. The sorption data were evaluated and the

otal sorbed quantities, maximum cadmium uptakes and removalercents with respect to flow rate are presented in Table 2. It is alsobserved that maximum cadmium uptake and cadmium removalercentage were obtained as 132.3 mg g−1 and 71.4% respectively,t 3 ml min−1 for xanthated chitosan and 40.1 mg g−1 and 56.3%espectively for plain flakes.

.1.1. Application of Thomas modelTo determine the maximum solid-phase concentration (Q0)

t different bed depth in column, the data were fitted to thehomas equation model by using linear regression analysis. Table 3llustrates the model constant (kth), maximum solid-phase con-entration (Q0) and correlation coefficient for plain and xanthatedhitosan. It is clear from Table 3 and Fig. 3, that the model gave aood fit for the experimental data obtained for both plain and xan-hated chitosan. Higher values of Q0 were obtained for xanthatedhitosan compared to plain chitosan indicating that uptake capac-ty of xanthated chitosan is higher than the plain chitosan. This wasn agreement with our earlier studies involving batch reactors [20].

.2. Effect of bed height

Using Bohart–Adams approach at least nine individual column

ests must be conducted to collect the required laboratory datahich is a time consuming task. A technique has been described

y Hutchins [28] which requires only three column tests to collecthe necessary data. In this technique, called the bed depth service

N. Sankararamakrishnan et al. / Separation and Purification Technology 63 (2008) 213–219 217

Table 2Column data and parameters obtained for plain and xanthated chitosan

Adsorbenta Bed height (cm) Flow rate (ml min−1) Uptake (mg g−1) tb (h) te (h) �t (h) (dc/dt)b (mg l−1h−1) Total removal (%)

Plain chitosan 9 3 40.1 ± 0.5 7.2 40.1 32.9 0.30 56.35 21.3 ± 0.3 4.5 18.0 13.5 0.72 39.98 17.2 ± 0.2 0.2 8.0 7.8 1.25 36.2

Xanthated chitosan 9 3 132.3 ± 1.5 43.2 102.0 58.8 0.17 71.45 87.5 ± 0.9 23.0 49.0 26.0 0.38 60.18 94.1 ± 1.8 13.0 40.0 23.0 0.43 49.5

Xanthated chitosan 15 3 169.8 ± 2.5 70.0 136.0 66.0 0.15 78.15 165.2 ± 2.8 40.0 72.0 32.0 0.30 76.28 159.8 ± 1.8 13.0 46.0 33.0 0.30 73.3

a Conditions: influent pH 8; initial Cd(II) conc. 10 mg l−1.b Slope of the breakthrough curve from tb to te.

Table 3Parameters predicted from Thomas model for plain and xanthated chitosan

Adsorbent Bed height(cm)

Flow rate(l h−1)

kth

(l mg−1 h−1)Q0

(mg g−1)R2

Plain chitosan 9.0 0.18 0.018 5.81 0.969.0 0.30 0.026 3.92 0.909.0 0.48 0.073 3.17 0.95

Xanthated chitosan 9.0 0.18 0.0134 19.04 0.949.0 0.30 0.0249 18.71 0.979.0 0.48 0.0278 14.76 0.94

Xanthated chitosan 15.0 0.18 0.0100 18.85 0.94

C

t

t

w

a

b

t0csr

Ft5

15.0 0.30 0.0290 17.94 0.9515.0 0.48 0.0234 16.63 0.94

onditions: initial Cd(II) conc. 10 mg l−1; Influent pH 8.

ime (BDST) approach the Bohart–Adams equation is expressed as

b = aZ + b (8)

here

= slope = N0

C0v(9)

= intercept = 1kaC0

ln(

C0

Cb− 1

)(10)

b, the time at which metal concentration in the effluent reached.1 mg l−1(h) and Z is the bed height (cm). The parameters N0 and ka

an be calculated from the slope of the linear plot of tb versus Z. Fig. 3hows the BDST plot for plain and xanthated chitosan for a volumet-ic flow rate of 5 ml min−1 (linear flow rate = 73.0 cm h−1). The linear

ig. 3. Bed depth service time plot for plain and xanthated chitosan. Condi-ions: initial Cd(II) conc. 10 mg l−1, linear flow rate 73 cm h−1 (volumetric flow rateml min−1) influent pH 8, breakthrough conc. 0.1 mg l−1.

Fwre

rco

t

t

TP

A

PX

Cl

ig. 4. (a) Breakthrough curves for biosorption of cadmium from electroplatingastewater during two regeneration cycles. Conditions: bed height 9 cm; flow

ate = 3 ml min−1; pH 8; influent Cd(II) 10.5 mg l−1. (b) Elution curves for two regen-ration cycles using 0.01 M H2SO4. Conditions: flow rate 10 ml min−1.

elationship obtained for cadmium sorption on xanthated and plainhitosan flakes from the bed depth plot for an initial concentration

f 10 mg l−1 is given in Eqs. (11) and (12), respectively:

b = 20Z + 11 (11)

b = 3Z − 0.8333 (12)

able 4arameters predicted from BDST model for plain and xanthated chitosan

dsorbent N0 (g l−1) ka (l mg−1 h−1)

lain chitosan 2.19 0.5514anthated chitosan 14.60 0.0418

onditions: initial Cd(II) conc. 10 mg l−1; bed height: 5, 9, and 15 cm; influent pH 8;inear flow rate 0.73 m h−1 (volumetric flow rate 5 ml min−1).

218 N. Sankararamakrishnan et al. / Separation and Purification Technology 63 (2008) 213–219

Table 5Sorption process parameters for two sorption–desorption cycles using electroplating wastewater

Cycle Uptake (mg g−1) tb (h) te (h) �t (h) (dc/dt)a (mg l−1 h−1) Total removal (%) Time for elution (h) Elution efficiency (%)

1 130.3 ± 2.5 37.2 114.0 76.8 0.334 69.3 22.0 84.52

C rate

ogcb

Z

tTp

3

acatawtettsutbmcmtmc

awstcipePwioiccihC[

Hma

3

dbap3wcirew

3

swobcipec

4

tappcccftbgro

Acknowledgements

100.5 ± 2.2 19.5 81.0 61.5 0.686

onditions: adsorption: influent Cd(II) conc. 10.5 mg l−1, pH 8, bed height 9 cm, flowa Slope of the breakthrough curve from tb to te.

The saturation concentrations (N0) and the rate constant (ka)btained from BDST for plain and xanthated chitosan flakes areiven in Table 4. The critical bed depth (Z) required for preventingadmium concentration exceed above 0.1 mg l−1, which is obtainedy substituting t = 0 in expression (6), is given below:

= V0

KaN0ln

(C0

Cb− 1

)(13)

The critical depth values obtained for plain and xanthated chi-osan from the BDST plot were 0.278 and 0.0075 cm, respectively.he BDST model parameters can be useful to further scale up therocess for other flow rates without further experimental run.

.3. Mechanism of adsorption and interference of other ions:

In the present condition, there exists two kinds of covalent inter-ction one between cadmium ions and xanthate group formingadmium xanthate bonding and the other between cadmium andmine group of chitosan forming cadmium amine bonding. Underhe prevailing pH conditions, (pH 8) cadmium predominantly existss Cd(OH)+ species. Hydroxy metal complexes are known to adsorbith a higher affinity than the completely hydrated metals because

he formation of an OH group on the metal reduced the freenergy requirement for adsorption [29]. Therefore, it seems thathe adsorption of cadmium ions can be related to the change inhe availability of Cd(OH)+. The pKa of xanthate-xanthic acid dis-ociation constant is reported to be 1.70. Thus, in the pH rangesed in the present study, the characteristics of surface group ofhe adsorbent are unlikely to change. According to hard soft acidase (HSAB) theory, soft bases tend to form stable complexes withetals such as Cd2+, Pb2+ and Cu2+ [30]. Since xanthate groups

an be classified as soft bases, xanthated chitosan will have auch higher affinity and sorption capacity when compared to

hat of plain chitosan. The strong affinity between sulfur and cad-ium may result in high affinity constant and hence high uptake

apacity.In wastewater streams, the metal of interest is usually found in

matrix containing several metal ions [31]. These ions can interactith heavy metals and thus modify their behavior towards the

orbent material used. The sorption performance of adsorbentowards a given ion in solution could therefore vary. Some anionsan have an affinity towards the metal, so that they form annsoluble or soluble complex, displaced with difficulty in theresence of the sorbent material [31]. In the earlier studies [20] theffect of various anions (Cl−, SO4

2−, CN−) and cations (Cu2+, Hg2+,b2+, Cd2+, Ni2+, Zn2+) on the adsorption capacity of cadmiumas investigated in batch reactors. As expected, only cyanide

ons interfered in the adsorption as the stability constants ofther anions were considerably lower than xanthate. However,t was found that with higher dose rate of the adsorbent, theyanide interference could be conveniently overcome. In the

ase of cations, the results indicated that none of the metal ionsnterferred in the adsorption. For a mixed-metal solutions theierarchy of xanthate precipitation is in the following order:u2+ − Hg2+ > Pb2+ > Cd2+ > Ni2+ > Zn2+ > Ca2+ − Mg2+ − Mn2+ � Na+

32]. Under alkaline conditions (pH 8), the metal ions Pb2+,

m(fl

69.2 19.0 82.8

3 ml min−1. Desorption: eluant 0.01 M H2SO4, flow rate 10 ml min−1.

g2+ and Cu2+ which have higher formation constant than cad-ium precipitate as hydroxides and does not interfere in the

dsorption.

.4. Treatment of electroplating wastewater

Electroplating wastewater obtained from the local industry wasiluted to obtain the working range of the present study. Columnreakthrough curves for removal of cadmium from the wastewaterre shown in Fig. 4. It is evident from Table 5, that for electro-lating wastewater, at a column height of 9 cm and a flow rate ofml min−1, cadmium uptake was found to be 130.3 ± 2.5 mg g−1,hereas for synthetic solution it was 132.3 ± 1.5 mg g−1. In the

ase of percentage cadmium removal, xanthated chitosan exhib-ted 69.3% and 71.4% for wastewater and synthetic solutions,espectively. These results prove that xanthated chitosan could beffectively used for the treatment of cyanide rich electroplatingastewater.

.5. Regeneration

In the present study, xanthated chitosan was reused for twoorption–desorption cycles at 3 ml min−1. The column was packedith 7.26 g of xanthated chitosan yielding an initial bed height

f 9 cm and bed volume of 37.4 ml. Table 5 summarizes thereakthrough time, exhaustion time and column uptake for twoycles. As shown in Fig. 4, a decreased breakthrough and inncreased exhaustion time was observed as the regeneration cyclesrogressed resulting in a broadened mass transfer zone. How-ver, a good metal sorption capacity was obtained in both theycles.

. Conclusion

Xanthated chitosan was found to be an effective biosorbent forhe removal of cadmium from cyanide rich electroplating wastew-ter. An increase in bed height resulted in improved sorptionerformance. The BDST model constants were determined androposed for the use in column design. The Thomas model was suc-essfully used to fit the column data at different flow rates and theonstants were evaluated. The sorption performance of xanthatedhitosan on electroplating wastewater was successfully evaluatedor two cycles. The sorption capacity exhibited by xanthated chi-osan was found to be high (169.8 ± 2.5 mg g−1) compared to otheriosorbents reported recently [S. mutium 102 mg g−1 [33]; bluereen algae 74.87 mg g−1 [34]. Finally, high sorption capacity, andeuse potential makes it very attractive material for the treatmentf cadmium in cyanide rich electroplating wastewater.

The authors are thankful for the funding provided by Depart-ent of science and Technology, New Delhi under the SERC scheme

SR/SI/IC-27/2006) and India Sea Food, Cochin for the Chitosanakes.

and P

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[[

[

[

[[

[

N. Sankararamakrishnan et al. / Separation

eferences

[1] M.P. Waalkes, Cadmium carcinogenesis in review, J. Inorg. Biochem. 79 (2000)241–244.

[2] W.J. McLay, F.P. Reinhard, Waste minimization and recovery technologies, Met.Finish. 98 (2000) 817–850.

[3] S.E. Kentish, G.W. Stevens, Innovations in separations technology for the recy-cling and re-use of liquid waste streams, Chem. Eng. J. 84 (2001) 149–159.

[4] S. Pivovarov, Adsorption of cadmium onto hematite: temperature dependence,J. Colloid Interface Sci. 234 (2001) 1–8.

[5] G. Mustafa, B. Singh, R.S. Kookana, Cadmium adsorption and desorptionbehaviour on goethite at low equilibrium concentrations: effects of pH andindex cations, Chemosphere 57 (2004) 1325–1342.

[6] E.A. Deliyanni, K.A. Matis, Sorption of Cd ions onto akaganeite-type nanocrys-tals, Sep. Purif. Technol. 45 (2005) 96–102.

[7] R. Weerasooriya, H.K.D.K. Wijesekara, A. Bandara, Surface complexation mod-eling of cadmium adsorption on gibbsite, Colloids Surf. A Physicochem. Eng.Aspects 207 (2002) 13–24.

[8] R.R. Bell, G.C. Saunders, Cadmium adsorption on hydrous aluminium (III) oxide:effect of adsorbed polyelectrolyte, Appl. Geochem. 20 (2005) 529–536.

[9] A. Saeed, M. Iqbal, Bioremoval of cadmium from aqueous solution by black gramhusk (Cicer arientinum), Water Res. 37 (2003) 3472–3480.

10] R. Leyva-Ramos, L.A. Bernal-Jacome, I. Acosta-Rodriguez, Adsorption of cad-mium(II) from aqueous solution on natural and oxidized corncob, Sep. Purif.Technol. 45 (2005) 41–49.

11] U. Kumar, M. Bandyopadhyay, Fixed bed column study for Cd(II) removalfrom wastewater using treated rice husk, J. Hazard. Mater. B129 (2006) 253–259.

12] P. Yin, Q. Yu, B. Jin, Z. Ling, Biosorption removal of cadmium from aqueoussolution by using pretreated fungal biomass cultured from starch wastewater,Water Res. 33 (1999) 1960–1963.

13] M.S.D. Erosa, T.I.S. Medina, R.N. Mendoza, M.A. Rodriguez, E. Guibal, Cadmiumsorption on chitosan sorbents: kinetic and equilibrium studies, Hydrometal-lurgy 61 (2001) 157–167.

14] B. Benguella, H. Benaissa, Cadmium removal from aqueous solutions by chitin:kinetic and equilibrium studies, Water Res. 36 (2002) 2463–2474.

15] S. Babel, T. Kurniawan, Low-cost adsorbents for heavy metals uptake from con-taminated water: a review, J. Hazard. Mater. B97 (2003) 219–243.

16] I.N. Jha, L. Iyengar, A.V.S. Prabhakara Rao, Removal of cadmium using chitosan,J. Env. Eng. 114 (1988) 962–975.

[

[

urification Technology 63 (2008) 213–219 219

17] T.-Y. Hsien, G.L. Rorrer, Heterogeneous cross-linking of chitosan gel beads:kinetics, modeling, and influence on cadmium ion adsorption capacity, Ind.Eng. Chem. Res. 36 (1997) 3631–3638.

18] B. Volesky, I. Prasetyo, Cadmium removal in a biosorption column, Biotechnol.Bioeng. 43 (1994) 1010–1015.

19] B. Volesky, J. Weber, J.M. Park, Continuous-flow metal biosorption in a regen-erable Sargassum column, Water Res. 37 (2003) 297–306.

20] N. Sankararamakrishnan, A.K. Sharma, R. Sanghi, Novel chitosan derivativefor the removal of cadmium in the presence of cyanide from electroplatingwastewater, J. Hazard. Mater. 148 (2007) 353–359.

21] N. Sankararamakrishnan, R. Sanghi, Preparation and characterization of a novelxanthated chitosan, Carbohyd. Polym. 66 (2006) 160–167.

22] T. Becker, M. Schlaak, H. Strasdeit, Adsorption of nickel(II), zinc(II) and cad-mium(II) by new chitosan derivatives, React. Funct. Polym. 44 (2000) 289–298.

23] APHA, American Public Health Association, Standard Methods for the Exam-ination of Water and Wastewater, AWWA-WPCF, 17th ed., Washington, USA,1995.

24] Z. Zulfadhly, M.D. Mashitah, S. Bhatia, Heavy metals removal in fixed-bed col-umn by the macro fungus Pycnoporus sanguineus, Environ. Pollut. 112 (2001)463–470.

25] G.S. Bohart, E.Q. Adams, Some aspect of the behavior of charcoal with respectto chlorine, J. Am. Chem. Soc. 42 (1920) 523–544.

26] L. Conter, R. Knox, Ground Water Pollution Control, Lewis, USA, 1986, p. 96.27] G. Yan, T. Viraraghavan, Heavy metal removal in a biosorption column by immo-

bilized M. rouxii biomass, Biores. Technol. 78 (2001) 243–249.28] R.A. Hutchins, Liquid Phase adsorption: maximizing performance, Chem. Eng.

87 (1980) 01–110.29] S. Schiewer, B. Volesky, Ionic strength and electrostatic effects in biosorption

of divalent metal ions and protons, Environ. Sci. Technol. 31 (1997) 247.30] M.J. Winter, d-Block Chemistry, Oxford University Press, New York, 1994.31] K.A. Matis, A.I. Zouboulis, I.C. Hancock, Waste microbial biomass for cadmium

ion removal: application of flotation for downstream separation, Biores. Tech-nol. 49 (1994) 253–259.

32] www.usace.army.mil/Publications/eng-manuals/em1110-1-4012/chap6.pdf.

33] J.L. Lodeiro, R. Barriada, M.E. Herrero, Sastre de Vicente, the marine macroalga

Cystoseira baccata as biosorbent for cadmium(II) and lead(II) removal: kineticand equilibrium studies, Environ. Pollut. 142 (2006) 264–273.

34] M. Iqbal, R.G.J. Edyvean, Loofa sponge immobilized fungal biosorbent: a robustsystem for cadmium and other dissolved metal removal from aqueous solution,Chemosphere 61 (2005) 510–518.