REMOVAL OE HEAVY METALS EROM WASTE WATERS BY

VERMICULITES
E, LVAIEZ-AYUSO AND A , G A R C A - S A N C H E Z *

Department of Environmental Geochemistry, IRNASA, CSIC, Apto 257. Salamanca, Spain

(Received 13 iiiie 2002; Accepied 22 Oelober 2002)

ABSTRACT
The sorbent behaviour of (natural and exfoliated) vermiculite minerals WDS studied wilh respect to metal cations (Ni, Cu,
Cd) commonly present in waste waters of [he metal finishing industry. The Langmuir model was found to describe Ihe
sorption processes well, showing the maximum sorption capadties of natural vermiculite to be high enough (b^^ = 26.0 mg g"
'; bfj = 23.5 mg g"'; b, = 19.3 mg g-') for ts use in metal wasle water purification systems to be feasible. The column studies
performed with this minL-ral show that a large volume of metal solutions with similar concentrations to those usually present
in wasle waters can be purified down to the legal limit of waste. These positive results are ratified when real waste waters
are used al lab-scale as well as at semi-industrial scaie, validating the use of this mineral as a cosl-effei
purify such waste waters.

Keywords: Waste waters, purification, heavy metals, sorption, vermiculite.

INTRODUCTION

commonly present in indusfirial effluents. Maximum sorption

capadfies of 0,17mgg'', 0,40 m g g ' , 1.9 mg g', 5,5 mg g"' and

The natural levels of heavy metals in the environment

35 mg g ' have been quoted for Cd on clinoptilolite [1], felspar

have been greatly increased, mainly in the last decades, due to

[2], kaolin, illite and monfinorillonite [3] respecfively. Similar

different human activities causing, in many cas^s, serious

capacifies are shown by these minerals for Cu with maximum

pollution problems in both aquafic and soil systems. Among

values of 1,4 m g g ' on kaolin, 2,5 mgg"' on illite and 23 mgg"'

ihe numerous sources of heavy metals, industries are at

on mcntmorillonite [3], Other minerals such as pyrophyllite

present, those which contribute most to this pollution,

[4] and wollastonite [5] have also proved their sorbent

particularly the metal firshing sector. This sector is integrated

properties showing respecfively capacifies of 3,3 mg g'' and

by a large number of enterprises which are widely spread

6.5 mg g'' for Ni, In general, the highest sorption capadHes

implying a great environmental risk. In addifion, the majority

correspond to minerals which perform most of the metal

of these enterprises are small-to-medican enterprises (SMHs)

sorpfion process by means of exchange processes. It is well

without the economic and technical capacity needed to

known that vermiculites have high cafion exchange capadfies

appropriately face up to the purification of waste waters

(100-150 cmol(+) kg-') [6, 7| representing, therefore, potential

produced in their industrial activity.

sorbents for metal cations. Some researches have been

purification

system

most

The waste waler

usually employed

by

these

industries is the precipitation technique, Tfus treatment is

performed studying the sorbent behaviour of this mineral

with respect to zinc, cobalt, cadmium and lead [8-liJ. These

generally incapable of attaining purificafion levels within the

studies have demonstrated the capadty of this mineral to sorb

legal limits of waste (0.1-3 mg 1') and also generates a large

heavy metals even in the presence of cafions such as calcium

amount of sludges that is difficult to handle. More eifecfive

and potassium, although the initial sorption capacity is

purificafion systems, such as ion-exchange columns, are only

decreased by these compefing cations, as much as their

suitable for the big enterprises of the sector because of their

fiigh cost,

concentrafion increases. However, in spite of this posifive

behaviour

the

actual

environmental

applications

of

An alternative promising opfion to purify these waste

vermiculites have been extremely limited, being focused on

waters could be the use of low-cost sorbent minerals. Many

agricultural, horficultural and industrial (fire protecfion,

researches have been performed on this matter studying the

plaster board manufacture, lightweight concrete, packing

sorbent behaviour of different minerals with respect to metals

materials, etc.) applicafions, mainly in the exfoliated form

[12-14],
The main objectives of the present work are: 1) to
determine the sorption capacity of ,some vermiculites with
respect to heavy metals (Ni, Cu, Cd) commonly present in the
waste waters of galvanic industries, 2) to explore the
behaviour of this mineral in a continuous flow system, 3) to
characterize the wastes generated in the sorption process and
4) bo explore the possibility of using this mineral to purify real
industrial waste waters.

Tabie 2,

Specific surface area, cation exchange capadty and

point of zero net proton charge of vermiculite

Sample

CEC

SE
1
14.6

Natural vermicul
lili te 5 30

(cmol(+)l<f'Y
76,2
41,7

PZNPC
6.7-7.2
8,3-8,4

'(cmol(4) kg-') - meq (100 g)''

MATERIALS AND METHODS

Sorbent Minerals

permanently charged sites is mainly characterized by

elecfrostatic interactions, therefore, metal cations bound in

In this study two kinds of vermiculite were used: a

such a way can readily be exchanged by other caHons. The

natural vermiculite coming from Askos (Greece) and an

sorption of heavy metal cations by sites of variable charge

exfoliated vermiculite coming from South Africa.

(surface hydroxyl groups) is governed by covalent binding

and therefore chemically specific. In titrations it is assumed
that H ' (or OH") are only reacting with variable charges

Chemicals

present on the surface of minerals, but, actually, other

The metal ions stiidied were Ni(II), Cu(n) and Cd(n),

reactions (dissolution, precipitation, formation of dissolved

Synthetic solutions were prepared from standards of 10002

metal complexes and proton/cation exchange) can occur that

mg r^ (Panreac) and pH was adjusted using HNOj and

compete with the surface for these ions [20], For this reason

NaOH,

the tifration curves and the PZNPC obtained from them are
more accurately a measure of H^ (or OH") adsorption

Char

zation of Mineral Samples

The mineral samples were characterized by X-ray

Batch Sorption Studies

diffraction (XRD) (Table 1), the semiquantitative analyses

were based on diffractometer traces of unoriented powder

Batch sorption studies were carried out in centrifuge

samples using the methods employed by Schultz [15] and by

tubes by subjecting a given dose of mineral to a period of

Islam and Lotse [16], The specific surface area (SE) (Table 2)

shaking with 25 ml of solution on a vertical rotary shaker (5D

was determined using the single point BET N2 adsorption

turns min-') in a chamber set at 22 "C. The mineral was

method, although the values obtained from this method only

separated using a centirifugation step (4000 rpm) over 10 min.

refer to the outer surface. The cation exchange capacity (CEC)

Metal concentrations in the supernatant were determined

was determined by the ammonium acetate method 17] (Table

with a VARAN SPECTRA AA-2002 Atomic Absorption

2), The point of zero net proton charge (PZNPC) (Table 2) was

Spectrophotometer, All assays were carried out in triplicate

determined using a batch titration technique following the

method used by Van Raij and Peech [18], This point of zero
charge theoretically refers to the pH at which proton charge

The stiidies with synthetic solutions were perfonned

using the following conditions:

(OH) developed from adsorption of H* or OH" is equal to zero,

considering the contribution of variable charge apart from the

Kinetic Sfridies

structural charge [19], The binding mechanism of cations on

Mineral dose: 10 g Y'; mineral particle size: < 0.1 mm;
initial metal concenfradon: 100 mg I"'; pH: 5 for Cu and 6 for
Ni and Cd; shaking time: 0,5, 1,2, 4 and 6 h.
Sorption Is other
Natural

Exfoliated vermiculite
Vermicubte*
100
' Transformed by a sudden heating (500 C) to a mini
A of basal spacing.

culite dose: 2,5 g I"' and exfoliated

initial metal concenfration; 10-100 mg 1'; pH: 5 for Cu and 6

for Ni and Cd; shaking time: 6 h. The dose used for each
metal-mineral system is appropriate to obtain the isotherms in
the range of concenfrations used.

their leachates obtained from DIN 38414-S4 leaching test [21]

Effect of pH

as the European Proposal for landfiiling of wastes [22]

Natural vermicuUte dose: 5 g Y' for Ni and 2,5 g V for

established. This leaching lest takes a liquid (deionized waler)

Cd and Cu; mineral particle size: < 0,1 mm; initial metal
concentration: 100 mg T'; pH range: 3-6 for Ni and Cd and 3-5

to solid ratio of 1 Utre:100 g, which is slowly stirred at 0,5 rpm

ysed.

for Cui shaking time: 6 h.

Analysis of m

s performed by AAS,

Effed of Mineral Particle Size

Natural vermicuUte dose: 5 g T' for Ni and 2,5 g T' for
Cd and Cu; mineral particle size: < 0,1, 0,2-0,6, 0.6-1.0 and 1,02,0 mm; initial metal concentration: 100 mg 1'; pH: 5 for Cu
and 6 for Ni and Cd; shaking time: 6 h,

The results (Figure 1) show that in both vermiculites the

majority of the sorption process (80-95%) takes place in the
first minutes (30 min) the remainder being completed in a

Contin

slower second step. The global process, mainly in the case of

s Flow Studies

exfoliated vermiculile, even lasts several hours. Taking into

ing
percolahng tubes filled with 40 g of mineral with a particle

account that ion exchange reactions at surface sites exposed to

si2e ranged between 0,2 and 1,0 m m . The solutions (initial

first, the main contributors for sorption on these minerals,

metal concentration: 50 mg 1'; pH: 5 for Cu and 6 for Ni and

diffusion through the mineral interlayer has to be the process

Cd) were p a s s e d t h r o u g h the columns (height: 6.8 cm;

which Umits the rate of the sorption process.

Contin

solution are extremely fast and that these reactions are, at

diameter: 2,9 cm; bed-volume, volume of vermiculite bed.

The cations implicated in the ion exchange processes

(BV):
pump. Effluent samples were collected periodically with a

are Mg and Ca, being respectively resposible for about 92%

fraction coUector (Foxy Jr., ISCO) and analysed by AAS.

and 16% of that of exfoliated vermiculite, the presence of K

and 8% of the CEC of natural vermicuUte and for about 84%

and Na as exchange cations is negUgible,

te Characterization
Sorption Isotherms
ated from contin
ding to the chem

N a t u r a l veimiculi!

The sorption isotherms are shown

Exfoliated vermiculite

Figure 1, Kinetic curves of vermicuUte samples (A Ni, B.Cu, O Cd).

in Figure 2,

:e samples {A Ni, ,Cu, O Gd).

'e 2, Sorption isother

The sorpfion data were fitted to the Langmuir equafion, X/M

= (KbCe)/(l+KCe), and to the FreundUch equation, X/M = K
Ce"^ where X/M is the amount of solute retained per unit
weight of the sorbent, Ge is the equilibrium concentration of
solute remaining in the solution, K (Langmuir) is the
equilibrium constant (affinity term), b represents the
maximum amount that can be sorbed, K (Freundlich) is a
measure of the sorpfion capacity for a concentrafion Ce equal
to 1 and n represents the extent to which sorption is
Table 3. Char

dependent on the equilibrium concentrafion.

The Langmuir and Freundlich characteristic parameters
and the degree of correlation of the sorpfion data with respect
to these equations are given in Table 3 and 4 respecfively.
Both models describe the sorption processes well.
For all sorpfion processes the n parameter values are above 1,
showing a reducfion in the sorpfion degree as the metal
equilibrium concentrafion increases, as is generally tound for
metal sorption on minerals.

ristic parameters and determinafion coeffident of the experimental data according to the Langmuir equation.
K
Ims"'

b
Sample
Natural
vermiculite

Exfoliated

Metal

mg s"'

R-

Ni

19,3

0,329

0,398

0.9776 ( p < 0.001)

Cd

23,5

0,209

0.403

0.9982 ( p < 0.00!)

Gu

26.0

0,409

0.294

0.9933 ( p < 0.001)

Ni

5,91

0,101

0,716

0,9975 ( p < 0,001)

Cd

8,42

0.075

0,179

0.9641 ( p < 0,001)

Cu

8,60

0,135

1.58

0.9953 ( p < 0,001)

Table 4. Characterisfic para

equafion.
Sample

t of the experimental data according to the Freundlich

K(B')

R-

Ni

5,14

9,01

0,9853 ( p < 0.001}

Cd

4,04

9,57

0,9734 ( p < 0.001)

Cu

3,27

8,60

0,8925 (p < 0,001)

Ni

5,65

3.15

0.9819 ( p < 0,001)

X oiae

Cd

2,77

2,18

0,9844 (p<: 0.001)

vtiiiiiLulili:

Cu

4,21

4,80

0.9586 (p < 0,001)

Natural vermiculite

Metal

According to b parameter (mmol g"') sorpfion on both

vermiculites occurs following the sequence: Cu > Ni > Cd,
This order does not agree accurately with that established by
the ionic potenfials (ZVr) of cations: Ni > Cu > Cd (ZVrNi =
5.71, ZVrt^ = 5.48, ZVrcj = 4,21; [23]), as would occur if only
electrostatic forces were Involved in sorption. Therefore, the
retention sequence suggests that in addition to cation
exchange process other mechanisms take part in the global
sorpfion process. The higher relenfion of Cu could be due to a
higher contribution of variable charges in its retention; these
type of sites react preferably with cafions which readily
hydrolyse as is the case of Cu (log Kc = 6.3, log K^, = 4.7,
logKcj = 4.1; K:firsthydrolysis constant).
For all the metal cations, the sorpfion capacity of natural
vermiculite is much higher than that shown by exfoliated
vermiculite, which has to be mainly related to its much higher
CEC,
The sorption capadfies shown by natural vermiculite
are high enough to consider its use in metal waste water
purificafion systems. These capacities are much higher than
those exhibited by cUnoptilolite, felspar, kaolin, illite,
pyrophyllite and wollastonite [1-5] showing, therefore, a
much greater potential to be effecfively used in purificafion
processes. Although montmorillonite has similar sorption
capacities [3] to that shown by natural vermiculite its
characteristics, low hardness of aggregates and swelling

phenomena, make its use impossible in continuous fiow

systems without being previously subjected to a heafing
treatment; a clear disadvantage with respect to vermiculite
sii^ce this treatment lessens the mineral sorption capacity [24]

Effect of pH
The results obtained studying tlie effect of pH on the
sorpfion capacity of natural vermiculite are shown in Figure
3. For all the metals a decrease in sorption is produced with
the decrease in pH. In general, the tendency of protons to
react with sorption permanent charge sites is very limited,
only monovalent cations can be substantially subjected to
hydrolyfic exchange, mainly at low electrolyte concentrafions
[7]. Therefore, in this mineral H* can mainly compete with
metal cafions for sorpfion variable charge sites. The effect of
pH is much more marked for Cu, this behaviour is related to
the greater contribufion of variable charges in its sorpfion,
due to its facility to hydrolyse.
Particle Siz
The results obtained studying the effect of miner;
particle size on the sorpfion capacity of natural vermiculil
are shown on Figure 4, A slight decrease in sorption i!

pH
Figure 3,

Effect of pH on sorpfion on natural vermiculite sample (A Ni, H.Cu, O Cd),

Figure 4, Effect of parficle size of nafijral i

ffl 10-2,0 mm).

produced with the mineral particle size increase. This

behaviour seems to be related to crystal breaking produced as
the mineral parficle size decreases, it causes a larger number
of broken bonds and consistently a larger number of variable
charge sites. The greater influence of parficle size on Cu
sorpfion corroborates this idea, suggesting again a greater
contribution of variable charges in its sorpfion,

iorpfion ( n<O,l mm, S 0.2-0,6 nr

being at 21 bV about U.I mg I , atier, ttie isii concentration

increases gradually to reach at 352 BV the most sfinngent limit
established by European environment legislation (2 mg I"'),
These results indicate that the vermicuiite column can purify
a great volume of Ni solution down to the legal limit of waste,
The total amount of Ni removed (14.6 mmo!) during the 385
BV percolated corresponds to a sorpfion capadty of 0,730 meq
g'. The total amount of Mg eluated was 23.2 mmol and the
total amount of Ca eluated was !.99 mmol which correspond
Continuous Row Studies
to a release of 1.16 meq g"' of Mg and 0,100 meq g"' of Ca, The
total amount of Mg + Ca released (1.26 meq g"') was more
The sorbent behaviour of natural vermiculite was also than the removed amount of Ni. This imbalance in the cafion
studied using a continuous flow system. This system appears,
exchange process couid be attributed to an addifional
at first, as the most appropriate to perform purification
exchange process related to Na (> 0,025 mol 1') present in the
processes. On the one hand, a continuous displacement of the percolated solufion, due to the addifion of NaOH canied out
sorption equilibrium (in the way of sorption) is produced by
to adjust the solution pH, The exchangeable Mg of vermiculile
this method, such displacement could allow higher metal is readily replaced by Na [25],
retentions to be achieved. On the other, the continuous
The chemical data obtained for percolating a Cd
method allows the global puriHcafion process to be performed
solufion through a vermiculite bed are shown in Figure 5. The
by a single step while the batch method needs several stages metal concentration also remains very low along a high
(shaking, settling and/or filtration) carried out repeatedly,
number of BV, attaining at 221 BV the most stringent limit (0,1
requiring, therefore, more time and space to perform the mg !') established by law. After, the Cd concentration
purification process in an effective way. The metal increases slightly reaching values of 1-2 mg 1"^ at about 300 BV
concentrafions used in the percolated solufions were similar
and undergoing subsequently a mudi more rnarked mcrease,
to those present in rinse waste waters of the electroplating
In spite of the low limit of waste fixed by law for Cd the
process,
vermiculite column is able to purify a large volume of Cd
The results obtained for percolating a Ni solution solufion down to this value, I ne total amount of Cd removed
through a vermiculite bed are shown in Figure 5, The melal (7,51 mmol) during the 389 BV percolated corresponds to a
concontrafion remains very low along a high number of B v, sorption capacity or 0.3/6 meq g , ine total amounts of tvlc

50

50

100

100

150

200

250

300

350

400

150

200

250

300

350

400

concentration in the effluent ( Ni, Cd or Cu (mg I''), O Mg (mmol I"'), A Ca (mmol 1"^

and Ca eluated were respectively 14,3 mmol and 1.60 mmol

which correspond to a release of 0,716 meq g'' of Mg and 0.080
meq g'' of Ca, The total amounf of Mg + Ca released (0.796
meq g^') doubles the removed amounf of Cd, In this case the
apparent imbalance in the cation exchange process could also
be explained by an additional exchange process related to Na
present in the percolated solution.
The results for the treatment of Cu solution with
5, The letal
bed
concentration surpasses the most stiringent limit established
by law (0,2 mg 1') at 85 BV, This purified volume is relatively
low with respect to the other metal cations studied. After this
volume, the Cu concentration increases more quickly. This

landfill Proposal [22) are shown in Table 5, According to this

characterization the wastes are classified as hazardous. The
wastes of the sorption process of Ni and Cd on vermiculite do
not comply with the eluate criteria (their eluate concenti^ations
are higher than the maximum values fixed for hazardous
wastes), therefore, these wastes have to be treated prior to
landfill. In the case of the waste of the sorption process of Cu
on vermicuUte its eluate concentration is in the range fixed for
hazardous wastes and could be disposed in a hazardous
waste landfiU without any previous treatment. Alternately
these wastes could be sent for joint disposal with munidpal
wastes, if the load index of waste (amount of hazardous waste
which can be disposed in relation to a fixed amount of
munidpa! waste) allows it.

lesser effectiveness in the purification process has to be related

Although the purification process pased on metaJ

to the suppression or reduction of Cu sorption on variable

charge sites at the particle size used in columns. The total
amount of Cu removed (10.0 mmol) during the 267 BV

EEC LandfiU Proposal [22] as hazardous, as are the majority

percolated corresponds to a sorption capacity of 0.500 meq g'.

of sludge wastes generated after the physicochemical

The total amount of Mg eluated was 13.9 mmol and the total

treatment of metal finishing rinse waste waters [26], the

amount of Ca eluated was 6,54 mmol which correspond to a

former are much easier to handle and also potentially easier

release of 0,696 meq g ' of Mg and 0,328 meq g-' of Ca, The

to stabiUse than sludges wastes.

total amount of Mg + Ca released (!,02 meq g"') also doubles

the removed amount of Cu, An additional exchange process
related to Na present in the percolated solution could explain

Real Waste Water Purification at Lab-scale

again the apparent imbalance in the cation exchange process.

The responsability of Na for the additional exchange

The sorbent behaviour of natural vermicuUte in a

process was verified by XRD, A conversion in the basal

continuous flow system was also studied with respect to real

spacing from 14 A (typical of Mg-vermicuUte) to 12 A (typical

waste waters. The waste waters chosen to carry out the

of Na-vermiculite) is produced in vermiculite samples after

purification process come from a Ni electroplating process

the sorption processes, nevertheless the conversion degree is

which uses the typical electroplating solution (Watts bath).

sUghtly different for each mineral sample.

The nickel plating is by far the most usual process in relation

When these spent columns are subjected to an

to copper or cadmium plating. The sample of Ni waste water

extraction treatinent with CatNOj)! 4H2O 0,5 M (operating

used was supplied by Industria Galvnica Dalla Torre S.p,a,

conditions: soUd/caldum nitrate solution: 1/2,5, shaking

(Treviso, Italy). The concentration and pH of this sample are

time: 24 h) about 80-95% of metal sorbed in each case is

the same as those of the synthetic solution previously used.

extracted, proving that at the mineral particle size used in

The results obtained are shown in Figure 6. The

columns the process dearly responsible for metal sorption is

purification level obtained for the real waste water is similar

the exchange process, even in the case of Cu.

to that obtained for the synthetic solution. At 225 BV the Ni

concentration in the effluent remains below 0,2 mg 1"', far
from the limit of waste established by law (2 mg I').

Waste Chai

Therefore, vermiculite can be used in a continuous system as

-imental results obtained from the DIN 38414[21] and the values established by the EEC

Table 5,

an effective treatment to purify waste waters produced from

the nickel electroplating process.

Waste characterizat
Chemical leachate parandeters (m^

;!')

Ni
Ve
WasI.p.

Cd

5,55
3,99

Ve

3.67

Ve
WasI:e dassification

Hilzardous wa.stc range

Inirt waste

Cu

0.4-2,0

0,1-0,5

2-10

25

50

75

100

125

150

175

200

225

Figure 6, Eiufion curve of an industrial waste water containing Ni filtered through the natural vermiculite sample; G:
concenfi-ation in the effluent.

and the sorbent capacity shown by the mineral at the

Waste Water Purificafion at Semi-industrial Scale

percolated volume (6,8 mg g'', tar from its maximum sorption

The posifive results obtained from the lab-scale study
need to be validated by m e a n s of semi-industrial tests to
definitively rafify the worth of the vermiculite column system
as waste water purificafion system.
Basically, the p r o t o t y p e used in this study w a s
composed of a tank where the waste waters are collected, a
tari for p H conditioning, when necessary, and a vermiculite
column with 0,40 m of height and 0,60 m of diameter. The
amount of vermiculite used in the column was 100 kg and its
parfide size ranged between 0,25 and 1,0 mm. The mineral in
the column w a s separated in 8 layers ot 0.05 m by geotextiles
with the aim of avoiding the preferential ways formation,
having instead a laminar flow through all the column.
The waste waters chosen to perform this semi-industrial
study came from the typical Ni electroplating process. The
waste water samples used in the purification tests had an
average concentrafion of 178 m g I ' , being ranged between 95
and 265 m g I ' , and a p H about 6,5. High and variable metal
concentrations were chosen to perform this study in order to
test the system in extreme conditions. The flow rate used to
percolate the waste waters through the mineral column was
85 1 h ' .
The results obtained are shown in Figure 7, After the
percolafior
the effluer

0 1 of V

/ater the Ni i

raHoi

remained below 0.4 mg T', Taking into account

ion level (still tar from the limit of waste: 2 mg 1-')

capacity), the vermiculite column system appears to be an

effective purificafion treatment. This treatment has clear
advantages with respect to the physicochemical treafinent of
purification level down to the legal limit of waste and with
the wastes being easy to handle,
CONCLUSIONS
Heavy metals are retained on vermiculites according to
the following sequence: Cu > Ni > Cd, differing from the
theoretical established according to the ionic potential of
cafions (Ni > Gu > Gd), The greater sorption of Cu suggests
that, apart from electrostatic forces between this metal cation
and permanent charges, variable charges also contribute to its
sorpfion in an important extent. The greater influence ot pH
and mineral particle size in Cu sorpfion is also consistent with
the larger participation of variable charges in its retention.
The sorpfion capacities of natural vermiculite sample (b,;^ =
26.0 mg g': b . 23.5 mg g': b, - 19.3 mg g') arc much
higher than those of exfoliated vermiculite sample (b^^ = 8.60
Tig g"'; bc = 8,42 mg g'; b^, = 5,91 mg g'), this behaviour is
related to its higher GEC,
Nat

;:uUte

:an purify, in an effecfive

way, a large volume of Cu solufif I and even greater of Cd

and Ni solutions with similar mel II concentrafions to those
present in rinse waste waters. This i

500

1000

1500

Figure 7. Elufion curve of the semi-industrial purificafion pro

sample; C: concentration in the effluent.

2000

2500

3000

3500

4000

of Ni waste water filtered through the n

purificafion level when real waste waters coming from thi

ACKNOWLEDGEMENTS

nickel electroplating process are tested. The positive result;

obtained from lab-scale sorpfion sfiidies are ratified wher
purification

processes at a semi-industrial

This work was carried out under Project BE-97-5093 of

scale ar(

Brite Euram m Programme, Directorate General Xn, U,E. The

performed, validating the use of this mineral as a cost-

authors thank Dr. M. Pafi-onis (IGME, Greece) for supplying

effecfive treatment to purify such waste waters.

the natural vermiculite sample and Industria Galvnica Dalla

The spent filters were characterized as hazardous according tc

Torre, S.p.a. (Treviso, Italy) for their help in obtaining the

the method DIN 38414-S4.

waste water samples.

Singh B,, Alloway B,J. and Bochereau F.J.M,, Cadmium sorption behavior of
Sd. Plant. Anal. 31, 2775-2786 (2000).
Pelte S., Flamant G., Flamand R,, Gauthier D,, Bche E, and Berjoan R., Effect

nd synthefic zeolites. Commun. Soil

rmal fi^eafinent on feldspar sorpfive

properfies: idenfificafion of uptake mechanisms. Min. Eng. 13, 609-622 (2000),

Fanah H,, Hatton D, and Pickering W,F The affinity of metal ions for clay surfaces. Chem. Geoi., 28, 55-68 (1980).
Scheidgger A,M Fendorf M, and Sparks D.L., Mechanisms of nickel sorption on pyrophyllite: macroscopic and
microscopic approaches. Soil Scl Soe. Am. j . . 60,1763-1772 (1996).
Sharma Y.C, Gupta G.S,, Prasad G, and Rupainwar D,C,, Use of wollastonite in the removal of Ni(n) from aqueous
solufions. lViT(irr,/lir nnd Soil Pollut.. 49, 69-79 (1990),
Grim R,E., Applied Clay Mineralogy. McGraw-Hill Book Company, Inc., New York, USA. (1962),
McBride M,B., Environmental Chemistry of Soils. Oxford Universitj' Press, Inc, New York, USA. (1994),
Tiller K.G, and Hodgson J,F., The specific sorpfion of cobalt and zinc by layer silicates. Clays Clay Miner., 9. 393-403 (1962)
Ziper C, Komameni S, and Baker D.E., Specific cadmium sorpfion in relation to the cr}'stal chemistry of clay minerals Soil
Scl Soc. Am. ; 52,49-53 (1988).

Das N,C. and Bandyopadhyay M., Removal of lead by vermiculite medium. Appl Clay Sei., 6,221-231 (1991),
van Bladel R., Halen H. and Cloos P., Calcium-zinc and calcium-cadmium exchange in suspensions of various types of
clays. Clay Miner.. 28, 33-38 (1593),
Dickson T Vermiculite; little growth envisaged, nd. Miner. (London), 179,47-50 (1982).
Liu I,J Vermiculile fit perlite for animal feedsfiiff and crop farming, nd. Miner. (London), 262, 43-49 (19B9).
Lin I., Perlite & vermiculite; crudely speaking the potential is good. bid. Miner. (London), 368, 55-57 (1998),
Schultz LG,, Quantitafivo interpretation of mineralogical composition from X-ray and chemical data for the Pierre Shale.
U.S. Gra/, Surv. Prof Paper, 391-C (1965).
Islam A.K.M,E, and Lotse E,G,, Quantitative mineralogical analysis of some Bangladesh soils with X-ray, ion exchange and
selective dissolution techniques. Clay Miner., 21,31-42 (1986),
Tan K,H Soil SampUng, Preparation and Analysis, Marcel Dekker Inc, New York, USA. (1996),
van Raij B. and Peech M., Electrod-iemical properfies of some oxisols and alfisols of the tropics. Soil Scl Soc. Am. Proc. 36,
587-593(1972).
Bolan N,S., Naidu R,, Syers ],K, and Tillman R.W,, Surface charge and solute interactions in soils. Adv. Agron., 67, 87-140
(1999).
Lewis-Russ A., Measurement of surface charge of inorganic geologic materials: Techniques and their consequences. Adv.
Agron.. 6,199-2A3 {1991).
DIN 38414-S4, Determination of leaehability by water (Si), group S: sludge and sediments, German standard methods for the
examination of water, waste water and sludge. Insfitut fur Normung, Berlin, (1984),
Proposal (93/C 212/02) for a Council Direcfive on the landfilUng of wastes, COM (93) 275 nal-SYN 335, DOCE 5-8-93,
European Economic Community, Brussels, (1993),
Huheey H,E,, Inorganic Chemistry: Principles ofStriieture and Reactivity, Harper and Row, New York, USA, (1972),
De Boodt M,F,, Application of the sorption theory to eliminate heavy metals from waste waters and contaminated soils. In:
Interactions at the Soil Colloid-Soil Solution Interface. Bolt G.H,, De Boodt M,F Hayes M.H.B. and McBride M.B, (eds,), NATO
ASI Series (Series E; AppUed Sdences-Vol 190), Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 293-320
(1991).
Walker G,F. and Milne A,, Hydrafion of vermiculite saturated with various cations. Trans. 4"* Int. Congr. Soil Scl, 2, 62-67
(1950).
Viguri ) . , tbae2 R., Andrs A., Ortiz I. and Irabien A,, Environmental characte riza fion of metal finishing sluges. Environ.
TeehnoL, 20,171-180 (1999),

Copyright of Environmental Technology is the property of Taylor & Francis Ltd and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.