Computational Water, Energy, and Environmental Engineering, 2013, 2, 16-25
doi:10.4236/cweee.2013.23B004 Published Online July 2013 (http://www.scirp.org/journal/cweee)
Silver Nanoparticle Adsorption to Soi l a nd Water
Treatment Residuals and Impact on Zebrafish in a
Lab-scale Constructed Wetland
Angela Ebeling, Victoria Hartmann, Aubrey Rockman, Andrew Armstrong,
Robert Balza, Jarrod Erbe, Daniel Ebeling
Biology, Chemistry, Biochemistry Departments, Wisconsin Lutheran College, Milwaukee, WI, USA
ReceivedApril, 2013
ABSTRACT
Nanoparticles (< 100 nm) are becoming more prevalent in residential and industrial uses and may enter the environment
through wastewater. Although lab studies have shown that nanoparticles can be toxic to various organisms, limited re-
search has been done on the effects of nanoparticles in the environment. Environmental conditions such as pH and ionic
strength are known to alter the biotoxicity of nanoparticles, but these effects are not well understood. The objectives of
this research were to determine the impacts of silver nanoparticles (AgNP) on zebrafish in the pseudo-natural environ-
ment of a lab-scale constructed wetland, and to investigate wastewater remediation through soil and water treatment
residual (WTR) adsorption of AgNPs. Concurrently, the effect of particle size on AgNP sorption was examined. Re-
searchers exposed adult zebrafish in a lab-scale constructed wetland to concentrations of AgNP ranging from 0 - 50 mg
AgNP/L and compared them to negative controls with no silver exposure and to positive controls with exposure to sil-
ver nitrate. The results suggest that aggregated AgNP do not impact zebrafish. Separately, sorption experiments were
carried out examining three media - a wetland soil, a silt loam soil, and a WTR - in their capacity to remove AgNPs
from water. The silt loam retained less AgNPs from solution than did the wetland soil or the WTR. In the WTR AgNPs
were associated with sand size particles (2 mm - 0.05 mm), but in the wetland soil and silt loam, approximately half of
the AgNPs were associated with the sand-sized particles, while the rest were associated with silt sized (~0.05 mm) or
smaller particles. The larger sorption capacity of the wetland soil and WTR was attributed to their higher carbon content.
The sorption data indicate that AgNPs adsorbed to soil and WTRs and support the idea that natural and constructed
wetlands can remove AgNPs from wastewater.
Keywords: Silver Nanoparticles; Soil; Water Treatment Residuals; Constructed Wetland; Zebrafish; Remediation
1. Introduction
Sufficient, clean, safe drinking water is increasingly
scarce in many parts of the world [1-4]. Pollutants such
as excess nutrients, particulates, and pathogens are
known to cause environmental as well as human health
problems [5-12]. Conventional and natural methods can
effectively remediate wastewater of these types of pol-
lutants [13-15]. Other pollutants such as pharmaceuticals
(hormones, anti-depressants) and nanomaterials (materi-
als less than 100 nm in at least one dimension [16]) in
wastewater are becoming more common [17-21]. For
example, nanoparticles are becoming more prevalent in
residential use (e.g. sunscreen, textiles), medical applica-
tions (e.g. wound treatment), industry (e.g. sensors and
solar cells), and environmental remediation [22-25]. As a
result, nanoparticles are able to enter the environment via
wastewater or improper disposal [16,19,25]. Unfortu-
nately, little is known about the removal of these new
types of pollutants using conventional or natural waste-
water treatment methods.
Numerous studies have shown that many nanomateri-
als, including silver nanoparticles (AgNPs), have toxic
effects on organisms (zebrafish, medaka, rainbow trout,
crucian carp, flathead minnow [26]; rainbow trout [27];
bacteria [28]; zebrafish embryos [29]; plants [30]; ze-
brafish, microalga, water flea [31]; bacteria [32]). Both
engineered and natural nanoparticles play important roles
in the health of terrestrial and aquatic ecosystems [33-35].
Toxicity of metal nanoparticles is well known in labora-
tory settings as indicated in the previously cited studies,
but the fate of nanoparticles in the environment and the
potential for bioaccumulation is unclear and complicated
[16,22,36].
Natural wetlands are known to have important roles in

A. EBELING ET AL. 17
wastewater remediation [37] and constructed wetlands
have been increasingly been used to remove pollutants
[13,15]. The use of natural media (e.g. soil) as well as
engineered or recycled media (e.g. water treatment re-
siduals, WTRs) could improve the effectiveness of the
natural wastewater treatment and these media could have
the potential to remove nanoparticles as well. The objec-
tives of this research were to determine the effectiveness
of several media (two soils and a water treatment residual)
on AgNP removal from solution and to investigate AgNP
effects on zebrafish in a (simulated) natural setting
(lab-scale constructed wetland).
2. Materials and Methods
2.1. General
In general, two main experiments were conducted. The
first experiment used three media to investigate the sorp-
tion of AgNPs out of wastewater: wetland soil, silt loam
soil, and WTRs (waste product of drinking water treat-
ment process). The second experiment investigated the
effect of AgNPs on zebrafish (Danio rerio) living in the
“natural” environment of a lab-scale constructed wetland.
2.2. Sorption Experiments
The sorption study was modeled after sorption experi-
ments used to investigate phosphorus sorption to soil or
water treatment residuals [38-40]. Three media were
shaken with various concentrations of AgNPs in solution
and the amount of AgNPs remaining in the solution after
an equilibration time was measured.
All three media were sent to the Soil and Plant Analy-
sis Lab in Madison, WI for elemental analysis (total
minerals by Inductively Coupled Plasma Optical Emis-
sion Spectrometry, total N by Total/Kjeldahl, and total C
by LECO CNS-2000 analyzer) and particle size analysis
(hydrometer method). A wetland soil was used because
this sorption experiment was designed to be compared
with the constructed wetland study explained in the fol-
lowing section. The wetland soil was used in both the
sorption study and in the constructed wetland and was
obtained from Certified Products, New Berlin, WI (Black
Topsoil: http://www.certifiedproductswi.com/). The sec-
ond medium, a silt loam soil (Plano silt loam, fine-silty,
mixed, superactive, mesic, typic arguidoll) was chosen
because it represents a typical soil in Wisconsin. The
third medium, WTR, was chosen because it has previ-
ously been shown to have the ability to remove phos-
phorus from wastewater and could be used in constructed
wetlands for the removal of that nutrient [41]. If this ma-
terial could also be used to remove AgNPs, it would pro-
vide an even greater incentive to beneficially reuse the
material as an amendment to help clean wastewater.
Little research has been done on the effects of WTRs on
organisms in the environment [38], so to establish that
these WTRs would not induce mortality, preliminary
laboratory experiments using Escherichia coli and native
soil bacteria were conducted and gave no evidence that
moderate levels of WTRs (0.05 g WTR/ml) negatively
impact bacterial growth (data not shown). The nanoparti-
cles used in the sorption experiment were 10 nm mono-
dispersed silver nanoparticles in 2 mM sodium citrate
buffer obtained from Nano Composix (10 nm citrate
NanoXact Silver, JMW1148).
To conduct the sorption experiment, 0.5 g (dry weight)
of each medium was weighed into 15 ml conical tubes.
Then 10 ml of AgNP solution was added to each tube.
Five concentrations of silver nanoparticles were used: 0,
6, 15, 30, and 60 mg AgNP/L. The conical tubes were
capped and shaken for 18 hours at 60 rpm on a Fotodyne
Orbit shaker (Lab-Line Instruments, Inc. Model Number
3520) at 23℃ (room temperature). Each trial was repli-
cated three times. After the 18 hour equilibration time,
the supernatant in each tube was sampled three times –2
ml was removed after 30 s of particle settling, after 2
hours of particles settling, and after centrifuging at 2500
rpm for 10 min (IEC Centra-7R Refrigerated Centrifuge,
S.N. 23601916). The supernatant was sampled at these
times, because one objective in this research was to in-
vestigate AgNPs affinity for adsorbing to different size
particles. According to Stoke’s Law
2
()
18
p
f
sp
g
d
vu
 
 (1)
where vs is the particle settling velocity, ρp is the density
of the particles, ρf the density of the fluid, g the accelera-
tion due to gravity, dp the diameter of the particle, and u
the fluid viscosity, this equation predicts that sand size
particles will settle after about 40 s (thus any AgNP
measured in the supernatant would be either soluble, or
adsorbed to silt or clay sized particles), silt size particles
will settle after approximately 2 hours (thus any AgNP
measured in the supernatant after 2 hours would be solu-
ble or adsorbed to clay size particles), and after centri-
fuging all particles would be removed from the super-
natant (and any AgNPs measured would be in solution)
[42,43]. Using this method to determine particle size is
based on an empirical method and cannot be used to ac-
curately define the particle size [43], but for this research
these sampling times give a rough estimation of the size
of the particles with which AgNPs were associated.
A modified digestion method was used to quantify the
amount of silver in the supernatants [44,45:EPA SW
846Method 3050B]. To each 2 ml supernatant sample, 1
ml of 6 M nitric acid was added using a repeat pipetter
(Eppendorf, Repeater Plus, 2849689). The samples were
placed into a water bath at 90℃ for one hour to digest

A. EBELING ET AL. 18
before measurement of elemental silver on a Perkin El-
mer AAnalyst 200 Atomic Absorption Spectrometer
(S/N 20054062503). Silver nitrate (AgNO3, Fischer Che-
micals, Lot number 041796) was used to make standards.
Samples and AgNO3 standards were analyzed using a
silver detection lamp (PerkinElmer Lumina Hollow Ca-
thode Lamp, P/N N305-0120, S/N 030211-020140).
2.3. Lab-Scale Constructed Wetland Experiment
In this experiment, AgNP effects on zebrafish were in-
vestigated in a lab-scale constructed wetland monitored
in a climate controlled greenhouse. Because most previ-
ous studies of AgNP impacts on organisms have taken
place in petri dishes or other aseptic environments, this
experiment was designed to investigate AgNP impacts in
a more natural setting. Each AgNP treatment was applied
in a separate constructed wetland that consisted of a five
gallon bucket in which wetland media was placed in a
polyvinyl chloride (PVC) column (to keep the media
separate from the zebrafish).
The five gallon bucket contained wetland media in a
11.4 cm (4.5 in) (diameter) by 38.1 cm (15 in) (height)
capped PVC column (Figure 1). To make the wetland
media, each PVC column was filled with 20 cm of 1.3
cm (0.5 inch) washed gravel and 8 cm of wetland soil
(~800 g). The gravel and wetland soil were obtained
from Certified Products, New Berlin, WI. Each bucket
also contained a circulating pump (Mini-Jet 404, Marine-
land) and a small amount of aquarium gravel to support
the column from tipping over. A 1.3 cm (0.5 inch) dia-
meter tube was directed from the pump to a split which
distributed the water pressure; the split was controlled
with a pinch clamp with one half dispersing oxygenated
water back to the fish and the other half entering the top
of the column. The column had multiple 1.3 cm (0.5 inch)
Figure 1. Photos of (a) the top view of the lab-scale con-
structed wetland: five gallon bucket with soil column in the
middle and tubing along the side connected to the circulat-
ing pump and (b) the side view of the soil column: 20 cm of
gravel on the bottom and 8 cm of wetland soil at the top;
mesh covering 1.3 cm holes.
holes covered with mesh to allow water to be circulated
from the bucket through the soil media (Figure 1(b)).
This allowed AgNPs to have contact with the soil media
(which is more similar to a natural environment than
having zebrafish and nanoparticles in isolation).
Each bucket contained 12 L of deionized (DI) water
with 3.8 g of ocean salt and 0.5 g of pH 7 buffer. There
were six treatments: negative control (ocean salt and no
AgNP), negative control with dispersing agent (ocean
salt with 2 mg/L Tide and no AgNP), and 15 mg AgNP/L
(< 90 nm powder, mKnano), 25 mg AgNP/L, 50 mg
AgNP/L, and a positive control (15 mg AgNO3/L). Each
AgNP treatment also had ocean salt and 2mg/L Tide. The
liquid laundry detergent Tide (Tide® Active with Fe-
breze) was used as the dispersing agent for the AgNP
powder, both because it was shown to be an effective dis-
persing agent in preliminary trials and because it mim-
icked one of the ways nanoparticles might enter waste-
water (i.e. through residential laundry). Tide is a com-
mercial laundry detergent commonly used in American
households. After the constructed wetland was prepared
(PVC column of wetland media, circulating pump, and
treatment addition), five zebrafish, three female and two
male, were placed into each bucket. The zebrafish were
sexually mature fish of at least 2 months of age and sup-
plied from Aquatics Unlimited (Greenfield, WI). The
constructed wetlands were placed in a temperature regu-
lated greenhouse room kept at 27 ± 5℃. The temperature,
pH, silver content of the water, and the health of the fish
were monitored daily. The fish were kept in the con-
structed wetland exposed to AgNP for one week. The
zebrafish were fed daily with Zeigler adult zebrafish diet
(Zeigler product #AH271). Institutional approval from
the on-campus Animal Care and Use Committee was re-
ceived before carrying out this research.
At the end of the week remaining live zebrafish were
euthanized and a soil sample taken from the wetland me-
dia column at 3 depths (surface, center, bottom). The
water and soil samples were digested with 6 M nitric acid
and analyzed with atomic absorption spectroscopy using
the method described above in the AgNP sorption ex-
periment section. For the water samples, 2 ml of sample
were digested with 1 ml of 6 M nitric acid. The soil sam-
ples were air dried after which 0.5 g of soil was digested
with 2 ml of 6 M nitric acid and 10 ml of 2 mM citrate
solution before elemental silver analysis on the AA. This
experiment was replicated three times in consecutive
weeks.
3. Results and Discussion
3.1. Physical and Chemical Characteristics of the
Media
Chemical and physical analysis of the media used in both

A. EBELING ET AL. 19
the sorption experiment and lab-scale constructed wet-
land experiment is reported in Table 1. The texture of the
wetland soil, silt loam soil, and WTR was sandy loam,
silt loam, and loamy sand, respectively. The wetland soil
and WTR had very similar sand content (71% and 75%,
respectively), the same silt content (23%), and corre-
spondingly different clay content (6% and 2%, respec-
tively). The wetland soil had the highest carbon content
(332,600 mg/kg) as would be expected. The WTR had
lower carbon content (76,800 mg/kg), and the silt loam
had the lowest (19,850 mg/kg). These media had varying
nutrient contents, e.g. the wetland soil had the highest
nitrogen and phosphorus content (24,320 mg N/kg and
967 mg P/kg) but the lowest magnesium and aluminum
content (2782 mg Mg/kg and 4035 mg Al/kg). The WTR
had aluminum content an order of magnitude higher than
the silt loam (110,532 vs. 15,480 mg Al/kg, respectively)
and two orders of magnitude higher than the wetland soil
(4035 mg Al/kg). This was not surprising because alu-
minum salts are used as coagulants in water treatment.
3.2. Sorption Experiments
The purpose of the sorption experiments was to deter-
mine if the media had the ability to remove AgNPs from
water as well as to investigate which soil particle size
(sand, silt, or clay) AgNPs adsorb too preferentially. The
horizontal axes in Figure 2 are the initial AgNP concen-
trations in the solution shaken with each medium (mg
AgNP/L solution). The vertical axes are the mass of
AgNP adsorbed per mass of medium (mg AgNP/kg me-
dium). Straight lines indicate that the medium still has
the ability adsorb more AgNP; a curve bending to the
right (becoming horizontal) indicates that the medium
Table 1. Selected properties of the three media used in the
sorption and constructed we tland experiments.
Parametera Silt Loam Soil Wetland Soil WTR
C (mg/kg) 19850 332600 76800
N (mg/kg) 1818 24320 4068
P (mg/kg) 389 967 788
Ca (mg/kg) 3459 23954 31010
Mg (mg/kg) 4263 2728 12901
S (mg/kg) 199 11705 1655
Fe (mg/kg) 18830 14504 6614
Al (mg/kg) 15480 4035 110532
Sand (%) 19 71 75
Silt (%) 63 23 23
Clay (%) 18 6 2
Soil Texture Silt Loam Sandy Loam Loamy Sand
aAll minerals are total elemental; C by LECO; N by Kjeldahl; P, Ca, Mg, S,
Fe, and Al by ICP-OES. Sand, silt, and clay percentages were determined by
the hydrometer method. All analyses were completed at the Soil and Plant
Analysis Lab, Madison, WI.
had a diminished capacity to adsorb more AgNP, thus
leaving more in solution. This occurs because the sorp-
tion sites gradually become filled. None of the media in
this experiment show much curve (Figures 2(a), (b), (c)),
indicating that they all have the potential to adsorb more
AgNPs from solutions with concentrations of AgNPs
higher than the highest used in this study (60 mg
AgNP/L). However, both the wetland soil and the WTR
adsorbed a much greater total mass of AgNPs per mass
of media than did the silt loam soil (1250 and 1000 mg
AgNP/kg vs. 220 mg AgNP/kg, respectively). The aver-
age percentage of AgNP adsorbed to all size particles at
the highest initial solution concentration (60 mg AgNP/L)
in the wetland soil, WTR, and silt loam soil (determined
by subtracting the concentration of AgNP in solution
after shaking from the initial concentration of AgNP in
the shaking solution and dividing by the initial concen-
tration) was 100%, 81%, and 18%, respectively (Table
2). This data also shows that the silt loam soil adsorbed
approximately half of the total AgNP from the lowest
initial concentration (49%) and progressively adsorbed a
smaller percentage as the initial AgNP concentration
increased. The WTR showed a similar phenomenon ad-
sorbing 93% of the total AgNP available at lowest ini-
tial concentration decreasing to adsorbing 81% of the
total AgNP in solution at the highest initial concentration.
These results suggest that the wetland soil and the WTR
are able to remove substantial amounts of AgNP from
water.
Looking more closely at the impact of the size of the
media particles in removing nanoparticles, the data show
that all of AgNPs were associated with sand size particles
in the WTR (Figure 2(c)). The data points for each sam-
pling time (30 s, 2 hrs, and after centrifuging) of the
WTR are very similar, indicating that little more AgNPs
were removed with the smaller sized particles. However,
data from the silt loam and wetland soils show that ap-
proximately half of the AgNPs were removed from the
60 mg AgNP solution after 30 s of settling (100 out of
220 mg AgNP/kg and 680 out of 1250 mg AgNP/kg,
respectively) (Figures 2(a) and (b)). A similar effect was
seen at the lower initial AgNP solution concentrations.
This indicates that about half of the AgNP in solution
were adsorbed to sand sized particles and about half of
the AgNP were adsorbed to silt and clay size particles.
The wetland soil and WTR differ from the silt loam
soil in that they both have very similar sand content (>
70% sand) and both have a higher total carbon value
compared to the silt loam soil (Table 1). The sand con-
tent cannot be responsible for the higher amount of
AgNP sorption in the wetland soil and WTR since only
half of the total AgNPs in solution were removed with
sand particles in the wetland soil (Figure 2(b)). However,
the trend correlates well with the increase in carbon con-

A. EBELING ET AL. 20
0
50
100
150
200
250
300
0 20406080
Mass of Ag NP Adsorbed to Mediu m
(mg AgNP/kg Soil)
Initial Nanopart icle Solut ion Concen tra tion (mg AgNP /L)
Silt Loam Soil
AgNP adsorbed to
sand*
AgNP adsorbed to
silt and sand
AgNP adsorbed to
sand, silt, and clay
(a)
0
200
400
600
800
1000
1200
1400
0 20406080
Mass of Ag NP Adsorbed to Medium
(mg AgNP/kg Soil)
Initial Nanop article Soluti on Concentration (mg AgNP/L)
Wetland Soil
AgNP adsorbed to
sand*
AgNP adsorbed to
silt and sand
AgNP adsorbed to
sand, silt, and clay
(b)
0
200
400
600
800
1000
1200
1400
0 20406080
Mass of Ag NP Adsorbed to Medium
(mg AgNP/kg Soil)
Initial Nanoparticle Solut ion Concentration (m g AgNP/L)
Water Treatmen t Residu al
AgNP adsorbed to
sand*
AgNP adsorbed to
silt and sand
AgNP adsorbed to
sand, silt, and clay
(c)
Figure 2. Silver nanoparticle (AgNP, 10 nm monodispersed) adsorption to three media (a) silt loam soil, (b) wetland soil, and
(c) water treatment residual (WTR). Each data point is the average of three trials, with vertical error bars indicating stan-
dard deviation. The three curves on each graph indicate nanoparticles adsorbed to different size classes of soil. “AgNP ad-
sorbed to sand” was measured from the solution concentration after 30 s (approximate time for sand to settle), “AgNP ad-
sorbed to silt and clay” was measured from the solution concentration after 2 hours (approximate time for silt to settle, and
“AgNP adsorbed to all media” was measured from the solution concentration after centrifugation (only dissolved nanoparti-
cles in soluti on).
tent. The wetland soil has the highest carbon content and
showed no reduction in ability to remove AgNP. The
WTR showed a slight reduction and had lower carbon
content, while the silt loam soil had the lowest amount of
carbon and was the least capable of removing AgNPs
(Tables 1 and 2, Figure 2(a)). Recently, researchers
found that dispersion and toxicity of AgNP were de-
pendent on the amount of humic acid present [46]. At
high humic acid concentrations (> 20 mg total organic
carbon/L), significant aggregation of AgNPs was ob-

A. EBELING ET AL. 21
Table 2. Percentage of the total silver nanoparticles (AgNP)
adsorbed by each medium at each initial solution concen-
tration. This was calculated from the average difference
between the initial solution concentration and the final
(equilibrium) solution concentration.
Percentage of AgNP Adsorbed
Initial AgNP Solution
Concentration Silt Loam SoilWetland Soil WTR
% % %
6 mg AgNP/L 49 100 93
15 mg AgNP/L 31 100 90
30 mg AgNP/L 16 100 89
60 mg AgNP/L 18 100 81
served. When studying the relative risk ratios for differ-
ent metallic nanomaterials other researchers found AgNPs
pose a greater environmental risk than either TiO2 or
ZnO nanoparticles, which highlights the importance of
studying their fate in the environment [47]. However,
they also note that their data overestimates the risk to the
terrestrial environment, because in ecotoxicity studies
there is an assumption that metallic silver is present,
when in fact, other studies have shown that often AgNPs
are converted to Ag2S during wastewater treatment [48-
50] and as such are much less soluble and therefore less
toxic. Aggregation size was not measured in the data
reported in this research, nor was the form of Ag after
equilibration. In Figure 2, all of the removal of AgNPs
from the solution is represented as adsorption to the me-
dia. Aggregation of AgNPs and settling from solution is
not distinguished from adsorption, so adsorption amounts
may be inflated. Future research that images the particles
and characterizes the adsorption to the media would help
determine this.
3.3. Lab-Scale Constructed Wetland Experiment
The purpose of the lab-scale constructed wetlands was to
examine the toxicity of AgNPs to zebrafish living in a
pseudo-natural environment rather than an aseptic, un-
natural environment of a petri dish. As the results of the
sorption experiment indicate, other environmental factors
may impact the fate of AgNPs that enter an ecosystem,
potentially rendering them less toxic or simply removed
from the environment.
The results of this experiment are inconclusive but do
shed light on the impact that environmental factors have
on the fate of AgNPs. Fish mortality (Figure 3) only
occurred in the positive control treatment (15 mg/L Ag-
NO3) (where in two out of the three weeks, two of the
five fish did not survive to the end of the seven day ex-
periment) and in the negative control (where one fish
during one of the three weeks did not survive to the end
of the experiment). There was no mortality in any of the
Figure 3. Adult zebrafish survival expressed as a % of fish
surviving after 7 day exposure. Each bar is the average of
three trials of 5 fish (2 male, 3 female) per lab-scale con-
structed wetland. Error bars indicate standard deviation;
Kruskal-Wallis test gave a p-value of 0.134 between treat-
ments.
AgNP treatments. At first, this data made sense in light
of the sorption experiments explained above. The wet-
land soil had been shown to be able to remove AgNPs
from water, so when analyzing the soil of the constructed
wetland after the seven-day exposure, it was expected
that it would contain AgNPs. However, after digesting
the soil at three depth levels, there were no AgNPs
measured at any depth (data not shown). Additionally,
the daily water samples taken from the middle of the
bucket also did not contain AgNPs (data not shown). If
the AgNPs were not in the circulating water or in the soil,
the conclusion was drawn that the dispersant (Tide de-
tergent) may not have been strong enough to keep the
AgNPs dispersed, and the AgNPs may have aggregated
and sunk to the bottom of the bucket and stayed in the
layer of pebbles. Preliminary trials indicated that the 2 ml
Tide/L could keep AgNP in solution, although not as
well as at higher concentrations. However, this level was
also considered the detergent level where fish may ab-
sorb twice the amount of chemicals than they would
normally absorb [51]. Tide was chosen as the dispersing
agent in this study because it not only acted as a dispers-
ant, but also is a likely way for AgNP to enter the envi-
ronment through residential laundry.
Other research has shown that AgNPs aggregate in sa-
line solutions [52], which support this conclusion. There
was no feasible way found to measure the AgNPs in that
layer of pebbles and confirm the conclusion. Recent re-
search is beginning to focus on the fate of nanoparticles
in a more natural setting, such as the study reported here.
Reference [53] found that plants in a system could help
decrease the toxicity of AgNPs because plants release
dissolved organic matter which can bind with Ag ions. In
a freshwater system, researchers found that sediments
accumulated most of the ceria nanoparticles used in their
aquatic system [54]. Additionally other researchers sug-
gest that the chemistry of the nanoparticle capping agent
plays an important role in the fate and transport of

A. EBELING ET AL. 22
AgNPs and environmental factors such as pH, ionic
strength, and electrolyte composition can help predict the
fate and transport of AgNPs [55]. Their data showed that
positively charged branched polyethyleneimine stabilized
AgNPs would most likely have limited mobility in soils,
groundwater, and other environments, but sterically sta-
bilized polyvinylpyrrolidone AgNPs may have the great-
est potential for mobility and transport. Another recent
study indicated that in a sandy loam soil AgNP concen-
trations eight times greater than AgNO3 concentrations
were needed to induce significant reproductive toxicity in
earthworms [56]. A study such as the one reported here,
although it investigates only a small fraction of the ques-
tions still remaining regarding the fate of nanoparticles in
the environment provides valuable new information to
help guide future studies. Some researchers [47] do not
lament the idea that each nanomaterial may react differ-
ently depending on the material and the environmental
properties and conditions, but instead they reinforce the
importance of continuing to study these materials under
many different conditions.
Although the lab-scale constructed wetland experiment
did not lead to conclusive results regarding the fate of
nanoparticles in this environment, it did indicate that the
toxicity levels shown in laboratory conditions are not the
same as in a more natural environment. Thus the need for
similar experiments simulating natural environments is
underscored. Simple, inexpensive experimental designs
such as this can be implemented to investigate parame-
ters that impact AgNP fate and biotoxicity. Simulated
natural environments may prove useful in determining
the mechanism of AgNP toxicity to fish. Is direct expo-
sure to suspended AgNPs more or less toxic than dietary
exposure to algae or zooplankton that has previously
internalized nanoparticles? This question is especially
relevant in light of the recent observation that a wide
variety of living organisms do take AgNPs out of the
water column [57]. Additionally, desorption experiments
investigating how tightly bound AgNPs are to environ-
mental media will help elucidate the effectiveness and
lifetime of various media in a constructed wetland setting.
Preliminary trials have indicated that AgNPs are bound
most tightly to the wetland soil, followed by the silt loam
soil, and least tightly to the WTR (data not shown), but
further research is necessary to investigate this further.
As is suggested by multiple groups of researchers [16,58,
and others], a multidisciplinary approach is crucial to
understanding nanoparticle risks in the environment and
will involve collaborations between chemists, biologists,
toxicologists, ecologists, engineers, and environmental
scientists.
4. Conclusions
The results of the constructed wetland study indicate that
silver nanoparticles appear to aggregate in a salt solution
rendering them less toxic to zebrafish than would be ex-
pected from previous studies using silver nanoparticles
and zebrafish in a pure media. This research did not
measure the size of the nanoparticles after they were
added to the lab-scale constructed wetland so the aggre-
gation of the nanoparticles cannot be known for sure, but
the lack of mortality in the zebrafish and the absence of
silver nanoparticles in the water media and wetland soil
after nanoparticle addition indicates that the particles
most likely sank to the bottom of the buckets. More im-
portantly, the sorption studies provide evidence that soil
and water treatment residuals have the ability to remove
silver nanoparticles from wastewater. This means that
natural and constructed wetlands could be sinks for silver
nanoparticles, removing them from wastewater. Using
water treatment residuals in a constructed wetland would
be a beneficial reuse of a waste product that would other-
wise need to be disposed of. It is important to remember
that after the silver nanoparticles are removed from the
water by sorption to soil or other media, they still remain
in the environment adsorbed to the media. Further studies
are needed to investigate how tightly and for how long
the silver nanoparticles are retained by soil and water
treatment residuals. These desorption studies would shed
light on the long-term sustainability of a wetland de-
signed to remove nanoparticles and the kind of engineer-
ing that would be needed to best manage constructed
wetlands.
5. Acknowledgements
The authors thank the Wisconsin Lutheran College Fac-
ulty Development committee for support of this research.
They also wish to thank Michael Reep, Amanda Wagner,
Krystal Weishaar, and Benjamin Tellier for their contri-
butions to the preliminary experimental designs and test
trials.
REFERENCES
[1] WHO: World Health Organization, “Small-scale Water
Supplies in the pan-European Region,” United Nations
Economic Commission for Europe, 2010.
[2] T. Brick, B. Primrose, R. Chandrasekhar, S. Roy, J. Mu-
liyil, G. Kang, “Water Contamination in Urban South In-
dia: Household Storage Practices and Their Implications
for Water Safety and Enteric Infections,” Journal of Hy-
giene and Environmental Health, Vol. 207, No. 5, 2004,
pp. 473-480.doi:10.1078/1438-4639-00318
[3] R. B. Levin, P. R. Epstein, T. E. Ford, W. H. Harrington,
E. Olson, E. G. Reichard, “U.S. Drinking Water Chal-
lenges in the Twenty-First Century,” Environmental
Health Perspect ives, Vol. 100, 2002, pp. 43-52.
doi:10.1289/ehp.02110s143
[4] A. J. Jowet, “China’s Water Crisis,” The Geographical

A. EBELING ET AL. 23
Journal, Vol. 152, No. 1, 1986, pp. 9-18.
doi:10.2307/632934
[5] P. Vonlanthen, D. Bittner, A. G. Hudson, K. A. Young, R.
Muller, B. Lundsgaard-Hansen, D. Roy, S. Di Piazza, C.
R. Largiader and O. Seehausen, “Eutrophication Causes
Speciation Reversal in Whitefish Adaptive Radiations”,
Nature, Vol. 482, 2012, pp. 357-362.
doi:10.1038/nature10824
[6] S. S. Kaushal, W. M. Lewis Jr., and J. H. McCutehan Jr.,
“Land Use Change and Nitrogen Enrichment of a Rocky
Mountain Watershed,” Ecological Applications, Vol. 16,
No. 1, 2006, pp. 299-312.
doi:10.1890/05-0134
[7] J. Fawell and M. J. Nieuwenhuijsen, “Contaminants in
Drinking Water,” British Medical Bulletin, Vol. 68, No. 1,
2003, pp. 199-208. doi:10.1093/bmb/ldg027
[8] S. N. Levine and D. W. Schindler, “Influence of Nitrogen
to Phosphorus Supply Ratios and Physicochemical Con-
ditions on Cyanobacteria and Phytoplankton Species
Composition in the Experimental Lakes Area, Canada,”
Canadian Journal of Fisheries and Aquatic Sciences, Vol.
56, No. 3, 1999, pp. 451-466.
doi:10.1139/f98-183
[9] D. L. Correll, “The Role of Phosphorus in the Eutrophi-
cation for Receiving Waters: A Review,” Journal of En-
vironmental Quality, Vol. 27, No. 2, 1998, pp. 261-266.
doi:10.2134/jeq1998.00472425002700020004x
[10] T. C. Daniel, A. N. Sharpley, and J. L. Lumunyon, “Ag-
ricultural Phosphorus and Eutrophication: A Symposium
Overview,” Journal of Environmental Quality, Vol. 27,
No. 2, 1998, pp. 251-257.
doi:10.2134/jeq1998.00472425002700020002x
[11] E. G. Srinath and S. C. Pillai, “Phosphorus in Sewage,
Polluted Waters, Sludges, and Effluents,” The Quarterly
Review of Biology, Vol. 41. No. 4, 1966, pp. 384-407.
doi:10.1086/405158
[12] C. N. Sawyer, H. B. Gotaas, and J. B. Lackey, “Factors
Involved in Disposal of Sewage Effluents to Lakes,” Se-
wage and Industrial Wastes, Vol. 26, No. 3, 1954, pp.
317-328.
[13] S. K. Liehr, “Natural Treatment and Onsite Processes,”
Water Environment Research, Vol. 77, No. 6, 2005, pp.
1389-1424. doi:10.2175/106143005X54416
[14] UN, “Waste-Water Treatment Technologies: A General
Review,” Economic and Social Commission for Western
Asia, New York, 2003.
[15] U. Mander and P. D. Jenssen Eds., “Constructed Wet-
lands for Waste Water Treatment in Cold Climates,” WIT
Press, Southampton, 2002.
[16] E. S. Bernhardt, B. P. Colman, M. F. Hochella, Jr., B. J.
Cardinale, R. M. Nisbet, C. J. Richardson, and L. Yin,
“An Ecological Perspective on Nanomaterial Impacts in
the Environment,” Journal of Environmental Quality, Vol.
39, No. 6, 2010, pp. 1-12.
doi:10.2134/jeq2009.0479
[17] A. B. A. Boxall, M. A. Rudd, B. W. Brooks, D. J. Cald-
well, K. Choi, S. Hickmann, E. Innes, K. Ostapyk, J. P.
Staveley, T. Verslycke, G. T. Ankley, K. F. Beazley, S. E.
Benlanger, J. P. berninger, P. Carriquiriborde, A. Coors,
P. DeLeo, S. D. Dyer, J. F. Ericson, J. Gagne, J. P. Biesy,
T. Gouin, L. Hallstrom, M. V. Karlsson and D. G. J.
Larsson, “Pharmaceuticals and Personal Care Products in
the Environment: What Are the Big Questions?” Envi-
ronmental Health Perspectives, Vol. 120, No. 9, 2012, pp.
1221-1229. doi:10.1289/ehp.1104477
[18] S. Rodrigues-Mozaz and H. S. Weinberg, “Meeting Re-
port: Pharmaceuticals in Water–An Interdisciplinary Ap-
proach to a Public Health Challenge,” Environmental
Health Perspectives, Vol. 118, No. 7, 2010, pp.
1016-1020. doi:10.1289/ehp.0901532
[19] J. Fabrega, S. N. Luoma, C. R. Tyler, T. S. Galloway and
J. R. Lead, “Silver Nanoparticles: Behaviour and Effects
in the Aquatic Environment,” Environmental Interna-
tional, Vol. 37, No. 2, 2011, pp.517-531.
doi:10.1016/j.envint.2010.10.012
[20] A. M. Comerton, R. C. Andrews and D. M. Bagley,
“Practical Overview of Analytical Methods for Endo-
crine-Disrupting Compounds, Pharmaceuticals, and Per-
sonal Care Products in Water and Wastewater,” Philiso-
phical Transacations: Mathematical, Physical, and En-
gineering Sciences, Vol. 367, No. 1904, 2009, pp.
3923-3939.
[21] C. G. Daughton and T. A. Ternes, “Pharmaceuticals and
Personal Care Products in the Environment: Agents of
Subtle Change?” Environmental Health Perspectives, Vol.
107, 1999, pp. 907-938.
doi:10.1289/ehp.99107s6907
[22] N. B. Golovina and L. M. Kustov, “Toxicity of Metal
Nanoparticles with a Focus on Silver,” Mendeleev Com-
munications, Vol. 23, No. 2, 2013, pp. 59-65.
doi:10.1016/j.mencom.2013.03.001
[23] C. You, C. Han, X. Wang, Y. Zheng, Q. Li, X. Hu and H.
Sun, “The Progress of Silver Nanoparticles in the Anti-
bacterial Mechanism, Clinical Application, and Cytotox-
icity,” Molecular Biology Reports, Vol. 39, No. 9, 2012,
pp. 9093-9201.
doi:10.1007/s11033-012-1792-8
[24] K. Kulthong, S. Srising, K. Boonpavanitchak, W. Kang-
wansupamonkon and R. Maniratanachote, “Determination
of Silver Nanoparticle Release from Antibacterial Fabrics
into Artificial Sweat,” Particle Fibre Toxicology, Vol. 7,
2010, pp. 8. doi:10.1186/1743-8977-7-8
[25] B. Karn, T. Kuiken and M. Otto, “Nanotechnology and in
Situ Remediation: A Review of the Benefits and Potential
Risks,” Environmental Health Perspectives, Vol. 117, No.
12, 2009, pp. 1823-1831.
[26] M. Yousefian and B. Payam, “Effects of Nanochemical
Particles on Some Histological Parameters of Fish,” Ad-
vances in Environmental Biology, Vol. 6, No. 3, 2012, pp.
1209-1215.
[27] F. Gagne, C. Andre, R. Skirrow, M. Gelinas, J. Auclair, G.
van Aggelen, P. Turcotte and C. Gagnon, “Toxicity of
Silvernanparticles to Rainbow Trout: a Toxicogenomic
Approach,” Chemosphere, Vol. 89, No. 5, 2012, pp.
615-622. doi:10.1016/j.chemosphere.2012.05.063
[28] S. W. Kim and Y. J. An, “Effect of ZnO and TiO2 Nano-
particles Preilluminated with UVA and UVB light on

A. EBELING ET AL. 24
Escherichia coli and Bacillus subtilis,” Applied Microbi-
ology and Biotechnology, Vol. 95, No. 1, 2012, pp.
243-253.doi:10.1007/s00253-012-4153-6
[29] D. A. Cowart, S. M. Guida, S. Ismat and A. G. Marsh,
“Effects of Ag Nanoparticles on Survival and Oxygen
Consumption of Zebrafish Embryos, Danio rerio,” Jour-
nal of Environmental Science and Health, Part A: Tox-
ic/Hazardous Substances and Environmental Engineering,
Vol. 46, No. 10, 2011, pp. 1122-1128.
[30] D. Stampoulis, S. K. Sinha and J. White, “As-
say-Dependent Phytotoxicity of Nanoparticles to Plants,”
Environmental Science and Technology, Vol. 43, No. 24,
2009, pp. 9473-9479.
doi:10.1021/es901695c
[31] R. J. Griffit, J. Luo, J. Gao, J. C. Bonzongo and D. S.
Barber, “Effects of Particle Composition and Species on
Toxicity of Metallic Nanomaterials in Aquatic Organ-
isms,” Nanomaterials in the Environment, Vol. 27, No. 9,
2008, pp. 1972-1978.
doi:10.1016/j.scitotenv.2006.11.007
[32] K. Y. Yoon, J. H. Byeon, J. H. Park and J. H. wang,
“Susceptibility Constants of Escherichia coli and Bacillus
subtilis to Silver and Copper Nanoparticles,” Science of
the Total Environment, Vol. 373, No. 2-3, 2007, pp.
572-575.doi:10.1016/j.scitotenv.2006.11.007
[33] A. S. Barnard and H. Guo, “Nature’s Nanostructures,”
Pan Stanford Publishing, Singapore, 2012.
[34] N. J. Kagengi and A. Thompson, “The Emerging Empha-
sis on Nanometer-Scale Processes in Soil Environments,”
Soil Science Society of America Journal, Vol. 75, No. 2,
2011, pp. 333-334. doi:10.2136/sssaj2011.000npsintro
[35] B. K. G. Theng and G. Yuan, “Nanoparticles in the Soil
Environment,” Elements, Vol. 4, No. 6, 2008, pp.
395-399. doi:10.2113/gselements.4.6.395
[36] J. M. Zook, M. D. Halter, D. Cleveland and S. E. Long,
“Disentangling the Effects of Polymer Coatings on Silver
Nanoparticle Agglomeration, Dissolution, and Toxicity to
Determine Mechanisms of Nanotoxicity,” Journal of
Nanoparticle Research, Vol. 14, 2012, pp. 1165-1572
doi:10.1007/s11051-012-1165-1
[37] K. R. Reddy, E. M. D’Angelo and W. G. Harris, “Bio-
chemistry of Wetlands,” In Handbook of Soil Science,
M.E. Sumner (Ed.), CRC Press, 1999.
[38] J. A. Ippolito, K. A. Barbarick and H. A. Elliott, “Drink-
ing Water Treatment Residuals: A Review of Recent
Uses,” Journal of Environmental Quality, Vol. 40, No.1,
2011, pp. 1-12. doi:10.2134/jeq2010.0242
[39] E. A. Dayton and N. T. Basta, “A Method for Determin-
ing the Phosphorus Sorption Capacity and Amorphous
Aluminum of Aluminum-Based Drinking Water Treat-
ment Residuals,” Journal of Environmental Quality, Vol.
34, No.3, 2005, pp. 1112-1118.doi:10.2134/jeq2004.0230
[40] P. S. Nair, T. J. Logan, A. N. Sharpley, L. E. Sommers,
M. A. Tabatabai and T. L. Yuan, “Interlaboratory Com-
parison of a Standardized Phosphorus Adsorption Proce-
dure,” Journal of Environmental Quality, Vol. 13, No. 4,
1984, pp. 591-595.
doi:10.2134/jeq1984.00472425001300040016x
[41] H. A. Elliot, G. A. O’Connor, P. Lu and S. Brinton, “In-
fluence of Water Treatment Residuals on Phosphorus
Solubility and Leaching,” Journal of Environmental
Quality, Vol. 31, 2002, pp. 1362-0.69.
[42] G. J. Bouyoucos, “Hydrometer Method Improved for
Making Particle Size Analysis of Soils,” Agronomy
Journal, Vol. 54, No. 5, 1962, pp. 464-465.
doi:10.2134/agronj1962.00021962005400050028x
[43] G. W. Gee and J. W. Bauder, “Particle-size Analysis,” In
A. Klute, Ed., Methods of soil analysis. Part 1. 2nd ed.
Agronomy Monograph 9. ASA and SSSA, Madison, WI,
1986, pp. 383-411.
[44] T. M. Benn and P. Westerhoff, “Nanoparticle Silver Re-
leased into Water from Commercially Available Sock
Fabrics,” Environmental Science and Technology, Vol. 42,
No. 11, 2008, pp. 4133-4139.
doi;10.1021/es7032718
[45] EPA. “Method 3050B (SW-846): Acid Digestion of Se-
diments, Sludges, and Soils,” Revision 2, 1996.
http://www.epa.gov/sam/pdfs/EPA-3050b.pdf
[46] J. Gao, K. Powers, Y. Wang, H. Zhou, S.M. Roberts, B.
M. Moudgil, B. Kooopman and D. S. Barber, “Influence
of Suwannee River Humic Acid on Particle Properties
and Toxicity of Silver Nanoparticles,” Chemosphere, Vol.
89, No. 1, 2012 pp. 96-101.
doi:10.1016/j.chemosphere.2012.04.024
[47] F. Gottschalk, E. Kost and B. Nowack, “Engineered Na-
nomaterials in Waters and Soils: A Risk Quantification
Based on Probabilistic Exposure and Effect Modeling,”
Environmental Toxicology and Chemistry, 2013, (In
Press). doi:10.1002/etc.2177
[48] B. C. Reinsch, C. Levard, Z. Li, R. Ma, A. Wise, K. B.
Gregory, G. E. J. Brown and G. V. Lowry, “Sulfidation of
Silver Nanoparticles Decreases Escherichia coli Growth
Inhibition,” Environmental Science and Technology, Vol.
46, No. 13, 2012, pp. 6992-7000.doi:10.1021/es203732x
[49] R. Kaegi, A. Boegelin, B. Sinnet, S. Zuleeg, H. Hagen-
dorfer, M. Berkhardt and H. Siegrist, “Behavior of Metal-
lic Silver Nanoparticles in a Pilot Wastewater Treatment
Plant,” Environmental Science and Technology, Vol. 45,
No. 9, 2011, pp. 3902-3908.
doi:10.1021/es1041892
[50] H. T. Ratte, “Bioaccumulation and Toxicity of Silver
Compounds: A Review,” Environmental Toxicology and
Chemistry, Vol. 18, No. 2, 1999, pp. 89-108.
doi:10.1002/etc.5620180112
[51] Lenntech BV, “Detergents Occurring in Freshwater,”
2012
http://www.lenntech.com/aquatic/detergents.htm.
[52] G. E. Batley, J. K. Kirby, M. J. McLaughlin, “Fate and
Risks of Nanomaterials in Aquatic and Terrestrial Envi-
ronments,” Accounts of Chemical Research, Vol. 46, No.
3, 2013, pp. 854-862.doi:10.1021/ar2003368
[53] J. M. Unrine, B .P. Colman, A. J. Bone, A. P. Gondikas
and C. W. Matson, “Biotic and Abiotic Interactions in
Aquatic Microcosms Determine Fate and Toxicity of Ag
Nanoparticles. Part 1. Aggregation and Dissolution,” En-
vironmental Science and Technology, Vol. 46, No.13,

A. EBELING ET AL. 25
2012, pp. 6915-6924. doi:10.1021/es204682q
[54] P. Zhang, X. He, Y. Ma, K. Lu, Y. Zhao and Z. Zhang,
“Distribution and Bioavailability of Ceria Nanoparticles
in an Aquatic Ecosystem Model,” Chemosphere, Vol. 89,
No. 5, 2012, pp. 530-535.
doi:10.1016/j.chemosphere.2012.05.044
[55] A. M. Badawy, T. P. Luxton, R. G. Silva, K. G. Scheckel,
M. T. Suidan and T. M. Tolaymat, “Impact of Environ-
mental Conditions (pH, Ionic Strength, and Electronlyte
Type) on the Surface Charge and Aggregation of Silver
Nanoparticles Suspensions,” Environmental Science and
Technology, Vol. 44, No. 4, 2010, pp. 1260-1266.
doi:10.1021/es902240k
[56] W. A. Shoults-Wilson, B. C. Reinsch, O. V. Tysusko, P.
M. Bertsch, G. V. Lowry and J. M. Unrine, “Role of Par-
ticle Size and Soil Type in Toxicity of Silver Nanoparti-
cles to Earthworms,” Soil Science Society of America
Journal, Vol. 75, No. 2, 2011, pp. 365-377.
doi:10.2136/sssaj2010.0127nps
[57] D. Cleveland, S. E. Long, P. L. Pennington, E. Cooper, M.
Fulton, G. I. Scott, T. Brewer, J. Davis, E. J. Petersen and
L. Wood, “Pilot Estuarine Mesocosm Study on the Envi-
ronmental Fate of Silver Nanomaterials Leached from
Consumer Products,” Science of the Total Environment,
Vol. 421-422, No. 5, 2012, pp. 267-272.
doi:10.1002/ieam.5630030316
[58] R. D. Handy and B. J. Shaw, “Ecotoxicity of Nanomate-
rials to Fish: Challenges for Ecotoxicity Testing,” Inte-
grated Environmental Assessment and Management, Vol.
3, No. 3, 2007, pp. 458-460.
doi:10.1002/ieam.5630030316