Fate and ecotoxicity of the new antifouling compound Irgarol 1051 in the aquatic environment

Article abstract of DOI:10.1016/S0043-1354(00)00095-6

Residue analyses and ecotoxicity assessment were conducted on the new antifouling compound Irgarol 1051 (2-methylthio-4-tert-butylamino-6-cyclopropylamino-s-triazine) and its degradation product M1 (2-methylthio-4-tert-butylamino-6-amino-s-triazine) in order to delineate the environmental fate and impact of Irgarol 1051 on the aquatic ecosystem. For the first time, the Irgarol degradation product (M1) was positively identified in environmental samples. During the 1998 Irgarol survey, concentrations of M1 (up to 1870 ng/l) were generally higher than those of Irgarol in the coastal waters of the Seto Inland Sea in Japan, suggesting a greater environmental persistence for M1 than for the parent compound Irgarol 1051 in the aquatic ecosystem. Ecotoxicity testing revealed that Irgarol 1051 and M1 were moderately toxic to a marine bacterium and the four crustaceans tested, but were highly toxic to some algae and higher plants. In the root elongation inhibition bioassay, M1 showed a phytotoxicity at least 10 times greater than that of Irgarol and six other triazine herbicides (terbutryn, terbutylazine, terbumeton, simetryn, atrazine and simazine). These results strongly suggest that both Irgarol 1051 and its persistent degradation product M1 may potentially affect and/or damage the primary producer community in aquatic ecosystems. To safeguard the aquatic ecosystem from the damaging impact of micro contaminants, it is recommended that, besides monitoring for the target parent compound, major degradation products should also be included in environmental surveys. Otherwise, there is a risk of underestimating the ultimate impact of a particular toxicant on the environment. Copyright (C) 2000 Elsevier Science Ltd. Residue analyses and ecotoxicity assessment were conducted on the new antifouling compound Irgarol 1051 (2-methylthio-4-tert-butylamino-6-cyclopropylamino-s-triazine) and its degradation product M1 (2-methylthio-4-tert-butylamino-6-amino-s-triazine) in order to delineate the environmental fate and impact of Irgarol 1051 on the aquatic ecosystem. For the first time, the Irgarol degradation product (M1) was positively identified in environmental samples. During the 1998 Irgarol survey, concentrations of M1 (up to 1870 ng/l) were generally higher than those of Irgarol in the coastal waters of the Seto Inland Sea in Japan, suggesting a greater environmental persistence for M1 than for the parent compound Irgarol 1051 in the aquatic ecosystem. Ecotoxicity testing revealed that Irgarol 1051 and M1 were moderately toxic to a marine bacterium and the four crustaceans tested, but were highly toxic to some algae and higher plants. In the root elongation inhibition bioassay, M1 showed a phytotoxicity at least 10 times greater than that of Irgarol and six other triazine herbicides (terbutryn, terbutylazine, terbumeton, simetryn, atrazine and simazine). These results strongly suggest that both Irgarol 1051 and its persistent degradation product M1 may potentially affect and/or damage the primary producer community in aquatic ecosystems. To safeguard the aquatic ecosystem from the damaging impact of micro contaminants, it is recommended that, besides monitoring for the target parent compound, major degradation products should also be included in environmental surveys. Otherwise, there is a risk of underestimating the ultimate impact of a particular toxicant on the environment.

Full text of DOI:10.1016/S0043-1354(00)00095-6

Wat. Res. Vol. 34, No. 14, pp. 3523±3530, 2000
7 2000 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
PII: S0043-1354(00)00095-6
0043-1354/00/$ - see front matter
www.elsevier.com/locate/watres
FATE AND ECOTOXICITY OF THE NEW ANTIFOULING
COMPOUND IRGAROL 1051 IN THE AQUATIC
ENVIRONMENT
H. OKAMURA1*, I. AOYAMA1M, D. LIU2, R. J. MAGUIRE2,
G. J. PACEPAVICIUS2 and Y. L. LAU2
1Research Institute for Bioresources, Okayama University, Chuo 2-20-1, Kurashiki, 710-0046, Japan
and 2National Water Research Institute, Environment Canada, Burlington, Ontario, Canada, L7R 4A6
(First received 1 March 1999; accepted in revised form 1 August 1999)
AbstractÐResidue analyses and ecotoxicity assessment were conducted on the new antifouling
compound Irgarol 1051 (2-methylthio-4-tert-butylamino-6-cyclopropylamino-s-triazine) and its
degradation product M1 (2-methylthio-4-tert-butylamino-6-amino-s-triazine) in order to delineate the
environmental fate and impact of Irgarol 1051 on the aquatic ecosystem. For the ®rst time, the Irgarol
degradation product (M1) was positively identi®ed in environmental samples. During the 1998 Irgarol
survey, concentrations of M1 (up to 1870 ng/l) were generally higher than those of Irgarol in the
coastal waters of the Seto Inland Sea in Japan, suggesting a greater environmental persistence for M1
than for the parent compound Irgarol 1051 in the aquatic ecosystem. Ecotoxicity testing revealed that
Irgarol 1051 and M1 were moderately toxic to a marine bacterium and the four crustaceans tested, but
were highly toxic to some algae and higher plants. In the root elongation inhibition bioassay, M1
showed a phytotoxicity at least 10 times greater than that of Irgarol and six other triazine herbicides
(terbutryn, terbutylazine, terbumeton, simetryn, atrazine and simazine). These results strongly suggest
that both Irgarol 1051 and its persistent degradation product M1 may potentially a€ect and/or damage
the primary producer community in aquatic ecosystems. To safeguard the aquatic ecosystem from the
damaging impact of micro contaminants, it is recommended that, besides monitoring for the target
parent compound, major degradation products should also be included in environmental surveys.
Otherwise, there is a risk of underestimating the ultimate impact of a particular toxicant on the
environment. 7 2000 Elsevier Science Ltd. All rights reserved
Key wordsÐantifouling compound, degradation product, ecotoxicity, fate, herbicide, Irgarol 1051, M1,
Seto Inland Sea Japan, transformation
INTRODUCTION
possibility that Irgarol 1051 may become a ubiqui-
tous contaminant of the aquatic environment.
The majority of Irgarol studies have been on its
environmental occurrence, and there is little infor-
Irgarol 1051 (2-methylthio-4-tert-butylamino-6-
cyclopropylamino-s-triazine) is a newly developed
herbicidal additive for use in copper-based antifoul-
mation in the open literature on its environmental
ing paints (Ciba±Geigy, 1995). It is intended as a
replacement for the widely used antifouling agent
dation product(s). The fate and transformation
tributyltin (TBT) which has been regulated interna-
tionally since 1990, primarily due to its severe
vironment are not yet fully understood. Recent stu-
impact on the aquatic ecosystem. Irgarol 1051 does
not appear to undergo rapid biodegradation, and
degradation, and photodegradation of Irgarol 1051
its residues have been found in European coastal
and lake waters (Ferrer et al., 1997; Gough et al.,
1994; Readman et al., 1993; Tolosa et al., 1996;
Toth et al., 1996; Zhou et al., 1996). Recently, this
new antifoulant has been detected in Japanese
coastal waters (Liu et al., 1999b), suggesting the
fate and transformation, particularly on its degra-
pathways of Irgarol 1051 in the ambient aquatic en-
dies have shown that biological and chemical
could lead to the formation of a stable degradation
product, ``M1''. Biological formation of M1 was
mainly via N-dealkylation of the cyclopropylamino
group (Liu et al., 1997), while the chemical degra-
dation of Irgarol 1051 by mercuric chloride
appeared to follow the reaction of a catalyzed hy-
drolysis (Liu et al., 1999a). Photodegradation of
Irgarol 1051 by sunlight could be greatly enhanced
by naturally occurring organics in river water and
seawater (Okamura et al., 1999), and photochemical
*Author to whom all correspondence should be addressed.
Tel.: +81-86-434-1235; fax: +81-86-434-1249; e-mail:
hokamura@rib.okayama-u.ac.jp
3523

3524
H. Okamura et al.
and then the solution was passed through an ODS column
(Sep-Pak Plus cartridge, tC18 ENV, Waters). Compounds
adsorbed on the resins were eluted with 5 ml of methanol
and the methanol eluate was concentrated to dryness by a
centrifugal evaporator (EC-57CS, Sakuma) at 408C. The
residue was re-dissolved in 1 ml of methanol and was sub-
jected to preparative HPLC for the isolation of M1. The
HPLC column used was a Develosil ODS 5 (5 mm, 20 mm
 250 mm, Nomura Chemical) and the mobile phase was
a mixture of (A) 10 mM phosphate solution (pH 2.6) and
process was likely responsible for the formation of
M1 in seawater samples of the Seto Inland Sea of
Japan in our present study. Thus, M1 could be a
signi®cant and major degradation product during
the degradation of Irgarol in the ambient aquatic
environment, and the possibility has not been
reported before.
This study was intended to delineate the environ-
mental fate and ecotoxicity of Irgarol 1051 and its (B) MeCN. A ¯ow rate of 10 ml/min was maintained
using a Hitachi HPLC pump (L-6250). Detection was
degradation product M1 in the aquatic environment
accomplished using a UV detector (L-4200H, Hitachi) at
by (a) determining the ambient concentration levels
of Irgarol 1051 and M1 in seawater and sediment
230 nm with a 1 mm light path cell. Fractions containing
M1 were collected using a fraction collector (SF-2120,
samples collected from the Seto Inland Sea (Japan),
and (b) by assessing the ecotoxicity of the two com-
pounds towards aquatic and terrestrial organisms.
Such information is useful in the development of a
management strategy for the control of Irgarol
Advantec) and were then concentrated to dryness. Phos-
phate in the residue was removed with an ODS column
cartridge. A total amount of 14.5 mg puri®ed M1 was
obtained from 21 mg Irgarol 1051. The M1 was utilized in
the ecotoxicity tests as well as the authentic standard for
residue analysis. The purity of the compound was judged
1051 and other new antifouling chemicals in the to be greater than 95% according to its total ion chroma-
togram obtained by the GC-MS analysis.
aquatic environment.
Sampling sites
MATERIALS AND METHODS
In the Irgarol survey of 1998 (No. 5 in Table 2), sea-
water samples were collected from sites 235-4 and the
Mizushima Port. Both sites had previously been selected
Test chemicals
for Irgarol 1051 analysis (Liu et al., 1999b). The exact lo-
cation of the sampling points at the two study sites is
shown in Figs 2 and 3. In brief, site 235-4 is located north-
east of Kojima Peninsula, which is in the southern part of
Okayama Prefecture. Two small ports are situated in this
area, one being a marina with a capacity for 50 pleasure
boats (sampling point 1) and the other a ®shery harbour
with a similar holding capacity for ®shing boats (sampling
points 2 and 3). A high Irgarol concentration (211 ng/l)
had previously been detected in seawater at sampling
point 1 of site 235-4 (Liu et al., 1999b). Sampling points
MIZ-1, MIZ-2, and MIZ-3 in the Mizushima Port were
located at the extreme back part of the port. Vessels for
the Marine Police, the Customs Oce, and the Maritime
Safety Board were all anchored around sampling points
MIZ-1 and MIZ-2, while a vessel for the Immigration
Oce was berthed at sampling point MIZ-3, where Irgarol
1051 was found in 1996 (Liu et al., 1999b).
Irgarol 1051 (2-methylthio-4-tert-butylamino-6-cyclopro-
pylamino-s-triazine, 95%) was gift of Ciba±Geigy
a
Canada (Mississauga, Canada). M1 was prepared by a
mercuric chloride-catalyzed hydrolysis of Irgarol 1051 (Liu
et al., 1999a). Six additional s-triazine herbicides (Fig. 1)
were also tested to compare their phytotoxicity: terbutryn
(2-methylthio-4-tert-butylamino-6-ethylamino-s-triazine,
98%, Riedel-de Haen), terbutylazine (2-chloro-4-tert-buty-
lamino-6-ethylamino-s-triazine, 99.1%, Riedel-de Haen),
terbumeton (2-methoxy-4-tert-butylamino-6-ethylamino-s-
triazine, 99%, Riedel-de Haen), simetryn (2-methylthio-
4,6-diethylamino-s-triazine, 99%, GL Science), atrazine (2-
chloro-4-isopropylamino-6-ethylamino-s-triazine,
98%,
Riedel-de Haen), simazine (2-chloro-4,6-diethylamino-s-
triazine, 99%, GL Science). Organic solvents (methanol,
99.8%, ethyl acetate, 99.8%, Kanto Chemicals; acetonitrile
(MeCN), 99.8%, Ishidzu Chemicals) used for the extrac-
tion and HPLC analysis were pesticide or HPLC grade.
Dimethyl sulfoxide (DMSO, 99.7%, Kanto Chemicals)
used to dissolve chemicals for toxicity tests was spec-
troscopy grade.
Sample collection and preparation
Surface seawater samples collected at sites 235-4 and the
Mizushima Port were ®ltered through a glass ®ber ®lter
(1 mm pore size, 90 mm, GS 25, Advantec). The ®ltered
water was then passed through an ODS column (Sep-Pak
Plus cartridge, tC18 ENV, Waters) pre-conditioned by
passage of methanol (5 ml) and water (10 ml) successively.
The materials adsorbed on the resins were eluted with
methanol (5 ml) and the methanol in the extract was
removed with a centrifugal evaporator at 408C. MeCN
(0.2 ml) was added to the residue and the solution was ®l-
Puri®cation of the degradation product M1
A solution of Irgarol 1051 in acetonitrile (10,000 mg/l)
was slowly added to 3 l of distilled water to a ®nal concen-
tration of 7 mg/l, and followed by the addition of a mer-
curic chloride solution (10,000 mg/l) until the mercuric
chloride in the reaction mixture reached 20 mg/l. The reac-
tion mixture was kept at room temperature (218C) for 2 h
Fig. 2. Sampling points at the sites (A) 235-4 and (B)
Mizushima Port.
Fig. 1. Structure of test chemicals.

Fate and ecotoxicity of irgarol
3525
tered through a ®lter cartridge (0.2 mm pore size, SamPrep its recovery rate calculated in comparison with the pre-
C02-LG, Millipore) by centrifugation. The MeCN extract vious data (surveys No. 1±4).
(1 ml) was then analyzed by GC-MS. The GC-MS analyses were performed as described pre-
Sediment samples were collected at the three sampling viously (Liu et al., 1999b). Peak areas of the molecular
points in site 235-4 on 2 September, 1998 using a grab
sampler, and they were sieved through a 2 mm-mesh strai-
ner immediately after the collection. Glass centrifuge tubes
ions m/z 253 and m/z 213 were used for quantifying the
concentrations of Irgarol 1051 and M1, respectively. The
synthetic sea water samples spiked with Irgarol and M1 to
containing sediments were centrifuged (2500 rpm  20 min) reach several concentrations ranging from 1 ng/l to 1000
and the supernatant discarded. The sediment pellets in the
tube were freeze-dried and then 20 ml of ethyl acetate
were added to a test tube containing 2 g of the lyophilized
sediment. The tube was sonicated for 15 min in a water
bath. After centrifugation the supernatant was collected
and the pellet was extracted once more with fresh solvent.
The combined ethyl acetate extract (40 ml) was evaporated
to dryness and 0.2 ml of MeCN was added to the residue.
After passing through a 0.2 mm membrane ®lter, 1 ml of
the MeCN extract was subjected to a GC-MS analysis.
ng/l were analyzed according to the above mentioned pro-
cedure and the previous report (Liu et al., 1999b). The
detection limit for each compound was regarded as 5 ng/l
under the conditions hired in this study.
Ecotoxicity test
Toxic e€ects of Irgarol 1051 and M1 were evaluated
towards ecologically important species such as bacteria,
crustaceans, microalgae, and terrestrial plants representing
di€erent trophic levels in the ecosystem. The phytotoxicity
of several common s-triazine herbicides (e.g., terbutryn,
terbutylazine, terbumeton, simetryn, atrazine and sima-
zine) was also evaluated for comparison. Stock solutions
of each test chemical were prepared at concentrations of
10,000 mg/l in DMSO. They were tested over at least four
concentrations in each toxicity test.
Residue analysis of Irgarol 1051 and its degradation product
The GC-MS spectra of the ®nal MeCN extracts from
the seawater samples collected during surveys No. 1, 2, 3,
and 4 in our previous 1996±1997 Irgarol ®eld study (Liu
et al., 1999b) were re-examined for Irgarol 1051 and its
degradation product M1. Extracts from 23 out of the total
The e€ect of test compounds on the marine bacter-
ium Vibrio ®sheri was evaluated using the Microtox
93 sampling sites yielded a same chromatographic peak test with 30 min-EC50 (Azur Environmental, USA). The
with mass spectrum and retention time that matched per-
fectly with the puri®ed M1 standard. For con®rming M1
in these extracts, the original 93 MeCN extracts which had
been stored in a refrigerator for about one year were re-
toxicity of the test chemicals to the four crustacean species
(Daphnia magna, Daphnia pulex, Thamnocephalus platyurus
and Artemia salina) was assessed using the commercially
available TOXKITs (Creasel, Belgium): Daphtoxkit
F
analyzed by GC-MS using the puri®ed M1 as a standard. magna, Daphtoxkit F pulex, Thamnotoxkit F, and Artox-
With such an approach, the concentration of Irgarol 1051
in the 23 samples could be simultaneously quanti®ed and
kit M. The cysts of the above four species are ready for
use in each kit and young neonates born in 24 h are used
Fig. 3. Spatial distribution of Irgarol 1051 and its degradation product M1 in the waters of the Seto
Inland Sea, Japan.

3526
H. Okamura et al.
in the toxicity testing. All tests were conducted in the dark
at 208C, following the test procedure provided by the sup-
plier. At least 40 animals were used for each test concen-
tration and the 24 h-LC50 values were calculated for all
the species (Persoone, 1998). A 48 h-LC50 value was also
calculated for D. magna.
in each treatment was also calculated. Toxicity tests were
replicated three times on separate days. The inhibition rate
at each concentration was calculated on the basis of the
average root length obtained from 100 seeds in ®ve petri
dishes prepared as control. The 120 h-EC50 value was cal-
culated from the inhibition rates.
EC50 or LC50 values with 95% con®dence intervals for
each test compound on crustaceans, algae and lettuce were
calculated by Probit analysis using the program ``EcoTox
Statistics Release 1.1'' developed by Dr Y. Yoshioka of
Oita University, Japan.
The algal growth inhibition test was performed using
the freshwater green alga Selenastrum capricornutum
(NIES-35) in accordance with the International Organiz-
ation for Standardization (1987) procedure except for
the culture volume. A typical test solution (3 ml) con-
taining test compound and algae in the medium was
prepared in a 10 ml test tube capped with a thin sili-
cone stopper. The initial number of algal cells was
adjusted to 104 cells/ml and the test tubes were then kept
at 258C in an incubator under continuous illumination.
The algal cell number was determined every 24 h until 72 h
using a cell counter (Celltac MEK4300, Nihon Kohden).
Three test tubes were used at each concentration tested.
The 72 h-EbC50 and ErC50 values were calculated using
the inhibition rates obtained from the area under the
growth curve and from the growth rate, respectively.
The inhibition e€ect of test chemicals on the root
elongation of a higher plant was investigated using lettuce
seeds (Lactuca sativa, OGR-326, Tohoku Seed). Lettuce is
ideal for phytotoxicity testing due to its sensitivity, high
germination rate and long shelf life (Wong and Keturi,
1990). The solution (4 ml) containing the test compound
dissolved in ultra-pure water was added to a ®lter paper
(Whatman No. 3) in a plastic petri dish (9 cm  1 cm
depth). Twenty seeds were placed on the ®lter paper for
each petri dish and the dishes were kept in a loosely sealed
polyethylene bag to reduce evaporation. Incubation was
made in an incubator at 258C in the dark. The root
lengths of the lettuce plants in each treatment were
measured after 5 days of incubation and the average root
RESULTS
The mass spectrum (GC-MS-EI, scanning mode)
of the puri®ed M1 standard prepared by mercuric
chloride-catalyzed hydrolysis of Irgarol 1051 was
found to match exactly with the one obtained from
Irgarol biodegradation (Liu et al., 1997), and from
Irgarol photodegradation (Okamura et al., 1999), as
well as with the one found in the waters of the Seto
Inland Sea in the present investigation (data not
shown). This strongly suggested that M1 was one
of the major degradation products during the en-
vironmental degradation of Irgarol 1051.
Irgarol degradation product M1 was found in 23
out of the total 93 samples analyzed (Table 1), and
this initial ®nding of M1 in waters of the Seto
Inland Sea through re-analysis of the GC-MS data
from our previous 1996±1997 Irgarol survey was
important. There has been no report in the open lit-
erature on the occurrence of Irgarol degradation
lengths of 20 seeds were calculated. The germination rate product(s) in the ambient environment, even though
Table 1. Dissolved Irgarol 1051 and M1 in the waters of Seto Inland Sea
Survey no.a
Sitea
Classi®cationa Sampling date Irgarol concentration Recovery of Irgarol M1 concentration
(d.m.y)a
(ng/l)
(%)
(ng/l)
3
4
HIROSHIMA Prefecture
256'
272'
OKAYAMA Prefecture
marina
®shery harbour
16.7.1997
15.7.1997
77.9
142
116
126
19.7
31.9
235
235-1
235-3
237-1
245
245'
HYOGO Prefecture
197-2
201'
209
marina
marina
®shery harbour
marina
marina
21.8.1997
21.8.1997
21.8.1997
17.8.1997
17.8.1997
17.8.1997
85.3
ND
ND
ND
ND
105
85
651
30.9
68.8
136
829
71.4
®shery harbour
79
72
marina
®shery harbour
marina
18.8.1997
17.8.1997
18.8.1997
ND
58.8
ND
1270
90.5
62.1
OSAKA Prefecture
194
marina
18.8.1997
ND
135
TOKUSHIMA Prefecture
307
310
311
KAGAWA Prefecture
291
296
297
marina
marina
®shery harbour
19.8.1997
19.8.1997
19.8.1997
ND
ND
81.3
29.4
31.9
33.3
110
88
marina
marina
marina
marina
20.8.1997
20.8.1997
19.8.1997
19.8.1997
ND
ND
73.5
ND
47.0
68.8
28.8
34.8
304
EHIME Prefecture
313'
318
3180
322
marina
marina
marina
®shery harbour
21.8.1997
20.8.1997
20.8.1997
21.8.1997
ND
ND
ND
ND
158
820
54.6
79.4
aRefer to Liu et al. (1999b). ND: not detected.

Fate and ecotoxicity of irgarol
3527
Irgarol 1051 has been widely found in the European new antifouling agent in southern Japan, the spatial
freshwater and marine environment. The re-analysis distribution of Irgarol 1051 and M1 in the waters
outcome was also supported by examining the of the Seto Inland Sea is shown in Fig. 3. It can be
stored MeCN extracts (48C for about 1 year) of seen that M1 constituted a major portion of total
those 23 samples by GC-MS (Table 1). Irgarol was Irgarol residues in the collected samples. Appar-
found in 7 MeCN extracts with recovery rates ran- ently, Irgarol 1051 has a wider distribution in the
ging from 72 to 116% compared with the previous northern part of the Seto Inland Sea, whereas M1
data (Liu et al., 1999b). Thus, the recovery demon- was more frequently detected in the northeastern
strates that no remarkable degradation or concen- part of the Inland Sea as well as in the Shikoku
tration occurred during the storage. Table
1
Island area. We have no explanation for this distri-
indicated that concentrations of M1 in the 23 sea- bution pattern. Figure 3 clearly shows that Irgarol
water samples ranged from 19.7 to 1270 ng/l with 1051 is widely used in the whole of the Seto Inland
the maximum concentration at site 197-2 where no Sea.
trace of Irgarol 1051 was detected. Therefore, M1
The toxicity of Irgarol 1051 and M1 towards
appeared to be more stable than Irgarol 1051 in the some ecologically relevant species is shown in
stored MeCN extracts, and possibly in the ambient Tables 3 and 4. Both compounds, up to a concen-
environment.
tration level of 50 mg/l, had little toxic e€ect on the
The existence of both Irgarol and M1 in the test marine bacterium V. ®sheri. The LC50 values
waters of the Seto Inland Sea was again con®rmed and the 95% con®dence intervals for both com-
in our 1998 Irgarol survey (Table 2). Irgarol 1051 pounds to D. magna were similar. In general,
was detected in all 18 seawater samples collected at Irgarol showed a slightly higher toxicity to D. pulex
the two study sites (235-4 and Mizushima Port), and T. platyurus than its degradation product M1.
with concentrations ranging from 54.7 to 296 ng/l. To the test marine crustacean A. salina, Irgarol
M1 was only detected in 14 samples, but with con- 1051 at the maximum test concentration (40 mg/l)
centration levels (20.3±1870 ng/l) which were about was found to cause only a 30% mortality, whereas
1.6±7.3 times higher than that of Irgarol. Seasonal M1 at the same concentration level had no appar-
e€ects on the occurrence of Irgarol and M1 at the ent e€ect on the test species.
two study sites were also investigated. Their levels
The phytotoxicity of Irgarol 1051, M1 and 6
were generally higher in May and lower in Novem- other related s-triazine herbicides to microalgae and
ber, probably re¯ecting seasonal boating activity. a higher plant is shown in Table 4. The EbC50
However, neither Irgarol nor M1 was detected in values for seven test chemicals for the microalgae
the three sediment samples collected at site 235-4. Selenastrum capricornutum were always lower than
The site had been dredged about two weeks prior the ErC50 values, suggesting that EbC50 could be
to sediment sample collection, and this may render considered as a more sensitive index for ranking the
the concentration of Irgarol and M1 in the samples toxicity of these chemicals. The 72 h-EbC50 values
too low for detection.
of Irgarol 1051 and M1 to the microalgae were
To illustrate the trend of marine pollution by this fairly low at 1.6 and 19 mg/l, respectively. Terbutryn
Table 2. Dissolved Irgarol 1051 and M1 in the waters and sedimentsa
Survey No.
5
Site
Sample
Sampling date (d.m.y)
Irgarol concentration (ng/l)
M1 concentration (ng/l)
OKAYAMA Prefecture
235-4-1
235-4-2
235-4-3
235-4-1
235-4-2
235-4-3
235-4-1
235-4-2
235-4-3
235-4-1
235-4-2
235-4-3
MIZ-1
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
seawater
sediment
sediment
sediment
31.5.1998
31.5.1998
31.5.1998
2.9.1998
2.9.1998
2.9.1998
9.10.1998
9.10.1998
9.10.1998
5.11.1998
5.11.1998
5.11.1998
16.6.1998
16.6.1998
16.6.1998
5.11.1998
5.11.1998
5.11.1998
2.9.1998
2.9.1998
2.9.1998
257
227
296
125
157
124
170
265
131
58.9
54.7
83.0
130
113
99.2
80.0
66.3
112
ND
ND
ND
1870
540
1210
509
346
441
316
416
ND
91.7
399
ND
418
244
ND
ND
20.3
366
ND
ND
ND
MIZ-2
MIZ-3
MIZ-1
MIZ-2
MIZ-3
235-4-1
235-4-2
235-4-3
aND: not detected.

3528
H. Okamura et al.
Table 3. E€ect of Irgarol 1051 and M1 on bacteria and crustaceansa
Crustaceans
Bacteria
Test species
endopoint
index
Vibrio ®sheri
bioluminescence
30 min-EC50 (mg/l)
Daphnia magna lethality
Daphnia pulex
lethality 24 h-LC50
(mg/l)
Thamnocepharus
platyurus lethality
214 h-LC50 (mg/l)
Artemia salina
lethality 24 h-LC50
(mg/l)
24 h-LC50
(mg/l)
48 h-LC50
(mg/l)
Irgarol 1051
M1
> 50
> 50
16 (14±18) 8.3 (6.7±10)
17 (15±21) 11 (9.0±12)
5.7 (5.1±6.3)
27 (21±38)
12 (11±13)
19 (17±21)
> 40
> 40
a95% con®dence intervals in parentheses.
was slightly less toxic than Irgarol with an EbC50 strates that Irgarol could undergo environmental
value of 2.0 mg/l. The toxicity potential of the seven transformation in aquatic ecosystems. Moreover,
compounds tested follows the order: Irgarol 1051 > the much higher concentration levels of M1 than
terbutryn > simetryn > terbutylazine > M1 > ter- that of Irgarol 1051 in seawater strongly suggests
bumeton > simazine > atrazine. The 120 h-EC50 that M1 is more stable and/or persistent than the
values of the eight test compounds for lettuce root parent Irgarol in the environment.
elongation are also presented in Table 4. All the
Studies from our laboratories have indicated that
test chemicals (with the exception of M1 and sime- Irgarol 1051 can undergo biological and chemical
tryn), up to a concentration of 50 mg/l, had little degradation, and photodegradation, with the for-
inhibitory e€ect on the germination of lettuce seeds. mation of M1 as the major degradation product
However, M1 was signi®cantly more toxic (Liu et al., 1997, 1999a; Okamura et al., 1999).
(EC50=4.3 mg/l) than Irgarol and ®ve other tria- Based on the results of our past (Liu et al., 1999b)
zine herbicides in terms of inhibiting root and present studies, it can be concluded that Irgarol
elongation.
1051 is widely used in antifouling formulations in
the Seto Inland Sea.
The toxicity of Irgarol 1051 to D. magna and
rainbow trout is shown by a 48 h-EC50 value of
DISCUSSION
The positive identi®cation of Irgarol 1051 and its 8.1 mg/l, and a LC50 value of 0.86 mg/l, respectively
degradation product M1 in the waters of the Seto (Toth et al., 1996). The 48 h-EC50 value of terbu-
Inland Sea (Japan) in the present study is signi®cant tryn, a triazine herbicide, to D. magna was reported
from the view point of managing antifouling chemi- as 7.1 mg/l (Marchini et al., 1988). Thus, most tria-
cals in the aquatic environment. This is the ®rst zine herbicides, including Irgarol 1051 and its
report in the open literature on the detection of a degradation product M1, appear to have only mod-
major Irgarol degradation product in the ambient erate toxicity to D. magna. Being a herbicide,
environment. Irgarol 1051 is considered to be non- Irgarol is much more toxic to algae and higher
biodegradable (Callow and Willingtham, 1996), and plant species, mainly due to the inhibition of photo-
its degradation in seawater and freshwater is slow, synthesis (Dahl and Blanck, 1996). The phytotoxi-
with half-lives of about 100 and 200 days, respect- city of Irgarol to S. capricornutum and Skeletonema
ively (Ciba±Geigy, 1995). However, the detection of costatum has been reported as 1.26 mg/l (120 h-
M1 in the waters of the Seto Inland Sea demon- EC50) and 0.45 mg/l (120 h-EC50), respectively. A
Table 4. Phytotoxicity of Irgarol 1051, M1 and related s-triazine compounds
Microalgaea
Higher plantb
Test species endopoint
index
Selenastrum capricornutum
Lactica sativa root elongation
5 d-EC50 (mg/l)
cell numberÐarea
72 h-EbC50 (mg/l)
cell numberÐgrowth rate
72 h-ErC50 (mg/l)
Irgarol 1051
M1
1.6 (1.5±1.7)
19 (18-21)
2.0 (1.9±2.1)
17 (16±18)
26 (24±27)
11 (10±12)
110 (110±120)
100 (94±110)
2.3 (2.1±2.4)
46 (42±50)
3.3 (3.1±3.6)
36 (33±40)
59 (54±64)
21 (19±22)
180 (170±190)
220 (200±240)
> 50
4.3 (3.9±4.6)
> 50
> 50
> 50
41 (36±49)
> 50
> 50
terbutryn
terbutylazine
terbumeton
simetryn
atrazine
simazine
a95% con®dence intervals in parentheses under continuous light.
b95% con®dence intervals in parentheses under dark.

Fate and ecotoxicity of irgarol
14-day EC50 value of 1.65 mg/l was also obtained
3529
CONCLUSIONS
for frond growth inhibition in the duckweed Lemna
gibba (Bard et al., 1994). An EC50 value of 5.4 mg/l
was determined for zoospores of the seaweed Enter-
omorpha intestinalis (Scarlett et al., 1997). The 72 h-
EbC50 value of Irgarol 1051 to S. capricornutum
obtained in this study was within the same magni-
tude as reported in the literature (Bard et al., 1994).
Although the EC50 value of M1 to S. capricornutum
was about 12 times higher than that of Irgarol, it
was still signi®cantly lower than that of atrazine
and simazine (Table 4), i.e., M1 is more toxic than
atrazine and simazine.
Speci®c conclusions derived from this study are
as follows.
1. For the ®rst time, the presence of an Irgarol
1051 degradation product (M1) was positively
identi®ed in ambient environmental samples. Our
results indicated that the new antifouling com-
pound Irgarol has been widely used in the Seto
Inland Sea of Japan.
2. Concentration levels of M1 found in the waters
of the Seto Inland Sea ranged from 0 to 1870 ng/
l. The origin of M1 in these seawater samples
was probably due to photodegradation of Irgarol
1051 in seawater by natural sunlight.
From the perspective of pesticide management
in the aquatic ecosystem, the ability of M1 to 3. During the 1998 Irgarol survey, concentrations
of M1 were mostly higher than those of Irgarol
1051 in the coastal waters of the Seto Inland
Sea, suggesting a greater environmental persist-
ence for M1 than for the parent compound
Irgarol 1051.
have a much stronger inhibitory e€ect on the
root elongation of lettuce seeds (EC50 value of
4.3 mg/l) than its parent compound Irgarol and the
other six s-triazine herbicides is interesting. This
means that M1 may have a di€erent toxicity mech-
anism from the other triazine compounds in inhibit-
ing root elongation. Atrazine does not a€ect plants
in the absence of light (Solomon et al., 1996), and
is probably not able to inhibit the root elongation
in darkness as observed in this experiment.
4. Both Irgarol 1051 and M1 had weak toxicity to
marine bacteria and marine crustaceans. How-
ever, both compounds were highly toxic to
microalgae and higher plant species, with a bioci-
dal potential greater than atrazine and simazine.
5. Irgarol 1051 and its degradation product M1 are
persistent biocides, and the extensive use of
Irgarol 1051 in antifouling formulations could
a€ect primary production in aquatic ecosystems.
To properly assess the impact of Irgarol 1051
and its degradation product M1 on the aquatic eco-
system, it is necessary to compare the toxicity data
with their ambient environmental concentrations.
The maximum concentration (0.296 mg/l) of Irgarol
1051 was found at site 235-4-3 in May 1998, and it
was about 20% of the 72 h-EbC50 value (1.6 mg/l)
to the test microalga S. capricornutum. Such a con-
centration is almost 12 times that of the no-
observed-e€ect-concentration (0.022 mg/l) to the
zoospores of the seaweed E. intestinalis (Scarlett et
al., 1997). At concentration levels of 63-250 ng/l in
seawater Irgarol 1051 was shown to be capable of
damaging sensitive periphyton communities (Dahl
and Blanck, 1996). Such concentration levels were
found within contaminated areas in marinas and
®shery harbours in the Seto Inland Sea. Therefore,
Irgarol 1051 at current levels in the Seto Inland Sea
may have the potential to a€ect primary pro-
ductivity. In May 1998, a maximum concentration
of 1.87 mg/l of M1 was detected at site 235-4-1.
This level was about 10% of the 72 h-EC50 (19 mg/
l) to the test microalga S. capricornutum, and was
approximately 0.04% of the 120 h-EC50 value
(4.3 mg/l) for lettuce root elongation. Since M1 is
likely to be more persistent than its parent com-
pound Irgarol 1051, and itself still possesses strong
biocidal activity against non-target microalgae and
rooted aquatic plant species, further research is
needed to assess the environmental fate and chronic
toxicity of M1 and other possible Irgarol degra-
dation products towards the aquatic community.
AcknowledgementsÐThis work was supported in part by
the Grant-in-Aid for Encouragement of Young Scientists
(the Ministry of Education, Japan, No. 09780494), the
Salt Science Research Foundation (No. 9816), the Ohara
Foundation for Agricultural Research, and the Great
Lakes 2000 Funds of the Environment Canada. The
authors wish to thank Ciba±Geigy Canada Ltd.
(Mississauga) for the free sample of Irgarol 1051 and the
reviewers for their comments and suggestions.
REFERENCES
Bard J., Pedersen A. and Tiberg E. (1994) Scienti®c Docu-
mentation and Research. KEMI National Chemicals
Inspectorate, Solna, Sweden.
Callow M. E. and Willingtham G. L. (1996) Degradation
of antifouling biocides. Biofouling 10, 239±249.
Ciba±Geigy (1995) Irgarol 1051 in antifouling paints. In
Technical Information Bulletin. Ciba±Geigy Ltd, Basel,
Switzerland.
Dahl B. and Blanck H. (1996) Toxic e€ects of the anti-
fouling agent Irgarol 1051 on periphyton communities
in coastal water microcosm. Marine Pollution Bulletin
32(4), 342±350.
Ferrer I., Ballesteros B., Marco M. P. and Barcelo D.
(1997) Pilot survey for determination of the antifouling
agent Irgarol 1051 in enclosed seawater samples by a
direct enzyme-linked immunosorbent assay and solid-
phase extraction followed by liquid chromatography-
diode array detection. Environmental Science and Tech-
nology 31, 3530±3535.
Gough M. A., Fothergill J. and Hendrie J. D. (1994) A
survey of southern England coastal waters for the s-tria-

3530
H. Okamura et al.
zine antifouling compound Irgarol 1051. Marine Pol-
lution Bulletin 28, 613±620.
International Organization for Standardization. (1987)
Water quality-Algal growth inhibition test. ISO/DIS
8692.
contamination from a triazine herbicide used in anti-
fouling paints. Environmental Science and Technology
27, 1940±1942.
Scarlett A., Donkin M. E., Fileman T. W. and Donkin P.
(1997) Occurrence of the marine antifouling agent
Irgarol 1051 within the Plymouth Sound locality: impli-
cations for the green macroalga Enteromorpha intestina-
lis. Marine Pollution Bulletin 34(8), 645±651.
Solomon K. R., Baker D. B., Richards R. P., Dixon K.
R., Klaine S. J., La Point T. W., Kendall R. J., Weiss-
kopf C. P., Giddings J. M., Giesy J. P., Hall Jr L. W.
and Williams W. M. (1996) Ecological risk assessment
of atrazine in north American surface waters. Environ-
mental Toxicology and Chemistry 15(1), 31±76.
Tolosa I., Readman J. W., Blaevoet A., Ghilini S., Bar-
tocci J. and Horvat M. (1996) Contamination of Medi-
terranean (Cote d'Azur) coastal waters by organotins
and Irgarol 1051 used in antifouling paints. Marine Pol-
lution Bulletin 32, 335±341.
Toth S., van Slooten B., Spack L., de Alencastro L. F.
and Tarradellas J. (1996) Irgarol 1051, an antifouling
compound in freshwater, sediment, and biota of lake
Geneva. Bulletin for Environmental Contamination and
Toxicology 57, 426±433.
Wong W. and Keturi P. (1990) Comparative seed germina-
tion tests using ten plant species for toxicity assessment
of a metal engraving e‚uent sample. Water, Air, and
Soil Pollution 52, 369±376.
Zhou J. L., Fileman T. W., Evans S., Donkin P., Man-
toura R. F. C. and Rowland S. J. (1996) Seasonal distri-
bution of dissolved pesticides and polynuclear aromatic
hydrocarbons in the Humber estuary and Humber
coastal zone. Marine Pollution Bulletin 32, 599±608.
Liu D., Maguire R. J., Lau Y. L., Pacepavicius G. J.,
Okamura H. and Aoyama I. (1997) Transformation of
the new antifouling compound Irgarol 1051 by Phanero-
chaete chrysosporium. Water Research 31(9), 2363±2369.
Liu D., Pacepavicius G. J., Maguire R. J., Lau Y. L.,
Okamura H. and Aoyama I. (1999a) Mercuric chloride-
catalyzed hydrolysis of the new antifouling compound
Irgarol 1051. Water Research 33(1), 155±163.
Liu D., Pacepavicius G. J., Maguire R. J., Lau Y. L.,
Okamura H. and Aoyama I. (1999b) Survey for the
occurrence of the new antifouling compound Irgarol
1051 in the aquatic environment. Water Research
33(12), 2833±2843.
Marchini S., Passerini L., Cesareo D. and Tosato M. L.
(1988) Herbicidal triazines: acute toxicity on daphnia,
®sh, and plants and analysis of its relationships with
structural factors. Ecotoxicology and Environmental
Safety 16, 148±157.
Okamura H., Aoyama I., Liu D., Maguire R. J., Pacepavi-
cius G. J. and Lau Y. L. (1999) Photodegradation of
Irgarol 1051 in water. Journal of Environmental Science
and Health B34(2), 225±238.
Persoone G. (1998) Development and validation of toxkit
microbiotests with invertebrates, in particular crus-
taceans. In Microscale testing in aquatic toxicology, eds
P. G. Wells, K. Lee and C. Blaise, pp. 437±449. CRC
Press, Boca Raton.
Readman J. W., Kwong L. L. W., Grondin D., Bartocci
J., Villeneuve J. P. and Mee L. D. (1993) Coastal water