Vol.2, No.1, 33-40 (2010) Natural Science
http://dx.doi.org/10.4236/ns.2010.21005
Toxicity evaluation of produced formation waters after
filtration treatment
Loredana Manfra, Chiara Maggi, Jessica Bianchi, Michela Mannozzi, Olga Faraponova, Livia
Mariani, Fulvio Onorati, Andrea Tornambè, Claudia Virno Lamberti, Erika Magaletti
ISPRA, Advanced Institute for Environmental Protection and Research, Rome, Italy;
Received 19 June 2009; revised 24 July 2009; accepted 28 July 2009.
ABSTRACT
During the last years many authors have char-
acterized the produced formation waters (PFWs)
with respect to chemical compounds and toxic-
ity. Most of data are related to PFWs collected on
offshore platform after treatment process. The
available results showed that the particulate
phase had an influence on PFW toxicity. As-
suming the toxicity of PFWs treated on platform,
the aim of this paper is to study the toxicity of
these PFWs after a further filtration treatment
carried out in laboratory. For this purpose PFWs
were sampled from three natural gas platforms
located in the Adriatic Sea (Italy) below treat-
ment system. The eco-toxicological bioassays
have been conducted on test-organisms be-
longing to different trophic levels such as bac-
teria, algae, crustaceans and fishes. The PFWs
resulted toxic according to an overall assess-
ment obtained through the bioassays. Further-
more, it has been possible to identify the spe-
cies that were more sensitive to the tested PFWs,
namely Tigriopus fulvus, Dicentrarchus labrax
and Vibrio fischeri. Besides, a chemical char-
acterization was reported related to the con-
taminants present in the PFWs to go with
eco-toxicological assessment. Barium, zinc and
manganese showed the most concentrations
among the metals and the lower molecular
weight components were common among the
organic compounds. Some differences among
PFWs were observed both for toxicity and
chemical composition. The highest toxicity was
recorded in PFWs (PFW1 and PFW2) containing
the highest concentrations of some metals (Ba,
Mn and Zn) and/or BTEX.
Keywords: Adriatic Sea (Italy); Offshore Platforms;
Natural Gas Production Fields; Produced Formation
Waters; Toxicity Assessment; Bacterium (Vibrio
Fischeri); Algae (Dunaliella Tertiolecta and
Phaeodactylum Tricornutum); Crustaceans
(Artemia Franciscana and Tigriopus Fulvus); Fish
(Dicentrarchus Labrax); Chemical Characterization
1. INTRODUCTION
Produced formation water (PFW) is water naturally pre-
sent in sedimentary formations from which oil and gas
are mined. It is piped to the surface during the produc-
tion process and may be discharged into the sea when
the rejection is not possible. Before discharge, PFWs are
treated directly on platform to reduce oil and solid sus-
pended content [1]. In spite of this treatment, PFWs still
include oil and particles.
In Italy, like in other Countries, the legislation binds
to control the oil content in the PFW when they are dis-
charged into the sea [2]; for this reason, the PFW char-
acterization has been limited for several years to meas-
urement of “oil in water”, which means analysis of non
polar aliphatic hydrocarbons. However, PFW contains a
variety of compounds such as metals (i.e. barium, copper,
zinc, and iron), volatile aromatic compounds (benzene,
toluene, ethylbenzene, xylenes, called BTEX), semi-
volatile compounds (naphthalene, phenanthrene, diben-
zothiophene and their C1-C3 alkyl homologues), phenols
alkylated up to C7, organic acids (C1-C6 compounds) and
some additives of possible employment (i.e. diethylene
glycol, called DEG) [3].
For this reason, during the last years many authors
have characterized PFWs with respect to these com-
pounds (metals, aromatic and aliphatic hydrocarbons,
phenols and additives) and to the toxicity of PFWs.
The most data-gathering is related to PFWs collected
on platform after treatment process and untreated in
laboratory later on [4-8]. Few data are referred to
PFWs collected on oil platforms below treatment sys-
tem and filtered in laboratory subsequently [3,9].
There are scattered data about PFWs originated from
Italian offshore gas installations (Adriatic Sea): some

L. Manfra et al. / Natural Science 2 (2010) 33-40
34
chemical analyses were made on seawaters and mussels
near Adriatic platforms [10]; a methodological approach
was proposed to study the environmental impact of oil
and gas offshore platforms [11]; preliminary results were
published on metal content and toxicity of PFWs coming
from gas Adriatic platforms [12-15].
In these studies, the PFWs generally showed to be toxic
and their toxicity was higher for samples unfiltered in
laboratory. This effect was observed probably because of
particulate phase influence. The effects of particulate could
be to mechanic level (oral ingestion and digestion) and/or a
chemical process (adsorption) [7,12].
The aim of this paper was to investigate the toxicity of
PFWs originated from three natural gas platforms in the
Adriatic Sea (Italy). These PFWs were collected below
treatment system occurred on platforms and then filtered
in laboratory. Their toxicity was evaluated using the in-
tegrated response of many species. The eco-toxicological
battery included six species belonging to different tro-
phic levels: a bacterium (Vibrio fischeri) representative
of the debris chain, two algae species (Dunaliella tertio-
lecta and Phaeodactylum tricornutum) as primary pro-
ducers, two crustaceans (Artemia franciscana and Ti-
griopus fulvus) as primary consumers and a fish (Dicen-
trarchus labrax) as representative of the highest trophic
level (secondary consumer).
Besides, a chemical characterization of PFWs was re-
ported to go with eco-toxicological assessment. We in-
vestigated the metals, BTEX, PAHs and DEG. Metals
and PAHs were analyzed both in filtered and particulate
PFWs, because these compounds are present above all as
particles [7,16]. BTEX and DEG were recorded directly
in the whole sample of PFWs. This choice was necessary
because BTEX are partitioned between gas and liquid
phases, therefore the particulate matter does not influ-
ence their concentration; moreover they are volatile
compounds and the filtration procedure causes loss of
analytes. DEG was also analyzed in unfiltered sample
because it is highly soluble in water, not much volatile
and it does not tend toward absorption on particles [17].
Analytical methods by themselves were not able to
give information on what happens when organisms are
exposed to PFWs, which concentrations are toxic and
which is ecological impact of a PFW discharge. The use
of bioassays, together with the classic chemical analyses,
can contribute to the understanding of these aspects.
2. MATERIALS AND METHODS
2.1. Sampling and Sample Treatment of
Produced Formation Water (PFW)
PFW samples were collected from three different gas
platforms situated at about 20 km off the Adriatic coast
(Pescara and Rimini, Italy): one of these (PFW1) was
collected in October 2005 and the other two (PFW3 and
PFW2) in June 2006.
On offshore platforms PFW is stored in a tank which
empties when it is full load. All PFWs were sampled
from a tap located on the platform, which receives the
PFW after this has had a physical-chemical treatment
(depressurization, gravity separation techniques, activa-
ted carbon filtration). The physical-chemical parameters
of PFWs (salinity, pH, conductibility, ORP and oxygen
dissolved) were measured in laboratory by multi-pa-
rameter probe (YSI, mod. 556MPS) (Table 1).
For the bioassays, about ten litres of PFW were im-
mediately filtered (Millipore®, 0.45 µm) and refrigerated
in polystyrene vessels at 4°C until their execution. The
bioassays were carried out in 72 hours.
The PFWs were stored in different containers accord-
ing to type of chemical analyses. For the metals two li-
tres were filtered (Millipore®, 0.45 µm), acidified with
high purity nitric acid and refrigerated at 4°C until
analysis; the filters were stored at -20°C. For BTEX 10
mL of PFW were stored in SPME dark vials (Varian S. p.
A); a magnetic stirrer bar was inserted in each vial prior
to sealing the vial by magnetic steel closures equipped
with Teflon septa. The vials were refrigerated at 4°C and
the samples were acidified at pH=2 with HCl and satu-
rated with NaCl. For the PAHs analysis, one litre of
PFW was immediately filtered (Millipore®, 0.45 µm)
and, together with the filters, stored at 4°C. For the DEG
analysis one litre of sample was collected in dark glass
bottles, saturated with mercury chloride and refrigerated
at 4°C to avoid photochemical and bacterial activity.
2.2. Bioassays
The bioassays were carried out on filtered samples, ac-
cording to the methods reported in Table 2 and summa-
rized for each taxon as follows:
Bacteria: Controls and different concentrations for each
PFW sampled (dilution ratio 1:2) were tested according to
the Basic Protocol [18] and the method ISO [19] with
Table 1. Physical-chemical parameters of production formation water and information on platforms.
PFW Salinity
(PSU) pH Conductibility (mS/cm2) ORP
Oxygen
dissolved
(%)
Volume flux of
PFW (mc/year)
Platform
Installation
(year)
Platform distance
from coast (Km)
Water
depth
(m)
PFW1 34 7 51 -10093 6000 1991 36 116
PFW2 37 7 56 -10596 3000 1972 15 18
PFW3 37 8 55 -7086 3000 1991 21 23

L. Manfra et al. / Natural Science 2 (2010) 33-40
35
35
bacteria coming from freeze-dried SDI. PFW salinity was
not adjusted prior to testing. The Software Microtox Om-
niTM v. 1.16 was utilized to calculate the EC50 and EC20
values (effect concentration of 50% and 20% respectively)
and the Dunnett test was used to calculate the NOEC
value (no observed effect concentration).
Algae: One control and some concentrations for
each PFW sampled were tested according to the ISO
method [20] with Phaeodactylum tricornutum strain
1090-1° and Dunaliella tertiolecta strain 13.86, ob-
tained from the Plant Physiology Institute of Gottingen
University (Germany). Algal growth medium was
prepared with artificial seawater [20]; for D. tertio-
lecta nutrients, according to the IRSA-CNR method
[21] and vitamins according to the ISO method [20]
were added. The algal inoculum had an initial density
of 10000 cells mL-1 ± 10% for P. tricornutum and 2000
cells mL-1± 10% for D. tertiolecta. Regression analysis
technique was performed for the determination of EC50
and EC20; the Dunnett test was used to calculate the
NOEC value.
Crustaceans: A control and some concentrations for
each PFW sampled were tested with nauplii of Artemia
franciscana , according to the APAT IRSA-CNR method
[22], and with nauplii of Tigriopus fulvus, according to
the ISO/FDIS method [23] as modified by Faraponova
et al. [24,25]. Reference cysts of A. franciscana were
obtained from the Quality Assurance Research Divi-
sion U. S. Environmental Protection Agency (Cincin-
nati OH 45268, USA) or from the Laboratory for Bio-
logical Research in Aquatic Pollution, University of
Ghent (Belgium). The eggs of A. franciscana were
hatched in synthetic seawater and the nauplii were
used within 48 hours of hatching [22]. Synchronized
nauplii (24-48h) of T. fulvus were collected from a
culture of two hundred females taken from a mass
laboratory culture originated from the Italian coast
(Calafuria, Livorno) and supplied with an algal mix-
ture (Tetroselmis suecica and Isochrysis galbana, ratio
1:2). Probit analysis was performed for the determina-
tion of EC50 and EC15, the Dunnett test was used to
calculate the NOEC value.
Fish: One control and some concentrations for each
PFW sampled were tested with juveniles of Dicen-
trarchus labrax (80 days old, length of 3.74±0.28 cm
and weight of 0.48±0.08 g), according to the EPA [26]
and OECD [27] methods. Organisms were supplied by
the hatchery production plant ASA (Rome), stabled in
synthetic seawater with salinity of 20±1 PSU for 15 days
and fed with granulated food until 24 hours before the
test. Probit analysis was performed for the determination
of EC50 and EC15, the Dunnett test was used to calculate
the NOEC value.
The results were compared to a toxicity scale reported in
Table 3. On the basis of this toxicity scale, the samples
were classified as follows: 1) toxic 10≤ EC50<100, 20≤
EC20<50, effect percentage ≥50; 2) weakly toxic EC50>100,
EC20>50, 20≤effect percentage<50; 3) no toxic EC50 no
calculable, EC20>100, effect percentage 20.
2.3. Analysis of Metals
Determination of iron (Fe), copper (Cu), zinc (Zn), lead
(Pb), chromium (Cr), manganese (Mn), nickel (Ni), bar-
ium (Ba), arsenic (As), cadmium (Cd) and mercury (Hg)
was carried on both filtered and particulate samples. The
filtered sample was directly analyzed. The metal disso-
lution of particulate fraction collected on the filters was
conducted using microwave-assisted digestion (Mile-
stone MLS Ethos TC) with 3 mL of HNO3 and 9 mL of
HCl. The metal concentrations were determined by a
graphite furnace atomic absorption with Zeeman back-
ground correction technique (SpectrAA-220Z, Varian)
and by coupled emission plasma ICP-OES (Liberty AX,
Varian). For Hg analysis a Direct Mercury Analyzer
(DMA-80, FKV) instrument was used (EPA Method
[28]). All samples were run in triplicate. The quantifica-
tion limits (LOQ) were: 0.0005 mg/l for Hg and Cd, 0.01
mg/L for the other metals.
2.4. Analysis of Organic Compounds
BTEX: The analyses were extracted and pre-concen-
trated by means Solid Phase Micro Extraction (SPME)
using a stable flex fiber of divinilbenzene-carboxen-
poly-dimethylsiloxane (film thickness: 55/30 µm) (Su-
pelco®) by head space sampling. The analytical deter-
minations of BTEX were carried out in unfiltered sam-
ples using a modified EPA method [29]. A gas chroma-
tography coupled with mass spectrometry (GC HP 5790
Agilent Technologies® and MS 5973 Network Agilent
Technologies®) were used. The method detection limits
were 1 µg/L for benzene and 0.1 µg/L for toluene, ethyl
benzene and xylenes.
PAHs: The analyses were investigated in filtered sam-
ples and on the particulate matter retained by the filter
(Millipore®, 0.45 µm). The analyses were extracted by
the filtered samples by means Solid Phase Extraction
technique. The treatment of filters was carried out by
ultrasonic extraction for 20 minutes with 10 mL of di-
chloromethane. Then, both the extraction phases were
evaporated at 1-2 mL with a gentle nitrogen flow. Af-
terward, 1 mL of toluene was added and the residual
dichloromethane was completely removed. All solvents
were capillary GC grade supplied by Sigma-Aldrich.
The analyses of PAHs were carried out in gas chroma-
tography coupled with mass spectrometry (GC HP
5790® and MS 5973 Network Agilent Technologies).
The LOQ was 1 µg/L for each analysis.
DEG: An extraction procedure of DEG was carried
out with 2 mL SPE cartridges packed with 200 mg of EN
V+stationary phase (International Sorbent Technology,

L. Manfra et al. / Natural Science 2 (2010) 33-40
36
Table 2. Experimental conditions of bioassays (*for PFW1 were tested the concentrations: 10-20-40-80%).
Vibrio
fischeri
Phaeodactylum
tricornutum
Dunaliella
tertiolecta
Artemia
franciscana Tigriopus fulvus Dicentrarchus labrax
Organisms/
Life stage cells unialgal
culture
unialgal
culture nauplii 48h nauplii 48h juveniles
Strain/origin commercially
available culture culture
commercially
available culture hatchery
Type of test static static static static static static
Time exposure 5-10-15min. 72h 72h 96h 96h 96h
Intensity of lux Not required 7000 7000 3000-4000 500-1200 500-800
Photoperiod
(L:D) Not required 24:0 24:0 14:10 16:8 16:8
Dilution water/
control
synthetic
seawater synthetic seawatersynthetic seawatersynthetic seawatersynthetic sea- water synthetic seawater
Salinity (PSU) 35 32 32 35 38 20
Temperature
(°C) 15±1 20 ± 2 20 ± 2 25 ± 2 18 ± 2 20  1
pH 8.0 – 8.2 8  0.5 8  0.5 6.5 – 8.5 8.0 ± 0.3 7.5  0.5
Vessel 5mL 100mL 100mL 50 mL
culture plates 12
wells 2000 mL
Volume/well 1mL 25mL 25mL 40mL 3mL 1800mL
N°organisms/well - 10000/mL 2000/mL 10 10 5
N°of concentra-
tions (Range)
5
(6-11-22-45-90%)
5
(6-12-25-50-100%)
5
*(6-12-25-50-100%)
5
(6-12-25-50-100%)
5
(5-10-20-40-80%)
5
(6-12-25-50-100%)
N°of replicates 3 3 3 3 3-4 3
Feeding during
the test absent absent absent D. tertiolecta absent absent
Endpoint/Effect bioluminescence
inhibition rate
growth inhibition
rate
growth inhibition
rate
immobilization
rate
mortality rate;
moult release rate mortality rate
Expression of
endpoint
EC50;EC20
NOEC
EC50;EC20
NOEC
EC50;EC20
NOEC
EC50;EC15
NOEC
EC50;EC15
NOEC
EC50;EC15
NOEC
Acceptability
(effect control) 10% >0.04/h >0.04/h 10%
10% (mortality)
20% (moult
release)
10%
Glamorgan, UK). The off-line solid-phase extraction/
pre-concentratio n technique was followed by a nano-
scale flow injection/direct-electron ionization (EI) mass
spectrometric analysis. A quadrupole mass spectrometer
(Palo Alto, CA) was coupled with a Direct-Electron
Ionization (EI) [30-32]. Using this approach, DEG was
detected within a concentration of 31 µg/L [33].
Table 3. Toxicity scale used in this paper to classify the toxic-
ity of Production Formation Water (PFW).
Effect
(%)
EC50
(%)
EC20 or
EC15 (%)
TOXICITY
ASSESSMENT
%<20 n.c. >100 no toxic
20≤%<50 >100 >50 weakly toxic
≥50 10≤%<100 20<%<50 toxic
3. RESULTS
3.1. Bioassays
The results of the eco-toxicological battery are reported
in Table 4. The three filtered PFWs resulted toxic ac-
cording to the overall assessment related to bioassays.
The species showed different sensitivity to PFW: the two
microalgae and Artemia franciscana showed higher val-
ues of EC50 and EC20 (EC15) than the other organisms,
indicating weak toxicity. In particular, Artemia did not
record toxicity for PFW1 (no efficient concentration was
calculable). The other crustacean T. fulvus showed toxic
effects for all PFWs (PFW2>PFW1≥PFW3) related to
both mortality and moult release. The sub-lethal effect
was already observed at PFW concentrations that were
not causing mortality of nauplii (20-80) %. The fish spe-
cies D. labrax showed a toxic response similar to the one
of T. fulvus, with a minimum value of EC50 equal to 15%
and maximum value of 47% as follows: PFW2>PFW1≥

L. Manfra et al. / Natural Science 2 (2010) 33-40
37
37
PFW3. The bacterium V. fischeri recorded lower toxicity
than T. fulvus and D. labrax but yet a toxic effect was
observed for all PFWs (PFW1>PFW3≥PFW2).
3.2. Analysis of Metals
In filtered sample, Ba, Mn and Zn showed detectable
concentrations. There were not significant differences
between PFW2 and PFW3 exclusive of Ba and Mn.
These two metals showed higher concentrations of one
order in PFW2 than PFW3. In particulate sample, all
metals were detectable except Pb and Hg. Ni, Cd and As
registered detectable concentrations only in PFW3. Ba,
Zn and Fe showed highest concentrations and some sig-
nificant differences among the PFWs analyzed. Ba con-
centration was higher of one order in PFW1 and PFW2
than PFW3; Zn was higher of two orders in PFW2 and
PFW3 than PFW1; Fe was higher of one order in PFW3
than PFW1 and PFW2.
3.3. Analysis of Organic Compounds
As reported in introduction section, BTEX and DEG
were analysed on unfiltered PFWs. The analyses of the
three platforms pointed out that the volatile organic
compounds (BTEX) were detected at very high concen-
trations by the following ranking: PFW1 (1281.8 μg/L)
>PFW3 (66.5 μg/L)>PFW2 (48.0 μg/L) (Table 6),
showing for PFW1 values almost twenty times higher
than ones of PFW2 and PFW3. The DEG showed con-
centrations ranging from 2400 to 13000 μg/L by the fol-
lowing ranking of the fields: PFW3>PFW2>PFW1 (Ta-
ble 6). PAHs were investigated both in filtered and par-
ticulate sample but they were lower than LOQ (1 μg/L)
in filtered PFWs. In particulate sample PAHs showed the
trend as follows: PFW1 (150.0 μg/L)>PFW3 (126.0
μg/L)>PFW2 (100.0 μg/L), with values about of the
same magnitude order. The congeners with two and
three rings recorded the following concentrations: 87
μg/L in PFW1 compared to 150 μg/L of the total PAHs
content; 54 μg/L in PFW2 compared to 100 μg/L of total
PAHs concentration, and 66 μg/L in PFW3 compared to
126 μg/L of total PAHs. The detected concentrations
were of the same order of magnitude for all three PFW
investigated.
4. DISCUSSIONS
The bioassays showed that the PFWs were toxic even if
filtered. EC50 values ranged between the minimum of
14.8 % and values higher than 100 %. Some of the sam-
ples did record no toxicity at the EC50 level observed
Table 4. Bioassay results related to three Italian Production Formation Waters and toxicity evaluation according to toxicity
scale (n.c. not calculable; n.d. not determined).
Species
(time of exposure, end point) % PFW1 PFW2 PFW3 PFW1 PFW2 PFW3
EC50 67 (59-75)> 90 > 90
EC20 20(18-23) 29(27-31)28(25-31)
Vibrio fischeri
(15 min. bioluminescence) NOEC - - -
toxic toxic toxic
EC50 > 80 > 100 > 100
EC20 > 80 52
(46-58)
68
(24-111)
Phaeodactylum tricornutum
(72h growth)
NOEC 40 25 25
weakly
toxic
weakly
toxic
weakly
toxic
EC50 > 80 n.d. n.d
EC20 > 80 n.d n.d
Dunaliella tertiolecta
(72h growth) NOEC 40 n.d n.d
weakly
toxic n. d n. d.
EC50 > 100 > 100 > 100
EC15 > 100 86
(55-259)
77
(42-462)
Artemia franciscana
(96h immobilization)
NOEC n.c. 50 25
no toxic weakly
toxic
weakly
toxic
EC50 29 15 44
EC15 19 8 29
Tigriopus fulvus
(96h mortality) NOEC 10 5 20
toxic toxic toxic
EC50 23 25 77
EC15 n.c. n.c. n.c.
Tigriopus fulvus
(96h moult release) NOEC 10 5 10
toxic toxic toxic
EC50 32
(27-39)
15
(n.c)
47
(n.c)
EC15 23
(15-28)
11
(n.c)
15
(n.c)
Dicentrarchus labrax
(96h mortality)
NOEC 13 6 6
toxic toxic toxic

L. Manfra et al. / Natural Science 2 (2010) 33-40
38
Table 5. Metal concentrations in filtered samples of two Produced Formation Waters, total suspended solids and metal concen-
trations in particulate samples of Produced Formation Water from three offshore gas platforms in the Adriatic Sea (Italy).
Parameters Unit PFW2 PFW3 PFW1 PFW2 PFW3
FILTERED PARTICULATE
Total suspended solid mg 177.42 367.05 398.10
Ba mg/L 1.63 0.13 309.65 237.75 13.50
Cr mg/L <0.01 <0.01 0.12 1.34 0.45
Cu mg/L <0.01 <0.01 0.05 0.55 0.42
Mn mg/L 0.34 0.04 0.84 1.05 0.77
Ni mg/L <0.01 <0.01 0.07 <0.01 0.35
Pb mg/L <0.01 <0.01 <0.01 <0.01 <0.01
Zn mg/L 0.18 0.37 0.14 61.92 59.77
Cd mg/L <0.0005 <0.0005 <0.0005 <0.0005 0.62
Fe mg/L <0.10 <0.10 242.40 775.70 1335.00
As mg/L <0.01 <0.01 0.01 <0.01 9.25
Hg mg/L <0.0005 <0.0005 <0.0005 <0.0005 <0.0005
Table 6. Organic compound concentrations in Produced Formation Water (PFW) samples from three offshore gas platforms in
the Adriatic Sea (Italy).
Analytes PFW1 (µg/L)PFW2 (µg/L) PFW3 (µg/L)
Benzene 256.0 10.4 20.4
Toluene 50.6 14.1 12.1
Ethyllbenzene 115.2 7.7 13.8
Xilenes (o,m,p-xylene) 860.0 14.8 20.2
BTEX
(unfiltered sample)
BTEX 1281.8 47.0 66.5
DEG (unfiltered sample) Diethylene glycol 2400 9600 13000
Naphtalene 14 8 11
Acenaphtylene 21 15 17
Acenaphthene 19 15 15
Fluorene 16 4 6
Phenanthrene 13 8 10
Anthracene 4 4 7
Fluorantrene 12 10 12
Pyrene 10 7 10
Benzo(a)anthracene 8 4 7
Crysene 5 6 7
Benzo(b)fluorantene 6 4 4
Benzo(k)fluorantene 6 5 6
Benzo(a)pyrene 5 3 3
Dibenzo(a,h)anthracene 4 4 4
Benzo(g,h,i)perylene 4 3 4
Indenopyrene 3 < 1 3
PAHs 2 - 3 ring congeners 87 54 66
PAHs
(particulate sample)
PAHs 150 100 126
butdid show evidence of a toxic response at the EC20
level. The EC50 data lied within or were higher than the
range (5.54-20.73%) previously reported for another
filtered PFW originated from an Italian gas platform and
assayed by bacteria and sea urchins [14]. Ours EC50 data
also were higher than values related to unfiltered PFWs
coming from the North Sea platforms [4-8]. This shows
that the filtered samples have generally lower toxicity
than the untreated samples.
The difference of sensitivity among the species has
been quite remarkable: T. fulvus and D. labrax showed
the highest toxicity (EC50<50%), followed by V. fischeri
(EC20<30%). Artemia and the two algae did not record
significant toxic effect (EC20>50%).

L. Manfra et al. / Natural Science 2 (2010) 33-40
39
39
In addition to the toxicity assessment, we analyzed
chemically PFW (metals, BTEX, PAHs and DEG). Ap-
preciable concentrations of Ba, Mn and Zn were re-
corded in filtered samples while also high quantities of
Fe were registered in particulate samples. Ba is probably
related to drilling fluid residuals of PFW [7], Zn may be
derived from corrosion or chipping of galvanized struc-
tures on the platform or in the oil/water separator system
[34] and Fe could have natural origin or derive from
corrosive events. The lower weight aromatic hydrocar-
bons (BTEX) were found by significant concentrations
in liquid phase, while the PAHs were recorded only on
particulate samples. DEG concentrations also were of
milligram order in liquid phase but very low compared
to the threshold of 3500 mg/L imposed by the PFW dis-
charge authorization decrees issued by the Italian Minis-
try of the Environment.
An integrated evaluation of the eco-toxicological and
chemical results showed that test-organisms were espe-
cially sensitive when exposed to PFWs containing Ba,
Mn, Zn and BTEX. T. fulvus and D. labrax showed the
highest toxicity in PFW2 containing high concentrations
of Ba, Mn and Zn. V. fischeri showed the highest toxic
effect in PFW1 that recorded the highest quantities of
BTEX. Nobody among the test-organisms indicated a
preference for PFW3 containing the highest value of
DEG. Moreover, DEG is not toxic alone but could de-
termine co-solvent effects with other chemical com-
pounds [35].
5. CONCLUSIONS
Our results indicate that a filtration treatment might di-
minish PFW toxicity. If a similar treatment was carried
out on the platform before the PFW discharge, the eco-
logical risk associated to the discharge would be proba-
bly reduced.
Besides, the results confirm the different sensitivity of
test-organisms belonging to different trophic levels. Be-
cause PFW chemical composition is so variable for type
and concentration of contaminants, test-organisms that
are tolerant to a type of PFW could be sensitive to others.
For this reason, we think that it is not correct to establish
a single species to investigate the PFWs but a battery of
species should always be applied in order to have inte-
grated responses.
6. ACKNOWLEDGEMENTS
The present work was funded by the Italian Ministry of
Environment, Land and Sea. We thank Lucio Lattanzi
for analysis of BTEX and PAHs, Achille Cappiello and
Giorgio Famiglini for DEG analysis, Manuela Dattolo,
Silvia Mariotti and Antonella Cozzolino for metal analy-
sis. The authors acknowledge B. Trabucco for construc-
tive comments. We thank the Aquaculture Plant Nuovo
Azzurro (Civitavecchia, Roma-Italy) for providing the
larvae of Dicentrarchus labrax.
REFERENCES
[1] Trefry, J.H., Naito, K.L., Trocine, R.P. and Metz, S.
(1995) Distribution and bioaccumulation of heavy metals
from produced water discharges to the Gulf of Mexico.
Water Science Technology, 32, 2, 31-36.
[2] Decree of Environmental Ministry 190 of July 28 (1994) G.
U. n. 190 16/08/1994.
[3] Brendehaugh, J., Johnsen, S., Bryne, K.H., Gjose, A.L.,
Eide, T.H. and Aamot, E. (1992) Toxicity testing and
chemical characterization of produced water: A prelimi-
nary study. In: Ray, J.P. and Engelhart, F.R. Eds, Produced
Water Technological/Environmental Issues and Solutions,
Plenum Press, New York, 245-256.
[4] Schiff, K.C., Reish, D.J., Anderson, J.W. and Bay, S.M.
(1992) A comparative evaluation of produced water tox-
icity. In: Ray, J.P. and Engelhart, F.R. Eds, Produced Wa-
ter Technological/Environmental Issues and Solutions,
Plenum Press, New York, 199-208.
[5] Sommerville, H.J., Bennett, D., Davenport, J.N., Holt,
M.S., Lynes, A. and Mahieau, A. (1987) Environmental
effect of produced water from North Sea oil operations.
Marine Pollution Bulletin, 18, 549-558.
[6] Stephenson, M.T. (1992) A survey of produced water
studies. In: Reed, M. and Johnsen, S. Eds. Produced Water
2 Environmental Issues and Mitigation Technologies, Ple-
num Press, New York, 1-11.
[7] Stromgren, T., Sorstrom, S.E., Schou, L., Kaarstad, I.,
Aunaas, T. and Brakstad, O.G. (1995) Acute toxic effects
of produced water in relation to chemical composition and
dispersion. Marine Environmental Research, 40, 147-169.
[8] Utvik, T.I.R. (1999) Chemical characterization of produced
water from four offshore oil production platforms in the
North Sea. Chemosphere, 39, 15, 2593-2606.
[9] Scott, K.A., Yeats, P., Wohlgeschaffen, G., Dalziel, J.,
Niven, S. and Lee, K. (2007) Precipitation of heavy metals
in produced water: Influence on contaminant transport and
toxicity. Marine Environmental Research, 63, 146-167.
[10] Cicero, A.M., Di Mento, R., Gabellini, M., Maggi, C.,
Trabucco, B., Astorri, M. and Ferraro, M. (2003) Monitor-
ing of environmental impact resulting from offshore oil
and gas installations in the Adriatic Sea: Preliminary
evaluations. Annali di Chimica, 93, 701-705.
[11] Maggi, C., Trabucco, B., Mannozzi, M., Di Mento, R.,
Gabellino, M., Manfra, L., Nonnis, O., Virno Lamberti, C.
and Cicero, A.M. (2007) A methodological approach to
study the environmental impact of oil and gas offshore
platforms. Rapp. Comm. Int. Mer Médit., 38.
[12] Mariani, L., Manfra, L., Maggi, C., Savorelli, F., Di Mento,
R. and Cicero, A.M. (2004) Produced formation waters: A
preliminary study on chemical characterization and acute
toxicity by using fish larve Dicentrarcus labrax L., 1758.
Fresenius Environmental Bulletin, 13, 1427-1432.
[13] Faraponova, O., Onorati, F., Magaletti, E. and Virno Lam-
berti, C. (2007a) Sensitivity of Tigriopus fulvus (Copepoda,
Harpacticoida) towards diethilene glycol (DEG) and pro-
duced formation water (PFW). International Meiofauna
Conference-Recife, Brazil, 89.
[14] Manfra, L., Moltedo, G., Virno Lamberti, C., Maggi, C.,
Finora, M.G., Gabellino, M., Giuliani, S., Onorati, F., Di

L. Manfra et al. / Natural Science 2 (2010) 33-40
40
Mento, R. and Cicero A.M. (2007) Metal content and tox-
icity of produced formation water (PFW): Study of the
possible effects of the discharge on marine environment.
Archives of Environmental Contamination and Toxicology,
53, 183-190.
[15] Manfra, L., De Nicola, E., Maggi, C., Zambianchi, E.,
Caramello, D., Toscano, A., Cianelli, D. and Cicero, A.M.
Submitted Exposure of rotifers, crustaceans and sea ur-
chins to produced formation water and seawaters collected
in a gas platform area. Journal of the Marine Biological
Association.
[16] Prego, R., Cottè, M.H., Cobelo Garcìa, A. and Martin, J.M.
(2006) Trace metals in the water column of the Vigo Ria:
Offshore exchange in mid-winter conditions. Estuarine
Costal and Shelf Science, 68, 289-296.
[17] Weyerhaeuser (2005) Material safety data sheet MSDS
WC 379-01 Rev. 8/26/2005, 1-7.
[18] Azur Environmental (1995) Microtox® acute toxicity basic
test procedures, 63.
[19] ISO (2004) Water quality: Determination of the inhibitory
effect of water samples on the light emission of Vibrio
fischeri (luminescent bacteria test)-Part 3: Method using
freeze-dried bacteria. ISO/CD, 11348-3.
[20] ISO (2006) Water quality: Marine algal growth inhibition
test with Skeletonema costatum and Phaeodactylum tri-
cornutum. Reference number ISO 10253:2006, 12.
[21] IRSA (1978) Metodologia di saggio algale per lo studio di
acque marine. Quaderni dell'Istituto di Ricerca sulle Acque,
IRSA-CNR, Milano, 39.
[22] APAT-IRSA-CNR (2003) Metodo 8060. Metodo di valu-
tazione della tossicità acuta con Artemia sp. In: APAT-
IRSA-CNR, Metodi analitici per le acque, Manuali e Linee
Guida 29/2003, Volume Terzo, 1043-1049.
[23] ISO/FDIS (1998) Water quality: Determination of acute
lethal toxicity to marine copepods (Copepoda, Crustacea).
ISO/FDIS, 16, 14669.
[24] Faraponova, O., De Pascale, D., Onorati, F. and Finoia,
M.G. (2005) Tigriopus fulvus (Copepoda, Harpacticoda) as
a target species in biological assays. Meiofauna Marina, 14,
91-95.
[25] Faraponova, O., Virno Lamberti, C. and Onorati, F. (2007b)
Study intensification of Tigriopus fulvus (Copepoda, Har-
pacticoida) as a target species in bioassays. International
Meiofauna Conference-ecife, Brazil, 90.
[26] EPA (1993) Methods for measuring the acute toxicity of
effluents and receiving waters to freshwaters and marine
organisms. EPA/600/4-90/027F, Environmental Protection
Agency, Cincinnati, Ohio, 71-91.
[27] OECD (1992) Guideline for testing of chemicals n° 203.
Fish, Acute Toxicity Test, 1-9.
[28] EPA (1998) Mercury in solids and solutions by thermal
decomposition, amalgamation, and atomic absorption
spectrophotometry Method 7473. Environmental Protec-
tion Agency, Cincinnati, Ohio.
[29] EPA (1996) Volatile organic compounds by gas chroma-
tography/mass spectrometry (GC/MS) Method 8260B
(Revision 2). Environmental Protection Agency, Cincinnati,
Ohio.
[30] Cappiello, A., Famiglini, G., Mangani, F. and Palma, P.
(2001) New trends in the application of electron ionization
to liquid chromatography-mass spectrometry interfacing.
Mass Spectrom. Rev., 20, 88-104.
[31] Cappiello, A., Famiglini, G., Mangani, F. and Palma, P.
(2002) A simple approach for coupling liquid chromatog-
raphy and electron ionization mass spectrometry. J. Am.
Soc. Mass. Spectrom., 13, 265-273.
[32] Cappiello, A., Famiglini, G. and Palma, P. (2003) Elec-
tron ionization for LC/MS. Analytical Chemistry, 75,
497A-503A.
[33] Cappiello, A., Famiglini, G., Palma, P., Pierini, E., Trufelli,
H., Maggi, C., Manfra, L. and Mannozzi, M. (2007) Ap-
plication of nano-FIA-Direct-EI-MS to determine diethyl-
ene glycol in produced formation water discharges and
seawater samples. Chemosphere, 69, 554-560.
[34] Neff, J.M., Sauer, T.C. and Maciolek, N. (1992) Composi-
tion, fate and effects of produced water discharges to
nearshore marine waters. In: Ray J.P. and Engelhart F.R.
Eds, Produced Water Technological/Environmental Issues
and Solutions, Plenum Press, New York, 371-385.
[35] Sorensen, J.A., Gallagher, J.R., Hawthorne, S.B. and Au-
lich, T.R. (2000) Gas Industry Groundwater Research
Program, Final Report for U.S. Department of Energy Na-
tional Energy Technology Laboratory Cooperative
Agreement No. DE-FC26-98FT40321; Energy & Envi-
ronmental Research Center Publication 2004-EERC-
07-01, Grand Forks, ND.