Process Biochemistry 39 (2004) 833–839

Use of a polarographic method to determine copper, nickel and zincconstants of complexation by extracellular polymers extracted

from activated sludge

G. Guibaud∗, N. Tixier, A. Bouju, M. Baudu

Laboratoire des Sciences de l’Eau et de l’Environnement, Faculté des Sciences et Techniques, 123, avenue Albert Thomas, 87060 Limoges Cedex, France

Received 25 November 2002; received in revised form 28 March 2003; accepted 6 May 2003

Abstract

The constants of complexation of extracellular polymers (ECPs) extracted from sludges for three metals, Cu, Ni and Zn at pH 7, weredetermined using a polarographic method (stripping mercury dropping electrode mode (SMDE)). The curves obtained were exploited accordingto Chau’s or Ruzic’s method. The influence of the biochemical composition of ECP on the binding of Cu, Ni and Zn was investigated on sixsolutions of ECP. Polarography in SMDE mode is a simple method to determine complexation constants of ECP extracted from activatedsludges. For each, the number of sites of complexation increased in the following order: Zn� Ni < Cu. The low number of binding sites ontoECP for Zn did not allow the determination of a constant of complexation. For ECP studied, the constant of complexation was always higherfor Cu than for Ni. A statistical study showed that the higher the content of proteins, humic acids and polysaccharides contained in ECP, themore they were able to bind high quantities of Cu. For Ni, the parameters of complexation were linked to the amount of uronic acid. Theseresults suggest that carboxylic groups play a major role in Ni and Cu complexation by ECP at pH 7.© 2003 Elsevier Ltd. All rights reserved.

Keywords: Activated sludge; Complexation; Polarography; Extracellular polymers; Heavy metal

1. Introduction

The activated sludge process is commonly used for thetreatment of wastewater. The microbial population of thisbiological process is organised in flocs which are composedof microorganisms, extracellular polymers (ECPs), mineralparticles (clays) and divalent cations[1]. Protozoa and fila-mentous bacteria are also closely associated with the flocs.

ECPs have three different origins. These are generatedeither by cellular metabolism or by cellular lysis of the mi-croorganisms; they can also originate from compounds car-ried in wastewater[2]. ECPs are compounds having a highmolecular mass (molecular weight: >10 000 Da)[3]. Thedifferences in the composition of ECPs depend on sludgeorigin and on the different processes of biological treatment[2,4], and the extraction yields of different fractions of ECP

∗ Corresponding author. Tel.:+33-5-55-45-74-28;fax: +33-5-55-45-72-03.

E-mail addresses: gguibaud@unilim.fr (G. Guibaud),mbaudu@unilim.fr (M. Baudu).

are influenced by the extraction protocol used[5]. However,five main biochemical compounds are found in polymers:polysaccharides, proteins, nucleic acids, humic acids andlipids.

Effluents are often contaminated by heavy metals whichoriginate from industries, rainwater or domestic effluents.Heavy metals are often responsible for failures in the bio-logical units of wastewater treatment plants (WWTPs)[6].The sludges play an important role in the extraction of heavymetals from wastewater and ECPs are well known as adsor-bent material towards metals[7,8].

A wide variety of techniques can be employed to deter-mine the complexing capacity of a macromolecular ligand[9]. Among these techniques, voltammetric methods havebeen most widely used. Electroanalysis is frequently consid-ered because of its intrinsic ability in distinguishing betweenthe free and the bound metal ions[10]. To our knowledge,such a method has never been used to determine complex-ation parameters of ECP extracted from activated sludges.

The aim of this study was to use polarography (strippingmercury dropping electrode mode (SMDE)) to determine

1 0032-9592/$ – see front matter © 2003 Elsevier Ltd. All rights reserved.2 doi:10.1016/S0032-9592(03)00190-0

834 G. Guibaud et al. / Process Biochemistry 39 (2004) 833–839

the constants of complexation of six ECP samples extractedfrom different WWTPs with three metals: Cu, Ni and Znat pH 7. The influence of the biochemical composition ofECPs on the binding of Cu, Ni and Zn was investigated.

2. Materials and methods

2.1. Activated sludge samples used for ECP extraction

2.1.1. Methods of characterisationMixed liquor suspended solids (MLSSs) were determined

according to Ref.[11] and mixed liquor volatile suspendedsolid (MLVSS) according to Ref.[12]. Sludge settleabilitywas investigated through the determination of sludge volumeindex (SVI), zone settling velocity (ZSV) and settling initialflow (Fo, determined as ZSV× MLSS). SVI was determinedin a 1-l glass column after 30 min of sedimentation and ZSVwas determined in a 2-l cylindrical column. It correspondsto the follow-up of the interface of water/sludge during thefirst minutes of sedimentation. A microscopic examinationof sludge samples (magnification: 100× ) was carried outto highlight an eventual presence of filamentous bacteriaproliferation.

Table 1shows the main characteristics of activated sludgesamples used for ECP extraction.

2.1.2. Origin and characteristicsActivated sludge samples A, B, D and E were sampled

at the exit of the aeration tank of different treatment plants.Samples A and B originated respectively from a 180 000and a 280 000 population equivalent (p.eq) plant, working atlow organic load and treating mainly domestic wastewater.Sample D originated from a 20 000 p.eq plant, working atmedium organic load and treating, in addition to domesticwastewater, effluents of a cardboard factory and of a dairyindustry. Sample E originated from a 4000 p.eq plant treatingonly domestic wastewater, working at low organic load andwith a high sludge age. For each sludge samples A, B, D andE, experiments were carried out immediately after sampling.

Sample C had the same origin as sample B but was keptfor 3 weeks after sampling at ambient temperature, with adissolved oxygen concentration maintained at 2 mg l−1 bycontinuous aeration and supplied in a discontinuous wayonce per day (10 ml (l of sludge)−1) with a mixture ofthe following composition: 0.175 g (l of glucose)−1, 0.019

Table 1Characterisation of sludge samples used for ECP extraction

Sludge sample MLSS (g l−1) MLVSS (%) SVI (ml g−1) Fo (g m−2 s−1) Remarks

A 2.40 72.0 60 –B 5.27 62.9 85 2.65C 5.75 64.1 79 3.09D 7.67 86.6 93 0.89 Small filaments, supernatant troubleE 3.73 61.9 166 1.11F 3.10 70.3 303 0.48 Thick mixing up filaments, cloudy supernatant

mol (l of H3PO4)−1, 0.140 g (l of NH4Cl)−1 and 3.25 ml(l of meat extract)−1. This solution had a chemical oxygendemand (COD) of 2.1 g (l O2)−1.

Sample F had the same origin as sample E but was con-ditioned to develop an important filamentous bulking. A 8-lreactor continuously supplied by the solution previously de-scribed and an aerator maintaining the dissolved oxygen ata constant value of 2 mg l−1 were used. The flow modechosen for supplying sludge was 0.25 l h−1.

2.2. Extraction and characterisation of ECP

Extraction of ECP from sludges was carried out usingsonication coupled to the use of a resin and centrifugationaccording to Ref.[13]. Prior to extraction, the sludges wereconcentrated (about 10 times) using an MR 23i (JOUAN)type centrifuge at 4500 rpm for 10 min. The residues wererecovered and suspended again in ultrapure water. 50 ml ofconcentrated sludge was then sonicated with a sonopuls GM70 (BANDELIN) probe, working at 40 W for 2 min. Thistime allowed a maximum of polymer to be extracted, withoutdeteriorating the integrity of bacteria cells. The sludge wasthen treated with a cation exchange resin, a DOWEX 50*8(FLUKA). 300 ml of sonicated sludge and 180 g of resinwere mixed in 500-ml nalgen flasks at 200 rpm on a digitalKS 501 (Kalabortechnik) oscillating table for 2 h and at4◦C. After centrifugation of the sludge–resin mixture (5000rpm for 30 min at 4◦C), the supernatant containing ECPwas recovered by aspiration with an automatic pipette. Thesolutions of ECP were stored at−18◦C before use. Theyield of extraction is determined by the following ratio: dryweight of ECP/dry weight of sludge. It gives about 2 or 3%according to the sample investigated.

Dry weight (at 105◦C), volatile dry weight (at 550◦C)and total organic carbon (TOC) (using a Dorhmann Apollo9000 TOC-meter) were determined for solutions of ECP.The composition of ECP was determined using colorimetricmeasurements: protein content[14], humic acid content[15],sugar content[16], lipid content[17], uronic acid content[18] and nucleic acid content[19].

2.3. Determination of the concentration in ligands and theconstants of complexation

The determination of the parameters of complexation at22± 2◦C and pH 7 was carried out by measuring the con-

G. Guibaud et al. / Process Biochemistry 39 (2004) 833–839 835

tent of free metal in solution of polymer after addition ofCu, Ni or Zn. Investigations on free metals were carried outby polarography, using SMDE. A 626 polarecord polaro-graph coupled to a 663 VA stand of METROHM was used(time of drop fall: 1 s) according to Perret et al.[20]. Themeasurements were carried out in a non-complexant buffer(HEPES buffer, 5× 10−2 M; Sigma Aldrich)[21].

Operating conditions were as follows:

– In the measuring cell at 22± 2◦C:– 10 ml of ultrapure water.– 10 ml of HEPES buffer (10−1 M) at pH 7 (Sigma

Aldrich).– 1 ml of KNO3 (1 M) as electrolyte.– 0.2–1 ml of solution of ECP according to the metal

investigated, which means from 0.4 to 7.5 mg of ECP(expressed as DW).

– The measurement parameters of the polarograph were asfollows:– SMDE mode.– Potential range scanned:

– Cu: from 0.10 to−0.20 V.– Ni: from −0.5 to−1.5 V.– Zn: from−0.8 to−1.2 V.

– Mercury drop size: intermediate.– Time of drop fall: 1 s.

After each metal addition, the solution was subjected toagitation and to nitrogen flow for 7 min followed by 7 minof stabilisation (time necessary to obtain the complexationequilibrium) before measurement.

Analysis of the curves of the polarographic titration al-lowed, from modelling, different parameters characteristicof adsorption, the concentration in ligands (or quantity ofpotentially active sites) and the conditional constants of for-mation of the complex to be determined[22]. The theoreti-cal study of the complexation equilibrium associated to thelaw of mass action can be described as follows, assumingthe formation of a 1:1 complex.

M + L ⇔ ML (1)

K = [ML]

[M][L](2)

where M is the free metal, L the free ligand and ML theligand–metal complex.

The complexation capacity is obtained from the quantityof ligand available.

The concentration of total organic ligand ([L]o) is ob-tained fromEq. (3):

[L] o = [L] + [ML] ′ (3)

where [ML]′ is the complex initially present in ECP, mea-sured from metal considered as adsorbed onto the ECPsample. Desorption of metal initially adsorbed on poly-mers in an acidic environment allowed the value of [ML]′to be obtained. The polymers were therefore shaken for

overnight in nitric acid (0.5%), then centrifuged for 20 minat 4500× g. The nickel, copper or zinc concentrations con-tained in the centrifuged solution were measured by FAAS.The modelling of the complexation capacity curves was car-ried out using Chau’s graphic method and Ruzic’s linearisa-tion method[22] (Appendix A).

The modelling of the curves was carried out accordingto Chau’s and Ruzic’s graphic methods. Chau’s graphicmethod allowed the determination of [L] and [L]o, andRuzic’s method allowed the determination of [L]o andK.

3. Results and discussion

3.1. ECP characterisation

The characterisation of the solutions of polymers is pre-sented inTable 2; it shows a qualitative and quantitativevariation in the composition according to the ECP solutionsinvestigated.

The composition depends on the origin of the activatedsludge from which they were extracted, according to Ur-bain et al.[2]. Proteins, humic substances, uronic acids andpolysaccharides constituted the largest fractions of the solu-tions of ECP extracted. The mineral fraction of the solutionof ECP represented from 19 to 44% of the total of ECPs’dry weight. The mineral fraction is made up of mineral ions(such as Ca2+, Mg2+, Na+, K+, Cl−, SO2−

4 , PO3−4 , etc.),

iron or aluminum hydroxides and metal precipitates whichpresent functional groups able to bind with metals.

3.2. Determination of the concentration in ligands and ofthe constants of complexation

Fig. 1 shows an example of the polarographic titrationcurve obtained with Cu, Ni and Zn for sample E.

The curves of polarographic titration obtained from Cu orNi show two parts. The first, in which part of the metal ionadded in solution is bound by ECP and the second in whichall metal added remain in solution, due to the saturation of

Table 2ECP characterisation

A B C D E F

Volatile DW (%DW) 81 57 56 71 69 69TOC (mg C (g of DW)−1) 355 460 380 272 298 526TOC (mg C (g of

volatile DW)−1)288 262 213 193 205 363

Biochemical composition (mg (g of DW)−1 )Protein 229 95 171 261 261 293Humic acid 206 76 151 245 241 275Polysaccharide 143 70 94 142 199 187Nucleic acid 54 23 18 6 76 44Uronic acid 188 272 247 184 377 267Lipid 12 5 7 13 19 23

DW, dry weight.

836 G. Guibaud et al. / Process Biochemistry 39 (2004) 833–839

Fig. 1. Polarographic titration curve for ECP sample E.

all binding sites of ECP. For Zn, the polarographic titrationshows a linear variation on the entire concentration rangeinvestigated, due to the high concentration of Zn alreadybound to ECP.

Fig. 2presents an example of Ruzic’s modelisation for Niobtained with samples B, E and F.

A good determination coefficient (> 0.96) was obtainedfrom polarographic curves by Ruzic’s method.

Table 3 shows the number of binding sites determinedwith Chau’s or Ruzic’s method, andTable 4presents theapparent constant of complexation determined with Ruzic’smethod.

Table 3shows a number of binding sites varying from 0.01to 0.25 mmol (g of ECP dry weight)−1 for Zn, 0.68–2.34mmol (g of ECP dry weight)−1 for Ni and 2.02–4.05 mmol(g of ECP dry weight)−1 for Cu. For the ECP studied, thenumber of sites of complexation increased in the followingorder: Zn� Ni < Cu. The number of sites determined withChau’s method is very close to that determined with Ruzic’smethod. The constant of complexation (expressed as logK)varied from 2.6 to 3.0 for Ni and from 3.0 to 4.4 for Cu. For

Fig. 2. Ruzic’s modelling for Ni obtained from ECP samples B, E and F.

Table 3Determination of the number of sites in mmol (g of dry weight of ECP)−1

at pH 7

ECP sample Method

Chau Ruzic

Cu Ni Zn Cu Ni Zn

A 3.66 0.70 0.14 3.64 0.68 –B 1.99 1.07 0.01 2.02 1.52 –C 2.60 0.82 0.03 2.89 1.06 –D 2.70 0.73 0.07 2.80 0.98 –E 3.44 2.23 0.14 3.72 2.34 –F 4.10 1.28 0.25 4.05 1.48 –

the same ECP sample, the complexation constant for Cu isalways higher than that for Ni (Table 4). These results arein accordance with those of Ref.[20] who have determinedsome complexation constants at pH 6 for Cu and Ni by a syn-thetic polymer (polyacrylic acid: molecular weight, 3× 106

g mol−1) using a polarographic method in SMDE mode.The low number of binding sites and the high concen-

tration initially bound onto ECP did not allow the determi-nation of a complexation constant with Ruzic’s method forZn. For ECP investigated, the constant of complexation wasshown to be always higher for Cu than for Ni.

3.3. Relationship between complexation parameters andECP composition

A matrix of correlation (determined with Statistica of Stat-soft software) between the biochemical fractions composingthe polymers (six polymers studied), their content in organicmatter (expressed as percentage of dry weight) and the con-stants of affinity with Cu or Ni is presented inTable 5.

The statistical investigations presented inTable 4showthat the higher the increase of proteins, humic acids andpolysaccharides in ECP, the more ECP were able to bindhigh quantities of Cu. For Ni, the complexation parame-ters are linked with the amount of uronic acid. Many au-thors have underlined the role of proteins, humic substances,polysaccharides and uronic acids in metal binding by sludges[23,24]. No relationship was found between the content oflipids and nucleic acids of ECP and their affinity for Cu orNi.

Table 4Determination of the apparent constant of complexation with Ruzic’smethod at pH 7

ECP sample Cu Ni Zn

A 3.5 3.0 –B 4.4 2.6 –C 3.8 2.8 –D 3.4 2.9 –E 3.3 2.6 –F 3.0 2.7 –

G. Guibaud et al. / Process Biochemistry 39 (2004) 833–839 837

Tabl

e5

Mat

rixof

corr

elat

ion

betw

een

the

para

met

ers

ofth

efix

ing

ofC

uan

dN

ian

dth

eco

mpo

sitio

nof

EC

P

Vola

tile

DW

(%)

Pro

tein

sH

umic

acid

sP

olys

acch

arid

esN

ucle

icac

ids

Uro

nic

acid

sLi

pids

Cu

Num

ber

ofsi

tes

(Cha

u)0.

698

(P

=0.

123)

0.83

2(P

=0.

040)

0.81

8(P

=0.

047)

0.84

8(P

=0.

033)

0.65

5(P

=0.

158)

0.10

5(P

=0.

843)

0.85

5(P

=0.

030)

Num

ber

ofsi

tes

(Ruz

ic)

0.63

9(

P=

0.17

2)0.

848

(P=

0.03

3)0.

833

(P=

0.04

0)0.

883

(P=

0.02

0)0.

698

(P=

0.12

3)0.

209

(P=

0.69

1)0.

857

(P=

0.02

9)lo

gK

−0.6

25(P

=0.

185)

−0.9

91(P

=0.

000)

−0.9

89(P

=0.

000)

−0.9

32(P

=0.

007)

−0.4

35(P

=0.

388)

−0.1

04(P

=0.

845)

−0.9

20(P

=0.

009)

Ni

Num

ber

ofsi

tes

(Cha

u)−0

.035

7(P

=0.

947)

0.28

29(P=

0.58

7)0.

2787

(P=

0.59

3)0.

6119

(P=

0.19

7)0.

7396

(P=

0.09

3)0.

9518

(P=

0.00

3)0.

5613

(P=

0.24

6)N

umbe

rof

site

s(R

uzic

)−0

.257

9(P

=0.

622)

0.10

63(P=

0.84

1)0.

109

(P=

0.83

7)0.

4348

(P=

0.38

9)0.

5589

(P=

0.24

9)0.

9676

(P=

0.00

2)0.

435

(P=

0.38

9)lo

gK

0.51

87(P

=0.

292)

0.17

96(P=

0.73

3)0.

1713

(P=

0.74

6)−0

.124

(P=

0.81

5)−0

.315

5(P

=0.

542)

−0.8

54(P

=0.

030)

−0.2

535

(P=

0.62

8)

Firs

tnu

mbe

r,lin

ear

coef

ficie

ntof

corr

elat

ion;P

,co

effic

ient

indi

catin

ga

sign

ifica

ntco

rrel

atio

nbe

twee

nth

etw

opa

ram

eter

sif

infe

rior

at0.

05.

The constants of complexation with Cu or Ni do not corre-late with the organic matter content of the solutions of poly-mers (% volatile dry weight). The polymers studied have amineral content between 19 and 44%. The mineral mattercontained in particulate form in the polymers should playa role in the binding of Cu or Ni. The ions such as SO2−

4or PO3−

4 , iron or aluminum hydroxides can strongly boundheavy metal[10].

4. Conclusions

This study shows that the complexation constant of metalion by ECP extracted from activated sludge can be deter-mined using a simple method using polarography in SMDEmode. The parameters obtained at pH 7 in HEPES buffershow the ability of ECP to bind Cu, Ni and Zn, even if thecapacity of ECP to bind Zn is difficult to quantify, due toa high amount of Zn initially bound to ECP samples. ForECP samples studied, the number of sites of complexationwas shown to increase in the following order: Zn� Ni < Cu,and the constant of complexation is always higher for Cuthan for Ni. A statistical study has shown that the more thecontent of proteins, humic acids and polysaccharides con-tained in ECP increased, the more they were able to bindhigh quantities of Cu. For Ni, the complexation parametersare related to the amount of uronic acid. These results sug-gest that carboxylic groups play a major role in Ni and Cucomplexation by ECP.

ECPs have shown their ability in complexing some heavymetals; it should offer interesting perspectives in removingheavy metals from effluents. It would need further investi-gations to study the evolution of complexation with multiplemetal addition.

Acknowledgements

The authors thank the “Conseil Régional du Limousin”for the financial support of this study.

Appendix A. Chau’s and Ruzic’s methods

The theoretical study of the complexation equilibrium as-sociated to the law of mass action can be described as fol-lows, assuming the formation of a 1:1 complex[22]:

M + L ⇔ ML (A.1)

K = [ML]

[M][L](A.2)

where M is the free metal, L the free ligand and ML theligand–metal complex.

The complexation capacity is obtained from the quantityof ligand available.

838 G. Guibaud et al. / Process Biochemistry 39 (2004) 833–839

Fig. 3. Example of determination of [L]o with Chau’s method.

The concentration of total organic ligand ([L]o) is ob-tained fromEq. (A.3):

[L] o = [L] + [ML] ′ (A.3)

where [ML]′ is the complex initially present in ECP, mea-sured from metal considered as adsorbed onto the ECP sam-ple. The modelling of the complexation capacity curves wascarried out using Chau’s graphic method and Ruzic’s lin-earisation method.

The modelling of the curves was carried out according toChau’s and Ruzic’s graphic methods. Chau’s graphic methodallowed the determination of [L] and [L]o. Ruzic’s methodallowed the parameters [L]o andK to be obtained.

The determination of [L] and [L]o with Chau’s method iscarried out according toFig. 3.

The curve is divided in two zones (Fig. 1). In the first one,a part of the metal added is bound by ECP. When all thebinding sites of ECP are occupied, the metal added remainsin solution and the second zone of the curve (Fig. 3) shows alinear draw. The slope of the linear part of the curve allowsthe determination of free metal concentration. The free metalconcentration is used to determine the parametersK and [L]ofrom Ruzic’s method.

For the exploitation of Ruzic’s method,Eq. (A.2) is re-placed byEq. (A.4):

K[L][M] −[ML] = 0 (A.4)

The total metal and ligand concentrations are respectivelygiven byEqs. (A.5) and (A.6)

[M] o = [ML] + [M] (A.5)

[L] o = [ML] + [L] (A.6)

The combination ofEqs. (A.5) and (A.6)givesEq. (A.7)

[M] o = [L] o−[L] + [M] (A.7)

Eq. (A.8)is obtained from the combination ofEqs. (A.4),(A.6) and (A.7).

K[M] 2 + [M] (K[L] o−K[M] o + 1)−[M] o = 0 (A.8)

with [M] o = x, total metal concentration and [M]= y, freemetal concentration.

Then

x = Ky2 + K[L] oy + y

Ky + 1(A.9)

Kxy + x = Ky2 + K[L] oy + y (A.10)

[L] o = x − y +[

x

y− 1

] (1

K

)(A.11)

y

x − y= 1

[L] oy + 1

K[L] o (A.12)

The linear curvey/(x−y) as a function ofy allows thedetermination of 1/[L]o (slope of the curve) and (1/K)[L] o(intersection withy-axis).

References

[1] Sobeck DC, Higgins MJ. Examination of three theories for mecha-nisms of cation-induced bioflocculation. Water Res 2002;36:527–38.

[2] Urbain V, Block JC, Manem J. Biofloculation in activated sludge:an analytic approach. Water Res 1993;27:829–38.

[3] Morgan JN, Forster CF, Evison L. A comparative study of the natureof biopolymers extracted from anaerobic and activated sludges. WaterRes 1990;24:743–50.

[4] Frolund B, Keiding K, Nielsen PH. A comparative study of biopoly-mers from a conventional and an advanced activated sludge treatmentplant. Water Sci Technol 1994;29:137–41.

[5] Karapanagiotis NK, Rudd T, Sterrit RM, Lester JN. Extraction andcharacterisation of extracellular polymers in digested sewage sludge.Newspaper Chem Tech Biotechnol 1989;44:1639–47.

[6] Madoni P, Davoli D, Gorbi G. Toxic effect of heavy metals onthe activated sludge protozoan community. Water Res 1996;30:135–41.

[7] Brown MJ, Lester JN. Role of bacterial extracellular polymers inmetal uptake in pure bacterial culture and activated sludge: effect ofmetal concentration. Water Res 1982;16:1539–48.

[8] Guibaud G, Baudu M, Dollet P, Condat ML, Dagot C. Role despolymères extracellulaires dans l’adsorption du cadmium par lesboues activées. Environ Technol 1999;20:1045–54.

[9] Batley G. Trace element speciation: analytical methods and problems.New York: Wiley, 1989.

[10] Buffle J. Complexation Reactions in Aquatic Systems: An AnalyticalApproach. Chichester, UK: Ellis Horwood, 1988.

[11] AFNOR NFT 90-105-2 Qualité de l’eau, tome 2 2ème édition, 1997[12] AFNOR NFT 90-029 Qualité de l’eau, tome 2 2ème édition, 1997[13] Dignac MF, Urbain V, Rybacki D, Bruchet A, Snidaro D, Scribe

P. Chemical description of extracellular polymers: implication onactivated sludge floc structures. Water Sci Technol 1998;38:45–53.

[14] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measure-ment Folin with phenol reagent. J Biol Chem 1951;193:265–75.

[15] Frolund B, Griebe T, Nielsen PH. Enzymatic activity in the activatedsludge floc matrix. Appl Microb Biotechnol 1995;43:755–61.

[16] Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F. Colorimetricmethod for determination of sugars and related substances. AnalChem 1956;28:350–6.

[17] Frings CS, Dunn RT. A colorimetric method for determination oftotal serum lipids based on the sulfo-phospho-vanillin reaction. AmNewspaper Clin Pathol 1970;53:89–91.

[18] Blumenkrantz N, Absoe-Hansen G. New method for quantitativedetermination of uronic acids. Anal Biochem 1973;54:484–9.

[19] Burton K. A study of the conditions and mechanism of the dipheny-lamine reaction for the colorimetric estimation of deoxyribonucleicacid. Biochem Newspaper 1956;62:315–23.

G. Guibaud et al. / Process Biochemistry 39 (2004) 833–839 839

[20] Perret S, Morly C, Cromer MR, Vittori O. Polarographic study ofthe removal of Cd(II) and Pb(II) from dilute aqueous solution by asynthetic flocculant. Comparison with Cu(II) and Ni(II). Water Res2000;34:3614–20.

[21] Meador JP. The interaction of pH, dissolved organic carbon, andtotal copper in the determination of ionic copper and toxicity. AquatToxicol 1991;19:13–32.

[22] Ruzic I. Trace metal complexation at heterogeneous binding sites inaquatic systems. Mar Chem 1996;53:1–15.

[23] Goodwin JA, Forster CF. An examination of the extracellular poly-mers by activated sludge. Microbios 1989;57:179–85.

[24] Fukushi K, Kato S, Antsuki T, Omura T. Isolation of copperbinding proteins from activated sludge culture. Water Sci Technol2001;44:453–9.