ARTICLE IN PRESS

Water Research 39 (2005) 97–106

www.elsevier.com/locate/watres

The solids retention time—a suitable design parameter to

evaluate the capacity of wastewater treatment plants to remove
micropollutants
M. Claraa,, N. Kreuzingera, B. Strenna, O. Gansb, H. Kroissa
a
Institute for Water Quality and Waste Management, Vienna University of Technology, A-1040 Vienna, Karlsplatz 13/226, Austria
b
Umweltbundesamt GmbH, Spittelauer Lände 5, 1090 Vienna, Austria
Received 20 February 2004; received in revised form 10 August 2004; accepted 26 August 2004

Abstract

Micropollutants as endocrine disrupting compounds (EDC) or pharmaceuticals are of increased interest in water
pollution control. Wastewater treatment plant (WWTP) effluents are relevant point sources for residues of these
compounds in the aquatic environment. The solids retention time (SRT) is one important parameter for the design of
WWTPs, relating to growth rate of microorganisms and to effluent concentrations. If a specific substance is degraded in
dependency on the SRT, a critical value for the sludge age can be determined. In WWTPs operating SRTs below this
critical value, effluent concentrations in the range of influent concentrations or a distribution according to the
adsorption equilibrium have to be expected, whereas in WWTPs operating at SRTs higher than the critical value
degradation will occur. Critical SRTs were determined for different micropollutants, indicating that the design criteria
based on the sludge age allows an estimation of emissions. Different treatment technologies as conventional activated
sludge systems and a membrane bioreactor were considered and no significant differences in the treatment efficiency
were detected when operated at comparable SRT. The results of the investigations lead to the conclusion that low
effluent concentrations can be achieved in WWTPs operating SRTs higher than 10 days (referred to a temperature of
10 1C). This corresponds to the requirements for WWTPs situated in sensitive areas according to the urban wastewater
directive of the European Community (91/271/EEC) in moderate climatic zones.
r 2004 Elsevier Ltd. All rights reserved.

Keywords: Wastewater treatment; Plant design; Solids retention time; Removal potential; Endocrine disrupting compounds;
Pharmaceuticals; Hazardous substances

1. Introduction al., 2003) and they are of increasing importance in water

Micropollutants are detected in surface waters in
several European countries (Ternes, 1998; Fromme et
al., 2002; Heberer, 2002a; Kreuzinger, 2002; Andersen et
Corresponding author. Tel.: +43 1 58801 22661; fax:
+43 1 58801 22699.
E-mail address: mclara@iwag.tuwien.ac.at (M. Clara).
pollution control. Especially chemicals, which are
suspected to have an impact on humans and wildlife
hormone systems, so-called endocrine disrupting com-
pounds (EDC) and pharmaceuticals, have to be
emphasised. Negative adverse health effects on aquatic
organisms, which could be attributed to EDC, are
documented in various studies (Sonnenschein and Soto,
1998; Sumpter, 1998). In regard to pharmaceuticals

0043-1354/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2004.08.036
 ARTICLE IN PRESS
98 M. Clara et al. / Water Research 39 (2005) 97–106

Nomenclature

v reaction rate (mg l1 d1) s soluble substance concentration (ng l1)
vmax maximum reaction rate (mg l1 d1) KD specific adsorption coefficient (l kg X1
SS )
m specific biomass growth rate (d1) kobs transformation rate (l g X1 1
VSS,R d )
mmax maximum specific biomass growth rate (d1)
K half-saturation constant (mg l1) Indices
S substrate concentration (mg l1)
Y biomass yield constant max maximum
Q volumetric flow rate (l d1 or m3 d1) R reactor
V reactor volume (l or m3) 0 influent
F mass flow rate (ng d1) e effluent
R reactor (aeration tank) (m3) ES excess sludge
X particulate material concentration (mg l1 SS total suspended solids
or g l1) VSS volatile suspended solids

some persistent compounds were tracked from munici-
pal sewage to drinking water (Heberer, 2002b).
Endocrine disrupting compounds as well as residues
of pharmaceuticals are suspected to enter rivers, streams
and surface waters through the effluents of wastewater
treatment plants (WWTPs). Therefore, the elimination
of these substances in WWTPs is of elementary interest.
In this context, it is important whether a relationship
exists between achievable removal efficiencies and
applied design criteria.
A parameter commonly used for WWTPs design is
the sludge retention time (SRT) or sludge age, indicating
the mean residence time of microorganisms in the
reactor. The SRT is related to the growth rate of
microorganisms. Only organisms which are able to
reproduce themselves during this time can be detained
and enriched in the system. According to this definition,
high SRTs allow the enrichment of slowly growing
bacteria and consequently, the establishment of a more
diverse biocoenosis with broader physiological capabil-
ities compared to WWTPs operating low SRTs. The
correlation between treatment efficiency (e.g. carbon
removal, nitrification) and SRT is based on this principle
(ATV-DVWK, 2000) and can be described by Monod
kinetics. In a system with biomass recirculation the
installed SRT corresponds approx. to the reciprocal of
the growth rate. Considering this relationship, the
achievable effluent concentration of a certain compound
is dependent on the selected/operated SRT and inde-
pendent of influent concentrations.
Monod-type kinetic describes the relationship be-
tween the growth rate of a microbial species and the
concentration of a critical substance (or substances)
sustaining its growth. More ‘‘food’’ means faster growth
until the maximum growth rate mmax is achieved. Beyond
this point, other factors become limiting. This implies
that the chemical must be sufficiently easy to metabolise
and it also must be available in sufficient amounts to
result in significant energy and/or biomass recovery
(Schwarzenbach et al., 2003). In biological wastewater
treatment usually, substrate utilization and not cellular
growth rates are measured. Therefore, the enzyme-based
Michaelis–Menten kinetics seems to be more suitable to
describe the concentration gradients of the investigated
substrates. The Monod equilibrium can be interpreted as
a specific case of Michaelis–Menten kinetics. The
degradation of a certain amount of chemical mass
enables a proportional enhancement in microbial
biomass. This proportionality factor is called yield (Y),
expressing the ratio of cell production and mass of
substrate (S) converted. Applying this factor, the
removal rate of a chemical of concern can be
related to the production rate of new cells (Schwarzen-
bach et al., 2003). Assuming a constant Y, the
Monod approach corresponds to Michaelis–Menten
kinetics
dc S
¼ v ¼ vmax
dt KMS
mX VSS S X VSS
¼ ¼ mmax : ð1Þ
Y KS þ S Y
If a compound is present only in trace levels (ng l1,
lower mg l1 range) the removal of this substance does
not result in any significant biomass growth. Such
chemicals are usually transformed by a cometabolism
and the concentration variation rates correspond to a
mixed substrate growth. This can be described by the
multisubstrate Monod growth relationship, where mtot is
the total specific growth rate and mi is the specific growth
rate on substrate i, and the summation is taken over the
n substrates (Guha et al., 1999; Schwarzenbach et al.,
2003)
X
n
m  Si
mtot ¼ mi ; mi ¼ Pnmax;i : (2)
K i;S þ j¼1 ðK i;S =K j;S Þ  S j
i¼1
 ARTICLE IN PRESS
M. Clara et al. / Water Research 39 (2005) 97–106
If a micropollutant is biodegradable and this degrada-
tion can be described by the mentioned kinetics, even in
the low-concentration range and in case of mixed
substrate growth (cometabolism) a specific SRT can be
associated to the substance. This specific SRT must be
higher than the reciprocal of mtot (SRT4m1 tot ) and mtot is
predominantly determined by mmax of the primary
substrate. In dependency on mmax and KS of a primary
substrate and the specific biomass decay rate a critical
value for the SRT can be identified. At SRTs operated
below this critical value, no removal due to biodegrada-
tion is expected and if a substance is not adsorbed to
sludge, effluent concentrations in the range of influent
concentrations are probable. In WWTPs operated at
higher SRTs than the critical value a complete removal
of the respective substances is to be expected. Such
behaviour can be observed for nitrification. If loading is
very constant over time a distinct value marks the
critical SRT, whereas strongly varying effluent concen-
trations are observed within a certain SRT range at
unsteady influent conditions.
If degradation according to the described kinetic can
be assumed, the achievable effluent concentrations of the
investigated substances are not related to influent
concentrations but depend only on the operated SRT,
whereas treatment efficiencies in percent depend notably
on inflow concentrations.
The maximum growth rate depends on temperature.
Therefore the SRT is temperature dependent, too. Every
increase of temperature of 10 1C leads to a duplication of
microbial activity. Regarding the influence of mmax on
the SRT a comparable dependency of the SRT on
temperature can be assumed. According to Eq. (3), the
SRT referred to a reference temperature (Tref) can be
calculated from the operated SRT and the actual
temperature (T). The correction describes temperature
dependency of mmax of heterotrophic biomass, and seems
suitable for the correction of the SRT as the SRT is
determined by the excess sludge abstraction and carbon
removal is the relevant influencing process
SRTTref ¼ SRTT 1:072ðT ref 10Þ : (3)
Different organic micropollutants were investigated in
lab-scale facilities operated at different SRTs as well as
in full-scale WWTPs in order to detect substances for
which a critical SRT can be defined. Additionally, to
conventional WWTPs a membrane plant equipped with
ultrafiltration membranes was studied. The investigated
substances include the xenohormone bisphenol-A
(BPA), the natural estrogens 17b-estradiole (E2), estrone
(E1) and estriole (E3) and the pharmaceuticals 17a-
ethinylestradiole (EE2), ibuprofen (IBP), bezafibrate
(BZF), diclofenac (DCF) and carbamazepine (CBZ).
As the natural estrogens are subjected to transforma-
tions among themselves (Ternes, 1999; Joss et al., 2004),
99
calculations (mass balances) were performed on a molar
basis in order to be able to sum up the three fractions.
2. Materials and methods
2.1. Analytical methods
The chemical analyses were performed by the
Austrian Umweltbundesamt GmbH and PharmAnalyt
Ltd. (Austria). Statistical evaluation of measurements
and determination of limits of quantification (LOQ)
occurred according to DIN 32645 (1994). Indicated
recoveries are mean recovery rates for the whole
analytical step, including sample preparation.
BPA samples were acidified with 1 N sulphuric acid
and percolated by Isolute-C18 solid phase extraction
(SPE) columns at a flow rate of approx. 10 ml min1,
previously conditioned by 5 ml methanol and 5 ml water.
Prior to extraction a surrogate standard was added.
After drying with nitrogen, the compounds were eluted
by 5 ml acetone and 5 ml methanol/methyl-tert.-butyl-
ether (1:9). The extracts were evaporated below 0.5 ml
by nitrogen and filled up to a final volume of 1 ml with
acetonitrile. Ten microlitres of internal standard were
added prior to analysis. Aliquots of 10 ml were used for
analysis with LC–MS/MS applying electrospray ionisa-
tion (EI). The recoveries amounted to 80%. The LOQ
was 20 ng l1.
CBZ was analysed by LC–MS/MS. Ionisation of the
analytes was done by electrospray ionisation in positive
mode. Prior to sample extraction, a surrogate standard
(dihydrocarbamazepine or meclofenamic acid) was
added to the samples. Two different solid-phase extrac-
tion phases (cyclohexane (CH)- and ENV+ phase)
were employed for LC–MS/MS sample preparation. The
samples were acidified in the case of the C18 and CH
SPE cartridges, whereas the addition of a neutral EDTA
buffer solution was necessary for the ENV+ phase.
Mean recoveries of 75% are achieved and the LOQ
amounts to 20 ng l1.
IBP, DCF and BZF were separated and analysed by
GC–(EI)MS in the multiple ion detection (MID) mode.
Analytes were enriched by solid-phase extraction on a
RP-18 column under acidic conditions. Prior to sample
extraction a surrogate standard (meclofenamic acid) was
added to the samples. After elution with ethyl acetate
and dichloromethane a derivatization with diazo-
methane occurred, followed by a silica gel clean up
step. Mean recovery rates of 97% for IBP, 98% for
DCF and of 99% for BZF were achieved. For those
substances the LOQ was 20 ng l1 in effluent samples
and 100 ng l1 in influent samples.
Sample preparation for the determination of the
steroid hormones included liquid–liquid extraction with
ether followed by SPE with ENV+ cartridges eluted by
 ARTICLE IN PRESS
100 M. Clara et al. / Water Research 39 (2005) 97–106
methanol. After evaporation to dryness a derivatization
with dansylchloride occurred. Prior to extraction surro-
gate standards were added. The analysis was performed
using LC–MS/MS using an injection volume of 20 ml.
Ionisation of the analytes was done by electrospray
ionisation in positive mode. Recovery rates of nearly
100% for the extraction with ether and of more than
80% for the SPE were achieved. The LOQ for E1, E3
and EE2 was 1 ng l1 and for E2 5 ng l1.
2.2. Lab-scale experiments
Laboratory-scale treatment plants (LP) were operated
at four different SRTs. Characteristic SRTs for waste-
water treatment design according to ATV-DVWK
(2000) were chosen (1, 5, 13 and approx. 26 days). The
temperature in the lab-scale reactors varied between
20.3 1C and 27.4 1C. The test facilities were fed with
synthetic wastewater according to EN-ISO-9887 (1994).
The spiked substances were purchased from Sigma-
Aldrich.
The lab-scale plant running with a SRT of approx.
one day was operated as a sequencing batch reactor
(SBR, VR=4 l), whereas the experiments with higher
SRTs were performed in compact lab-scale plants with a
total volume of 10 l (bioreactor 8 l, final sedimentation
2 l). The plants were inoculated with sludges from
WWTPs with comparable SRTs as the installed ones.
All test facilities were operated with inflows of about
4.5 l d1. Hence, the hydraulic retention time was about
one day for the SBR and approx. two days for the other
experimental plants. While the SBR was aerated
continuously, the other bioreactors were intermittently
aerated, thus allowing nitrification and denitrification.
Excess sludge was removed once a day directly from the
reactors. In order to establish a biocoenosis typical for
the installed SRTs and to obtain equilibrium in the
system, the plants were operated for adaptation two to
three SRTs prior to sampling. The lab-scale experiments
are described in Clara et al. (2003).
Table 1
Characterisation of the investigated STPs
Plant characterisation
WWTP 1 PS, AS C, P 2,500,000
WWTP 2 AS+AS, AD C, P, N, DN
WWTP 3 PS, AS, AD C, P, N, DN
WWTP 4 AS, SSS C, P, N, DN
WWTP 5 MBR C, N, DN
PS—primary settling, AS—activated sludge system, MBR—membrane bioreactor, AD—anaerobic sludge digestion, SSS—
simultaneous sludge stabilisation, C—carbon removal, N—nitrification, DN—denitrification, P—phosphorus precipitation, pe—
population equivalent, SRT10 1C–SRT related to 10 1C (see Eq. (7)), VR—biological treatment reactor volume.
2.3. Full-scale wastewater treatment plants
Sampling campaigns were performed at selected full-
scale municipal wastewater treatment plants of different
sizes and different treatment efficiencies. A characterisa-
tion of the investigated plants is summarised in Table 1.
WWTP 4 and WWTP 5 were sampled three times at
different loading conditions. Daily composite samples of
influent and effluent were taken. In addition to the
conventional WWTPs (WWTP 1–WWTP 4) a pilot
plant (WWTP 5) equipped with membrane technology
and operated at different SRTs, was sampled. In the
membrane bioreactor (MBR) the separation of treated
wastewater and sludge is achieved with an ultrafiltration
membrane.
2.4. Mass balances
A plausibility analysis of the measured values was
made, using mass balances of phosphorus, nitrogen and
chemical oxygen demand (COD) according to Nowak et
al. (1999). Based on that data, an evaluation of process
characteristics (e.g. SRT, specific excess sludge produc-
tion, etc.) was performed.
Describing the removal of micropollutants different
possible pathways during wastewater treatment were
taken into account. The most important processes to
which an organic compound is subjected are volatilisa-
tion, adsorption and degradation, determining the
relevant removal pathways. According to Struijs et al.
(1991) volatilisation is negligible for the selected
substances. The removal is dependent on adsorption
and degradation/transformation processes. The flux
adsorbed to the sludge was calculated from the
measured dissolved effluent concentrations using equili-
brium solid/liquid partition coefficients KD (Clara et al.,
2004; Ternes et al., in press).
F 0 ¼ Q0 s0 ð1 þ K D :X SS;0 Þ; (4)
F e ¼ Qe se ð1 þ K D :X SS;e Þ; (5)
Design capacity (pe) SRT10 1C (d) VR (m3)
2 42,000
167,000 19 644+10,456
135,000 48 22,800
6,000 4100/42 3,200
50 22/82/40 4,2
 ARTICLE IN PRESS
M. Clara et al. / Water Research 39 (2005) 97–106 101

Contradictory results were obtained for diclofenac and

F ES ¼ QES se ð1 þ K D :X SS;R Þ;
removal ¼ ðF 0  F e Þ=F 0 ;
biotransformation ¼ F 0  F e  F ES
¼ kobs X VSS;R se V R :
To compare the results obtained for the different
treatment plants, total sludge retention times related to
10 1C (SRT10 1C) were calculated for all plants (Eq. (3)).
The reference temperature of 10 1C was chosen accord-
ing to the design guideline for municipal wastewater
treatment plants of the German Association for Water,
Wastewater and Waste (ATV-DVWK, 2000). The
calculation is based on the SRTs resulting from COD
mass balances, mean measured temperatures (T) in the
bioreactors and a correction coefficient (fP=1.072) for
the temperature.
3. Results and discussion
Different removal efficiencies were calculated for the
investigated substances. With increasing sludge age,
enhanced removal is observed for most substances.
(6)
ethinylestradiole. The antiepileptic drug carbamazepine
(7) is not removed during wastewater treatment and effluent
concentrations within the range of the corresponding
influent concentrations were measured. The measured
influent and effluent concentrations (ng l1) of the
ð8Þ investigated compounds in the different sampled treat-
ment facilities are shown in Table 2.
The results of the mass balances and the calculated
mean removal rates (%) for the investigated substances
are qualitatively summarised in Table 3. The removal
indicates the elimination from the liquid phase including
adsorption and degradation/transformation (Eq. (7)).
The mass balances for the steroid hormones do not
include metabolites.
Bisphenol-A, the natural estrogens, ibuprofen and
bezafibrate were removed nearly completely in depen-
dency on the applied sludge retention time. Fig. 1
illustrates the calculated removal rates (%) and the
measured effluent concentrations (ng l1) of these
compounds in the different treatment facilities in
dependency on the SRT10 1C (d).
At SRTs10 1C higher than 10 days removal rates of
more than 80% were observed for bisphenol-A. These
results confirm the ‘‘ready biodegradability’’ of BPA,

Table 2
Mean measured influent and effluent concentrations (ng l1) of the investigated substances at the sampled treatment systems

BP-A E2 E1 E3 EE2 IBP BZF DCF CBZ

LP 1 in 1895 57 35 n.d. 28 1900 4700 4189 3670

out 1840 18 38 n.d. 106 2000 3700 3480 3828
LP 2 in 1895 57 35 n.d. 28 1900 4700 4189 3670
out 410 n.d. 1 n.d. 8 63 362 2700 3872
LP 3 in 1895 57 35 n.d. 28 1900 4700 4189 3670
out 166 n.d. 2 n.d. 11 19 33 3267 3725
LP 4 in 1895 57 35 n.d. 28 1900 4700 4189 3670
out 164 n.d. 2 n.d. 10 18 108 3320 3462
WWTP 1 in 1710 54 34 336 8 2300 7600 1400 670
out 1530 30 72 275 5 2400 4800 1300 690
WWTP 2 in I 1255 35 51 23 4 3650 1525 1040 453
in II 1153 14 69 24 3 3675 1625 1440 517
out 723 n.d. 8 17 3 300 1030 1300 758
WWTP 3 in 720 46 670 143 70 1200 1550 905 325
out 125 n.d. n.d. n.d. n.d. 24 715 780 465
WWTP 4 in 2025 67 71 326 20 1480 1960 3250 1850
out 26 n.d. 4 n.d. 4 n.d. n.d. 1536 1594
WWTP 4 in 2376 82 29 372 3 2679 2014 4114 1200
out 35 8 n.d. 1 n.d. n.d. n.d. 1533 1337
WWTP 4 in 2151 125 81 660 3 2448 6840 3190 704
out 76 4 2 n.d. n.d. 20 692 1680 952
WWTP 5 in 2025 67 71 326 20 1480 1960 3250 1850
out 28 n.d. 2 n.d. 4 22 103 3464 1619
WWTP 5 in 2376 82 29 372 3 2679 2014 4114 1200
out 16 3 21 1 n.d. 22 73 2033 1147
WWTP 5 in 2151 125 81 660 3 2448 6840 3190 704
out 158 6 n.d. n.d. 2 69 1550 2140 794
 ARTICLE IN PRESS
102 M. Clara et al. / Water Research 39 (2005) 97–106

Table 3
Qualitative removals rates calculated from the mass balances according to Eq. (4)–Eq. (7) and related to the correspondent influent
mass fluxes

Lab-scale WWTPs Full-scale WWTPs

L1 L2 L3 L4 1 2 3 4 5

SRT10 1C (d) 2 10 34 68 2 0.6 19 48 42 182 550 22 40 82

BPA  + ++ ++  + + ++ ++ ++ ++ ++ ++ ++
E1+E2+E3 + ++ ++ ++   + ++ ++ ++ ++ ++ ++ ++
IBP  ++ ++ ++   ++ ++ ++ ++ ++ ++ ++ ++
BZF  ++ ++ ++    + ++ ++ ++ ++ + ++
EE2  + + + +   ++ + ++ + ++  +
DCF         + + +  + +
CBZ              

removal o20%; , 20%premovalo40%; +, 40%premovalo60%; +, 60%premovalo80%; ++, 80%premovalo100%.

documented by various authors (Staples et al., 1998;
Langford and Lester, 2003). In addition, the achievable
effluent concentrations reflect the dependency of the
removal efficiency on the SRT10 1C. At sludge ages
higher than 20 days, effluent concentrations near or
below the LOQ of 20 ng l1 were measured. According
to prior explanations and Fig. 1 a critical SRT10 1C of
about 10 days can be identified from the presented
results. However, the wide variation range of this critical
value has to be mentioned.
The natural estrogens 17b-estradiole (E2), estrone (E1)
and estriole (E3) are eliminated from the wastewater in
dependency on the SRT10 1C. At SRTs10 1C higher than
10 days nearly complete removal of those compounds
was achieved and effluent concentrations in the range of
the LOQ were measured. Comparable results are
reported by Joss et al. (2004). The critical SRT10 1C
determined for the natural estrogens amounts between 5
and 10 days.
Similar results were obtained for the pharmaceuticals
ibuprofen and bezafibrate. Both substances were re-
moved to a large extent during wastewater treatment
and treatment efficiencies of more than 95% were
calculated. Comparable results are reported by several
authors (Stumpf et al., 1998; Buser et al., 1999; Zwiener
et al., 2001). A critical value for the SRT10 1C can be
defined for those compounds, amounting to about 5
days for ibuprofen and to about 10 days for bezafibrate.
Whereas the range of variation is small for ibuprofen,
stronger fluctuations of the effluent concentration were
observed for bezafibrate.
Summarising these results it can be stated, that high
removal rates and low effluent concentrations are
achieved at SRTs10 1C higher than 10 days. Compared
to the design guideline of the German Association for
Water, Wastewater and Waste (ATV-DVWK, 2000) this
corresponds to the design criteria for nitrogen removal.
According to the presented results it can be deduced,
that WWTPs designed for nitrogen removal achieve
high eliminations for certain micropollutants through-
out the year.
Contradictory results were obtained for 17a-ethiny-
lestradiole and diclofenac. In some of the investigated
treatment facilities high removal rates were observed,
whereas in other WWTPs at comparable SRTs10 1C no or
only slight removal rates were obtained (see Table 3).
The calculated removal rates (%) and the measured
effluent concentrations (ng l1) of the different treatment
systems in relation to the SRT10 1C are shown in Fig. 2.
For none of the two substances a correlation between
achieved effluent concentrations or removal efficiencies
and the SRT10 1C is detectable and therefore, no critical
value for the SRT10 1C can be identified. For diclofenac
the effluent concentrations vary within a wide range,
whereas the effluent concentrations of EE2 are between 1
and 5 ng l1, except for the lab-scale experiments
resulting in effluent concentrations of approx. 10 ng l1.
Similar contradictory results are documented in litera-
ture for the analgesic drug diclofenac. Whereas Buser et
al. (1999) and Heberer (2002a) report no significant
elimination of diclofenac, Ternes et al. (1998) observed
elimination rates of up to 70%. Further contradictory
observations were documented with regard to EE2.
Whereas Ternes et al. (1999) observed no significant
removal of EE2 during batch experiments, Baronti et al.
(2000) and Joss et al. (2004) report eliminations of more
than 85% in full-scale wastewater treatment plants.
According to prior explanations, it has to be
mentioned, that treatment efficiencies in percent are
dependent on the influent concentrations. Especially for
ethinylestradiole influent concentrations in the range of
5 ng l1 were measured. Since the measured effluent
concentrations were within this range, no or only slight
removal was observed. In regard to this observation it
 ARTICLE IN PRESS
M. Clara et al. / Water Research 39 (2005) 97–106 103

120 2000
BPA BPA
100
1500
80
removal [%]

se [ng l-1]
60 1000

40
500
20

0 0
0.5 1 2 5 10 20 50 100 200 0.5 1 2 5 10 20 50 100 200
SRT10°C [d] SRT10°C [d]

120 1.4
E1 + E2 + E3 E1 + E2 + E3
100 1.2

1.0
80
removal [%]

se [nmol l-1]
0.8
60
0.6
40
0.4
20 0.2

0 0.0
0.5 1 2 5 10 20 50 100 200 0.5 1 2 5 10 20 50 100 200
SRT10°C [d] SRT10°C [d]

120 4000
IBP IBP
100
3000
80
removal [%]

se [ng l-1]

60 2000

40
1000
20

0 0
0.5 1 2 5 10 20 50 100 200 0.5 1 2 5 10 20 50 100 200
SRT10°C [d] SRT10°C [d]

120 5000
BZF BZF
100
4000
80
removal [%]

se [ng l-1]

3000
60
2000
40

20 1000

0 0
0.5 1 2 5 10 20 50 100 200 0.5 1 2 5 10 20 50 100 200
SRT10°C [d] SRT10°C [d]

Fig. 1. Calculated removal efficiencies (%) and measured dissolved effluent concentrations se (ng l1, nmol l1) in relation to the
SRT10 1C for BPA, E1+E2+E3, IBP and BZF in the different treatment facilities (K, LP1-LP4, J, WWTP 1; m, WWTP 2; n, WWTP
3; ’, WWTP 4; &, WWTP 5).
 ARTICLE IN PRESS
104 M. Clara et al. / Water Research 39 (2005) 97–106

100 4000
DCF DCF
80
removal [%] 3000
60

se [ng l-1]
40 2000

20
1000
0

-20
0.5 1 2 5 10 20 50 100 200 0.5 1 2 5 10 20 50 100 200
SRT10°C [d] SRT10°C [d]

100 14
EE2 EE2
80 12
10
removal [%]

60

se [ng l-1]
8
40
6
20
4
0 2
-20 0
0.5 1 2 5 10 20 50 100 200 0.5 1 2 5 10 20 50 100 200
SRT10°C [d] SRT10°C [d]

Fig. 2. Calculated removal efficiencies (%) and measured effluent concentrations (ng l1) for DCF and EE2 in the sewage treatment
facilities (K, LP1-LP4; J, WWTP 1; m, WWTP 2; n, WWTP 3; ’, WWTP 4; &, WWTP 5) in relation to the SRT10 1C.

Table 4
Observed degradation/transformation rates kobs (l g X1 1
VSS,R d ) for the different investigated WWTPs, according to Eq. (8)

Lab-scale WWTPs Full-scale WWTPs

L1 L2 L3 L4 1 2 3 4 5

SRT10 1C (d) 2 10 34 68 2 0.6 19 48 42 182 550 22 40 82

XSS,R (g l1) 0.5a 0.9a 1.8a 3.2a 4.0 5.1 5.9 3.1 4.9 4.0 4.0 6.3 4.5 11.8
XVSS,R (g l1) 0.4a 0.7a 1.5a 2.7a 3.2 3.5 4.1 2.0 3.2 2.4 2.7 4.7 3.2 9.3
BPA 1.5 2.2 4.1 2.6 0.7 — 0.8 2.1 1.6 3.6 4.4 45.2 0.5 73.7
E1+E2+E3 0.9 12.0 4.6 7.6 0.7 0.7 1.3 51.5 4.5 1.6 1.4 23.1 3.3 6.0
IBP 1.0 37.9 47.5 44.6 0.1 0.1 6.5 20.1 3.2 2.7 9.2 29.0 1.0 42.6
BZF 0.2 22.7 79.5 17.4 2.2 0.5 0.3 0.5 0.2 4.3 7.1 7.7 0.1 10.2
EE2 1.8 1.8 0.6 1.2 2.8 0.5 0.1 29.1 0.1 0.1 0.1 1.9 0.0 0.8
DCF 0.4 0.8 0.1 0.0 0.8 1.0 0.1 0.1 0.0 0.0 0.1 0.0 0.0 0.4
CBZ 0.0 0.1 0.1 0.0 0.1 0.1 0.3 0.1 0.0 0.0 0.0 0.1 0.0 0.0
a
Values valid for the reactors with BPA and the steroid hormones. The reactors with IBP, BZF, DCF and CBZ were operated with
lower volumetric loading rates resulting in lower suspended solids concentrations: XSS,R (XVSS,R) in L1 0.2 (0.15) g l1; L2 0.6 (0.4)
g l1; L3 1.0 (0.8) g l1; L4 1.2 (1.0) g l1.

has to be indicated, that if a substance is degraded
according to the kinetics described, a nearly complete
removal associated to negligible effluent concentrations
can be anticipated at correspondent SRTs10 1C. A
possible explanation for the observed behaviour of
EE2 is that the measured concentrations are lower or
within the range of the respective KS value.
Degradation/transformation rates kobs were calcu-
lated according to Eq. (8) and summarised in Table 4.
Negative values result due to negative mass balance
performance.
The presented biotransformation rates have to be
interpreted as rough estimates, as they were derived
from average values assuming a completely mixed and
 ARTICLE IN PRESS
M. Clara et al. / Water Research 39 (2005) 97–106
steady state system. Different redox conditions were not
considered. The calculation assumes comparable rates
for aerobic, anoxic and anaerobic conditions, as none or
only poor data on degradation pathways and required
redox conditions are known for the investigated
substances. Joss et al. (2004) observed a dependency of
the removal of E1 and EE2 on redox conditions,
maximum rates occurring under aerobic conditions.
Additional information would allow an adjustment of
the coefficients presented in Table 4. Furthermore it is
annotated, that if a plant operates SRTs10 1C higher than
the critical one and degradation is observed, kobs
depends on the influent concentrations as calculated
removal rates do.
Regarding the different investigated treatment tech-
nologies (conventional activated sludge systems versus
MBR) no significant difference is observed if run at
comparable SRT10 1C. However, MBRs allow achieving
high SRTs10 1C within a compact reactor, resulting in
higher volumetric reaction rates as compared to
conventional treatment systems.
4. Conclusions
The investigated micropollutants showed different
behaviours during the wastewater treatment process.
Some of the compounds were eliminated in dependency
on the solids retention time, whereas the antiepileptic
drug carbamazepine was not affected during the
treatment. For diclofenac and 17a-ethinylestradiole
contradictory results were obtained and beside the
SRT other influences seem to be of importance.
With regard to bisphenol-A, ibuprofen, bezafibrate
and the natural estrogens a strong correlation between
achievable effluent concentrations and the SRT10 1C was
observed. For these substances a critical value was
identifiable for the SRT10 1C amounting to approx. 10
days. The widest variation was determined for bi-
sphenol-A reaching approx. 5 days. Regarding different
treatment technologies, no differences in the perfor-
mance between conventional systems and the MBR was
detected.
The presented observations lead to the conclusion
that for WWTPs operated at SRTs10 1C higher than 10
days low effluent concentrations can be achieved,
leading to a significant reduction of emissions. Com-
pared to design guidelines the recommended minimum
sludge age corresponds to the requirements for nitrogen
removal (nitrification, denitrification). According to the
urban wastewater directive of the European Community
(91/271/EEC) this recommended treatment efficiency
corresponds to the standards for WWTPs situated in
sensitive areas. It can be concluded that WWTPs with
nitrogen removal also efficiently remove degradable
micropollutants. WWTPs designed for BOD removal
105
only show a high optimisation potential in regard to the
removal of conventional wastewater parameters (e.g.
nitrogen) as well as of micropollutants. Sensitive area
requirements according to 91/271/EEC not only
strongly reduce oxygen consumption for C and N
oxidation in receiving waters and contribute to eutro-
phication abatement by N and P elimination, but also
achieve efficient removal of micropollutants, partly
target compounds (e.g. BPA) included in the European
Water Framework Directive (2000/60/EC).
Acknowledgments
Results are derived from POSEIDON Project ‘‘As-
sessment of Technologies for the Removal of Pharma-
ceuticals and Personal Care Products in Sewage and
Drinking Water Facilities to Improve the Indirect
Potable Water Reuse’’ supported by the European
Union, Project number EVK1-CT-2000-00047 and the
Project ARCEM ‘‘Austrian Research Cooperative on
Endocrine Modulators’’ funded by Austrian Ministry of
Agriculture, Forestry, Environment and Water, Kom-
munalkredit Austria AG and the nine Austrian Federal
States.
References
Andersen, H., Siegrist, H., Halling-Sorensen, B., Ternes, T.,
2003. Fate of estrogens in a municipal sewage treatment
plant. Environ. Sci. Technol. 37 (18), 4021–4026.
ATV-DVWK, 2000. ATV-DVWK-Regelwerk, Arbeitsblatt
ATV-DVWK-A131: Bemessung von einstufigen Belebung-
sanlagen. ATV-DVWK Standard A131: design of biological
wastewater treatment plants. In: Deutsche Vereinigung für
Wasserwirtschaft Abwasser und Abfall e.V. (Eds.), GFA-
Gesellschaft zur Förderung der Abwassertechnik. Hennef,
Germany, ISBN 3-933707-41-2, http://www.gfa-verlag.de.
Baronti, C., Curini, R., D’Ascenzo, G., Di Corcia, A., Gentili,
A., Samperi, R., 2000. Monitoring natural and synthetic
estrogens at activated sludge sewage treatment plants and in
a receiving river water. Environ. Sci. Technol. 34 (24),
5059–5066.
Buser, H.-R., Poiger, T., Müller, M.D., 1999. Occurrence and
environmental behavior of the chiral pharmaceutical drug
ibuprofen in surface waters and in wastewater. Environ. Sci.
Technol. 33 (15), 2529–2535.
Clara, M., Strenn, B., Gans, O., Kreuzinger, N., 2003. The
elimination of selected pharmaceuticals in wastewater
treatment—lab scale experiments with different sludge
retention times. In: Brebbia, C.A. (Ed.), Water Resources
Management II, International Series on Progress in Water
Resources. WIT Press, Southampton, pp. 227–236.
Clara, M., Strenn, B., Saracevic, E., Kreuzinger, N., 2004.
Adsorption of bisphenol-A, 17b-estradiole and 17a-ethiny-
lestradiole to sewage sludge. Chemosphere 56 (9), 843–851.
 ARTICLE IN PRESS
106 M. Clara et al. / Water Research 39 (2005) 97–106
DIN 32645, 1994. Chemische Analytik; Nachweis-, Erfassungs-
und Bestimmungsgrenze; Ermittlung unter Wiederholbedin-
gungen; Begriffe, Verfahren, Auswertung. Chemical analy-
sis; decision limit; detection limit and determination limit;
estimation in case of repeatability; terms, methods, evalua-
tion. Beuth Verlag GmbH, Berlin-Wien-Zürich, http://
www.din.de.
EN-ISO-9887 1994. Water quality—evaluation of the aerobic
biodegradability of organic compounds in an aqueous
medium—semi-continuous activated sludge method (SCAS)
(ISO 9887:1992), German version EN ISO 9887:1994.
Deutsche Einheitsverfahren L28, Beuth Verlag GmbH and
VCH Verlagsgesellschaft Weinheim, Germany.
Fromme, H., Küchler, T., Otto, T., Pilz, K., Müller, J., Wenzel,
A., 2002. Occurrence of phthalates and bisphenol A and F
in the environment. Water Res. 36 (6), 1429–1438.
Guha, S., Peters, C.A., Jaffé, P.R., 1999. Multisubstrate
biodegradation kinetics of napthalene, phenanthrene and
pyrene mixtures. Biotechnol. Bioeng. 65 (5), 491–499.
Heberer, T., 2002a. Occurrence, fate and removal of pharma-
ceutical residues in the aquatic environment: a review of
recent research data. Toxicol. Lett. 131 (1–2), 5–17.
Heberer, T., 2002b. Tracking persistent pharmaceutical residues
from municipal sewage to drinking water. J. Hydrol. 266
(2–3), 175–189.
Joss, A., Andersen, H., Ternes, T., Richle, P.R., Siegrist, H.R.,
2004. Removal of estrogens in municipal wastewater
treatment under aerobic and anaerobic conditions: con-
sequences for plant optimization. Environ. Sci. Technol. 38
(11), 3047–3055.
Kreuzinger, N., 2002. Occurrence of highly discussed Pollutants
in the Stretch of the Austrian Danube related to the
Catchment Area. Oral presentation at SETAC Europe 12th
annual meeting, Vienna, Austria.
Langford, K.H., Lester, J.N., 2003. Fate and behaviour of
endocrine disruptors in wastewater treatment processes. In:
Birkett, J.W., Lester, J.N., eds.), Endocrine Disrupters in
Wastewater and Sludge Treatment Processes. Lewis Pub-
lisher & IWA Publishing, London, UK, pp. 103–144.
Nowak, O., Franz, A., Svardal, K., Müller, V., Kühn, V., 1999.
Parameter estimation for activated sludge models with the
help of mass balances. Water Sci. Technol. 39 (4), 113–120.
Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2003.
Environmental Organic Chemistry. Wiley, Hoboken, New
Jersey, USA.
Sonnenschein, C., Soto, A.M., 1998. An updated review of
environmental estrogen and androgen mimics and antago-
nists. J. Steroid Biochem. Mole. Biol. 65 (1–6), 143–150.
Staples, C.A., Dorn, P.B., Klecka, G.M., O’Block, S.T., Harris,
L.R., 1998. A review of the environmental fate, effects, and
exposures of bisphenol A. Chemosphere 36 (10), 2149–2173.
Struijs, J., Stoltenkamp, J., Meent, D.v.d., 1991. A spreadsheet-
based box model to predict the fate of xenobiotics in a
municipal treatment plant. Water Res. 25 (7), 891–900.
Stumpf, M., Ternes, T.A., Haberer, K., Baumann, W., 1998.
Isolierung von Ibuprofen-Metaboliten und deren Bedeutung
als Kontaminanten der aquatischen Umwelt. Isolation of
Ibuprofen-Metabolites and their Importance as Pollutants
of the Aquatic Environment. In: Fachgruppe Wasserchemie
in der Gesellschaft Deutscher Chemiker (Ed.), Vom Wasser,
vol. 91, VCH Verlagsgesellschaft mbH, pp. 291–303.
Sumpter, J.P., 1998. Xenoendocrine disrupters—environmental
impacts. Toxicol. Lett. 102–103, 337–342.
Ternes, T.A., 1998. Occurrence of drugs in German sewage
treatment plants and rivers. Water Res. 32 (11), 3245–3260.
Ternes, T.A., Kreckel, P., Müller, J., 1999. Behaviour and
occurrence of estrogens in municipal sewage treatment
plants—II. Aerobic batch experiments with activated
sludge. Sci. Total Environ. 225 (1–2), 91–99.
Ternes, T., Hermann, N., Bonerz, M., Knacker, T., Siegrist, H.-
R., Joss, A. A rapid method to measure the solid-water
distribution coefficient (Kd) for pharmaceuticals and musk
fragrances in sewage sludge. Water Res. in press.
Zwiener, C., Gremm, T.J., Frimmel, F.H., 2001. Pharmaceu-
tical residues in the aquatic environment and their
significance for drinking water production. In: Klaus,
Kümmerer (Ed.), Pharmaceuticals in the Environment.
Springer, Berlin, Heidelberg New York, pp. 81–89.