ARTICLE IN PRESS

Available at www.sciencedirect.com

WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 9 6 9 – 9 7 6

0043-1354/$ - see frodoi:10.1016/j.watres

�Corresponding auE-mail address:

journal homepage: www.elsevier.com/locate/watres

Degradation of phthalate esters in an activated sludgewastewater treatment plant

Peter Rosleva,�, Katrin Vorkampb, Jakob Aarupa, Klavs Frederiksena, Per Halkjær Nielsena

aSection of Environmental Engineering, Department of Biotechnology, Chemistry and Environmental Engineering, Aalborg University,

Sohngaardsholmsvej 57, DK-9000 Aalborg, DenmarkbDepartment of Environmental Chemistry and Microbiology, Danish Environmental Research Institute, Frederiksborgvej 399, DK-4000

Roskilde, Denmark

a r t i c l e i n f o

Article history:

Received 10 January 2006

Received in revised form

28 November 2006

Accepted 29 November 2006

Available online 26 January 2007

Keywords:

DMP

DBP

BBP

DEHP

Activated sludge

WWTP

Degradation

nt matter & 2006 Elsevie.2006.11.049

thor. Tel.: +45 9635 8505; fpr@bio.aau.dk (P. Roslev).

a b s t r a c t

Efficient removal of phthalate esters (PE) in wastewater treatment plants (WWTP) is

becoming an increasing priority in many countries. In this study, we examined the fate of

dimethyl phthalate (DMP), dibutyl phthalate (DBP), butylbenzyl phthalate (BBP), and di-(2-

ethylhexyl) phthalate (DEHP) in a full scale activated sludge WWTP with biological removal

of nitrogen and phosphorus. The mean concentrations of DMP, DBP, BBP, and DEHP at the

WWTP inlet were 1.9, 20.5, 37.9, and 71.9 mg/L, respectively. Less than 0.1%, 42%, 35%, and

96% of DMP, DBP, BBP, and DEHP was associated with suspended solids, respectively. The

overall microbial degradation of DMP, DBP, BBP, and DEHP in the WWTP was estimated to be

93%, 91%, 90%, and 81%, respectively. Seven to nine percent of the incoming PE were

recovered in the WWTP effluent. Factors affecting microbial degradation of DEHP in

activated sludge were studied using [U–14C-ring] DEHP as tracer. First order rate coefficients

for aerobic DEHP degradation were 1.0� 10�2, 1.4� 10�2, and 1.3� 10�3 at 20, 32, and 43 1C,

respectively. Aerobic degradation rates decreased dramatically under aerobic thermophilic

conditions (o0.1� 10�2 h�1 at 60 1C). The degradation rate under anoxic denitrifying

conditions was 0.3� 10�2 h�1, whereas the rate under alternating conditions (aerobic–a-

noxic) was 0.8� 10�2 h�1. Aerobic DEHP degradation in activated sludge samples was

stimulated 5–9 times by addition of a phthalate degrading bacterium. The phthalate

degrading bacterium was isolated from activated sludge, and maintained a capacity for

DEHP degradation while growing on vegetable oil. Collectively, the results of the study

identified several controls of microbial PE degradation in activated sludge. These controls

may be considered to enhance PE degradation in activated sludge WWTP with biological

removal of nitrogen and phosphorus.

& 2006 Elsevier Ltd. All rights reserved.

1. Introduction

The industrial use of phthalate esters (PE) has resulted in an

ubiquitous presence of these xenobiotic compounds in the

environment (Staples et al., 1997; Fromme et al., 2002). PE are

used mainly as additives in plastics to impart flexibility but

they are also used in the production of paints, glues,

lubricants, pharmaceutics, cosmetics, and pesticides. Most

r Ltd. All rights reserved.

ax: +45 9635 0558.

PE are not chemically bound to the products, and they may

subsequently escape into the environment during manufac-

turing, during product use, and/or after product disposal. The

global production of PE is in millions of tons per year (Nielsen

and Larsen, 1996).

The environmental fate of PE has received increasing

attention because of potential health effects in animals

and humans. Several studies have suggested that PE may

ARTICLE IN PRESS

WA T E R R E S E A R C H 4 1 ( 2 0 0 7 ) 9 6 9 – 9 7 6970

bioaccumulate in aquatic organisms (Staples et al., 1997).

Although the acute toxicity of many PE is relatively low, PE

metabolites may show toxic effects in biotests (Nalli et al.,

2002; Horn et al., 2004). Exposure to some PE may also result

in carcinogenic and teratogenic effects (Morgenroth, 1993;

Nielsen and Larsen, 1996). Furthermore, some PE show

endocrine effects, e.g., they may act as xenoestrogens (Blom

et al., 1998).

A major environmental source of PE is release from

industrial and municipal wastewater treatment plants

(WWTP). PE may be found in relatively high concentrations

(mg/L) in municipal wastewater due to urban runoff and

discharges from industry and households (Vikelsøe et al.,

1998). Common PE in municipal wastewater are dimethyl

phthalate (DMP), dibutyl phthalate (DBP), butylbenzyl phtha-

late (BBP), di-(2-ethylhexyl) phthalate (DEHP), diisononyl

phthalate (DNP), and dioctyl phthalate (DOP). Abiotic hydro-

lysis of PE in wastewater appears negligible under most

environmental conditions (Staples et al., 1997). In contrast, PE

may be enzymatically cleaved and subsequently degraded by

microorganisms in wastewater and activated sludge (O’Grady

et al., 1985; Jianlong et al., 1996; Wang et al., 1997; Cheng et al.,

2000). As a result, biodegradation may remove a significant

fraction of the PE entering activated sludge WWTP (Fauser et

al., 2003; Marttinen et al., 2003). However, the biotic and

abiotic mechanisms regulating the degradation efficiency in

most WWTP are not well understood.

The aim of this study was to investigate the fate of DMP,

DBP, BBP, and DEHP in a full scale activated sludge WWTP

with biological removal of nitrogen and phosphorus. A better

understanding of PE removal in WWTP requires precise

measurements of the concentrations in wastewater and

sludges. Thus, a procedure for extraction of PE with different

aqueous solubilities was developed, and used to establish a

simple mass balance for the WWTP. In addition, biotic and

abiotic factors affecting microbial DEHP degradation in

activated sludge were investigated in laboratory experiments

using [U–14C-ring] DEHP as tracer.

2. Materials and methods

2.1. Chemicals and glassware

High-purity (499%) DMP [CAS 131-11-3], DBP [CAS 84-74-2],

BBP [CAS 85-68-7], and DEHP [CAS 117-81-7] were purchased

from VWR-Merck (Copenhagen, Denmark). Dimethyl iso-

phthalate (DMiP, purity 99%), dioctyl terephthalate (DOTP,

purity 98%) and [U–14C-ring] DEHP (188.7 MBq/mmol, 499%

purity) were obtained from Sigma-Aldrich (Copenhagen,

Denmark). All solvents used for extraction and dilution of

phthalates were of HPLC quality or better. All glassware was

rinsed with hexane–acetone (1:1), and then heated at 430 1C

prior to use. Filters used in the extraction procedure were

heated at 430 1C before use.

2.2. Wastewater and sludges

Samples were collected from the influent, effluent, aeration

tank, and the digester at Aalborg East municipal WWTP

(Aalborg, Denmark). The wastewater at Aalborg East WWTP

originates from households (80%) and local industries (20%) and

corresponds to 100,000 individuals. The activated sludge plant

has biological nitrogen and phosphorus removal and operates

with the Biodenipho configuration (Henze et al., 2002) with an

anaerobic tank followed by alternating aerobic nitrifying and

anoxic denitrifying conditions. Some chemical phosphorus

removal takes place by dosage of ferrous sulfate. The hydraulic

retention time for the wastewater is about 1 day, the sludge

concentration in the process tanks is 4–7 g SS/L (equivalent to

2–4 g VSS/L with a content of 0.5–1.0�1012 bacteria/g VSS), the

sludge age is 21–28 days, the aerobic sludge age is 6–8 days, and

the sludge production is 5–6000kg SS/day. The surplus sludge is

concentrated and treated in a mesophilic digester followed by

dewatering to a water content of 20–25%. In the present study,

flow proportional wastewater samples (24 h) were collected

from the influent and effluent of the treatment plant in dry

weather conditions. Activated sludge samples and digested

sludge samples were collected from the aeration tanks and

after dewatering, respectively.

2.3. PE Extraction

An outline of the procedure used for extraction of PE is shown

in Fig. 1. PE extraction was generally based on 0.2 L influent,

1 L effluent, and 2 g of digested sludge. Wastewater samples

were homogenized with a blender (Braun, Germany) for 2 min

prior to extraction of PE. Samples were then filtered through

0.7 mm glass fiber filters (GF 75, Advantec, Toyo Roshi Kaisha

Ltd., Japan) to separate water and particles. As the filtrate

likely contains particles and colloids o0.7 mm, the terms

‘‘particles’’ and ‘‘filtrate’’ have to be considered as operational

parameters. During the initial method development, the

recovery of the extraction procedure was evaluated by spiking

activated sludge samples with known concentrations of PE

(100–150mg/L). Non-spiked wastewater samples were used for

final evaluation of the extraction procedure.

PE were extracted from the filtrate (Fig. 1) by solid phase

extraction (SPE) using reversed phase 10 mL C18 ISOLUTE

columns with 0.5 g sorbent mass (IST, Glamorgan, UK). The

columns were conditioned sequentially with 4 mL hexane,

4 mL acetone, and 8 mL distilled water. The columns were not

allowed to run dry after conditioning. Water samples were

passed through the columns under vacuum (10 mL/min).

Before elution, the SPE columns were dried by passing air

through the columns for 30 min. PE were eluted with 10 mL of

hexane:acetone (1:1).

Four solvent extraction procedures were initially compared

for recovery of PE from sludge solids: (1) extraction on a rotary

shaker, (2) extraction in an ultrasonic water bath, (3) extrac-

tion by a Soxhlet-type procedure (Redeker, 1997), and (4)

extraction by a ‘‘hot solvent’’ procedure. All sludge particles

and filters (Fig. 1) were homogenized in an analytical mill

prior to extraction (IKA, Wilmington, NC, USA). PE extraction

on the rotary shaker was carried out in 35 mL Pyrex vials at

150 rpm and 20 1C with four subsequent extractions of 20 min

using 20 mL hexane:ethyl acetate (1:1). PE extraction in the

ultrasonic water bath was carried out in 35 mL Pyrex vials at

40 1C with four subsequent extractions of 20 min using 20 mL

hexane:ethyl acetate (1:1). The Soxhlet-type extraction was

ARTICLE IN PRESS

Sample

Membrane filtration

(0.7 µm glass fibre filters)

Particles (> 0.7µm)

Hot Solvent Extraction

(hexane:ethylacetate)

Filtrate

Solid Phase Extraction

(C18 columns)

Elution(n-hexane:acetone)

GC analysis

Concentration under N2

Clean up

(aminopropyl columns)

Elution

(n-hexane)

Concentration under N2GC analysis

Fig. 1 – Outline of the procedure used for PE extraction.

WAT ER R ES E A R C H 41 (2007) 969– 976 971

carried out using a fexIKA 50 (FEXTRA) apparatus (IKA,

Staufen, Germany). Samples were extracted 4 times for 2 h

using 20 mL hexane:ethyl acetate (1:1), and cyclic heating to

100 1C followed by cooling to 60 1C (6 cycles per 2 h extraction).

PE were also extracted with a modified version of the FEXTRA

procedure in which solid samples were placed directly in

boiling hexane:ethyl acetate (1:1). This modified procedure is

referred to as ‘‘hot solvent extraction’’. The four extraction

procedures described above were initially evaluated by

comparing PE recovery from non-spiked dewatered sludge

samples, and from activated sludge spiked with DMP, DBP,

and DEHP.

2.4. Clean-up of PE extracts

To facilitate PE analysis by gas–liquid chromatography,

interfering sludge organics were removed by adsorption

chromatography using 3 mL aminopropyl columns with 0.5 g

sorbent mass (J.T. Baker, Phillisburgh, NJ, USA). In general,

only extracts from sludge solids required this clean-up (Fig.

1). PE extracts were concentrated to 1 mL under N2 before

clean-up. The aminopropyl columns were conditioned with

4 mL methanol followed by 4 mL hexane before loading the

PE extracts. PE were eluted with 10 mL of hexane under

vacuum. The columns were not allowed to run dry after

conditioning and during loading and elution. Recovery of

DMP, DBP, and DEHP was X80% in extracts after clean-up

and concentration under N2.

2.5. GC analysis of PE

Samples concentrated to 0.5–1 mL under N2 were analyzed on

a HP 5890 series II GC equipped with a flame ionization

detector and a HP Ultra 2 capillary column (50 m long, 0.2 mm

inner diameter, 0.33mm film thickness). Samples (1mL) were

injected in the splitless mode. DMiP and DOTP were added to

all samples as internal standards. The GC temperature

program was as follows: 1 min at 60 1C, from 60 to 270 1C at

15 1C/min, from 270 to 300 1C at 3 1C/min, and finally 300 1C for

5 min. H2 was used as carrier gas (1.5 mL/min). The phthalates

were identified on the basis of their retention times relative to

the internal standards. Quantification was based on peak

areas in relation to internal standards, and linear 5-point

calibration curves.

2.6. Degradation of [U–14C-ring] DEHP

Degradation of [U–14C-ring] DEHP to 14CO2 was measured

partly as described previously (Roslev et al., 1998). Degrada-

ARTICLE IN PRESS

WA T E R R E S E A R C H 4 1 ( 2 0 0 7 ) 9 6 9 – 9 7 6972

tion activity was studied under mesophilic (20–43 1C) and

thermophilic conditions (60 1C), and with different oxygen

regimes. Activated sludge samples (10 mL) were diluted 1:1

with sludge supernatant, and then incubated on a shaker at

120 rpm in serum bottles (120 mL) sealed with teflon lined

butyl rubber stoppers. [U–14C-ring] DEHP (100,000 dpm) was

added to each bottle dissolved in 10mL acetone. The acetone

was evaporated for 30 min under air prior to the incubation.

Anaerobic conditions were established by flushing bottles for

30 min with N2 (99.999% purity). Denitrifying conditions were

established in N2 flushed bottles by injecting a solution of

oxygen free KNO3 to a final concentration of 2 mM.

Aerobic thermophilic degradation was investigated by

incubating activated sludge samples at 60 1C, and by using

sludge samples from a 5 L aerobic thermophilic sludge reactor

permanently operated at 60 1C. The thermophilic sludge

reactor had been operated for 430 days at 60 1C before

sampling. During this period, the reactor was fed every other

day with 0.25 volumes of fresh activated sludge.

In all experiments, degradation of [U–14C-ring] DEHP was

estimated as the sum of 14CO2 recovered in the gas and liquid

phases at each time point. Subsamples from the gas (5 mL)

and liquid phases (1 mL) were withdrawn with needle and

syringe. The 14CO2 in gas samples was fixed in 10 mL

ethyleneglycol:monomethylether (1:7), whereas 14CO2 in li-

quid samples was fixed in 2 mL 1 M NaOH. Fixed 14CO2 was

subsequently recovered in a stripping chain, and measured in

a scintillation counter as described previously (Roslev et al.,

1998). The time course of [U–14C-ring] DEHP degradation was

fitted by exponential and fractional power equations as

described by Madsen et al. (1999). Initial first order rate

coefficients were calculated based on triplicate samples and

the first 4–10 data points from each time course.

2.7. Inoculation of sludge samples with PE degraders

Addition of PE degraders to sludge samples was evaluated as

a mean to stimulate DEHP degradation. An aerobic DEHP

degrading bacterium was isolated from activated sludge after

spread plating samples onto an inorganic minimal medium

containing 0.1 g/L DEHP (Roslev et al., 1998). The isolated

strain SDE 3 was capable of utilizing DEHP as sole carbon and

energy source, but could also grow on complex substrates

such as trypticase soy broth and various vegetable oils. Strain

SDE 3 was grown in minimal media with rapeseed oil (0.1 g/L)

prior to use in experiments with DEHP degradation. SDE 3 was

characterized based on its fatty acid composition using the

Sherlock Microbial Identification System (MIDI Inc., Newark,

MD, USA). Degradation of [U–14C-ring] DEHP in sludge

samples spiked with SDE 3 was carried out as described

above for non-spiked samples (Section 2.6).

3. Results and discussion

3.1. Extraction of PE from wastewater and activatedsludge

A prerequisite for studying the environmental fate of PE is the

availability of methods that allow accurate quantification of

PE in different environmental matrices. In the present study,

wastewater samples were separated into liquid and solid

subsamples by membrane filtration followed by SPE extrac-

tion of the liquid fraction, and hot solvent extraction of the

solid phase (Fig. 1). A partial clean-up of PE extracts on

aminopropyl SPE columns decreased the concentration of

organic contaminants including humic acids that may other-

wise interfere with the analysis of PE.

The recovery of DMP, DBP, and DEHP from spiked aqueous

samples by SPE (Fig. 1) using C18 columns and hexane:acetone

(1:1) as eluent was 87.3%73.0, 82.3%73.6, and 77.3%77.2,

respectively (mean7SD). The PE recovery was comparable or

slightly better than values obtained by Holadova and Hajslova

(1995) who recommended use of C18 columns and ethyl

acetate as the eluent. Jara et al. (2000) obtained a recovery of

less than 40% for polystyrene columns at high concentration

of DEHP (E100mg/L), which corresponds to the concentration

used in this study. At lower concentrations of DEHP (E1 mg/L),

these authors reached a recovery of about 100% for DEHP.

These results suggest that the relative recovery by SPE of the

most hydrophobic PE such as DEHP will likely increase with

decreasing aqueous concentration. In the proposed extraction

procedure (Fig. 1), PE with low water solubility will primarily

be present in the particle phase and not in the filtrate due to

sorption to organic and inorganic particles. In the investi-

gated WWTP, 95.6% and 90.3% of the measured DEHP was

associated with particles 40.7 mm in the influent and effluent,

respectively, corresponding to concentrations of approxi-

mately 7–29mg/L. As a result, any concentration effects on

the SPE recovery will likely be negligible when measuring

wastewater concentrations of DEHP with the method outlined

in Fig. 1.

Four extraction methods were compared for recovery of PE

from sludge solids: (1) solvent extraction on a rotary shaker,

(2) solvent extraction in an ultrasonic water bath, (3) Soxhlet-

type extraction (FEXTRA), and (4) ‘‘hot solvent extraction’’.

DMP was not detected in sludge solids with any of the four

methods tested because this compound was recovered

exclusively in wastewater filtrates (Fig. 1). Previous studies

focusing on PE extraction from sludge solids have suggested

that ultrasonic extraction is more convenient than Soxhlet

extraction (Zurmuhl, 1990), however, the present results

suggest that simple extraction directly in hot solvent may

be even more advantageous (Fig. 2). The overall recovery of

DEHP from non-spiked sludge samples was significantly

greater (Po0.045) for the hot solvent extraction procedure

compared to the other three methods tested (17–24% greater

recovery). Furthermore, the DEHP recovery was greater in the

first extract with the hot solvent procedure (fraction 1 in Fig.

2). The hot solvent procedure also resulted in the greatest

recovery of DBP and BBP from sludge solids (data not shown).

As a result, the hot solvent procedure was used for

subsequent measurements of PE in sludge solids.

3.2. Fate of PE in Aalborg East WWTP

The fate of PE in Aalborg East WWTP was examined by

analyzing flow proportional samples (24 h) from 3 to 5

different dates. The mean, maximum, and minimum con-

centrations of DMP, DBP, BBP, and DEHP in the influent,

ARTICLE IN PRESS

WAT ER R ES E A R C H 41 (2007) 969– 976 973

effluent, and in dewatered sludge are shown in Table 1. DEHP

was the most abundant PE in the incoming wastewater as

well as in the effluent and in the dewatered sludge (Table 1).

The mean concentration of DEHP at the WWTP inlet was

71.9mg/L (1.585 kg/day) (Tables 1 and 2). At the inlet, o0.1%,

42%, 35%, and 96% of DMP, DBP, BBP, and DEHP, respectively,

0

20

40

60

80

100

Rotaryshaker

Ultrasonic FEXTRA Hot SolventExtraction

DE

HP

(m

g/k

g T

S)

Fraction 1 Fraction 2

Fraction 3 Fraction 4

Fig. 2 – Recovery of DEHP from dewatered sewage sludge

measured with four extraction procedures: (a) solvent

extraction on a rotary shaker at 20 1C, (b) solvent extraction

in a ultrasonic water bath at 40 1C, (c) Soxhlet-type

extraction (FEXTRA), and (d) hot solvent extraction (boiling

solvent). Sludge samples were extracted sequentially

(fractions A–D) with n-hexane:ethyl acetate (1:1). Data are

the means of quadruplicate samples.

Table 2 – Mass balance for DMP, DBP, BBP, and DEHP in Aalbor

Influent (kg/day) Effluent (kg/day)

Mean SD Mean SD

DMP 0.034 0.024 0.003 0.003

DBP 0.602 0.249 0.053 0.011

BBP 0.805 0.645 0.070 0.026

DEHP 1.585 0.512 0.109 0.097

SD: standard deviation.

ND: not detectable.

Table 1 – Concentrations (mean, maximum, and minimum) andinfluent, effluent, and dewatered sludge in Aalborg East WWT

Influent concentrations (mg/L) Effluent con

Mean SD Max. Min. Mean SD

DMP 1.88 1.61 4.31 0.269 0.115 0.11

DBP 20.48 4.74 24.67 15.34 2.38 0.48

BBP 37.87 28.82 80.74 9.41 3.13 1.17

DEHP 71.89 13.64 84.10 53.23 4.92 4.36

ND: not detectable.

was associated with suspended solids retained by the 0.7 mm

filters used for PE extraction (see Fig. 1).

Less than 1.1% of the DBP and BBP entering Aalborg East

WWTP was recovered in the digested dewatered sludge

whereas DMP could not be detected in any of the dewatered

sludge samples tested (Tables 1 and 2). In contrast, 11.7% of

the incoming DEHP was recovered in the dewatered sludge

(Table 2). In these samples, the DEHP concentration varied

between 61.4 and 77.9 mg/kg (Table 1).

On the basis of the measured removal of PE, it was

estimated that 0.031, 0.546, 0.726, and 1.291 kg/day of DMP,

DBP, BBP, and DEHP was degraded biologically at Aalborg East

WWTP (Table 2). Volatilization and abiotic hydrolysis of PE

was considered negligible as suggested by others (Saeger and

Tucker, 1976; O’Grady et al., 1985; Staples et al., 1997). The

main biodegradation activity was likely associated with the

activated sludge process and the anaerobic mesophilic

digestion process as observed in other studies (Cheng et al.,

2000; Marttinen et al., 2003; Gavala et al., 2003). The estimated

biodegradation at Aalborg East WWTP corresponds to an

overall removal of 90–93% of the DMP, DBP, and BBP entering

the WWTP whereas the removal of DEHP was only 81% (Table

2). The estimated biodegradation of DEHP is slightly greater

than the values obtained by Fauser et al. (2003) who

calculated that 70% of the inlet flow of DEHP was degraded

in a Danish activated sludge WWTP with a Biodenipho

configuration similar to Aalborg East WWTP. Marttinen et al.

(2003) reported that only 61% of the inlet flow of DEHP was

degraded in a Finnish activated sludge WWTP with biological

nitrogen removal (nitrification/denitrification) and chemical

phosphorus removal. The process configuration of the

g East WWTP

Dewatered sludge (kg/day) Degraded (kg/day %)

Mean SD Mean Mean

ND ND 0.031 91.2

0.003 0.001 0.546 90.7

0.009 0.003 0.726 90.2

0.185 0.026 1.291 81.4

standard deviation (SD) for DMP, DBP, BBP, and DEHP in theP

centrations (mg/L) Dewatered sludge (mg/kg dw)

Max. Min. Mean SD Max. Min.

9 0.237 ND ND ND ND ND

1 2.73 1.83 1.19 0.27 1.37 1.00

4.33 1.99 3.41 1.26 4.30 2.52

9.93 2.08 67.18 9.28 77.88 61.37

ARTICLE IN PRESS

-0.8

-0.6

-0.4

-0.2

0

0 20 40 60 80

ln (

C/C

0)

Time (h)

Aerobic + strain SDE 3

Aerobic

Aerobic

thermophilic

Denitrifying

Fig. 3 – Degradation of [U–14C-ring] DEHP in activated sludge

samples incubated under aerobic thermophilic conditions

(60 1C) (diamonds), anoxic denitrifying conditions (20 1C)

(triangles), aerobic conditions (20 1C) (circles), and aerobic

conditions after addition of the DEHP degrading bacterium

SDE 3 (squares). Data points represent the means of

triplicate experiments.

WA T E R R E S E A R C H 4 1 ( 2 0 0 7 ) 9 6 9 – 9 7 6974

Finnish plant is largely the same as for the two Danish plants,

except for the anaerobic tanks. Based on the available data, it

is difficult to evaluate why the removal of DEHP was lower in

the Finnish plant. In this plant, approximately 29% of the

DEHP was degraded in the activated sludge process whereas

32% appeared to be degraded during anaerobic sludge

digestion (Marttinen et al., 2003).

The aqueous solubility for DMP, DBP, BBP, and DEHP has

been estimated to be 4200, 4.45, 4.59, and 0.003 mg/L,

respectively (Staples et al., 1997). Despite these notable

differences in solubility, a comparable relative proportion of

the PE entering the WWTP could be detected in the effluent

(7–9%). These values are comparable to data obtained by

Marttinen et al. (2003) who estimated that 6% of the DEHP

entering a Finnish WWTP was not removed.

3.3. Microbial degradation of DEHP

A series of laboratory experiments were conducted to study

factors that may affect microbial DEHP degradation in

activated sludge. Increased knowledge about factors that

limit biodegradation could potentially be used to optimize PE

removal in activated sludge WWTP. The studies focused on

DEHP because this is the most abundant and recalcitrant PE in

wastewater. [U–14C-ring] DEHP was added to sludge samples

as a tracer, and degradation of the aromatic ring was

measured as recovery of 14CO2. Decreases in [U–14C-ring]

DEHP concentrations corresponded to proportional increases

in 14CO2 concentrations suggesting that accumulation of

metabolites such as mono-2-ethylhexyl phthalate (MEHP),

phthalic acid (PA), and protochatechuate (PCA) was less

important. However, the data does not exclude that degrada-

tion products such as 2-ethylhexanoic acid and 2-ethylhex-

anol may have been produced as suggested previously (Nalli

et al., 2002; Horn et al., 2004).

DEHP degradation was generally biphasic with an initial

phase that followed first order kinetics succeeded by a phase

with relatively slower degradation activity (Fig. 3). A shift in

degradation kinetics was most pronounced for samples with

relatively high initial degradation activity. The apparent shift

in degradation kinetics is in agreement with observations

from other environments with microbial DEHP degradation

including soils and sludge-amended soils (Madsen et al., 1999;

Roslev et al., 1998). In these studies, the biphasic degradation

curves were described by an initial phase that followed

first order kinetics followed by a second phase with relatively

slower degradation activity that was described best

by fractional power kinetics. This shift in degradation

activity with time may be attributed to several factors

including changes in bioavailability of DEHP due to altered

sorption–desorption kinetics (Madsen et al., 1999; Roslev et

al., 1998). In the present study, we focused on the initial

degradation phase, and used first order rate coefficients

measured during the initial hours as a convenient means to

compare degradation activity.

The effects of different incubation conditions on the initial

first order rate coefficients for [U–14C-ring] DEHP degradation

in activated sludge are shown in Table 3. Aerobic incubation

of activated sludge at 32 1C increased DEHP degradation 1.4-

fold relative to controls at 20 1C, whereas incubation at 43 1C

decreased rates dramatically. The aerobic rate coefficient at

20 1C (0.0104 h�1) corresponded to an initial DEHP degradation

of 0.86 mg/(kg dw sludge h). The lowest DEHP degradation

activity was observed under aerobic thermophilic conditions

(60 1C) using activated sludge directly or sludge from an

aerobic thermophilic sludge reactor (Table 3 and Fig. 3). This

result is in contrast to reports by Banat et al. (1999) suggesting

that degradation of DEHP in municipal sewage sludge may be

increased under aerobic thermophilic conditions. These

authors reported an approximate 2-fold increase in DEHP

degradation in sludge reactors at 63 1C compared to the

removal at 20 1C.

Degradation of DEHP by aerobic microorganisms in acti-

vated sludge may be attenuated by the presence of competing

substrates including lipids that contain ester bonds. Mixed

results were obtained by preincubating activated sludge for

24–48 h under aerobic conditions to facilitate degradation of

potentially competing substrates prior to [U–14C-ring] DEHP

addition (Table 3). Aerobic preincubation for 24 h resulted in a

slight increase in subsequent DEHP degradation relative to

aerobic controls whereas preincubation for 48 h resulted in a

noticeable decrease (Table 3).

The availability of oxygen appeared to be a major regulator

of DEHP degradation in activated sludge. Anoxic denitrifying

conditions decreased [U–14C-ring] DEHP degradation to about

one-third of aerobic controls whereas alternating conditions

(changes between aerobic and anoxic denitrifying conditions

every 2 h) resulted in rates that were one-fourth lower than

aerobic controls. These results support previous findings

suggesting that DEHP degradation is attenuated in the

absence of oxygen (O’Connor et al., 1989; Ejlertsson et al.,

1997; Jianlong et al., 2000; Alatriste-Mondragon et al., 2003;

Gavala et al., 2003). However, significant differences in

degradation activity may exist between different electron

acceptor regimes. For example, degradation rates measured

in the present study under anoxic denitrifying conditions are

ARTICLE IN PRESS

Table 3 – First order rate coefficients for initial degradation of [U–14C-ring] DEHP in activated sludge

Incubation conditions Temperature (1C) First order rate coefficient

h�1 R2

Aerobic 20 0.0104 0.99

Aerobic 32 0.0143 0.99

Aerobic 43 0.00134 0.98

Aerobic 60 0.0002 0.96

Aerobic (sludge from a thermophilic reactor) 60 0.0007 0.99

Aerobic (after preincubation under aerobic conditions for 24 h) 20 0.0117 0.99

Aerobic (after preincubation under aerobic conditions for 48 h) 20 0.0063 0.97

Anoxic denitrifying 20 0.0032 0.99

Alternating (aerobic–anoxic denitrifying) 20 0.0077 0.99

Aerobic (after aging in the presence of DEHP for 18 h under anaerobic conditions) 20 0.0052 0.97

Aerobic (after aging in the presence of DEHP for 42 h under anaerobic conditions) 20 0.0048 0.97

Aerobic (+strain SDE 3 (20:1 vol:vol)) 20 0.0488 0.97

Aerobic (+strain SDE 3 (1:1 vol:vol)) 20 0.0905 0.77

WAT ER R ES E A R C H 41 (2007) 969– 976 975

19–21 times greater than DEHP degradation rates measured by

Gavala et al. (2003) under anaerobic methanogenic conditions.

These findings suggest that most degradation of DEHP took

place in the aerobic and anoxic tanks of Aalborg East WWTP

compared with degradation in the digester. Degradation

during mesophilic treatment of sludge in anaerobic digesters

at WWTP will likely require somewhat longer retention times

to obtain efficient DEHP degradation compared to activated

sludge treatment under aerobic and anoxic conditions.

The potential effects of pollution age on DEHP degradation

were investigated by aging activated sludge samples under

anaerobic conditions (no nitrate) at 5 1C in the presence of

[U–14C-ring] DEHP prior to incubation under aerobic condi-

tions at 20 1C. This preincubation may result in partial

sorption of the added DEHP to sludge particles resulting in a

potentially lower bioavailability. No degradation of [U–14C-

ring] DEHP to 14CO2 was observed during the cold anaerobic

aging step which is in accordance with several studies

suggesting that DEHP degradation under anaerobic methano-

genic conditions is low or absent (Ejlertsson et al., 1997;

Alatriste-Mondragon et al., 2003; Gavala et al., 2003). Aging for

18 or 42 h reduced the subsequent aerobic DEHP degradation

by about 50% (Table 3) suggesting that bioavailability and

perhaps also DEHP degraders will be affected by the sludge

history (age and oxygen regimes). These findings also suggest

that the age and history of the wastewater entering a WWTP

may have an effect on the subsequent PE degradation rates

within the plant. For example, sewer lines with mainly

aerobic conditions (gravity sewers) may ensure a better

degradation of DEHP in the treatment plant (and within the

sewer), than anaerobic sewer lines (pressure mains).

One aim of the present study was to identify conditions that

may increase PE degradation in activated sludge. A 5- to 9-fold

increase in initial DEHP degradation was observed after

seeding activated sludge samples with different concentra-

tions of the PE degrading bacterium SDE 3 (Table 3 and Fig. 3).

The increase in DEHP degradation relative to controls

occurred during the initial 10 h of incubation. Strain SDE 3

was originally isolated from sludge using DEHP as the only

carbon and energy source. Interestingly, strain SDE 3 main-

tained a high capacity for PE degradation when grown on

various vegetable oils including rapeseed oil suggesting that

the hydrolytic enzymes used for oil hydrolysis were relatively

non-specific. This capacity for PE degradation was main-

tained after repeated transfer and growth of SDE 3 in minimal

media containing rapeseed oil (months), and the strain was

subsequently grown on rapeseed oil prior to its addition to

sludge samples.

Strain SDE 3 was identified as a Rhodococcus spp. Rhodococci

are filamentous gram positive bacteria belonging to the

actinomycetes. Strain SDE 3 displayed strong hydrophobic

characteristics supported by membrane lipids with

many saturated fatty acids. Several Rhodococcus species

including Rhodococcus erythropolis (Nocardia erythropolis) and

Rhodococcus rhodochrous have been shown to possess a

potential for PE metabolism (Kurane, 1986; Nalli et al., 2002).

Further work is needed to evaluate whether Rhodococci or

microorganisms with similar potential for PE metabolism

may be exploited to increase PE degradation in activated

sludge WWTP.

4. Conclusions

�

An activated sludge WWTP with biological removal of

nitrogen and phosphorus removed 91–93% of the PE in

wastewater.

�

Microbial degradation of PE in activated sludge decreased

with the water solubility of the esters. The relative

degradation of DMP, DBP, BBP, and DEHP varied between

91.2% and 81.4%.

�

Despite differences in aqueous solubility, a comparable

fraction of DMP, DBP, BBP, and DEHP entering the WWTP

was recovered in the effluent. Low amounts of DMP, DBP,

and BBP were also recovered in treated dewatered sludge.

In contrast, non-trivial amounts of DEHP were recovered in

the treated sludge with concentrations varying between

61.4 and 77.9 mg/kg.

�

The relatively high concentrations of DEHP in the treated

wastewater and the dewatered sludge are important for

ARTICLE IN PRESS

WA T E R R E S E A R C H 4 1 ( 2 0 0 7 ) 9 6 9 – 9 7 6976

risk assessments and evaluations of environmental

sources of PE.

�

DEHP degradation in activated sludge was severely atte-

nuated by aging and/or oxygen limitation. In contrast,

moderate temperature increases and selection for or

addition of specialized bacteria may stimulate DEHP

degradation. Hence, these factors should be considered

to enhance PE degradation in activated sludge WWTP.

Acknowledgments

We thank Kirsten Maagaard for excellent technical assis-

tance. This work was supported by a grant from the Danish

Technical Research Council (‘‘Activity and Diversity in Com-

plex Microbial Systems’’).

R E F E R E N C E S

Alatriste-Mondragon, F., Iranpour, R., Ahring, B.K., 2003. Toxicityof di-(2-ethylhexyl)phthalate on the anaerobic digestion ofwastewater sludge. Water Res. 37, 1260–1269.

Banat, F.A., Prechtl, S., Bischof, F., 1999. Experimental assessmentof bio-reduction of di-2-ethylhexyl phthalate (DEHP) underaerobic thermophilic conditions. Chemosphere 39, 2097–2106.

Blom, A., Ekman, E., Johannisson, A., Norrgren, L., Pesonen, M.,1998. Effects of xenoestrogenic environmental pollutants onthe proliferation of a human breast cancer cell line (MCF-7).Arch. Environ. Contam. Toxicol. 34, 306–310.

Cheng, H.F., Chen, S.Y., Lin, J.G., 2000. Biodegradation of di-(2-ethylhexyl)phthalate in sewage sludge. Water Sci. Technol. 41,1–6.

Ejlertsson, J., Alnervik, M., Jonsson, S., Svenson, B.H., 1997.Influence of water solubility, side-chain degradability, andside-chain structure on the degradation of phthalic acid estersunder methanogenic conditions. Environ. Sci. Technol. 31,2761–2764.

Fauser, P., Vikelsøe, J., Sørensen, P.B., Carlsen, L., 2003. Phthalates,nonylphenols and LAS in an alternately operated wastewatertreatment plant—fate modelling based on measured concen-trations in wastewater and sludge. Water Res. 37, 1288–1295.

Fromme, H., Kuchler, T., Otto, T., Pilz, K., Muller, J., Wenzel, A.,2002. Occurrence of phthalates and bisphenol A and F in theenvironment. Water Res. 36, 1429–1438.

Gavala, H.N., Alatriste-Mondragon, F., Iranpour, R., Ahring, B.K.,2003. Biodegradation of phthalate esters during the mesophi-lic anaerobic digestion of sludge. Chemosphere 52, 673–682.

Henze, M., Harremoes, P., La Cour Jansen, J., Arvin, E., 2002.Wastewater Treatment. Biological and Chemical Processes,third ed. Springer, Berlin.

Holadova, K., Hajslova, J., 1995. A comparison of different ways ofsample preparation for the determination of phthalic acidesters in water and plant matrices. Issue Series Title: Int. J.Environ. Anal. Chem. 59, 43–57.

Horn, O., Nalli, S., Cooper, D., Nicell, J., 2004. Plasticizermetabolites in the environment. Water Res. 38, 3693–3698.

Jara, S., Lysebo, C., Greibrokk, T., Lundanes, E., 2000. Determina-tion of phthalates in water samples using polystyrene solid-phase extraction and liquid chromatography quantification.Anal. Chim. Acta 407, 165–171.

Jianlong, W., Ping, L., Yi, Q., 1996. Biodegradation of phthalicacid esters by acclimated activated sludge. Environ. Int. 22,737–741.

Jianlong, W., Lujun, C., Hanchang, S., Yi, Q., 2000. Microbialdegradation of phthalic acid esters under anaerobic digestionof sludge. Chemosphere 41, 1245–1248.

Kurane, R., 1986. Microbial degradation of phthalate esters.Microbiol. Sci. 3, 92–95.

Madsen, P.L., Thyme, J.B., Henriksen, K., Møldrup, P., Roslev,P., 1999. Kinetics of di-(2-ethylhexyl)phthalate mineralizationin sludge-amended soil. Environ. Sci. Technol. 33,2601–2606.

Marttinen, S.K., Kettunen, R.H., Sormunen, K.M., Rintala, J.A.,2003. Removal of bis(2-ethylhexyl)phthalate at a sewagetreatment plant. Water Res. 37, 1385–1393.

Morgenroth, V., 1993. Scientific evaluation of the data-derivedsafety factors for the acceptable daily intake. Case study:diethylhexylphthalate. Food Addit. Contam. 10, 363–373.

Nalli, S., Cooper, D.G., Nicell, J.A., 2002. Biodegradation ofplasticizers by Rhodococcus rhodochrous. Biodegradation 13,343–352.

Nielsen, E., Larsen, P.B., 1996. Toxicological evaluation and limitvalues for DEHP and phthalates other than DEHP. DanishEnvironmental Protection Agency.

O’Connor, O.A., Rivera, M.D., Young, L.Y., 1989. Toxicity andbiodegradation of phthalic acid esters under methanogenicconditions. Environ. Toxicol. Chem. 8, 569–576.

O’Grady, D.P., Howard, P.H., Werner, A.F., 1985. Activated sludgebiodegradation of 12 commercial phthalate esters. Appl.Environ. Microbiol. 49, 443–445.

Redeker, J., 1997. Solid–fluid Extraction with Solvents. SpecialIssue from LaborPraxis. IKA, Staufen, Germany, pp. 1–5.

Roslev, P., Madsen, P.L., Thyme, J.B., Henriksen, K., 1998.Degradation of phthalic acid and di-(2-ethylhexyl)phthalate(DEHP) by indigenous and inoculated microorganisms insludge amended soil. Appl. Environ. Microbiol. 64,4711–4719.

Saeger, V.W., Tucker, E.S., 1976. Biodegradation of phthalic acidesters in river water and activated sludge. Appl. Environ.Microbiol. 31, 29–34.

Staples, C.A., Peterson, D.R., Parkerton, T.F., Adams, W.J., 1997. Theenvironmental fate of phthalate esters: a literature review.Chemosphere 35, 667–749.

Vikelsøe, J. Thomsen, M., Johansen, E., 1998. Sources of phthalatesand nonylphenols in municipal waste water. NERI TechnicalReport No. 225, National Environmental Research Institute,Denmark.

Wang, J., Liu, P., Shi, H., Qian, Y., 1997. Kinetics of phthalic acidester degradation by acclimated activated sludge. ProcessBiochem. 32, 567–571.

Zurmuhl, T., 1990. Development of a method for the determina-tion of phthalate esters in sewage sludge including chroma-tographic separation from polychlorinated biphenyls,pesticides and polyaromatic hydrocarbons. Analyst 115,1171–1175.