ArticlePDF Available

Abstract and Figures

Over the course of 78 days, 9 outdoor mesocosms, each with 1350 L capacity, were situated on a pontoon platform in the middle of a lake and exposed to 0 µg L−1 TiO2, 25 µg L−1 TiO2, or 250 µg L−1 TiO2 nanoparticles in the form of E171 TiO2 human food additive five times a week. Mesocosms were inoculated with sediment, phytoplankton, zooplankton, macroinvertebrates, macrophytes, and fish before exposure, ensuring a complete food web. Physicochemical parameters of the water, nutrient concentrations, and biomass of the taxa were monitored. Concentrations of 25 µg L−1 TiO2 and 250 µg L−1 TiO2 caused a reduction in available soluble reactive phosphorus in the mesocosms by 15% and 23%, respectively, but not in the amount of total phosphorous. The biomass of Rotifera was significantly reduced by 32% and 57% in the TiO2 25 µg L−1 and TiO2 250 µg L−1 treatments, respectively, when compared to the control; however, the biomass of the other monitored groups—Cladocera, Copepoda, phytoplankton, macrophytes, chironomids, and fish—remained unaffected. In conclusion, environmentally relevant concentrations of TiO2 nanoparticles may negatively affect certain parameters and taxa of the freshwater lentic aquatic ecosystem. However, these negative effects are not significant enough to affect the overall function of the ecosystem, as there were no cascade effects leading to a major change in its trophic state or primary production.
Content may be subject to copyright.
http://informahealthcare.com/nan
ISSN: 1743-5390 (print), 1743-5404 (electronic)
Nanotoxicology, 2016; 10(7): 902–912
!2016 Taylor & Francis. DOI: 10.3109/17435390.2016.1140242
ORIGINAL ARTICLE
Food web effects of titanium dioxide nanoparticles in an outdoor
freshwater mesocosm experiment
Boris Jovanovic
´
1,2
, Gizem Bezirci
3
, Ali Serhan C¸ag
˘an
3
, Jan Coppens
3
, Eti E. Levi
3
, Zehra Oluz
4
, Eylu¨ l Tuncel
4
,
Hatice Duran
4
, and Meryem Bekliog
˘lu
3,5
1
Faculty of Veterinary Medicine, Chair for Fish Diseases and Fisheries Biology, Ludwig Maximilian University of Munich (LMU), Munich, Germany,
2
Center for Nanoscience (CeNS), LMU, Munich, Germany,
3
Department of Biology, Middle East Technical University, Ankara, Turkey,
4
Department of
Materials Science and Nanotechnology Engineering, TOBB University of Economics and Technology, Ankara, Turkey, and
5
Kemal Kurda¸s Ecological
Research and Training Stations, Lake Eymir, Middle East Technical University, Ankara, Turkey
Abstract
Over the course of 78 days, nine outdoor mesocosms, each with 1350 L capacity, were situated
on a pontoon platform in the middle of a lake and exposed to 0 mgL
1
TiO
2
,25mgL
1
TiO
2
or
250 mgL
1
TiO
2
nanoparticles in the form of E171 TiO
2
human food additive five times a week.
Mesocosms were inoculated with sediment, phytoplankton, zooplankton, macroinvertebrates,
macrophytes and fish before exposure, ensuring a complete food web. Physicochemical
parameters of the water, nutrient concentrations, and biomass of the taxa were monitored.
Concentrations of 25 mgL
1
TiO
2
and 250 mgL
1
TiO
2
caused a reduction in available soluble
reactive phosphorus in the mesocosms by 15 and 23%, respectively, but not in the amount of
total phosphorus. The biomass of Rotifera was significantly reduced by 32 and 57% in the TiO
2
25 mgL
1
and TiO
2
250 mgL
1
treatments, respectively, when compared to the control;
however, the biomass of the other monitored groups—Cladocera, Copepoda, phytoplankton,
macrophytes, chironomids and fish—remained unaffected. In conclusion, environmentally
relevant concentrations of TiO
2
nanoparticles may negatively affect certain parameters and taxa
of the freshwater lentic aquatic ecosystem. However, these negative effects are not significant
enough to affect the overall function of the ecosystem, as there were no cascade effects
leading to a major change in its trophic state or primary production.
Keywords
E171, environmental toxicology, mesocosms,
plankton, titanium dioxide
History
Received 22 March 2015
Revised 23 December 2015
Accepted 5 January 2016
Published online 11 February 2016
Introduction
The application of emerging nanotechnology in various fields is
releasing nanoparticles into the environment at an increasing
pace. Therefore, ecotoxicological findings are needed in order to
carry out adequate risk assessment. Risk assessment is currently
governed by European Government and Council regulations
concerning REACH: the Registration, Evaluation, Authorization,
and Restriction of Chemicals (European Parliament, 2006).
Recent legislation has further clarified how REACH applies to
nanomaterials (European Commission, 2008). In 2006, the
Chemicals Committee of the Organization for Economic
Co-operation and Development (OECD) formed a special
Working Party on Manufactured Nanomaterials (WPMN). Its
purpose was to examine the potential impact of nanoparticles on
the environment and on human health, focusing predominantly on
testing and assessment methods. In 2007, the WPMN launched a
special program and agreed on a priority list of nanomaterials and
relevant endpoints for environmental safety testing. One of the
nanomaterials in the OECD WPMN priority list is titanium
dioxide (TiO
2
).
Between 1916 and 2011, an estimated total of
165 050 000 metric tonnes of TiO
2
was produced worldwide,
including both nano and bulk forms (Jovanovic
´, 2015a). Nano-
TiO
2
is used as a constituent in sunscreens, soaps, shampoos,
toothpastes and other cosmetics, as well as in the paper, building
materials, plastics, ink, pharmaceuticals and food industries. As a
colorant ingredient in food products, TiO
2
is often listed as E171
and used in a variety of common food products, with an estimated
human consumption of 1 mg kg
1
body weight per day (Weir
et al., 2012). E171 is a European Union designation for a white
food color additive known elsewhere by designations such as
CI77891 or Pigment White 6. A significant portion of TiO
2
in
E171 is in the nano-form (Weir et al., 2012; Yang et al., 2014).
The estimated environmental concentration of nano-TiO
2
in
surface water is 0.7–16 mgL
1
(Mueller and Nowack, 2008); in
the case of treated wastewater, effluent concentration is
25 mgL
1
(Westerhoff et al., 2011). In urban runoff, concen-
tration can be as high as 600 mgL
1
(Kaegi et al., 2008), and in
the case of raw sewage, up to 3000 mgL
1
(Kiser et al., 2009;
Westerhoff et al., 2011). Recently, it was suggested that a major
source of nano-TiO
2
in the environment may in fact be E171,
rather than coming from textile or other industry emissions
(Windler et al., 2012). In a large European coastal metropolis such
Correspondence: Dr. Boris Jovanovic
´, Faculty of Veterinary Medicine,
Chair for Fish Diseases and Fisheries Biology, Ludwig Maximilian
University of Munich (LMU), Munich, Germany. E-mail:
nanoaquatox@gmail.com
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
as Istanbul, which has 14 million people, the daily excretory
contribution of both nano, micro and bulk TiO
2
to municipal
wastewater via raw sewage would be 980 kg (1 tonne), as on
average a person consumes 70 mg of TiO
2
per day or 1 mg/kg
body weight (Weir et al., 2012). It has been estimated that 96% of
all TiO
2
is removed from raw sewage in wastewater treatment
plants, but the remaining 4% is released to the aquatic environ-
ment (Westerhoff et al., 2011) predominantly in the nano size. In
the case of Istanbul, this dispersal would result in an approximate
discharge of 40 kg of TiO
2
per day (almost 15 tonnes per year)
into the aquatic environment in the form of E171 TiO
2
food color
alone. A significant portion of these 15 tonnes may be in nano-
form, since smaller particles pass through filters more easily.
Concentration of TiO
2
in wastewaters from production and
refinement factories are even higher, in the range of 1 g L
1
.
For example, the concentration of TiO
2
in the wastewater
supernatant of the TiO
2
producing factory in Finland was
determined to be 1.3 gL
1
(Lehtinen et al., 1984). However, no
information was provided on particle size.
So far, other than standard toxicity texts, very little data have
been acquired about the potential impacts of TiO
2
on the
ecosystem (either nano or microsized), rather than on a species
or population. Furthermore, existing data often conflict. For
example, the addition of TiO
2
nanoparticles to algae culture
medium directly increases the biomass of many freshwater algae
species (Kulacki & Cardinale, 2012). Inversely, nano and micro
TiO
2
is photo sensitive and can express bactericidal and
algaecidal effects by producing reactive oxygen species (Cheng
et al., 2008; Coleman et al., 2005; Huang et al., 2000; Miller et al.,
2012). TiO
2
nanoparticles can also affect the oxidation rate of
ammonia, promoting its conversion to nontoxic form and thus
potentially playing an important role in the reduction of
eutrophication (Altomare & Selli, 2013). Nano-TiO
2
impacts the
pore water surface properties of freshwater sediments and
increases sediment phosphorus adsorption capacity to its max-
imum (Luo et al., 2011). Similarly, among all of the tested
nanomaterials, TiO
2
has the highest removal rate of phosphorous
in water, with an adsorption coefficient of 28.3 mg g
1
, and the
lowest desorption capacity; thus, it could potentially be used for
controlling and preventing eutrophication (Moharami & Jalali,
2014). Nano-TiO
2
retains phosphorous impurities from its
production process (Liu et al., 2013; Yang et al., 2014), delivering
extra phosphorus to aquatic ecosystems. In a study conducted
with hydroponic tomatoes, it was concluded that titanium (Ti)
may compensate for nitrogen (N) deficiencies on plant growth and
metabolism, probably because Ti enhances both the bioavailabil-
ity of N and the N root uptake in these terrestrial plants (Haghighi
et al., 2012). However, it is unknown whether nano-TiO
2
can exert
any similar effects on macrophytes. In fact, a standard OECD No.
221 toxicity test demonstrated that nano-TiO
2
can inhibit the
growth of duckweed (Lemna sp.) with a lowest observed effect
concentration of 125 ppm (Kim et al., 2011). Furthermore, long-
term exposure of wastewater-activated sludge to nano-TiO
2
can
significantly reduce total nitrogen (TN) removal efficiency and
reduce diversity of the microbial community (Zheng et al., 2011).
It is unclear whether this effect can be manifested in an aquatic
ecosystem.
Despite such conflicting reports, some progress has been made
with higher tier toxicity testing. For example, a study performed
in a paddy microcosm showed that nano-TiO
2
had bioaccumu-
lated in hydrobiota and was transferred from prey (biofilm, water
dropwort) to predator (nematodes and mudsnails) in a trophic
food chain (Yeo & Nam, 2013). However, this experiment was
oversimplified, short (17 days), and without many of the recom-
mended OECD monitoring endpoints (OECD, 2006), thus
reducing the relevance of conclusions other than the effects of
bioaccumulation and trophic transfer. In another study, micro-
cosms were again employed to study the effect of nano-TiO
2
bioaccumulation in aquatic organisms (Kulacki et al., 2012).
Results were similar to the previous study, showing nano-TiO
2
accrual in biofilms and bioaccumulation in freshwater snails after
the consumption of biofilms.
Still, no studies have examined the effects of TiO
2
(nano or
micro) from the food web perspective in an outdoor freshwater
mesocosm experiment set up according to OECD guidelines for
higher tier testing. This article aims to fill this major gap by
investigating the following hypothesis: Exposure of lentic fresh-
water mesocosms to E171 TiO
2
will cause measurable changes in
biomass production, from primary producers to zooplankton and
invertebrates, through changes in availability of phosphorous,
nitrogen and sunlight.
Methodology
Mesocosms setup
The experiment was conducted in METU Golet (395201100 N,
324602900 E), a small lake belonging to Middle East Technical
University of Ankara, Turkey. The area of the lake is 2 ha, with a
maximum depth of 11 m. It is situated 998 m above sea level, is
surrounded by hills and forest, and does not receive any municipal
waste water. Swimming and other outdoor activities are forbidden.
The general public, outside of university personnel, is not allowed
to visit, and the experimental area is guarded 24 h per day.
Setup included a floating pontoon platform for storing all
mesocosms together (Supplementary Figure S1). It held nine
identical cylindrical (1.2 mD) fiberglass (4 mm thick) tanks (1.2 m
high), as described in our previous study (Landkildehus et al.,
2014). The platform was anchored in the middle of the lake. After
anchoring, 0.113 m
3
of lake sediment (10 cm of the tank height)
was added to each mesocosm. The sediment was dug from the
lake with a shovel, sieved through 1 cm
2
mesh, and dried by
exposure to sunlight for one week to prevent the hatching of any
fish eggs. Following sediment addition, the mesocosms were
filled with 500 mm filtered lake water (1100 L each). The nutrient
concentration of the lake water at the time was 0.045mg L
1
of
total phosphorous (TP), 0.006 mg L
1
of soluble reactive phos-
phorous (SRP) and 0.37 mg L
1
of TN. Two weeks later, the
mesocosms were inoculated with zooplankton. For this purpose,
10 vertical and 10 horizontal hauls (approximate horizontal haul
was 100 m each) were performed with zooplankton nets on two
nearby lakes, Eymir and Mogan. The hauls were mixed with 20 L
of water, and 750 mL of the zooplankton sample from each lake
was added to each mesocosm. The next day, aquatic macro-
invertebrates of various taxonomic groups (Chironomidae,
Trichoptera, Ephemeroptera, Gammaridae, Odonata, Hirudinea
and Lymnea) were added in equal proportions. The invertebrates
were collected from Lakes Mogan and Eymir with a benthic grab,
as well as with kick-nets on a nearby stream that feeds Lake
Mogan. Three flat pebbles with an approximate sediment contact
surface area of 20–25 cm
2
were added to each tank to provide
additional cover for invertebrates. A week later, seven shoots of
Potamogeton pectinatus and seven shoots of Potamogeton
perfoliatus (approximate length of 5–10 cm) macrophytes were
planted into each mesocosm. Immediately following, eight
topmouth gudgeons (Pseudorasbora parva) were added to each
mesocosm. Great care was taken to ensure the same biomass of fish
(equal distribution of size) in each tank, which was 9–10 gm
3
.
The sampling procedure began 4th May 2014 (Day 0), one day
after fish addition. The experiment lasted for 78 days, concluding
20 July 2014. A control group of three random mesocosms was
selected, while the other six, in two clusters of three random
tanks, were experimental groups. Sample size of mesocosms
DOI: 10.3109/17435390.2016.1140242 Aquatic ecotoxicology of titanium dioxide nanoparticles 903
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
(N¼three per group) was selected based on OECD recommen-
dation as a standard operating procedure for mesocosms studies
(OECD, 2006). In the experimental groups, E171 commercial
food grade TiO
2
was added five times a week (Monday to Friday
at 11:00 AM) at a concentration of either 25 mgL
1
or 250 mgL
1
per treatment, which was triplicated. For this purpose, a stock
suspension of 25 g L
1
was prepared with deionized water. A
commercial sample of human food grade E171 TiO
2
C.I. 77891
manufactured by Fiorio Colori Spa of Italy was obtained through
Pharmorgana GmbH in Eppstein, Germany. According to the
manufacturer, the product was of 99% purity. Control mesocosms
received 10 mL of deionized water five times a week. Water depth
was checked weekly, and the TiO
2
suspension was adjusted
accordingly to water volume to maintain a constant addition of
25 mgL
1
or 250 mgL
1
. These two concentrations were selected
based on the literature. A concentration of 25 mgL
1
is the daily
dose that aquatic ecosystems may receive from wastewater
treatment plants after filtration (Westerhoff et al., 2011) and is
close to the highest predicted concentration in surface water based
on a high emission scenario (Mueller & Nowack, 2008). A
concentration of 250 mgL
1
was selected as a worst case scenario
by applying a 10factor. This concentration could potentially
enter aquatic environments through urban runoff or in cases where
wastewater treatment facilities are not working or not present
(Kaegi et al., 2008; Kiser et al., 2009; Westerhoff et al., 2011). We
assumed a very high sedimentation rate of TiO
2
based on
published data of nanoparticle sedimentation in conditions that
simulate shallow lakes (Keller et al., 2010; Velzeboer et al.,
2014); thus, the addition of TiO
2
was performed five times per
week in order to simulate nanoparticle ‘‘snowing’’ effect which
would normally occur in the environment. At Days 0, 30 and 60,
0.0105 mg L
1
of TP and 0.18 mg L
1
of TN were added to each
mesocosm in the form of Na
2
HPO
4
2(H
2
O) or Ca(NO
3
)
2
4(H
2
O),
respectively. These values are 30% of the nutrient concentration
from day minus 10. Nutrients were added to maintain primary
production and avoid nutrient limitation through biological or
sediment retention. In a previous similar mesocosm setup,
monthly nutrient retention was 20–50% of the available pool
(Landkildehus et al., 2014).
Nanoparticle characterization
A Brunauer–Emmett–Teller (BET) surface area analysis of E171
TiO
2
powder was performed with Autosorb-iQ Station 1 in an N
2
atmosphere. Thermo Scientific K-Alpha was used to carry out
X-ray photoelectron spectroscopy (XPS) analyzes. The Mg Ka
(1253.6 eV) X-ray source was operated at 300 W. A pass energy of
117.40 eV was used for the survey spectra. The spectra were
recorded using a 60take-off angle relative to the normal surface.
X-ray diffraction measurements were made using a Pananalytical
X’pert Pro multipurpose X-ray diffractometer in reflection
geometry. CuKaradiation (¼0.154 nm) was used by operating
at 40 kV and 40 mA. Measurements were made in the 2yrange
from 1to 80in steps of 0.05. Transmission electron
microscopic (TEM) images were obtained for E171 using an
FEI Tecnai G2 F30. Samples were prepared by drop casting 1–2
drops of particle dispersions in ethanol onto a carbon-coated
copper grid. Atomic force microscopy (AFM) was performed with
PSIA XE-100E force spectroscopy with microfabricated Silicon
cantilevers (Olympus OMCL-AC160TS-W2) having a spring
constant of 40 N/mm in tapping mode. The hydrodynamic radius
and the zeta potential of the TiO
2
nanoparticles were measured
with Malvern ZetaNano ZS at 25 C. TiO
2
suspensions were
prepared with 25 mgL
1
and 250 mgL
1
concentrations in two
different media (deionized water and lake water). These suspen-
sions were not sonicated and were prepared in a same manner as
daily made suspensions for mesocosms exposure. Average
hydrodynamic radius was measured every 5 min for f irst 60 min
after preparation. An additional measurement after 24 h was
executed as well. Such measurements were performed in order to
look at suspension instability over the course of time as high
sedimentation was expected. Zeta potentials were measured after
15 min of suspension preparation. An U-shaped capillary cell
DTS1060 was used to estimate both the zeta potential and the
hydrodynamic radius using a He–Ne laser source of 5 mW at
633 nm wavelength.
Sedimentation and effective exposure concentrations were
measured by inductively coupled plasma mass spectrometry (ICP-
MS). For this purpose, one meter long glass sedimentation tube
with 5 cm diameter was used. Sedimentation tubes (N¼3) were
filled with lake water and spiked with 250 mgL
1
of TiO
2
. With a
long syringe water samples were taken at 1, 4, 8 and 24 h after
spiking. Each time, water was collected from 10, 50 and 90 cm of
the column depth. In addition, a control sample consisting only of
lake water was included in order to determine background
concentration of titanium ions in the lake water. Water samples
were prepared for ICP-MS analysis according to the standard
methodology—Method EPA6020A ICP-MS. Concentrations of
titanium isotopes 47 and 49 were measured by Perkin Elmer
NexION 300, corrected for natural abundance, averaged and
later converted to TiO
2
concentration using the following formula:
TiO
2
conc. ¼(Ti conc. of the experimental sample back-
ground Ti conc. of the lake water)mass ratio of TiO
2
/Ti.
Sampling procedure
During the first 3 weeks, samplings and measurements were
performed weekly, starting with Day 0. After that, a bi-weekly
schedule was followed. All samples were collected and all
measurements were performed at the same time of day, before the
daily TiO
2
addition. Each time, water temperature, conductivity,
total dissolved solids, salinity, percentage of oxygen saturation,
total dissolved oxygen and pH were measured in each mesocosm
at a depth of 10 cm, 50 cm, and just above the bottom.
Measurements were performed with a YSI 556 MPS multiprobe
field meter (YSI Incorporated, Yellow Springs, OH). Water
transparency was measured with a Secchi disk, while depth was
measured with a Laylin SM5 depthmeter portable sounder (Laylin
Associates, Unionville, VA).
During each sampling, photosyntethically active radiation
(PAR light) was measured in each mesocosm with a Li-250A light
meter (LI-COR, Lincoln, NE), starting just above the water
surface and continuing until 70 cm water depth in intervals of
10 cm. All data were expressed as a percentage of surface PAR
intensity for the given mesocosm to compensate for different
weather conditions. PAR light was only measured until Day 51,
after which it became impossible due to macrophyte growth.
At each sampling, 20 L was collected from the middle of the
water column of each mesocosm with an improvised PVC pipe
device (5 cm diameter) with a valve on one end and then mixed in
a bucket. Of those 20 L, 1 L total was taken for nutrient,
suspended solids, and Chl-a analyzes, while 6 L were filtered
through a 20 mm mesh filter to collect zooplankton. Both the
filtered water and remaining unfiltered water were returned to the
corresponding mesocosm after sampling. Zooplankton was
preserved in 4% Lugol’s solution. Chl-a pigment and carotenoid
content were determined in phytoplankton by ethanol extraction
in triplicate (Jespersen & Christoffersen, 1987). Measurements
were performed with a Lambda 35 UV/VIS spectrophotometer
(PerkinElmer Inc., Waltham, MA). In addition to Chl-a and
carotenoids, 480/663 and 430/410 ratios were calculated, and
phytoplankton was grouped into four categories: (i) healthy
904 B. Jovanovic
´et al. Nanotoxicology, 2016; 10(7): 902–912
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
phytoplankton (480/66351.3 and 430/41041.2), (ii) heavily
grazed phytoplankton containing degradation products (480/
66351.3 and 430/41051.2), (iii) nitrogen deficient phytoplank-
ton (480/66341.3 and 430/41041.2) and (iv) heavily contami-
nated phytoplankton containing excessive suspended solids
(480/66341.3 and 430/41051.2). Phytoplankton wet biomass
was estimated from the concentration of Chl-a, assuming one unit
of phytoplankton wet weight equaled 0.505% of Chl-a (Kasprzak
et al., 2008), and converted to dry biomass with a factor of 0.2
(Parparov et al., 2014). The ratio of zooplankton to phytoplankton
dry biomass was calculated as an indicator of zooplankton grazing
pressure on phytoplankton.
A periphyton growth experiment was also undertaken in all
mesocosms. Transparent polypropylene strips (21 297 mm
each) with a slightly textured surface (IBICOÕ, Germany) were
placed 30 cm from the mesocosm wall, 50 cm below the water
surface. Three strips were introduced to each mesocosm for
30 days to allow periphyton colonization, and then replaced with a
new set. Both wet and dry biomass were calculated after scraping
the periphyton from the strips.
Total suspended solids were quantified at each sampling by
filtering a known amount of water from each mesocosm and
quantifying the dry mass of residue on the filter.
NH
4+
-N, NO
2
+NO
3
and TN were measured after each
sampling using the Skalar Autoanalyzer (San++ Automated Wet
Chemistry Analyzer, Skalar Analytical, B.V., Breda, the
Netherlands) according to manufacturer protocol. To determine
TP, the acid hydrolysis method was used (Mackereth et al., 1978).
To determine SRP, filtered water was processed using the
molybdate reaction method (Mackereth et al., 1978).
Zooplankton specimens were counted under the microscope
(LEICA MZ 16 stereomicroscope). Length measurements were
taken, and the zooplankton dry biomass was calculated for each
taxon independently according to previously published formulae
for Alona sp. (Rosen, 1981), Bosmina sp. (Michaloudi, 2005),
Daphnia sp. (Bottrell et al., 1976; McCauley, 1984) and copepods
(Bottrell et al., 1976). The mass of each nauplii was considered to
be 0.25 mg (Culver et al., 1985). The volume of rotifers was
calculated from their length for each taxon independently
(Ruttner-Kolisko, 1977), and for unidentified rotifers (51% of
all rotifers by count), the formula V ¼0.124a
3
was used to
calculate volume as an average of all rotifer taxa where awas the
length. Rotifer volume was later transformed to wet weight,
assuming a specific gravity of rotifer body size equal to 1
(1 mg¼10
6
mm
3
(Bottrell et al., 1976). The dry weight of rotifers
is estimated to be 4% of the wet weight for Asplanchna spp. and
10% for all others (Bottrell et al., 1976).
Starting at Day 36, when the growth of the macrophytes
became visible, the percent plant volume inhabited (PVI%) was
calculated using plant surface coverage, height, and water depth.
At the end of the experiment, all macrophytes were harvested
with a hand rake and taken to the laboratory, where they were
cleaned and dried at 105C for 24 h to determine dry mass.
Surface sediment samples (0–5 cm) were also retrieved with a
KC-Denmark Kajak Corer (5.2 cm internal diameter) to collect
chironomids. Six cores were taken from each mesocosm.
Chironomid larvae were preserved in ethanol and their abundance
and biomass were determined.
Statistics
Initial values of all parameters prior to the start of the experiment
(Day 0 samples) and values for parameters sampled only at the
end of the experiment (Day 78 sample of macrophytes, fish and
chironomids) were analyzed with a one-way analysis of variance
(ANOVA). For significant differences, a post hoc comparison of
means between a single control and two experimental groups was
performed using Dunnett’s test. All other data were tested for
treatment, time, and treatment time interaction using split-plot
repeated measures ANOVA (split-plot rANOVA), followed by
Dunnett’s test if significant difference was detected. Data were
log-transformed before analysis where appropriate to reduce
skewness and to approximate to normal distribution. pvalue
0.05 was considered statistically significant, unless otherwise
noted. Post hoc observed power analysis was also performed
(Supplementary Table S1). Statistica 12.0 software (StatSoft Inc.,
Tulsa, OK ) was used for all analyzes. In some cases, for better
visual representation of data, figures are presented both in the
mgL
1
and percentage of control units.
Results
TiO
2
characterization
According to BET analysis, the specific surface area of the TiO
2
was calculated to be 6.137 m
2
g
1
, while pore volume was
0.123 cc g
1
and pore diameter was 2.968 nm.
All detected diffraction peaks in XRD analysis were well
defined and can be perfectly assigned to the anatase crystal
structure. The sharp peaks corresponded to the (101), (004),
(200), (105), (211), (204), (116), (220) and (215) crystal planes of
TiO
2
particles. No characteristic peaks referring to other crystal-
line forms were detected (Figure 1A).
The XPS survey scan of the E171 sample is shown in
Figure 1(B–D). The Ti2p spectra exhibit a Ti2p3/2 peak at
463.8 eV and a 2p1/2 peak at 458.0 eV, characteristic of TiO
2
(Figure 1C). The O1s spectra show a main peak at 529.2 eV,
assigned to oxygen that is bound to tetravalent Ti ions, and a
shoulder at 532.5 eV, which implies that the surface is partially
covered with hydroxide OH groups (Figure 1D). The titan-
ium:oxygen ratio also indicates that the TiO
2
is of anatase crystal
structure.
The TEM investigation revealed that E171 TiO
2
has broad size
distributions with sharp, clean, and well-defined edges. Particle
size varies between 50 and 300 nm (Figure 1E), and the particles
form aggregates in the ethanol suspension. AFM analysis revealed
more precisely the diameter of E171 TiO
2
particles. According to
the AFM mean, particle size ± standard error of the mean (SEM)
was 167 ± 50 nm.
According to DLS measurement in pure water, the particle
diameter is below 100 nm for 30% of TiO
2
particles. The results
in Table 1 show the average hydrodynamic diameters of TiO
2
nanoparticles in two different media and at two different
concentrations over the course of time. TiO
2
suspension in lake
water was making larger aggregates than in deionized water and
the aggregate size in the lake water was reduced when the
nanoparticle concentration increased from 25 to 250 mgL
1
. The
suspension was highly polydispersed with a PDI of 0.9–1. The
z-potential values measured for the dispersions of the TiO
2
NPs in
deionized water and lake water were slightly different. z-potential
ranged from 12.2 ± 0.4 to 20.2 ± 0.4 mV (mean ± SD) for
TiO
2
in deionized water and 6.8 ± 0.3 to 7.4 ± 0.3 mV for
TiO
2
in lake water, respectively. Both dispersions are unstable
because the z-potential value is higher than 30 mV and lower
than +30 mV, which are considered the lowest and the upper
limits for a stable colloidal dispersion. The average hydrodynamic
diameter of the aggregates (Table 1) was high, and it was evident
that the aggregates were settling down fast to the bottom of the
testing chamber within first hour. Such rapid sedimentation
resulted in a decline of the average hydrodynamic diameter of the
remaining aggregates present in the suspension.
Sedimentation of 250 mgL
1
TiO
2
particles in the lake water
occurred, as expected. As determined by ICP-MS at least 48.5%
DOI: 10.3109/17435390.2016.1140242 Aquatic ecotoxicology of titanium dioxide nanoparticles 905
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
of the initial 250 mgL
1
TiO
2
concentration added to the test
system was lost from the suspension after 24 h; most likely by
settling to the bottom (Supplementary Figure S2). This loss of
concentration was most likely achieved through fast sedimentation
of the largest aggregates (responsible for the considerable share of
mass %) as evident by a decline of suspension average
hydrodynamic diameter in the first hour determined by DLS
(Table 1). There was a statistically significant difference in
sedimentation related to time span (rANOVA p50.05). As the
time elapsed the sedimentation become more rapid and the
average TiO
2
concentration ± SEM was: 238.3 ± 13.8;
230.7 ± 18.0; 182.5 ± 9.5 and 128.8 ± 12.5 mgL
1
after 1, 4, 8
and 24 h, respectively. The biggest decrease in concentration
per hour due to sedimentation occurred between 4 and
8 h (12.05 mgL
1
h
1
), as opposed to the first hour
(11.70 mgL
1
h
1
); between 1 and 4 h (2.53 mgL
1
h
1
)or
between 8 and 24h (3.58 mgL
1
h
1
). No depth related stratifi-
cation of TiO
2
suspension occurred in the system (rANOVA
p40.05), although, on average over all time points, concentration
of TiO
2
increased with the water column depth and was:
178.6 ± 17.2; 187.7 ± 16.6 and 229.5 ± 17.1 (mean ± SEM)
mgL
1
for 10, 50 and 90 cm water column depth, respectively.
Physicochemical effects
No statistical differences were found in any of the measured
parameters between any tanks on Day 0 before the start of the
experiment. Thus all groups had similar conditions. During the
experiment, based on the split-plot rANOVA output, no signifi-
cant difference was found between the treatments for dissolved
Figure 1. Characteristics of E171 TiO
2
particles. (A) XRD patterns of the crystal structure, (B) XPS spectra survey scan, (C) XPS spectra of the Ti2p
peak, (D) XPS spectra of the O1s peak and (E) TEM images of E171 TiO
2
particles.
906 B. Jovanovic
´et al. Nanotoxicology, 2016; 10(7): 902–912
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
oxygen, water temperature, salinity, conductivity, water transpar-
ency, pH, total dissolved solids and total suspended solids at any
of the measured water column depths (Table 2). The total water
volume per mesocosm was reduced by 30% toward the end of
experiment due to evaporation, in accordance with established
norms (OECD, 2006). PAR light intensity (Figure 2) increased 5%
on average throughout the water column in the TiO
2
250 mgL
1
treatment compared to the control (Dunnett’s test p50.01). There
was no significant difference between the TiO
2
25 mgL
1
treatment and the control.
TN, NH
4+
N, NO
2
+NO
3
and TP levels were also unaffected
by the treatments (Table 2). However, there was a significant and
dose-dependent reduction of SRP levels in TiO
2
treatments
compared to the control (Figure 3A–C). Overall, the concentra-
tion was reduced by 15 and 23% in the TiO
2
25 mgL
1
and TiO
2
250 mgL
1
treatments, respectively, compared to the control
(Figure 3C).
Biological effects
The treatment of mesocosms with TiO
2
did not induce change in
the concentration of Chl-a, carotenoids, or 480/663 and 430/410
ratios (rANOVA p40.1) when compared to the control (Figure
4A–C).
Analysis of the zooplankton biomass revealed that TiO
2
did
not induce any change in the biomass of Cladocera or Copepoda,
separately or combined (Figure 5A). However, the biomass of
Rotifera was significantly reduced in the TiO
2
25 mgL
1
(by 32%)
and TiO
2
250 mgL
1
(57%) treatments (Dunnett’s test p50.01)
when compared to the control (Figure 5B and C). The ratio of
zooplankton to phytoplankton dry biomass was not significantly
different when TiO
2
treatments were compared with the control;
thus, there was no difference in zooplankton grazing pressure on
phytoplankton.
There was no statistical difference in macrophytes
PVI% between the treatments and control at any single point
in time. At the very end of the experiment, the average PVI%
was 86. Also, no difference was found between the treatments
and control in the biomass of P. pectinatus or P. perfoliatus at
the end of the experiment. After the macrophytes were
harvested, washed, and separated, two additional species were
detected in each mesocosm: Chara sp. and Najas sp. These
species were not initially planted but grew from seeds in the
sediment. There was no difference in the biomass of either
Chara sp. or Najas sp. between the treatments and control.
Also, there was no difference in the total biomass of all of the
macrophytes combined or periphyton biomass between the
treatments and control.
Analysis of abundance and biomass of Chironomus
plumosus did not reveal any statistical difference among treat-
ments (Table 2).
During the experiment, two out of 72 fish died. One died in the
TiO
2
250 mgL
1
group; the second died in the control group.
There were no visible signs of lesions or infection on any
deceased or living fish. The majority of the fish were recovered
with a net after the experiment, and neither the average mass per
fish nor the estimated biomass were statistically different between
the treatments and control (Table 2).
Discussion and conclusion
E171 from Fiorio Colori Spa has been previously characterized as
having (i) an average particle size of 117 nm, with at least 20% of
the particles by number having a diameter5100 nm; (ii) anatase
crystal structure; (iii) 0.13% of Al
2
O
3
impurities and51% of SiO
2
by dry weight and (iv) an isoelectric point52.5 (Yang et al.,
2014). The same producer’s sample has been partially character-
ized elsewhere as having an average particle size of 110 nm, with
at least 36% of the particles by number having a diam-
eter5100 nm (Weir et al., 2012). Although the results of the
present study are not exactly the same as those of previous studies,
they are fairly similar. The present study found the average
primary particle size of the sample to be 167 nm, with pure
anatase crystal structure partially covered with hydroxide OH
groups. Previously, it was determined that E171 characteristics
can vary significantly among producers (Yang et al., 2014);
however, the present study indicates that even different batches
from the same producer may differ. However, methodology and
measuring instruments may also contribute to observed
discrepancies.
The ability of E171 TiO
2
to reduce SRP concentrations was
significant in the present study. It is known that nano-TiO
2
has a
high adsorption rate of phosphorous (28.3 mg g
1
) and a very
low desorption capacity; thus, obtained results are not unex-
pected (Moharami & Jalali, 2014). However, because a total of
1.42 mg L
1
or 14.25 mg L
1
of E171 TiO
2
was added to two
experimental mesocosm groups over 78 days of exposure and the
observed decrease in SRP concentration was 1 and 2 mgL
1
,
respectively, when compared to the control, the TiO
2
efficiently
removed a maximum of 0.7 mg of SRP per g of TiO
2
. This is
30less than the previously determined adsorption coefficient.
The previous study (Moharami & Jalali, 2014) was performed
Table 1. Estimated average hydrodynamic diameter (d
H
) of E171 TiO
2
over time.
Time
TiO
2
25 mgL
1
Lake water d
H
(nm)
(mean ± SD)
TiO
2
25 mgL
1
Deionized water d
H
(nm)
(mean ± SD)
TiO
2
250 mgL
1
Lake water d
H
(nm)
(mean ± SD)
TiO
2
250 mgL
1
Deionized water d
H
(nm)
(mean ± SD)
0 min 12550 ± 2 2494 ± 5 7110 ± 2 5540 ± 4
5 min 8488 ± 2 1796 ± 6 4414 ± 1 3622 ± 3
10 min 4199 ± 2 1635 ± 15 3266 ± 1 2991 ± 8
15 min 6489 ± 2 1513 ± 17 2033 ± 6 1744 ± 11
20 min 3915 ± 2 1441 ± 11 1411 ± 10 3058 ± 2
25 min 3740 ± 1 1150 ± 19 1532 ± 12 2791 ± 3
30 min 3532 ± 1 1320 ± 20 1421 ± 1 3974 ± 4
35 min 1545 ± 11 1261 ± 11 4168 ± 4
40 min 5980 ± 2 1457 ± 21 2655 ± 6 1943 ± 10
45 min 3920 ± 1 1112 ± 15 4271 ± 1 2135 ± 9
50 min 4261 ± 3 1154 ± 15 3817 ± 2
55 min 4895 ± 3 1282 ± 22 3195 ± 7
1 h 3319 ± 1 1296 ± 10 1636 ± 7
24 h 2415 ± 17 437 ± 54 433 ± 19 1148 ± 63
SD refers to standard deviation from six auto-repeated scans by the Malvern ZetaNano ZS.
DOI: 10.3109/17435390.2016.1140242 Aquatic ecotoxicology of titanium dioxide nanoparticles 907
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
Table 2. Summary of all the measurements.
Before the start of the experiment (Day 0) Throughout the experiment (average of all samplings: Days 8–78)
Control TiO
2
25 mgL
1
TiO
2
250 mgL
1
Control TiO
2
25 mgL
1
TiO
2
250 mgL
1
Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM
TN,mgL
1
0.68 0.08 0.69 0.06 0.60 0.02 0.55 0.05 0.62 0.06 0.55 0.05
NO
2
+NO
3
,mgL
1
0.01 0.00 0.01 0.00 0.01 0.00 0.01 0.00 0.02 0.00 0.01 0.00
NH
4+
N, mg L
1
0.11 0.01 0.10 0.01 0.12 0.01 0.05 0.02 0.05 0.01 0.05 0.02
DIN,mgL
1
0.12 0.01 0.11 0.01 0.13 0.01 0.06 0.01 0.07 0.01 0.06 0.02
TP, mgL
1
31.78 2.33 32.88 2.80 29.41 2.69 45.59 4.38 46.97 3.34 41.36 4.21
SRP, mgL
1
6.22 0.14 6.34 0.34 5.69 0.23 7.96 1.13 6.82 0.80 6.14 1.03
Water temperature, C 16.95 0.03 16.89 0.01 16.91 0.01 21.35 0.35 21.38 0.35 21.19 0.34
Conductivity, mS cm
1
0.41 0.01 0.40 0.00 0.41 0.00 0.37 0.01 0.38 0.01 0.37 0.01
PAR light, %* ––––– –35.74 1.47 36.57 1.58 40.45 1.72
Total suspended solids, mg L
1
25.75 2.72 25.35 3.52 41.45 22.67 32.26 6.13 30.91 5.50 28.11 6.10
Total dissolved solids, g L
1
0.27 0.00 0.26 0.00 0.27 0.00 0.24 0.01 0.25 0.01 0.24 0.01
Salinity, g L
1
0.20 0.01 0.19 0.00 0.20 0.00 0.18 0.01 0.18 0.01 0.18 0.01
O
2
,mgL
1
10.79 0.24 11.52 0.11 11.31 0.37 14.26 0.57 13.14 0.35 13.90 0.42
pH 8.40 0.01 8.39 0.06 8.35 0.08 8.68 0.12 8.55 0.10 8.65 0.11
Water column depth, cm 89.33 0.66 88.67 1.33 88.67 1.33 80.24 1.41 82.67 1.31 82.38 1.26
Secchi depth, cm 89.33 0.66 88.67 1.33 88.67 1.33 73.81 3.28 73.43 3.19 74.52 2.70
Chl-a, mgL
1
0.39 0.14 0.22 0.05 0.48 0.11 2.44 0.35 2.78 0.39 2.10 0.34
Periphyton wet biomass, mg cm
2
–––– – – 2.19 0.72 1.61 0.48 3.06 1.99
PVI, % ––––– –47.91 10.11 46.12 9.61 45.21 8.58
Cladocera/Copepoda, mgL
1
153.72 42.38 69.13 10.06 83.07 39.78 137.52 41.37 157.59 34.96 141.18 32.69
Rotifera, mgL
1
2.12 0.32 3.37 1.63 1.79 0.41 5.16 1.18 3.52 0.67 2.19 0.25
Before the start of the experiment (Day 0) At the end of the experiment (Day 78)
Chironomid biomass, mg cm
2
–––– – – 1.18 0.10 1.94 0.83 0.76 0.29
Fish biomass, g m
3
&9–10 &9–10 &9–10 10.79 1.25 9.74 0.83 8.80 0.09
P. pectinatus dry mass, g –––– - 75.41 52.82 47.34 32.72 11.74 7.09
P. perfoliatus dry mass, g –––– - 328.93 144.43 281.67 16.93 371.69 122.51
Chara sp. dry mass, g –––– - 210.63 88.79 207.75 57.71 399.02 9.18
Najas sp. dry mass, g –––– - 49.55 5.49 60.38 16.07 46.45 45.72
All macrophytes dry mass, g –––– - 664.54 110.44 597.14 25.24 828.90 82.25
*These are average results for all measurements at the various water column depth.
908 B. Jovanovic
´et al. Nanotoxicology, 2016; 10(7): 902–912
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
under optimum conditions for adsorption time, temperature, pH,
and adsorbent dosage, while the present results were obtained
under natural conditions. Although the TiO
2
reduced SRP
concentration in the mesocosms, this change had no biological
consequences since the biomass production capacity of the
phytoplankton and macrophytes was not limited by the
phosphorous.
TiO
2
nanoparticles are photoactive and are significantly more
toxic under natural sunlight to a variety of aquatic species
(Jovanovic
´, 2015b). Aggregation of TiO
2
nanoparticles and bio-
logical surface coating both of phytoplankton (Miller et al., 2012)
and zooplankton (Dabrunz et al., 2011) has been described as the
reason for this toxicity expression (Jovanovic
´, 2015b). TiO
2
phototoxicity is manifested by the particle production of reactive
ControlTiO2 250 µg/LTiO2 25 µg/L
40
50
60
70
80
90
100
110
120
130
% of average control
P = 0.1
P < 0.01
0
20
40
60
80
100
120
140(A)
(B)
(C)
0 8 16 22 36 51 65 77
0 8 16 22 36 51 65 77
Days of exposure
% of control
TiO2 25 ug/L
TiO2 250 ug/L
Control
0
3
6
9
12
15
18
21
24
Days of exposure
SRP ug/L
TiO2 25 ug/L
TiO2 250 ug/L
Control
Figure 3. Effect of TiO
2
on SRP levels. (A and B) time series; (C) combined data over multiple sampling times and expressed as % of the average
control from the corresponding control in time. Dashed lines represent point in time when 30% of the nutrients were added to the system based on
average nutrient concentration from day minus 10. A and B: whiskers represent SEM; C: boxes refer to SEM and whiskers to standard deviation.
0
10
20
30
40
50
60
70
80
90
100
0 102030405060
Depth (cm)
% of PAR light penetration
TiO2 25 ug/L
TiO2 250 ug/L
Control
*
Figure 2. Effect of TiO
2
on intensity of PAR light in the water column. Data were averaged over multiple sampling times. Asterisk (*) indicates that the
effect is statistically significant at a level of p50.01.
DOI: 10.3109/17435390.2016.1140242 Aquatic ecotoxicology of titanium dioxide nanoparticles 909
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
oxygen species, which cause oxidative damage (Li et al., 2014;
Miller et al., 2012). Additionally, the inhibition of molting,
reproduction, swimming, growth, or available food reduction for
zooplankton (Campos et al., 2013; Dabrunz et al., 2011;
Jacobasch et al., 2014) is yet another mode of action. All of
these toxic effects would essentially reduce the available biomass
of the target taxon in an ecosystem. However, such toxic effects
have been demonstrated only in laboratory settings using
0
20
40
60
80
100
120
140
160
180
200
Rotifera biomass % of average control
P < 0.01
P < 0.01
0
50
100
150
200
250
300
350
400
450
500
(A) (C)
(B)
Cladocera/Copepoda ug/L
TiO2 25 ug/L
TiO2 250 ug/L
Control
0
2
4
6
8
10
12
14
Rotifera ug/L
TiO2 25 ug/L
TiO2 250 ug/L
Control
Days of exposure
0 8 16 22 36 51 65
Days of exposure
0 8 16 22 36 51 65
ControlTiO2 250 µg/LTiO2 25 µg/L
Figure 5. Effect of TiO
2
on zooplankton biomass. (A) Cladocera + Copepoda time series, (B) Rotifera time series, (C) Rotifera combined data over
multiple sampling times and expressed as % of average control from corresponding control in time. Dashed lines represent point in time when 30% of
the nutrients were added to the system based on average nutrient concentration from day minus 10. A and B: whiskers represent SEM; C: boxes refer to
SEM and whiskers to standard deviation.
0
1
2
3
4
5
6
7
Days of exposure
Chl-a ug/L
TiO2 25 ug/L
TiO2 250 ug/L
Control
0
1
2
3
4
5
6
7
8
9
10
(B)
(A)
(C)
Carotenoids ug/L
TiO2 25 ug/L
TiO2 250 ug/L
Control
0 8 16 22 36 51 65 77
Days of exposure
0 8 16 22 36 51 65 77
Figure 4. (A) Time series of Chl-a concentration, (B) carotenoids concentration and (C) 480/663 and 430/410 ratios across TiO
2
treatments and control
group. Dashed lines represent point in time when 30% of the nutrients were added to the system based on average nutrient concentration from day
minus 10. Whiskers represent SEM.
910 B. Jovanovic
´et al. Nanotoxicology, 2016; 10(7): 902–912
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
standardized water media with a single species environment and
may be lost in a multispecies ecosystem environment. Nekton
organisms may actively seek shelter from sunlight, minimizing
the phototoxicity effects of TiO
2
. Food partitioning of consumer
organisms in the presence of TiO
2
to avoid increased competition
may be another mechanism to maintain ecosystem balance.
In an aquatic ecosystem, TiO
2
particles can settle at the bottom
rapidly due to presence of natural organic matter or can even be
covered by sediment. In fact, it was previously suggested that nano-
TiO
2
has a significant sedimentation rate (Keller et al., 2010;
Velzeboer et al., 2014). The present study also demonstrated that
concentration of suspended nano-TiO
2
in lake water decreased
50% within the 24 h period due to sedimentation. Concomitantly,
rapid aggregation of the particles was detected. Such behavior of
nanoparticles introduces a new variable—the ‘‘nanoparticle
snowing effect’’. The snowing effect normally occurs in the
aquatic environment within the proximity of nanoparticle pollution
sources due to the persistent input. While it is difficult to simulate
such effect in the conventional laboratory toxicity tests without
multispecies environment, the present outdoor mesocosm study
provides conditions close to reality. This is especially important
since concentration of nano-TiO
2
is not unequivocal in the aquatic
ecosystem due to the sedimentation and aggregation; and it is
changing dynamically on a spatiotemporal scale. As a result,
different species/individuals, even different parts of a single
individual (e.g. macrophytes) will be exposed to a different
concentration of nano-TiO
2
based on their biological traits and
ecological roles. Individuals from two different ecosystem com-
partments may be exposed to equal effective exposure concentra-
tions expressed as g L
1
. At the same time, they may also be
exposed to two drastically unequal concentrations if expressed as
gm
2
or particle #L
1
. Although the outdoor mesocosm studies
are more realistic compared to laboratory toxicity tests in terms of
risk assessment, quantification and characterization of nanoparti-
cles are much more difficult. Real time quantification and
characterization of nanoparticles in a mesocosm on a spatiotem-
poral scale are currently impossible due to technical restrictions,
while available snapshot analyzes provide only limited data
regarding effective exposure concentration.
The present study could not demonstrate any effects of
environmentally relevant concentrations of E171 TiO
2
on biomass
changes of phytoplankton, Cladocera, Copepoda, macrophytes, C.
plumosus or P. parva in mesocosms over a prolonged period of
exposure. The only apparent effect was for Rotifera, which
experienced a significant reduction in biomass compared to the
control. Very little literature exists on the effect of nanoparticles
or TiO
2
on rotifers. Previously, it was demonstrated that rotifers
are able to ingest plastic nanoparticles of various sizes (Snell &
Hicks, 2011). Nano-TiO
2
toxicity has been investigated in only
one euryhaline rotifer species, Brachionus plicatilis, and it caused
growth inhibition at a concentration far exceeding those of the
present study (Cle
´ment et al., 2013). The five most common
rotifer taxa observed in the present study were Hexarthra,
Polyarthra, Keratella, Asplanchna and Lecane, accounting for
490% of all counted individuals. Thus, there is not enough
scientific information to explain observed effects. A possible
reason for the rotifer biomass reduction may be biological surface
coating with nano-TiO
2
, which was previously described for other
zooplankton organisms (Dabrunz et al., 2011) and led to impaired
food filtering or reproduction. In addition, rotifers are inefficient
swimmers, spending up to 60% of their metabolism energy for
locomotion (Epp & Lewis, 1984). Thus, coating with a material of
high specific gravity such as TiO
2
(3.77–4.23 g cm
3
) may induce
starvation and exhaustion or may cause rotifers to sink. Reduced
swimming capability likely increases the risk of falling prey to
copepods or other zooplankton species. Although rotifers play an
important role in many freshwater plankton communities, they are
not considered a keystone species (Waltz, 1997). They may,
however, play a significant role in the microbial web (Arndt,
1993).
In conclusion, environmentally relevant concentrations of
E171 TiO
2
nanoparticles may negatively affect certain parameters
and taxa of the freshwater lentic aquatic ecosystem. In particular,
treatments of 25 mgL
1
TiO
2
and 250 mgL
1
TiO
2
caused a
reduction in the amount of available soluble reactive phosphorus
in experimental mesocosms by 15 and 23%, respectively. The
biomass of Rotifera was significantly reduced by 32 and 57% in
the TiO
2
25 mgL
1
and TiO
2
250 mgL
1
treatments, respectively,
when compared to the control. Finally, the intensity of PAR light
increased by 5% throughout the water column in the TiO
2
250 mgL
1
treatment. However, none of these negative effects
were significant enough to affect the overall function of the
ecosystem, as there were no cascade effects leading to a major
change in its trophic state or primary production.
Declaration of interest
The present research was partially supported by B. Jovanovic
´’s
Marie Curie FP7 Career Integration Grant within the 7th
European Union Framework Program (Project No. PCIG13-GA-
2013-618006). B. Jovanovic
´also extends acknowledgement to the
Scientific and Technological Research Council of Turkey
(TUBITAK) for partial support through the visiting scientist
fellowship program (TUBITAK 2221). M. Bekliog
˘lu was sup-
ported by the MARS project (Managing Aquatic ecosystems and
water Resources under multiple Stress), funded under the 7th EU
Framework Program, Theme 6 (Environment including Climate
Change), Contract No. 603378 (http://www.mars-project.eu). We
would also like to extend our gratitude to the TUBITAK Marmara
Research Center of Turkey, Gebze–Kocaeli, for the help with ICP-
MS analysis.
References
Altomare M, Selli E. 2013. Effects of metal nanoparticles deposition on
the photocatalytic oxidation of ammonia in TiO
2
aqueous suspensions.
Catalysis Today 209:127–33.
Arndt H. 1993. Rotifers as predators on components of the microbial web
(bacteria, heterotrophic flagellates, ciliates) — a review. Hydrobiologia
255256:231–46.
Bottrell HH, Duncan A, Gliwicz ZM, Grygierek E, Herzig A, Hillbricht-
Ilkowska A, et al. 1976. A review of some problems in zooplankton
production studies. Norw J Zool 24:419–56.
Campos B, Rivetti C, Rosenkranz P, Navas JM, Barata C. 2013. Effects of
nanoparticles of TiO
2
on food depletion and life-history responses of
Daphnia magna. Aquatic Toxicol 130–131:174–83.
Cheng TC, Chang CY, Chang CI, Hwang CJ, Hsu HC, Wang DY, Yao
KS. 2008. Photocatalytic bactericidal effect of TiO
2
film on fish
pathogens. Surf Coat Tech 203:925–7.
Cle
´ment L, Hurel C, Marmier N. 2013. Toxicity of TiO(2) nanoparticles
to cladocerans, algae, rotifers and plants - effects of size and crystalline
structure. Chemosphere 90:1083–90.
Coleman HM, Marquis CP, Scott JA, Chin SS, Amal R. 2005.
Bactericidal effects of titanium dioxide-based photocatalysts. Chem
Eng J 113:55–63.
Culver DA, Boucherle MM, Bean DJ, Fletcher JW. 1985. Biomass of
freshwater crustacean zooplankton from length–weight regressions.
Can J Fish Aquat Sci 42:1380–90.
Dabrunz A, Duester L, Prasse C, Seitz F, Rosenfeldt R, Schilde C, et al.
2011. Biological surface coating and molting inhibition as mechanisms
of TiO
2
nanoparticle toxicity in Daphnia magna. PLoS One 6:e20112.
Epp RW, Lewis WM. 1984. Cost and speed of locomotion for rotifers.
Oecologia 61:289–92.
European Commission, 2008. Follow-up to the 6th Meeting of the
REACH Competent Authorities for the implementation of Regulation
(EC) 1907/2006; (REACH). European Commission Report number:
DOI: 10.3109/17435390.2016.1140242 Aquatic ecotoxicology of titanium dioxide nanoparticles 911
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
CA/59/2008 [Online]. Available at: http://ec.europa.eu/environment/
chemicals/reach/pdf/nanomaterials.pdf Accessed on September 2014.
European Parliament, 2006. Regulation (EC) No. 1907/2006 of the
European Parliament and of the Council of 18 December 2006
concerning the Registration, Evaluation, Authorisation and Restriction
of Chemicals (REACH) (OJ L 396, 30.12.2006).
Haghighi M, Heidarian S, Teixeira da Silva J. 2012. The effect of titanium
amendment in N-withholding nutrient solution on physiological and
photosynthesis attributes and micronutrient uptake of tomato. Biol
Trace Elem Res 150:381–90.
Huang Z, Maness PC, Blake DM, Wolfrum EJ, Smolinski SL, Jacoby
WA. 2000. Bactericidal mode of titanium dioxide photocatalysis.
J Photochem Photobiol A Chem 130:163–70.
Jacobasch C, Vo
¨lker C, Giebner S, Vo
¨lker J, Alsenz H, Potouridis T, et al.
2014. Long-term effects of nanoscaled titanium dioxide on the
cladoceran Daphnia magna over six generations. Environ Pollut 186:
180–6.
Jespersen AM, Christoffersen K. 1987. Measurements of chlorophyll a
from phytoplankton using ethanol as extraction solvent. Archiv Fur
Hydrobiologie 109:445–54.
Jovanovic
´B. 2015a. Critical review of public health regulations of
titanium dioxide, a human food additive. Integr Environ Assess Manag
11:10–20.
Jovanovic
´B. 2015b. Review of titanium dioxide nanoparticle photo-
toxicity: developing a phototoxicity ratio to correct the endpoint values
of toxicity tests. Environ Toxicol Chem 34:1070–7.
Kaegi R, Ulrich A, Sinnet B, Vonbank R, Wichser A, Zuleeg S, et al.
2008. Synthetic TiO
2
nanoparticle emission from exterior facades into
the aquatic environment. Environ Pollut 156:233–9.
Kasprzak P, Padisa
´k J, Koschel R, Krienitz L, Gervais F. 2008.
Chlorophyll a concentration across a trophic gradient of lakes: an
estimator of phytoplankton biomass? Limnologica 38:327–38.
Keller AA, Wang H, Zhou D, Lenihan HS, Cherr G, Cardinale BJ, et al.
2010. Stability and aggregation of metal oxide nanoparticles in natural
aqueous matrices. Environ Sci Technol 44:1962–7.
Kim E, Kim SH, Kim HC, Lee S, Lee S, Jeong S. 2011. Growth inhibition
of aquatic plant caused by silver and titanium oxide nanoparticles.
Toxicol Environ Health Sci 3:1–6.
Kiser MA, Westerhoff P, Benn T, Wang Y, Pe
´rez-Rivera J, Hristovski K.
2009. Titanium nanomaterial removal and release from wastewater
treatment plants. Environ Sci Technol 43:6757–63.
Kulacki KJ, Cardinale BJ. 2012. Effects of nano-titanium dioxide on
freshwater algal population dynamics. PLoS One 7:e47130.
Kulacki KJ, Cardinale BJ, Keller AA, Bier R, Dickson H. 2012. How do
stream organisms respond to, and influence, the concentration of
titanium dioxide nanoparticles? A mesocosm study with algae and
herbivores. Environ Toxicol Chem 31:2414–22.
Landkildehus F, Søndergaard M, Beklioglu M, Adrian R, Angeler DG,
Hejzlar J, et al. 2014. Climate change effects on shallow lakes: design
and preliminary results of a cross-European climate gradient
mesocosm experiment. Estonian J Ecol 63:71–89.
Lehtinen KJ, Larsson A
´, Klingstedt G. 1984. Physiological disturbances
in rainbow trout, Salmo gairdneri (R.), exposed at two temperatures to
effluents from a titanium dioxide industry. Aquatic Toxicol 5:155–66.
Li S, Pan X, Wallis LK, Fan Z, Chen Z, Diamond SA. 2014. Comparison
of TiO
2
nanoparticle and graphene-TiO
2
nanoparticle composite
phototoxicity to Daphnia magna and Oryzias latipes. Chemosphere
112:62–9.
Liu X, Chen G, Erwin JG, Adam NK, Su C. 2013. Release of
phosphorous impurity from TiO
2
anatase and rutile nanoparticles in
aquatic environments and its implications. Water Res 47:6149–56.
Luo Z, Wang Z, Wei Q, Yan C, Liu F. 2011. Effects of engineered nano-
titanium dioxide on pore surface properties and phosphorus adsorption
of sediment: Its environmental implications. J Hazard Mater 192:
1364–9.
Mackereth FGH, Heron J, alling JF. 1978. Water analysis: some revised
methods for limnologists. Freshwater Biological Association:
Ambleside, Scientific Publication No. 36.
McCauley E. 1984. The estimation of the abundance and biomass of
zooplankton in samples. In: Downing JA, Rigler FH, eds. A Manual on
Methods for the Assessment of Secondary Productivity in Freshwaters.
Oxford: Blackwell Scientific Publications, 228–65.
Michaloudi E. 2005. Dry weights of the zooplankton of Lake Mikri
Prespa (Macedonia, Greece). Belg J Zool 135:223–7.
Miller RJ, Bennett S, Keller AA, Pease S, Lenihan HS. 2012. TiO
2
nanoparticles are phototoxic to marine phytoplankton. PLoS One 7:
e30321.
Moharami S, Jalali M. 2014. Effect of TiO
2
,Al
2
O
3
, and Fe
3
O
4
nanoparticles on phosphorus removal from aqueous solution. Environ
Prog Sustain Energy 33:1209–19.
Mueller NC, Nowack B. 2008. Exposure modeling of engineered
nanoparticles in the environment. Environ Sci Technol 42:4447–53.
OECD, 2006. Guidance document on simulated freshwater lentic field
tests (outdoor microcosms and mesocosms). OECD Environment
Health and Safety Publications Series on Testing and Assessment No.
53, Paris, France. ENV/JM/MONO(2006)17.
Parparov A, Zohary T, Berman T, Gal G. 2014. Seston and organic
matter. Lake Kinnaret: ecology and management. In: Zohary T,
Sukenik A, Berman T, Nishri A, eds. Aquatic Ecology Series. Vol. 6.
New York: Springer, 473–84.
Rosen RA. 1981. Length-dry weight relationships of some freshwater
zooplankton. J Freshwater Ecol 1:225–9.
Ruttner-Kolisko A. 1977. Suggestions for biomass calculations of
plankton rotifers. Arc Hydrobiol Beih Ergebn Limnol 8:71–6.
Snell TW, Hicks DG. 2011. Assessing toxicity of nanoparticles using
Brachionus manjavacas (Rotifera). Environ Toxicol 26:146–52.
Velzeboer I, Quik JTK, van de Meent D, Koelmans AA. 2014. Rapid
settling of nanoparticles due to heteroaggregation with suspended
sediment. Environ Toxicol Chem 33:1766–73.
Waltz N. 1997. Rotifer life history strategies and evolution in freshwater
plankton communities. In: Streit B, Stadler T, Lively CM, eds.
Evolutionary Ecology of Freshwater Animals: Concepts and Case
Studies. Vol. 82. Switzerland: Springer, 119–149.
Weir A, Westerhoff P, Fabricius L, Hristovski K, von Goetz N. 2012.
Titanium dioxide nanoparticles in food and personal care products.
Environ Sci Technol 46:2242–50.
Westerhoff P, Song G, Hristovski K, Kiser MA. 2011. Occurrence and
removal of titanium at full scale wastewater treatment plants:
implications for TiO
2
nanomaterials. J Environ Monit 13:1195–203.
Windler L, Lorenz C, von Goetz N, Hungerbu
¨hler K, Amberg M,
Heuberger M, Nowack B. 2012. Release of titanium dioxide from
textiles during washing. Environ Sci Technol 46:8181–8.
Yang Y, Doudrick K, Bi X, Hristovski K, Herckes P, Westerhoff P, Kaegi
R. 2014. Characterization of food-grade titanium dioxide: the presence
of nanosized particles. Environ Sci Technol 48:6391–400.
Yeo MK, Nam DH. 2013. Influence of different types of nanomaterials on
their bioaccumulation in a paddy microcosm: a comparison of TiO
2
nanoparticles and nanotubes. Environ Pollut 178:166–72.
Zheng X, Chen Y, Wu R. 2011. Long-term effects of titanium dioxide
nanoparticles on nitrogen and phosphorus removal from wastewater
and bacterial community shift in activated sludge. Environ Sci Technol
45:7284–90.
Supplementary material available online
Supplementary Table S1 and Figures S1–S2
912 B. Jovanovic
´et al. Nanotoxicology, 2016; 10(7): 902–912
Downloaded by [The University of British Columbia] at 02:14 10 June 2016
1
Supporting information
Food web effects of titanium dioxide nanoparticles in an outdoor freshwater
mesocosm experiment
Boris Jovanović a,b, Gizem Bezircic, Ali Serhan Çağanc, Jan Coppensc, Eti E. Levic,
Zehra Oluzd, Eylül Tunceld, Hatice Durand, Meryem Beklioğluc,e
aChair for Fish Diseases and Fisheries Biology, Faculty of Veterinary Medicine, Ludwig
Maximilian University of Munich (LMU), Munich, Germany
bCenter for Nanoscience (CeNS), LMU, Munich, Germany
cDepartment of Biology, Middle East Technical University, Ankara, Turkey
dDepartment of Materials Science and Nanotechnology Engineering, TOBB University of
Economics and Technology, Ankara, Turkey
eKemal Kurdaş Ecological Research and Training Stations, Lake Eymir, Middle East
Technical University, Ankara, Turkey
* Corresponding author:
Boris Jovanović
Chair for Fisheries Biology and Fish Diseases
Department of Veterinary Sciences
Faculty of Veterinary Medicine
Ludwig Maximilian University of Munich
Kaulbachstrasse 37, 80539 Munich, Germany
nanoaquatox@gmail.com
2
Supporting Figure S1
Image of pontoon platform and mesocosms used in the study. All tanks are equally
spaced and have the same environmental conditions.
3
0.0
50.0
100.0
150.0
200.0
250.0
300.0
350.0
10 cm 50 cm 90 cm
Water column depth
TiO
2
µg L
-1
1 h
4 h
8 h
24 h
Supporting Figure S2
Estimated change of TiO2 concentration in the test system over time at different water
column depths. Concentrations were measured with ICP-MS (N=3 for each data point).
Initial concentration of TiO2 added to the system was 250 µg L-1. Bars represent standard
error of the mean.
4
Supporting table 1. Statistical output of investigated parameters with rANOVA.
F statistics p value Partial eta
squared
Observed
power*
Rotifera biomass 26.71 0.004853 0.930339 0.988408
SRP 12.661 0.007029 0.808445 0.934759
PAR light 4.796 0.014238 0.210386 0.761581
TN 1.387 0.319786 0.316162 0.197972
NO2+NO3 2.6571 0.149133 0.469697 0.342883
NH4+-N 0.5776 0.589622 0.161458 0.108313
DIN 0.5776 0.589622 0.161458 0.108313
TP 2.24 0.254502 0.598399 0.201072
Water temperature 0.21 0.810172 0.027678 0.077421
Conductivity 0.488 0.636515 0.139793 0.098806
Total suspended solids 0.2513 0.785606 0.077283 0.074544
Total dissolved solids 0.381 0.698324 0.112805 0.087785
Salinity 0.434 0.666929 0.126305 0.093188
O2 0.905 0.435088 0.153328 0.164604
pH 0.71 0.528417 0.191540 0.122580
Water column depth 1.20 0.363585 0.286268 0.177130
Secchi depth 0.056 0.945687 0.018443 0.055379
Chl-a 0.5916 0.582769 0.164720 0.109798
Carotenoids 1.0534 0.415189 0.296444 0.150146
Periphyton wet biomass 0.52547 0.616185 0.149051 0.102793
PVI 0.6524 0.560080 0.206951 0.110442
Cladocera/Copepoda biomass 0.6984 0.533718 0.188845 0.121249
Zooplankton to phytoplankton dry biomass ratio 0.2298 0.801367 0.071154 0.072394
Chironomid biomass 1.15490 0.386957 0.315986 0.160397
Fish biomass 1.3055 0.338300 0.303212 0.188701
P. pectinatus dry mass 0.780962 0.499525 0.206551 0.130201
P. perfoliatus dry mass 0.16823 0.849013 0.053099 0.066276
Chara sp. dry mass 3.19010 0.113834 0.515355 0.401577
Najas sp. dry mass 0.06743 0.935491 0.021983 0.056444
All macrophytes dry mass 2.1754 0.194778 0.420332 0.288299
*Observed power is based on the post hoc power analysis utilizing the effect size and as such is inversely
related to the observed p value.
... The results of the current experiment show minimal impacts of an environmentally realistic concentration of nTiO 2 on all measured macroinvertebrate community parameters, which is in concordance with risk characterizations derived from available toxicity data and modelled environmental concentrations (Hong et al., 2021;Wigger & Nowack 2019, Coll et al., 2016. To our knowledge, only a single study assessed impacts of nTiO 2 on freshwater macroinvertebrate communities to date (Jovanović et al., 2016). In contrast to the present study, Jovanović et al. (2016) however focused on colonization rates of benthic macroinvertebrates in nTiO 2 spiked sediment (25 mg kg −1 ), and found impacts on abundances of taxa and composition of colonizing communities. ...
... To our knowledge, only a single study assessed impacts of nTiO 2 on freshwater macroinvertebrate communities to date (Jovanović et al., 2016). In contrast to the present study, Jovanović et al. (2016) however focused on colonization rates of benthic macroinvertebrates in nTiO 2 spiked sediment (25 mg kg −1 ), and found impacts on abundances of taxa and composition of colonizing communities. Crucially, exposure pathways and concentrations differed between both experiments, and the current study considered impacts on both benthic and pelagic taxa which were already present prior to application of nTiO 2 . ...
Article
Recently, the delivery of pesticides through novel controlled-release (nano-)formulations has been proposed intending to reduce (incidental) pesticide translocation to non-target sites. Concerns have however been raised with regards to the potentially enhanced toxicity of controlled-release (nano-)formulations to non-target organisms and ecosystems. We evaluated long-term (i.e. 1 and 3 month-) impacts of a novel controlled-release pesticide formulation (nano-TiO2-coated carbendazim) and its individual and combined constituents (i.e. nano-sized TiO2 and carbendazim) on naturally established freshwater macroinvertebrate communities. In doing so, we simultaneously assessed impacts of nano-sized TiO2 (nTiO2), currently one of the most used and emitted engineered nanomaterials world-wide. We determined ecological impacts on diversity (i.e. β-diversity), structure (i.e. rank abundance parameters), and functional composition (i.e. feeding guilds & trophic groups) of communities and underlying effects at lower organizational levels (i.e. population dynamics of individual taxa). Freshwater macroinvertebrate communities were negligibly impacted by nTiO2 at environmentally realistic concentrations. The controlled-release (nano-)formulation significantly delayed release of carbendazim to the water column. Nevertheless, conventional- (i.e. un-coated-) and nTiO2-coated carbendazim induced a similar set of adverse impacts at all investigated levels of ecological organization and time points. Our findings show fundamental restructuring of the taxonomic- and functional composition of macroinvertebrate communities as a result of low-level pesticide exposure, and thereby highlight the need for mitigating measures to reduce pesticide-induced stress on freshwater ecosystems.
... Microplastics (MPs) are identified as having particle sizes under 5 mm, whereas particle sizes <1000 nm are defined as NPs [5]. NPs are considered more harmful to aquatic animals than MPs owing to their higher surface area, cell affinity, and adsorption ability, thereby allowing them to penetrate biological barriers and accumulate in different tissues more easily, including gut, liver, and kidney [6][7][8]. ...
Article
Full-text available
Nanoplastic (NP) has become a global environmental concern due to its potentially wide distribution and unknown impacts on organisms. Many laboratory studies have reported the short-term toxicity of NPs but their long-term effects are unclear. Here, the chronic hepatoxic effects of NPs (diameter 100 nm; 0, 10, and 104 items/L) in medaka (Oryzias latipes) were evaluated after three-month exposure, by investigating the alterations in enzymatic biomarkers (digestion, oxidative status, and immunity) and histopathology in the liver. The trypsin and chymotrypsin activities of digestive enzymes were induced at the low concentration of NPs (10 items/L) but inhibited at high concentration (104 items/L); Only the amylase activity was significantly decreased in all NP-treated medaka. Oxidative stress was also induced by NP exposure, which was indicated by the responsive superoxide dismutase and catalase. Lysozyme was significantly reduced after exposure to the high concentration of NPs, while alkaline phosphatase was significantly induced at the low concentration. Exposure to NPs also caused liver damage (e.g., congestion and dilated sinusoids). Taken together, our finding shows that chronic exposure to NPs at low particle concentrations can impair the liver health of medaka by affecting liver enzymatic functions and causing histological damage. Hence, this implies potentially long-term threats of NPs to aquatic organisms, which call for more long-term aquatic toxicological studies on various species using low environmental concentrations.
... Aquatic animals, particularly zooplankton, are frequently used as test organisms to evaluate the ecological safety and health risks of pollutants. Previously, nanoparticles have been suggested to cause rapid toxicity mortality in Daphnia magna and D. galeata (Cui et al. 2017), reduction of Rotifera biomass (Jovanović et al. 2016), significant inhibitory effect on the heartbeat rate (Fekete-Kertész et al. 2016) and moulting inhibition in D. magna (Nasser et al. 2016). Moreover, metal nanoparticles may affect the rotifers across the generations (Martins et al. 2020). ...
Article
Full-text available
The increasing production of nano-TiO2 has attracted extensive concerns about the ecological consequence and health risk of these compounds in natural ecosystem. However, little is known about its toxicity on zooplankton, especially its possibility to access to the food chain via dietary exposure. To address this concern, the toxic and cumulative effects of nano-TiO2 on an aquatic food chain were explored through two trophic levels independently or jointly including producer and consumer. The results revealed that exposure to suspensions of nanomaterials had negative effects on both producers and consumers. Specifically, nanoparticles reduced the density of algal cells in a concentration-dependent way, and hatching life expectancy, average lifespan, net reproductive rate, and population intrinsic growth rate of rotifers decreased significantly with the concentration of nanomaterials increased (P < 0.05). Notably, nanoparticles accumulated in algal cells and were transferred to consumers through dietary exposure. Biomagnification of nano-TiO2 was observed in this simplified food chain, as many of the biomagnification factor (BMF) values in this study were >1. Exposure concentration, exposure time and their interactions play a strong part in the accumulation of nanoparticles in algae and rotifers. Overall, the present findings confirmed that nano-TiO2 was deleterious to plankton, posing a significant environmental threat to aquatic ecosystems. Graphical abstract
... The interest in these nanometer-scale materials is due to their unique properties, which differ from those of the chemically identical bulk materials. Moreover, the outstanding properties of NMs are highly dependent on the NM nature and size distribution [6,7]. The increasing use of NMs, and hence their release into the environment and the potential risk to humans and ecosystem health, concerns the scientific community. ...
Article
Intensive production of nanomaterials, especially metallic nanoparticles (MNPs), and their release into the environment pose several risks for humans and ecosystem health. Consequently, high-efficiency analytical methodologies are required for control and characterization of these emerging pollutants. Single-particle inductively coupled plasma – mass spectrometry (SP-ICP-MS) is a promising technique which allows the determination and characterization of MNPs. However, several elements or isotopes are hampered by spectral interferences, and dynamic-reaction cell (DRC) technology is becoming a useful tool for free interference determination by ICP-MS. DRC-based SP-ICP-MS methods using ammonia as a reaction gas (either on-mass approach or mass-shift approaches) have been developed for determining titanium dioxide nanoparticles (TiO2 NPs), copper oxide nanoparticles (CuO NPs), copper nanoparticles (Cu NPs), and zinc oxide nanoparticles (ZnO NPs). The effects of parameters such as ammonia flow rate and dwell time on the peak width (NP transient signal in SP-ICP-MS) were comprehensively studied. Influence of NP size and nature were also investigated.
... The chemical methods involve the toxic compounds including ethylenediaminetetraacetic acid (EDTA), triethyl amine, polyvinyl alcohol (PVA), thioglycerol and oleic acid that are normally used as a stabilizing and capping agent to protect agglomeration among NPs and production of controlled size and shape of NPs [89][90][91][92][93][94]. In addition, these hazardous chemicals may bound with plants that leads to NPs accumulation in the food chain, thus leading to human health risks [95]. The versatility and application of NPs have revolutionized the agricultural, industrial and health sectors, but at the same time, their occurrence in the environment may pose an adverse risk to ecological and human health. ...
Article
In recent years, engineered nanomaterials have gained enormous attention due to their beneficial applications in various areas of science including industrial, environmental and life sciences. Importantly, the use of nano- technology in agriculture will result in the release of nanoparticles (NPs) into the environment and may contaminate the food chain by entering into the plant system. However, many barriers to implementation must be addressed for technology to be adopted, including safety concerns, consumer acceptance and efficient delivery at field scale. Moreover, NPs interactions with the ecosystem and their toxicity in living systems have become a serious global concern. Additional intricate knowledge is still required about the long-term toxicological fate of NPs released in the environment and their biodegradation. The objective of this review is to provide the various sources of NPs release in the environment, exposure routes and their toxicity in agro-environment subsystems along with potential health risks to human. It also explores the possible mechanism of NPs biodegradation, emerging developments, and potential pitfalls into the research on biodegradable NPs to achieve environmental safety as well as global food security and sustainability.
... iv. Set exposures in mesocosms setting [120,121]. ...
Article
Full-text available
Marine nano-ecotoxicology has emerged with the purpose to assess the environmental risks associated with engineered nanomaterials (ENMs) among contaminants of emerging concerns entering the marine environment. ENMs' massive production and integration in everyday life applications , associated with their peculiar physical chemical features, including high biological reactivity , have imposed a pressing need to shed light on risk for humans and the environment. Environmental safety assessment, known as ecosafety, has thus become mandatory with the perspective to develop a more holistic exposure scenario and understand biological effects. Here, we review the current knowledge on behavior and impact of ENMs which end up in the marine environment. A focus on titanium dioxide (n-TiO2) and silver nanoparticles (AgNPs), among metal-based ENMs massively used in commercial products, and polymeric NPs as polystyrene (PS), largely adopted as proxy for nanoplastics, is made. ENMs eco-interactions with chemical molecules including (bio)nat-ural ones and anthropogenic pollutants, forming eco-and bio-coronas and link with their uptake and toxicity in marine organisms are discussed. An ecologically based design strategy (eco-design) is proposed to support the development of new ENMs, including those for environmental applications (e.g., nanoremediation), by balancing their effectiveness with no associated risk for marine organisms and humans.
... The chemical methods involve the toxic compounds including ethylenediaminetetraacetic acid (EDTA), triethyl amine, polyvinyl alcohol (PVA), thioglycerol and oleic acid that are normally used as a stabilizing and capping agent to protect agglomeration among NPs and production of controlled size and shape of NPs [89][90][91][92][93][94]. In addition, these hazardous chemicals may bound with plants that leads to NPs accumulation in the food chain, thus leading to human health risks [95]. The versatility and application of NPs have revolutionized the agricultural, industrial and health sectors, but at the same time, their occurrence in the environment may pose an adverse risk to ecological and human health. ...
Article
In recent years, engineered nanomaterials have gained enormous attention due to their beneficial applications in various areas of science including industrial, environmental and life sciences. Importantly, the use of nanotechnology in agriculture will result in the release of nanoparticles (NPs) into the environment and may contaminate the food chain by entering into the plant system. However, many barriers to implementation must be addressed for technology to be adopted, including safety concerns, consumer acceptance and efficient delivery at field scale. Moreover, NPs interactions with the ecosystem and their toxicity in living systems have become a serious global concern. Additional intricate knowledge is still required about the long-term toxicological fate of NPs released in the environment and their biodegradation. The objective of this review is to provide the various sources of NPs release in the environment, exposure routes and their toxicity in agro-environment subsystems alongwith potential health risks to human. It also explores the possible mechanism of NPs biodegradation, emerging developments, and potential pitfalls into the research on biodegradable NPs to achieve environmental safety as well as global food security and sustainability.
Article
Nanotechnology is a ground-breaking multidisciplinary field across a broad spectrum of basic and applied sciences for producing and applying nano-sized materials for innovative solutions. The use of nanomaterials in industrial applications, medical products, and consumers has increased repeatedly over the last few years, and these applications will likely continue to grow. In an aquatic ecosystem, nanoparticles take entry through a direct route that includes industrial discharge, disposal of wastewater treatment effluents, and indirect runoff from the soil. After reaching the aquatic environment, the nanomaterials are highly affected by their backdrops and subsequently go through various conversions like agglomeration, aggregation, dissolution, sulfidation, etc. The fate and the behavior of nanomaterials in the aquatic system not only depend on their physical-chemical properties but also on the pH, temperature, salinity, water hardness, and concentration of natural organic matter present in receiving water. In this review, emphasis has been given to the toxicological properties and potential risks of nanomaterials in terms of factors contributing to their toxicology, bioavailability, and accumulation in aquatic organisms as well as the environment. Furthermore, we summarize the published data on engineered nanoparticles’ effect on aquatic organisms. The issues related to the accumulation and penetration of nanoparticles in the aquatic organism, their toxic effect, and biotransformation along with the food web are also discussed. Since nanomaterials are being increasingly released into aquatic bodies, it is important to pay greater attention to their toxicity and how it affects the aquatic ecosystem. The nanomaterials are bioavailable to plants, resulting in trophic transfer, and they impact other organisms through biomagnification, as discussed in this review. To close the enormous information gap, extensive research on the interactions and impacts of NPs on different species belonging to different trophic levels of the aquatic environment and the destiny of NPs along the food chain of the ecosystem is urgently needed.
Article
Impacts of widespread release of engineered titanium dioxide nanoparticles (nTiO2) on freshwater phytoplankton and phytobenthic assemblages in the field, represents a significant knowledge gap. Using outdoor experiments, we quantified impacts of nTiO2 on phytoplankton and periphyton from UK rivers, applied at levels representative of environmentally realistic concentrations (0.05 mg/L) and hot spots of accumulation (5.0 mg/L). Addition of nTiO2 to river water led to rapid temporal size changes in homoagglomerates and many heteroaggregates of nTiO2 with algae in the phytoplankton, including green algae, pennate and centric diatoms, increasing settlement in some cells. Changes in phytoplankton composition were evident after 72 h resulting from a significant decline in the relative abundance of very small phytoplankton (1–3 μm), often accompanied by increases in centric diatoms at both concentrations. Significant changes detected in the composition of the phytobenthos after 12 days, were not evident when using benthic diatoms alone. A lack of inhibition in the maximum quantum yield (Fv/Fm) in phytobenthos after 72-h exposures contrasted with a significant inhibition in Fv/Fm in 75% of phytoplankton samples, the highest recorded in Rutile nTiO2 exposures at both concentrations of nTiO2. After 12 days the strong positive stimulatory responses recorded in the maximum relative electron transport rate (rETRmax) and the maximum non-photochemical coefficient (NPQmax), in phytoplankton and phytobenthos samples exposed to the higher Anatase nTiO2 concentration, were not measured in Rutile exposed biota. Collectively, these results indicate that the Rutile phase of nTiO2 has more negative impacts on freshwater algae than the Anatase form, at specific time scales, and phytoplankton may be more impacted by nTiO2 than phytobenthos. We caution that repeated release of nTiO2, could lead to significant changes in biomass and algal composition, dependent on the phase and concentration of nTiO2.
Article
Full-text available
Titanium dioxide nanoparticles are photoactive and under natural sunlight, produces reactive oxygen species. Reactive oxygen species can be detrimental to many organisms, causing oxidative damage, cell injury, and death. Most studies investigating TiO2 nanoparticle toxicity did not consider photoactivation and performed tests either in dark conditions or under artificial lighting that did not simulate natural irradiation. This paper summarizes the literature and derives a “phototoxicity ratio” (PR) between the results of nano-TiO2 experiments conducted in the absence of sunlight and conducted under solar or simulated solar radiation (SSR) for aquatic species. Therefore, PR can be used to correct endpoints of the toxicity tests with nano-TiO2 which were performed in absence of sunlight. Such corrections may also be important for regulators and risk assessors when reviewing previously published data. A significant difference was observed between the PRs of two distinct groups: (a) aquatic species belonging to order Cladocera and (b) all other aquatic species. Order Cladocera appeared very sensitive and prone to nano-TiO2 phototoxicity. On average nano-TiO2 was 20 times more toxic to non-Cladocera and 1867 times more toxic to Cladocera (median values 3.3 and 24.7 respectively) after illumination. Both median values and 75% quartile of PR are chosen as the most practical values for the correction of endpoints of nano-TiO2 toxicity tests which were performed in dark conditions, or in the absence of sunlight. This article is protected by copyright. All rights reserved
Article
Full-text available
From 1916-2011, an estimated total of 165,050,000 metric tonnes of titanium dioxide pigment were produced worldwide. Current safety regulations on the usage of the TiO2 pigment as an inactive ingredient additive in human food are based on legislation from 1969 and are arguably outdated. This paper compiles new research results to provide fresh data for potential risk re-assessment. However, even after 45 years, few scientific research reports have provided truly reliable data. For example, administration of very high doses of TiO2 is not relevant to daily human uptake. Nevertheless, since dose makes the poison, the literature provides a valuable source for understanding potential TiO2 toxicity after oral ingestion. Numerous scientific papers have observed that TiO2 can pass and be absorbed by the mammalian gastrointestinal tract; can bioconcentrate, bioaccumulate, and biomagnify in the tissues of mammals and other vertebrates; has a very limited elimination rate; and can cause histopathological and physiological changes in various organs of animals. Such action is contrary to the 1969 decision to approve the use of TiO2 as an inactive ingredient in human food without an established acceptable daily intake, stating that neither significant absorption nor tissue storage following ingestion of TiO2 was possible. Thus, relevant governmental agencies should reassess the safety of TiO2 as an additive in human food and consider establishing an acceptable maximum daily intake as a precautionary measure. Integr Environ Assess Manag © 2014 SETAC
Article
Climate change is expected to profoundly affect both temperature and net precipitation, with implications for lake water level. We describe the design of a harmonized, simultaneous, cross-European mesocosm experiment to elucidate the effects of climate change on community structure, functioning, and metabolism in shallow lakes at low and high nutrient levels with contrasting depths (1 and 2 m). We used cylindrical (D = 1.2 m) tanks that were either 1.2 or 2.2 m high, each having a 10-cm sediment layer. We inoculated the mesocosms with a mixed sample of sediment and plankton from lakes with contrasting nutrient concentrations and added macrophytes and planktivorous fish. Sediment was pre-equilibrated to the required experimental nutrient concentration. During the experiment the water level decreased with increasing temperature (up to 90 cm in the Mediterranean mesocosms) while conductivity increased. The average chlorophyll a concentration increased with temperature in the deep mesocosms but was more variable in the shallow mesocosms. Macrophyte
Article
Titanium dioxide (TiO2) is widely used in food products, which will eventually enter wastewater treatment plants and terrestrial or aquatic environments, yet little is known about the fraction of this TiO2 that is nano-scale, or the physical and chemical properties of TiO2 that influence its human and environmental fate or toxicity. Instead of analyzing TiO2 properties in complex food or environmental samples, we procured samples of food-grade TiO2 obtained from global food suppliers and then, using spectroscopic and other analytical techniques, quantified several parameters (elemental composition, crystal structure, size, surface composition) that are reported to influence environmental fate and toxicity. Another sample of nano-TiO2 that is generally sold for catalytic applications (P25) and widely used in toxicity studies, was analyzed for comparison. Food-grade and P25 TiO2 are engineered products, frequently synthesized from purified titanium precursors, and not milled from bulk scale minerals. Nano-sized materials were present in all of the food-grade TiO2 samples, and TEM showed that Samples 1-5 contained 35%, 23%, 21%, 17%, and 19% of nano-sized primary particles (diameter below 100 nm) by number, respectively (all primary P25 particles were < 100 nm). Both types of TiO2 aggregated in water with an average hydrodynamic diameter above 100 nm. Food-grade samples contained phosphorous (P), with concentrations ranging from 0.5 to 1.8 mg P/g TiO2. The phosphorous content in P25 was below ICP-MS detection limits. Presumably owing to a P-based coatings detected by XPS, the zeta potential of the food-grade TiO2 suspension in deionized water ranged from -10 to -45 mV around pH 7, and iso-electric point for food grade TiO2 (< pH 4) were significantly lower than that for P25. The presence of other elements in/on the TiO2 (Si content was 0.026%-0.062% and Al content was 0.0006%-0.810%) were also unique from P25 and would influence the environmental fate of TiO2. X-ray diffraction analysis confirmed the presence of anatase and/or rutile in the food-grade materials, and although the presence of amorphous TiO2 could not be ruled out, it is unlikely on the basis of Raman analysis. The food grade TiO2 was solar photo-active. Cationic dyes adsorbed more readily to food-grade TiO2 than P25, indicating very different potentials for interaction with organics in the environment. This research shows that food-grade TiO2 contains engineered nanomaterials with properties quite different from P25, which has previously been used in many eco-toxicity studies to date, and because food-grade TiO2 is more likely than P25 to enter the environment (i.e., potentially higher exposure levels) there is a need to design environmental (and human) fate and toxicity studies comparing food-grade to catalytic TiO2.
Article
Sedimentation of engineered nanoparticles (ENPs) has been studied mainly in artificial media and stagnant systems mimicking natural waters. This neglects the role of turbulence and heteroaggregation with sediment. We studied the apparent sedimentation rates of selected ENPs (CeO2, PVP-Ag and SiO2-Ag) in agitated sediment-water systems resembling fresh, estuarine and marine waters. Experiments were designed to mimic low energy and periodically resuspended sediment water systems (14 days), followed by a long term aging, resuspension and settling phase (6 months), as would occur in receiving shallow lakes. ENPs in systems with periodical resuspension of sediment were removed with sedimentation rates between 0.14 and 0.50 m/d. The sedimentation rates did not vary much among ENP type, salinity and aging time, which is attributed to the capture of ENPs in sediment flocks. The sedimentation rates were one to two orders of magnitude higher than those reported for aggregation-sedimentation in stagnant systems without suspended sediment. Heteroaggregation rates were estimated and ranged between 0.151 and 0.547 L/mg/d, which is up to 29 times higher than those reported for natural colloids under quiescent settling conditions. We conclude that rapid scavenging and sedimentation drives removal of ENPs from the water column. Environ Toxicol Chem © 2014 SETAC
Article
Phosphorus (P) removal from aqueous solutions was investigated using TiO2, Al2O3, and Fe3O4 nanoparticles (NPs). Adsorption study was performed to determine the optimum operation conditions such as adsorption time, temperature, pH, and adsorbent dosage. Sorption isotherms were well described by linear, Freundlich and Langmuir models. The maximum adsorption capacity of P was 28.3, 24.4, and 21.5 mg g−1 for TiO2, Fe3O4 and Al2O3, respectively. Desorption analysis showed that the desorption capacities were in an order of Al2O3 > Fe3O4 > TiO2. Kinetic data were best fitted with pseudo-second-order and intra-particle diffusion kinetic models. Scanning electron microscopy (SEM), energy dispersive X-ray (XRD), and NPs solution saturation indices (SI) before and after P sorption showed that the main mechanism of P sorption by TiO2 was adsorption, whereas P sorption by Al2O3 and Fe3O4 were due to adsorption and precipitation. Results showed that double layer model (DLM) could be modeled P adsorption onto adsorbents over a wide range of pH. Thermodynamic parameters confirmed the endothermic and not spontaneity nature of the P adsorption. These NPs have potential for use as efficient sorbents for the removal of P from aqueous solutions and TiO2 NPs were identified as the most promising sorbent due to their high metal uptake. © 2014 American Institute of Chemical Engineers Environ Prog, 2014