Evaluation of an adsorbent prepared by treating coconut ...

Indian Journal of Engineering & Materials Sciences Vol. 6, October 1999, pp 279-285

Evaluation of an adsorbent prepared by treating coconut husk with polysulphide for the removal of mercury from wastewater

* M K Sreedhar, A Madhukumar & T S Anirudhan

Department of Chemistry, University of Kera1a, Kariavattom, Thiruvananthapuram 695 581 , Indi a.

Received 9 November 1998; accepted 11 August 1999

Polysulphide treated coconut husk can be used as an effective adsorbent for the removal of Hg(II) from aqueous systems. The parameters affecting Hg(II) adsorption such as Hg(II) concentration, agitation time , pH , sorbent dose, ionic strength and temperature were studied to estimate optimum conditions. Maximum adsorption capacity was observed in the pH range 5.5 - 10.0. The adsorption isotherm was a lso affected by temperature since the adsorption capacity was increased by raising the temperature form 30 to 60°C. Lagergren equation and Langmuir and Freundlich isotherm models have been used to determine rate constants of adsorption and capacity of treated coconut husk. Different reagents were tested for extracting Hg(I1) from the spent adsorbent. High extraction efficiency was obtained with 0.5M Hel solution. Hg(II) can be successful ly removed from synthetic and industrial wastewaters. The results obtained point towards viable adsorbe nt which is both effective as well as economically attractive for Hg(II) removal from wastewater.

The presence of heavy metals in the aquatic environment can be detrimental to a variety of living species. Therefore, the elimination of heavy metals from water and wastewater is important to protect public health. Several inexpensive inde­genous materials have been studied for their utility as an adsorbent of _ heavy metals in water and

. I ? . wastewater' -.These adsorbents were mostly derIved from naturally occurring lignocellulosic materials. Cellulosic material s were found to behave as weakly acidic cationic exchangers due to substances inherently associated with cellulose such as pectins, lignins and tannins which contain polyphenolic hydroxyl groups and carboxylic groups'.

Coconut husk is a widely available and abundant natural material. Coconut husk from coir industry is considered as a waste material since it does not find any use as such and causes disposal problems. Coconut husk basically contains cellulose and lignin4. It is an effective adsorbent in removing Cd, Cu, Pb, Mg, Ni and Zn from aqueous solutions5

-7

. Studies on mercury adsorption using sulphur contammg materials were initiated by Knocke and Hemphill8

.

Meng et ai.9 studied the ability of used tyre rubber to immobilize mercury from solution . They speculated that a reaction between mercury and sulphur groups

* Author for correspondence

present in the tyre rubber accounted for high adsorption capacity. Moreover, we have found that the natural adsorption properties of cellulos ic materials can be further enhanced by polysulphide treatment lO

. In the present paper polysulphide treated coconut husk (PTCH) is evaluated for the removal of mercury from water under kinetic and equilibrium conditions. Removal of mercury by PTCH was testified using chloro - alkali manufacturing industry wastewater.

Experimental Procedure Coconut husk (CH) used as the starting material

was collected from a local coir factory . It was washed several times with distilled water to remove the surface adhered particles and water so lubl e materials and dried at 80 - 100°C. 10 9 of this sample (-80 +230 mesh size) was mixed with 200 mL of 5% sod ium polysulphide which was prepared by using wet method of Klemm et ai. 11 and the mixture was refluxed for I h. The product namely polysulphide treated coconut husk (PTCH) was fi Itered and washed many times with distilled water until no sulphide was released in the washing. The filtrate and a ll washings were collected and the amount of unadsorbed polysulphide was determined using the method of Feher and Berthold l2

. The amount of polysulphide retained by the coconut husk was then calcul ated and found to be 47.8 mg/g. The material was dried and

280 INDIAN J. ENG. MATER. SCI. , OCTOBER 1999

sieved to get -80 +230 mesh size. The physical and surface properties of ITCH were carefully determined by standard methods 10. The characteristics of the sorbent are: apparent density , 1.18 g/mL; cation exchange capacity, 1.03 meq/g; BET surface area, 87 .8 m2/g; moisture content, 7 .62 %; ash content, 3.82 %; porosity, 0.38 mL/g and pHzpc 4.5.

The adsorption experiments were carried out by agitating 100 mg of adsorbent with 50 mL of Hg(II) solution of desired concentration in 100 mL stoppered conical flask at 200 rpm for the predetermined time intervals using a temperature controlled waterbath Shaker. The pH of the solution was adjusted using 0 . 1 M HNO) and 0.1 M NaOH. Isotherm studies were conducted by varying the initial concentration from 40 to 400 mg/L at different temperatures. At the end of the predetermined time intervals the contents were centrifuged and the sup~rnatant was analysed. Chemical analysis of Hg(II) was done spcctrophotometrically using eosin method \3. Batch operations were also carried out for the removal of Hg(ll) from synthetic and actual industrial wastewaters collected from one of the chloro - alkali industries in Cochin city (Kerala). The effluent was characterised using standard methods

l4.

Table I-Adsorption capacity of CH and PTCH for Hg(II) (Initi al concn: 25 mglL; adsorbent dose: 2 giL).

Adsorbent

CH

PTCH

CH

PTCH

CH

PTCH

CH

PTCH

Time, h

3

3

8

8

24

24

Adsorption capacity

mglg %

4.32 34.51

9.87 78.95

5.94 47.52

12.29 98.3 1

6.17 49.34

12.27 98.19

6.13 49.01

12.26 98.07

Table 2-Effect of sodium polysulphide concentration on the surface modification of CH (Hg(II) concn: 25 mglL; adsorbent dose, 2 giL; agitation time, 3h; pH , 5.5) .

Sodium polysulphide Amount adsorbed by the concn ., % sample

mglg %

4.65 37.23

2 7.69 61 .52

4 9.99 79.93

5 12.29 98.35

6 12.28 98.28

8 12.2 1 97.71

10 12.1 I 96.93

After the attainment of equilibrium, the supernatants were carefully decanted and desorption experiments were carried out using diffe rent reagents. The desorbed Hg(II) was ana lysed as be fore.

Results and Discussion

The 0.1 g of coconut husk w ith and without polysulphide treatment was brought into contact with Hg(ll) solution for I, 3, 8 and 24 h. After it was equilibrated, the sorbed amount of Hg(I l) was detennined by measuring decreas ing Hg( II) concentration in the solution. The pre liminary investigation shows a high uptake of metal ions on ITCH in comparison with CH (Table I ). Also, it is found that Hg(II) desorbs if time is ex te nded to 24 h and so adsorption efficiency decreases .

The adsorption efficiency of Hg(II) was investigated by using different amount of sodium polysulphide. The concentration of sodium polysulphide varied form I to 10 %. The results are shown in Table 2. The highest effi c iency was shown by coconut husk treated with 5 % sod ium polysulphide; above and be low thi s concentrat ion the efficiency of Hg(II) removal was decreased.

The effect of initial Hg(II) concentration and agitation time on the adsorption of Hg(II) on PTCH is shown in Fig. 1. The uptake of Hg(II) inc reases with increase in contact period and reaches saturation po int within 3 h after which the uptake is found to become apparently constant. The saturation period of the adsorption IS extreamly independe nt of initial

.80

70 . ~.

"iil!60 e 'Ii 50 -~ :; 40 .., '0 '" 30 Q iii 20 .

= 10 .

Adsorbent dose pH Ionic strength Initial Hg(lJ) concn

: 2 gIL : 5.5 : 0.01 M : + 25 mg/L • 50 mglL ... 100 mglL x 150 mglL

__ _e--e-----.---.

OIL--~----~--------~--~ o 50 100 150 200 250

Time, min

Fig. I-Effect of initial concentration and agi tation time on Hg(l l) removal by PTC H.

'.

SREEDHAR et al. : POLYSULPHIDE TREATED COCONUT HUSK 28 1

concentration. With the increase in initial concentration of Hg(II) from 25 to 150 mg/L, the uptake of Hg(II) decreases from 98.3% (12.29 mg/g) to 71.65% (53.74 mg/g). This obviously shows that the percentage uptake of Hg(IJ) is dependent on initial concentration. This is because at high initial concentration the ratio of the initial number of moles of Hg(IJ) to the available surface area is high, hence the fractional adsorption becomes dependent on initial concentration.

The two important aspects for parameter evaluation of the adsorption study are the kinetic and the equilibria of adsorption. The adsorption of Hg(II) from liquid phase to solid phase can be considered as a reversible reaction with an equilibrium being established between two phases. The rate constants k ad

of adsorption of Hg(IJ) on PTCH were determined from the following first-order rate expression given by Lagergren 15

k log(qe -q)=log q -~t

e 2.303 ... (I)

where q and qe (both in mg/g) are the amounts of Hg(IJ) adsorbed on PTCH at time t and at equilibrium

1.6

1.4 ..

1.2

a.B

0.6

0.4

0.2

Initial Hg(II) concn :. 25 mglL • 50 mglL i. 100 mglL x 150 mglL

o . Adsorbent dose : 2 giL pH: 5.5

,..., -0.2 . I' h 0 0 (A) cr OOIC strengt : . I M I .. -0.4 . I I . I

,:;; 1.6 Initial Hg(ll) concn : 150 mg/L 1:>'1 .2 Adsorbent dose : 2 gIL

1.4 pH : 5.5

1.2

0.8

0.6 Temperature : + 30''C

0.4 .40"C

0.2 ... 50"C x 60°C

:0.01 M

(B) Ol--_____ -+-__ --+---~-__!

o 25 50 75 100 125 150

Time, min

Fig. 2-Lagergren plots for the adsorption of Hg(I1) on PTCH at different (A) concentrations and (8) temperatures.

respectively. The linear plots of log (qe-q) VS I

(Fig. 2A) for different concentrations indicate the validity of the above equation for the present system. At 30°C and pH 5.5 the kad values at initi al concentrations of 25, 50, 100 and 150 mg/L were found to be 2.483xI0-2

, 2.388xI0-2, 2 .365xI0-1 and

2.218xI0-2 min- I, respectively . The linear plots of log(qe-q) vs t (Fig 2B) for different temperatures further confirmed to first-order kinetics. The va lues of kad were calculated and found to be 2 .0 18x 10-

2,

2.464x I 0-2, 2.957xI0-2 and 3.035x I0-2 min- I at 30,

40, 50 and 60°C, respective ly. These va lues also suggest that adsorption IS faster at higher temperatures. The activation energy, E" was determined using Arrhenius equati on If>. A plot of

In kad vs ~ was found to be linear. The value of E" as . T

calculated from the slope of the plot is equal to 11 .88 kl/mol which was comparable with reported values l6

, 17 indicating the existence of intrapart ic le

diffusion in the adsorption process. The pore diffusion coefficients Di (m2/s ) for

adsorption of Hg(II) on PTCH at various temperatures and shaking speed were calculated by the method of Vermeulan l8 as applied by Streat et 0/

19.

2 0.030 ~J

Di = ... (2) t~

where ro is the particle radius and 11 12 is time for half adsorption. The values Di for Hg(lI ) adsorption onlo PTCH (ro=0.096 mm) were found to be 2.004x I 0-13

,

2.560xI0-1" 2.993xI0- u and 3.372x I0- 1.I m2/s at 30,

40, 50 and 60°C, respect ive ly . The Di values for Hg(II) adsorption for shaking speed of 100, 200. 300 and 400 rpm were 1.707x I0- 13

, 2.304x I0- 1\

2.980xI0- 13 and 3.445x I 0- 11 m2/s, respectively .

100 . • 'Adsorption

80 . ·Pltclpilation

60

Oi 40 > o e ~ 20

o .

Initial Hg(lI) concn : 150 mg/L Adsorbent dose : 2 gi L Ionic strength : 0.01 M Agitation time : 3 h

,20 ~_--+-__ +--_--+-__ +--_---'

o 2 4 6 8 10

pH

Fig. 3-Effect of pH on Hg( lI ) removal by PTCH .

282 INDIAN 1. ENG. MATER. SCI., OCTOBER 1999

According to earlier workers20 if pore diffusion is to be rate limiting then the pore diffusion coefficients should be in the range of 10-12 - 10- 14 m2/s. As per this, the rate limiting step appears to be pore diffusion for Hg(II) - PTCH system.

The pH of Hg(II) solution was adjusted to a desired value by adding HN03 or NaOH solution. The effect of pH on the adsorption of Hg(II) by PTCH is shown in Fig.3. For comparison mercuric .flydroxide precipitated by NaOH is also given in Fig.3. The precipitation curve shows a sharp decrease in concentration of Hg(II) ions in solution which suggests that mercury is precipitating from solution at this concentration well before adsorption is complete. However, at any pH Hg(II) removal by adsorption is greater than that of hydroxide precipitation. At a pH of 2.0, only 29.4% of the total Hg(II) is adsorbed by PTCH and at a pH of 5.5 approximately 96.8% of Hg(II) is removed. The extend of Hg(II) removal between a pH range of 5.5 and 10.0 increased only marginally. So this pH range of 5.5 and 10.0 was determined to be optimum. Adsorption of metal ions from solutions by solid phase can occur with formation of surface complex between the adsorbent ligand and the metal. At lower pH (below pH zpc) the adsorbent surface is protonated and the net PTCH surface charge becomes positive. This leads ' to unfavorable condition for the adsorption of cations. Significant adsorption was observed above pH zpc at which sorbent surface is negative and sorbate species Hg2+ and Hg(OHt are positively charged and therefore the interaction is that of electrostatic attraction. The marginal increase in adsorption at higher pH range indicates that the Hg(OHh species formed in this range retained in the micropores of the PTCH particles.

65 .r-----------------------------~ Initial Hg(II) concn : 200 mg/L Adsorbent dose : 2 gIL pH : 5.5 Agitation time : 3 h

50~--__ ~ __ ~ ____ ~ ____ ~ ____ ~

o 0.05 0.1 0.1 5 0.2 0.25

Ionic strength, M

Fig. 4-Effec( of ionic strength on Hg(JI) removal by PTCH.

Fig. 4 shows the effect of ionic strength on the adsorption efficiency of Hg(II) onto PTCH. A twentyfold increase in sodium nitrate concentration resulted in a decrease of 7. 15% in adsorption. This effect may be due to changes in mercury activity or in the properties of the electrical double layer. The decrease in adsorption with increase in ionic strength is due to the expansion of thickness of double layer21 . The adsorption is sens iti ve to changes in concentrations of the supporting electro lyte if the electrostatic attraction is the sign ificant mechanism for sorbate removal 21 . The results show that e1ctrostatic attraction plays a significant role in the removal of Hg(II) by PTCH.

The effect of variation in PTCH dose for an initial concentration of 250 mglL is shown in Fig. 5. It is apparent that by increasing the adsorbent dose from 1.0 to 8.0 gIL the removal efficiency increases but adsorption density decreases . The decrease in adsorption density may be attributed to the fact that some of adsorption sites remain unsaturated during adsorption process; whereas the number of available adsorption sites increases by increasing the adsorbent dose and that results in the increase of remova l efficiency. The particle interaction brought about by high sorbent concentration may also desorb some of the sorbate which is only loosely and revers ibl y bound to the sorbent surface.

Adsorption equil ibrium data are conveniently represented by adsorption isotherm which are helpful in determining the adsorption capacity of the adsorbent. Equilibrium data was adjusted according to the Langmuir and Freundlich isotherm mode ls which are represented mathematically as follows .

Ce = __ I_+~ ... (3) qe Q(Jb Q(J

-K C l i n qe - F e

100

90

80 ":!. 0

... :-70 .. > ~80 .. a: 50

40 .

30

20 0

Initial Hg(lI) conco : 250 mg/L Agitation time : 3 h pH : 5.5 Ionic strength · 0.01 M

2 4 6 8

Adsorbent dose, giL

... (4)

100 OJ)

90 eo E

80 " c:r 70 S .;;;

c: 60 " "0

c: 50 .2

e-40 ~ "0

30 « 20

10

Fig. 5-Effect of adsorbent dose on HgCI I) remo\'a l hy PTCH .

SREEDHAR et al. : POLYSULPHIDE TREATED COCONUT HUSK 283

where Ce is the equilibrium concentration (mgIL) qe is the amount adsorbed at equilibrium (mg/g). QO is the amount of Hg(II) adsorbed per unit weight of adsorbent and b is the energy of adsorption,

determined from linear plot of Ce vs Ce shown in qe

Fig. 6. KF and n are Freundlich constants which are the measures of adsorption capacity and intensity of adsorption. The values of KF and n are determined from the linear plot of log qe vs log Ce shown in Fig. 7 . The values of QO and b at temperatures 30, 40, 50 and 600C were computed using linear regression analysis and were found to be 88.17 mg/g and 0.314 Llmg, 95.45 mg/g and 0.373 Llmg, 105.74 mg/g and 0.437 Llmg and 117.21 mg/g and 0.557 Llmg respectively . QO and b values for the adsorption of Hg(II) on photofilm waste sludge22 at 30°C was

3.5 Adsorbent dose : 2.0 giL pH : 5.5

3 Agitation time : 3 h

~2.5

~ .. 2 ~ .. U 1.5

0.5

Temperature : + 3O"C x4O"C A 50''C • 6O"C

o~ __ ~ __ ~ __ ~~~~ __ ~ o 50 100 150 200 250 300

Ce, mg/L

Fig. 6-Langmuir isotherm for Hg(lI ) adsorption on PTCH .

Adsorbent dose: 2 giL 2.1 . pH : 5.5

2.05 Agitation time : 3 h 2 Temperature : + 3O"C

1.95

.. 1.9 c:r Ili 1.85 o - 1.8

1.75

1.7

1.65

• 40"C A 50"C

x

1.6 '1---4---+----'-__ -_~ _ ___.J

o 0.5 1.5

log,Ce

2 2.5 3

Fig. 7-Freundlieh isotherm for Hg(JI ) adsorption on PTCH.

reported to be 11 .76 mg/g and 0.490 Llmg respectively and on used tyre rubber9

, QO and b were 14.60 mg/g and 0.471 Llmg respectively. The QO value for the adsorption of Hg(II) has been reported by Servno et al. 23 is 74.02 mg/g for activated carbon from Merck.

The values of KF and n calculated from the Freundlich plots for 30, 40, 50 and 60°C were found to be 26.75 and 5.24; 28.19 and 5.07 ; 31.25 and 5.15 and 35.69 and 5.34 respectively. The increase in values of KF at higher temperatures shows that the adsorption rate increases with a raise in temperature.

. I The values of 0.1 < - < 1.0 shows that adsorption of

n Hg(II) on PTCH is favourabl e.

Thermodynamical parameters were calculated using the foJIowing equations

f).G" = -RT lnb ... (5)

!:!S" /).jf " lnb=----

R RT ... (6)

I The Van't Hoff plot of In b vs - was found to be

T linear and using the linear regression analysis, tlW and f).SO were computed from the slope and intercept. The values of thermodynamic parameters (Table 3) indicate that the overaJI process of adsorpt ion is spontaneous and endothermic in nature. The pos iti ve value of f).SO suggests the increased randomness at the

Table 3- Thermodynami c Parameters

Temperature, !:lGo, !:lH", !:lS", ,oC kllmol kll mol l lmol/K

30 -14.48

40 -15.41 15.73 99 .56

50 -16.33

60 -1 7.5 1

Table 4-Extrac tion of i-I g( 11) from mercury adsorbed PTCH

Extractant Conen. Quantity clesorbecl. %

NaNO, 0.50 M 40.7

NaCI 0.50M 413

Na2S04 O. IOM 29.5

HCI 0.50 M 94.S

HNO, 0.50 M 89.2

H2SO4 0.25 M 83.4

CH)COOH 0.50 M 37.2

NaCI-HCI 0.25 M 67.1

NH4NOr HNO, 0.25 M 529

284 INDIAN 1. ENG. MATER. SCI., OCTOBER 1999

100

90

80

70

60

t.s 50 > o E 11.1

cz: 100

90

80

70

60

50

40

20

• Synthetic wastewater (A) • Industrial wastewater

25 50 75 100 200 300 400

Adsorbent dose, mg/SO m L

Fig. 8-Removal of Hg(lI ) from synthetic and industrial wastewaters - (A) effect of pH (B) effect of adsorbent dose.

solid-solution interfaces during the adsorption . Solutions of NaN03, NaCI , Na2S04, HCI , HN03,

H2S04, CH)COOH, NaCI-HCI, NH4NO)-HNO) were evaluated for the extraction of Hg(II) from the PTCH into the aqueou s phase again. The desorption results are presented in Table 4. The hydrogen ions from HCI easily displaces Hg(II) ions bounded to PTCH during the extraction stage. An efficiency of 94.8% was obtained by using 0.5M HCI solution and is therefore suitable for the extraction of Hg(II) into the aqueous phase.

The effect of pH and adsorbent dosage on Hg(II) adsorption by PTCH from synthetic and industrial wastewaters are shown in Fig. 8. Composition of synthetic and industrial wastewaters are given in Table 5. Since the industrial wastewater contains very low amount of Hg(II) ( 1.6 mg/L) , it was spiked with Hg(NO)2 so that the total concentration of Hg(II) was

Table 5-Composition of syntheti c and indust ri al was tewaters

Synthetic wastewater

Industrial wastewater

Composition, mgt L

Hg : 25; Mg : 25; Ca : 40; Na : 25 ; K : 40: NH. : 20; S04 : 100; H2POJ : 100 ; CI : I DO: CH,COO : 60.

Hg : 1.6; Pb :2.7; Cd : 0.5; Mg : 25.6: Ca : 41.2 : Na : 280.8; P04 : I 0.9: 0 , : 16.5: HJ :20.7: CI :398.39; BOD :58.4; COD : 138.6: SS : 351\ .7

10 mgiL which re fl ects the Hg(II) content of certain industrial effluent such as vinyl chl oride and battery manufacturing wastewaters24. The percentage adsorption increases with increase of pH . In both cases, the maximum adsorption was observed at pH 10.0. The amounts of adsorbent dosage for the complete removal of Hg(II) from 50 mL synthetic and industrial wastewaters were found to be 200 and 100 mg, respectivel y; which are in good agreement with those obtained form the batch experiments menti oned above.

Conclusions Polysulphide treated coconut husk is a suitab le

adsorbent for the remova l of Hg(Il) from aqueous system/industrial effluents. Sorption of Hg(I J) is pH dependent and the best results are obtained in the pH range 5.5 - 10.0. The equilibrium data f it Langmuir and Freundlich adsorption isotherms and thermodynamic parameters have been calcul ated. The overall process of adsorption reported here, is shown to be endothermic and spontaneous in nature. The extraction of Hg(II) is easy us ing 0.5 M HC!. Quantitative removal of Hg( Il) from syn th eti c and industrial wastewaters confirms the va lidity of results obtained in batch mode studies. Thi s adsorbent would be useful for an economic treatment of wastewaters containing Hg(U) as coconut hu sk is a cheap and easily available support material (w ith no cos t) for polysulphide.

Acknowledgement The. authors are thankful to the Head, Department

of Chemistry, University of Kerala, Thiruvananthapuram for providing laboratory facilities.

References I Ferro- Gareia M A, Rivera V J, Rodriguez G J & Bauti sta T I,

Carbon, 26 ( 1986) 363. 2 Srivastava S K, Tyagi R & Pant N. Water Res. 23 ( 1989)

1161. 3 Nerell T P & Zeroni an S H, Cellulose ChemistI'.\' al/d its

applications (Ellis Herwood Ltd . Chichester. Engl and ). 1985.

SREEDHAR et al. : POLYSULPHIDE TREATED COCONUT HUSK 285

4 Benzon J A & Valase J R, Coconut production and utilization (PCRDF, Pasig, Philippines) , 1982.

5 Manas A E, Romova S & Torres A S, Utilization of coconut husk for water .-!reatment and recovery of heavy metals (Terminal Report , FORD PC) 1985.

6 Manju G N, Raji C & Anirudhan T S, Water Res , 32 (1998) 3062.

7 Anirudhan T S, Manju G N & Raji C, J ind Pollution Control, 13 (1997) 19.

8 Knocke W R & Hemphill H, Water Res, 15 (198 1) 275. 9 Meng X, Hua Z, Dermates D, Wang W & Yukuo H, J Hazard

Mater, 57 (1998) 231 . 10 Anirudhan T S & Sreedhar M K, Indian J Chem Technol, 5

(1998) 41. II Klemm W, Sodomann H & Langmesser P, Handbook of

preparative inorganic chemistry, edited by G Breuer (Academic Press, New York) 1963.

12 Feher F & Berthold H J, Z Analyst Chem, 138 (1953) 245 . 13 Jayateerth R M, Ana/l09 ( 1984) 1577.

14 APHA, Standard methods for the examination (~r water and wastewater, 18th edn (APHA, A WWA and WEF, Washington DC) 1992.

15 Lagergren S, Bil K Svenska Ventenskapsekad Handle. 24 (1898), as cited by Wasay et ai, Water Res, 30 ( 1<)96) 1143.

16 Baba Y, Inoue K, Goto A & Nakamori I. Bull Soc Sea WaleI' Sci Jpn, 38 (1984) 137.

17 Kuo S & Lotse E G, Soil Sci Soc Am Proc. 38 (1974) 50. 18 Vermeulen T , lnd Eng Chem , 45 (1953) 1664. 19 Streat M, Patrick J W & Perez M J C. WaleI' Res, 29 (1995)

467. 20 Bhattacharya A K & Venkobecher C. J En llil'OlI Ellg Di ll.

ASCE, 110 (1984) 110. 21 Muller G, Radke C J & Prausnitz J M. J PhI'S Chell/. 84

(1980) 369. 22 Selvaraj K & Chandramohan V, J Sci Illd Res. 57 (1998) 271. 23 Servno V G, Garcia A M, Mansilla A E & Calahorro C V,

Water Res, 32 (1998) I . 24 Harrison RM , Pollution ,' Causes, Effecls allli COlllrol. (The

Royal Society of Chemistry, Cambridge), 1990.