ORIGINAL PAPER

Anaerobic treatment of low-strength synthetic TCF effluentsand biomass adhesion in fixed-bed systems

Anuska Mosquera-Corral Æ Angela Belmar ÆJacqueline Decap Æ Katherine Sossa ÆHomero Urrutia Æ Gladys Vidal

Received: 13 November 2007 / Accepted: 21 December 2007 / Published online: 12 January 2008

� Springer-Verlag 2008

Abstract Toxicity effects produced by kraft mill efflu-

ents are due to the productive process. New bleaching

processes have been proposed (e.g. total chlorine free,

TCF) to reduce the production of toxic chlorine com-

pounds. In the TCF processes large amounts of chelating

compounds like the ethylenediaminetetraacetic acid

(EDTA) and diethylenetriaminepentaacetic acid (DPTA)

are used. The aim of this work is to research the feasi-

bility of the degradation of low-strength synthetic TCF

effluents in a anaerobic filter reactor (AF) and the bio-

mass adhesion. The effects on the operation of the AF at

different EDTA loading rates were tested in the range

from 0.07 to 0.51 g EDTA l-1 days-1. The maximum

EDTA removal percentage achieved was of 27%. Acute

toxicity (measured as 24 h-LC50) with Daphnia magna

was reduced from 14.23 to 54.53% before and after

anaerobic treatment, respectively. Observations of bio-

mass samples from the AF under the scanning microscope

verified the attached biomass.

Keywords Anaerobic � Biomass adhesion �Ethylenediaminetetraacetic acid (EDTA) �Total chlorine free (TCF) effluent

Introduction

In recent years, important technological innovations have

been developed worldwide in the forest industry, aimed at

reducing water consumption and generation of toxic con-

taminants. Thus, water circuit closures, new pulping and

bleaching processes have been introduced in order to

reduce water consumption and production of chlorinated

phenolic compounds. The TCF (total chlorine free)

bleaching process is one of these processes, which aims to

completely remove the discharges of chlorinated com-

pounds [1]. Furthermore, it has been stated that TCF

bleaching enhances the anaerobic COD biodegradability of

the effluent generated up to 75% [2].

However, the introduction of this TCF pulp is associated

to an increment on the amounts of chelating compounds used

to remove the metals present in the obtained cellulose pulp

[3]. The most commonly used chelating compounds are

ethylenediaminetetraacetic acid (EDTA) and diethylenetri-

aminepentaacetic acid (DPTA) [4]. No studies have been

completed within the pulping industry that demonstrates the

feasibility of removal of these chelating compounds by

anaerobic biological treatment. In some cases after a chem-

ical oxidation it has been observed the aerobic degradation of

both compounds [3]. Furthermore, the recalcitrant properties

of the DPTA have been already studied [5]. EDTA has been

also used to decrease the inhibitory effects on the methano-

genic activity of metals like cadmium, copper, nickel [6, 7].

All this factors make it interesting the study of these com-

pounds in the anaerobic processes.

A. Mosquera-Corral

Department of Chemical Engineering, School of Engineering,

University of Santiago de Compostela,

Lope Gomez de Marzoa s/n, 15782

Santiago de Compostela, Spain

A. Belmar � J. Decap � G. Vidal (&)

Environmental Science Center EULA-Chile,

University of Concepcion, P.O. Box 160-C, Concepcion, Chile

e-mail: glvidal@udec.cl

K. Sossa � H. Urrutia

Biotechnology Center, University of Concepcion,

P.O. Box 160-C, Concepcion, Chile

123

Bioprocess Biosyst Eng (2008) 31:535–540

DOI 10.1007/s00449-007-0194-0

In the half 60 s the anaerobic digestion was established as

an alternative to treat effluents with high organic loads (1–

40 g COD l-1 days-1), but also for low-strength (\1 g

COD l-1 days-1). It was in the early 90 s when these

technologies were extensively applied to the forest industry

[8]. The use of these technologies allows the production of

effluents which can be reused in the production process

provoking a decrease of fresh water consumption [9].

Technologies based on anaerobic biomass grown in the form

of biofilms have been found to provide biomass more

resistant to the presence of toxic compounds as those usually

contained in the effluents of the pulp and paper industry.

The aim of the present work was to study the effects of

EDTA present in the low-strength synthetic TCF effluents

on the operation of an anaerobic filter reactor (AF) and the

biomass adhesion to PVC corrugated rings.

Methods and materials

Wastewater

Effluent was prepared similar to the model media from

Rodrıguez et al. [3] as detailed in Table 1. The COD

concentration was between 0.38 and 0.48 g COD l-1 tak-

ing into account the COD supply from the different

concentrations of organic compounds including the EDTA

tested (107–214 mg l-1).

Inoculum

The reactor was inoculated with 5 g VSS l-1 of anaerobic

flocculent sludge treating effluents from a kraft process and

with a specific activity of 0.9 g COD g-1 VSS days-1.

Reactor

An anaerobic filter reactor with a working volume of

200 ml (Fig. 1) was operated in a continuous flow mode

during 370 days. The AF was filled with corrugated PVC

Raschig rings (1.4 cm internal diameter) acting as carrier

material and with a specific surface area of 449 m2 m-3.

The temperature was maintained at 37�C by using a ther-

mostatic chamber and pH was maintained in the range

7.0–8.5 by means of bicarbonate addition.

Operational conditions

The reactor was operated in five different periods with two

different operational strategies: (a) decrease of HRT from

1.25 to 0.46 days (I–IV) and (b) doubling of the EDTA

concentration from 107 to 214 mg EDTA�l-1 keeping the

HRT on 0.46 days (V).

The EDTA loading rates (LREDTA) applied to the reactor

ranged from 0.07 to 0.51 g EDTA l-1 days-1.

Adhesion experiments

The adhesion of the anaerobic biomass to the corrugated

PVC rings was tested in batch experiments performed by

Table 1 TCF synthetic effluent composition

Feeding media Nutrient solution

Compound g l-1 Compound g l-1

Sodium formate 0.5 H3PO4 0.314 9 10-6

Sodium acetate 0.1 MgSO4�7H2O 0.54

Sodium bicarbonate 4 CaCl2�2H2O 0.54

EDTA 0.107–0.214 FeCl3�6H2O 1.05

FeCl3�6 H2O NH4Cl 2.085

Nutrient solution 5 ml l-1 KH2PO4 0.285

Glyoxal 0.3

Vanillin 0.8

INFLUENT EFFLUENT

CH4

2

3

INFLUENT EFFLUENT

CH4

1

4

5

6

7

Fig. 1 Anaerobic filter reactor (AF) layout: 1 influent, 2 feeding

distributor, 3 sludge bed, 4 separator system solid/liquid/gas, 5 biogas

measurement, 6 carrier material, 7 effluent

536 Bioprocess Biosyst Eng (2008) 31:535–540

123

duplicate. Assays were carried out in dark vials of 100 ml

of volume. The same inoculum as in the reactor was used

in the tests and an amount of 2 g VSS l-1 were intro-

duced in each vessel. Around 30 pieces of cut corrugated

rings (surface of 2 9 10-4 m2 by piece) were added to

each vial. Different concentrations of EDTA (0, 0.05,

0.10 and 0.20 g l-1) were tested. No organic substrate,

except for the EDTA, was added to the tests. A propor-

tional concentration of iron in the vessels medium, as that

indicated in Table 1, was used. The vials were placed in

a thermostated bath where temperature was fixed at 37�C.

The specific activity of the adhesion experiments was

monitored according to the procedure described by Soto

et al. [10] by measuring the volumes of liquid displaced

due to biogas production. In order to quantify the number

of microorganisms attached to the supports three pieces

of corrugated PVC were removed from each vial along

the time. The pieces were sonicated in 10 ml of water

during 30 min; the liquid was filtered through filters of

0.2 lm pore size. The retained biomass on the filters was

treated with the DAPI dye (0.2 ml of a solution of

0.1 mg l-1). The filters were deposited on a slide covered

with vaseline and placed in the epifluorescence chamber

where the number of microorganisms was counted by

hand.

Toxicity experiments

The acute toxicity of the EDTA was determined in the

liquid samples by exposing juveniles (\24 h old) Daph-

nia magna and Daphnia obtusa during 24–48 h to this

compound. The mortality of organisms was monitored at

the end of the exposure time and it was defined as the

lack of organism motility when the vessel is shaken. The

average lethal concentrations at 24 and 48 h (LC50) were

estimated using the methods of Probit and Spear-

man–Karber [11, 12].

Analytical methods

Volatile suspended solid (VSS), chemical oxygen demand

(COD), biological oxygen demand (BOD5) were measured

according to the ‘‘Standard Methods’’ [13]. Total alkalinity

(TA) was determined by titration [14]. EDTA was deter-

mined according Bhattacharya and Kundu [15]. Samples

for COD, BOD5 and EDTA analysis were filtered through a

membrane of 0.45 lm (pore size).

Biomass samples for scanning electron microscopy were

analyzed according to the technique developed by Nation

[16]. The identification and quantification of the microbial

populations were performed by means of the Dot-Blot

technique [17]. The 16S rRNA of a biomass sample was

analyzed on a membrane using the specific oligonucleotide

probes detailed in Table 2 using the hybridization proce-

dure described by Stahl et al. [18].

Results and discussion

Process operation

The reactor was operated during 370 days (periods I–V)

fed with a synthetic TCF effluent and operated at different

HRT values from 1.4 to 0.46 days (see Table 3; Fig. 2).

The EDTA concentrations in the feeding media to the AF

were chosen in the range they are usually found in the

wastewater produced in TCF processes (0.06–0.10 g l-1)

[5]. The chosen values did not cause inhibition on the

Table 2 Oligonucleotide probes used for the Dot-Blot analysis

Probe Target organism Probe sequence (50–30) Ref.

EUB338 Bacteria GCTGCCTCCCGTAGGAGT [19]

ARC915 Archaea GTGCTCCCCCGCCAATTCCT [20]

MS1414 Methanosarcinaspecies and

close relatives

CTCACCCATACCTCACTCGGG [21]

Table 3 Operational conditions of the AF

Parameter Period (days)

I (0–67) II (68–130) III (131–194) IV (195–340) V (341–370)

LR (g EDTA l-1 days-1) 0.08 0.13 0.07 0.22 0.51

HRT (days) 1.25 ± 0.10 0.84 ± 0.06 1.44 ± 0.47 0.46 ± 0.19 0.46 ± 0.19

Removal (%) Media Range Media Range Media Range Media Range Media Range

COD 52.2 41.2–71.3 54.7 42.2–60.5 69.0 49.5–81.7 78.4 64.0–84.2 50.8 42.4–65.3

BOD5 89.0 83.0–92.0 93.0 82.0–98.0 97.5 95.2–98.8 97.8 96.4–98.7 94.3 91.2–98.1

EDTA 27.3 8.0–40 13.5 5.0–28.0 15.0 5.0–26.0 5.6 6.0–15.9 19.1 5.7–39.3

Bioprocess Biosyst Eng (2008) 31:535–540 537

123

methanogenic activity of the biomass, as it was determined

from previous batch experiments that an inhibition of 50%

of this activity occurred at concentrations of 0.4 g

EDTA l-1 [22].

The AF was started up with an organic load lower than

the maximum load feasible to be treated by the inoculated

amount of anaerobic sludge (around 14.5 times lower than

the maximum specific activity of the biomass) to overcome

the possible detrimental effects of the presence of EDTA.

The organic load was increased to reach a value of 1.06 g

COD l-1 days-1 (period V). The maximum COD removal

percentages were obtained during periods III and IV

reaching values of 78.4% (Fig. 2a) while in terms of BOD5

percentages were close to 100% (Fig. 2b). The total alka-

linity was always above 2 g CaCO3 l-1 to guarantee the

stability of the pH value from 7.3 to 8.5.

The average EDTA removal percentage obtained was

during the whole operation of the AF in the range from 5.6

to 27.3%. A balance calculation was performed to establish

the route of disappearance of the EDTA from the media.

The assumption that the readily biodegradable compounds

(acetate and formate) were removed in the reactor, as the

BOD5 concentration in the effluent was almost neglected,

was made. From the COD measurements it was obtained

that the percentage of COD remaining in the media

corresponded to the not degraded EDTA due to the fact that

this compound contributes to the 26% (periods I–IV) and to

the 52% (period V) of the theoretical COD in the feeding

which almost fits to the 80 and 50% of COD removal

percentages obtained, respectively. These observations are

similar to those from Alarcon et al. [5] working with and

effluent containing DPTA (0.1 g l-1) in a similar reactor.

The apparent disappearance of EDTA in the AF was then

attributed to the errors associated to sampling and analyt-

ical determinations. The complete depletion of the BOD5

was an indicative of the stable performance of the metha-

nogenic process indicating the absence of negative effects

on this activity due to the presence of EDTA in the feeding.

Similar behavior was observed by Alarcon et al. [5]

working with DPTA.

The toxicity of the influent and effluent to the AF in

period I, determined with D. obtusa and D. magna, indi-

cated a reduction of the toxicity of the media after the

anaerobic degradation. With D. obtusa the value of the

24 h-LC50 (acute toxicity) was reduced from 23.94 to

67.71%. In the experiments performed with D. magna

results indicated a reduction from 14.23 to 54.53%,

slightly different due to the higher sensitivity of the latter

compared to the D. obtusa. The results obtained from the

influent can be, in a certain way, compared to those from

Martins et al. [23] who indicate that the concentration of

acetate corresponding to the 24 h-LC50 for D. magna is

of 10.79 mg l-1 which is similar to that of the influent of

14.23 mg l-1 expressed in concentration. The results must

be interpreted with care because the feeding media to the

AF contained other major compounds like the formate

which could contribute to the acute toxicity at different

extent as it has been found by Cooman et al. [12] to

happen with other toxic compounds of the tannery efflu-

ents. The EDTA was not expected to cause toxic effect

due to the fact that it has been added to the media when

the acute toxicity of metals was determined and it has

been stated that its presence reduces the toxic effect of

metals like copper, cadmium, etc [6]. Taking into account

that the biodegradable substrates (acetate and formate)

were degraded during the anaerobic process the residual

toxicity in the effluent can be attributed to other salts like

ammonia present in the media [12].

Bacteria attachment

The reactor was inoculated with flocculent biomass char-

acterized by a wide diversity (Fig. 3a). With the objective

to determine the formation of bacteria attachment, after

33 days of operation samples of corrugated rings were

collected from the AF. The observation under the electron-

scanning microscope of the surface of the support material

Time (d)

LREDTA (gEDTA/L·d) 0.08 0.13

0

20

40

60

80

100

0.07 0.22 0.51

0 100 200 300 4000

20

40

60

80

100

Biofilm formation Treatment stage

I II III IV VPERIOD IV

CO

D r

emo

val (

%)

ED

TA

an

d B

OD

5 re

mo

val (

%)

a

b

Fig. 2 Anaerobic filter reactor (AF) performance. a COD (filledcircle) removal, b BOD5 (filled circle) and EDTA (filled diamond)

removal percentages, respectively

538 Bioprocess Biosyst Eng (2008) 31:535–540

123

positively indicated that the growth of bacteria attachment

occurred and it was found to be mainly composed by

bacillus-type bacteria (Fig. 3 b, c). In principle it could be

established that the presence of EDTA in the feeding media

did not exert any prejudicial effect on the development of

the bacteria attachment.

Adhesion experiments, performed with biomass with the

same origin as that inoculated to the reactor, indicated that

no EDTA degradation occurred in the test. The amount of

organisms adhered to the surface of the carrier material

indicated that the bacteria attachment was formed during

the initial 12 h (Fig. 4). No significant effect on the bio-

mass adhesion was observed with the different tested

EDTA concentrations compared to the results obtained

with the blank experiment (0 g l-1). These results cor-

roborated those obtained from the AF operation.

Furthermore, the microbial populations present in the

biomass adhesion were identified and quantified by means of

Fig. 3 Scanning microscope observation of the biomass: inoculum, 910,000 a, microorganisms attached to the support material on day 33 of the

AF operation, 92,500 (b) and 95,000 (c)

0 10 20 30 40 500

25

50

75

100

125

150

175

Time (d)

Mic

roor

gani

ms

num

ber

/ PV

C c

orru

gate

d su

ppor

t (cm

-2)

(x10

3 )

Fig. 4 Adhesion experiments: blank 0 g EDTA l-1 (filled diamond),

0.05 g EDTA l-1 (open circle), 0.1 g EDTA l-1 (open diamond),

0.2 g EDTA l-1 (filled circle)

Bioprocess Biosyst Eng (2008) 31:535–540 539

123

the Dot-Blot technique. Results indicated the presence of

65% of the microorganisms belonging to the domain Bac-

teria and 35% belonging to the domain Archaea. In the latter

35% of the population was found to belong to the Methan-

osarcina species. These are common archaea organisms

present in anaerobic reactors fed with relatively high acetate

concentrations and which easily accomplish with sudden

load changes [24]. Once more the appropriate operation of

the AF is corroborated by the grown microbial populations.

Conclusions

The biodegradation of a low-strength synthetic TCF

effluent in an AF was successfully accomplished in the

presence of different concentrations of EDTA (0.1–

0.2 g l-1). The maximum removal percentages reached

were close to 100% for BOD5 and of 78.4% for COD when

loads up to 0.83 g COD�l-1�days-1 were treated. Also,

anaerobic degradation of the synthetic effluent reduced the

acute toxicity measured with D. magna and D. obtusa in 64

and 74%, respectively.

The maximum EDTA removal percentage achieved was

of 27%.

No detrimental effect on the attachment of the biomass

to the carrier surface was observed caused by the presence

of the EDTA in the media. An amount of 35% of the

bacteria attachment corresponded to methanogenic bacteria

from the family Methanosarcinaceae, which conferred to

the biofilm a high resistance to operational changes.

Acknowledgments This work was financially supported by

FONDECYT 1070509. A. Mosquera-Corral thanks to the European

ALFA N� II-0311-FA-FCD-FI-FC by supporting her stay at the

Environmental Science Center EULA-Chile, Universidad de Con-

cepcion (Chile).

References

1. Thompson G, Swain J, Kay M, Foster CF (2001) The treatment of

pulp and paper mill effluent: a review. Bioresour Technol

77:275–286

2. Vidal G, Soto M, Field J, Mendez-Pampın R, Lema JM (1997)

Anaerobic biodegradability and toxicity of wastewaters from

chlorine and total chlorine-free bleaching of eucalyptus kraft

pulps. Water Res 31(10):2487–2494

3. Rodrıguez J, Mutis A, Yeber MC, Freer J, Baeza J, Mansilla HD

(1999) Chemical degradation of EDTA and DTPA in a totally

chlorine free (TCF) effluent. Water Sci Technol 40:267–272

4. Sillanpaa M, Pirkanniem K (2001) Recent developments in

chelate degradation. Environ Technol 22:791–801

5. Alarcon E, Decap J, Vidal G (2005) Resistance of diethylene-

triaminepentaacetic acid to anaerobic biodegradation. Electron J

Biotechnol 8(3):308–313 doi:10.2225/vol8-issue3-fulltext-5

6. Guilhermino L, Diamantino TC, Ribeiro R, GonCalves F, Soares

AMVM (1997) Stability of test media containing EDTA for the

evaluation of acute metal toxicity to Daphnia magna Straus.

Ecotoxicol Environ Saf 38:292–295

7. Chen Y, Cheng JJ, Creamer S (2007) Inhibition of anaerobic

digestion process: a review. Bioresour Technol. doi:10.1016/

j.biortech.2007.01.057

8. Lettinga G, Field J, Sierra-Alvarez R, van Lier J, Rintala J (1991)

Future perspective for the anaerobic treatment of forest industry

wastewater. Water Sci Technol 24(3/4):91–102

9. Van Lier LB, Lens P, Hushoff Pol LW (2001) Anaerobic treat-

ment for C and S removal in ‘‘zero-discharge’’ paper mills:

effects of process design on S removal efficiencies. Water Sci

Technol 44(4):189–194

10. Soto M, Mendez R, Lema JM (1993) Methanogenic and non-

methanogenicactivity tests. Theoretical basis and experimental

set up. Water Res 27(8):1361–1376

11. Finney D (1978) Statistical method in biological assay. Charles

Griffin & Co. Ltd (eds), London, pp 508

12. Cooman K, Gajardo M, Nieto J, Bornhardt C, Vidal G (2003)

Tannery wastewater characterization and toxicity effects on

Daphnia spp. Environ Toxicol 17:45–51

13. APHA-AWWA-WPCF (1985) Standard methods for examination

of water and wastewater, 16th eds, Washington

14. Ripley LE, Boyle W, Converse J (1986) Improved alkalimetric

monitoring for anaerobic digestion of high strength wastes. J

WPCF 58(5):406–411

15. Bhattacharya SN, Kundu KP (1971) Spectrophotometric deter-

mination of EDTA. Talanta 18:446–448

16. Nation JL (1983) A new method for using hexamethyldisilazane

for preparation of soft insect tissue for scanning electron

microscopy. Stain Technol 58:347–351

17. Sambrook J, Fritsch EF, Miniatis T (1989) Dot and Slot

hybridization of RNA. In: Molecular cloning a laboratory man-

ual, 2nd edn. Cold Spring Harbord Laboratory Press, USA pp

7.35–7.55

18. Stahl DA (1997) Application of phylogenetically based hybrid-

ization probes to microbial ecology. Mol Ecol 4:535–542

19. Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereu R, Stahl

DA (1990) Combination of 16S ribosomal-rRNA-targeted oli-

gonucleotide probes with flow cytometry for analyzing mixed

microbial-populations. Appl Environ Microbiol 56(6):1919–1925

20. Stahl DA, Amann R (1991) Development and application of

nucleic acid probes. In: Stackebrandt E, Goodfellow M (eds)

Nucleic acid techniques in bacterial systematics. Wiley, Chich-

ester, pp 205–248

21. Raskin L, Poulsen LK, Noguera DR, Rittmann BE, Stahl DA

(1994) Quantification of methanogenic groups in anaerobic bio-

logical reactors by oligonucleotide probe hybridization. Appl

Environ Microbiol 60:1241–1248

22. Vidal G, Gajardo M, Belmar A, Diez M (2001) Methanogenic

toxicity of ethylenediaminotetraacetic acid. XXIII Microbiology

Chilean Congress, November 28–30, Tome, Chile, pp 61

23. Martins J, Oliva-Teles L, Vasconcelos V (2007) Assays with

Daphnia magna and Danio rerio as alert systems in aquatic

toxicology. Environ Int 33:414–425

24. Conklin A, Stensel HD, Ferguson J (2006) Growth kinetics and

competition Between Methanosarcina and Methanosaeta in

mesophilic anaerobic digestion. Water Environ Res 78(5):486–

496

540 Bioprocess Biosyst Eng (2008) 31:535–540

123