Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil

Environmental Pollution xxx (2018) 1e8

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil* Joanna Makselon a, *, Nina Siebers a, b, Florian Meier c, Harry Vereecken a, Erwin Klumpp a a

Institute Agrosphere (IBG-3), Institute of Bio- and Geosciences, Forschungszentrum Jülich GmbH, Germany Ernst Ruska-Centre (ER-C), Forschungszentrum Jülich GmbH, Germany c Postnova Analytics GmbH, Germany b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 November 2017 Received in revised form 31 January 2018 Accepted 7 February 2018 Available online xxx

Undisturbed outdoor lysimeters containing arable loamy sand soil were used to examine the influence of either heavy rain events (high frequency of high rain intensity), steady rain (continuous rainfall of low rain intensity), and natural rainfall on the transport and retention of surfactant-stabilized silver nanoparticles (AgNP). In addition, the AgNPesoil associations within the Ap horizon were analyzed by means of particle-size fractionation, asymmetrical flow field-flow fractionation coupled with UV/Vis-detection and inductively coupled plasma mass spectrometer (AF4-UV/Vis-ICP-MS), and transmission electron microscopy coupled to an energy-dispersive X-ray (TEM-EDX) analyzer. The results showed that AgNP breakthrough for all rain events was less than 0.1% of the total AgNP mass applied, highlighting that nearly all AgNP were retained in the soil. Heavy rain treatment and natural rainfall revealed enhanced AgNP transport within the Ap horizon, which was attributed to the high pore water flow velocities and to the mobilization of AgNPesoil colloid associations. Particle-size fractionation of the soil revealed that AgNP were present in each size fraction and therefore indicated strong associations between AgNP and soil. In particular, water-dispersible colloids (WDC) in the size range of 0.45e0.1 mm were found to exhibit high potential for AgNP attachment. The AF4-UV/Vis-ICP-MS and TEM-EDX analyses of the WDC fraction confirmed that AgNP were persistent in soil and associated to soil colloids (mainly composed of Al, Fe, Si, and organic matter). These results confirm the particularly important role of soil colloids in the retention and remobilization of AgNP in soil. Furthermore, AF4-UV/Vis-ICP-MS results indicated the presence of single, homo-aggregated, and small AgNP probably due to dissolution. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Silver nanoparticles Transport Retention Rain events Soil colloids

1. Introduction Silver nanoparticles (AgNP) are one of the most widely used engineered nanoparticles due to their antimicrobial activity. Through the rising consumption of products containing AgNP, such as cosmetics and health care products, textiles, and plastics, the release of AgNP into the wider environment is becoming ever more likely and, as such, AgNP can also end up in soils via discharge from industry and waste water treatment plants (Anjum et al., 2013; Batley et al., 2013; McGillicuddy et al., 2017; Sun et al., 2014). In soils, AgNP can end up in soil organisms and plants and have toxicity effects either due to the AgNP themselves or ionic silver

*

This paper has been recommended for acceptance by B. Nowack. * Corresponding author. E-mail address: [email protected] (J. Makselon).

(Langdon et al., 2015; Makama et al., 2016; Pappas et al., 2017). It is therefore important to gain a better understanding of the fate and behavior of AgNP in soil. Numerous batch and column studies point to the fact that the transport and deposition of AgNP is controlled by the multiple chemical and physical soil and soil solution properties (Wang et al., 2014a, 2015, 2016), such as clay and organic matter content (Hoppe et al., 2014; Klitzke et al., 2015; Yang et al., 2014), soil surface heterogeneity (Lin et al., 2011), pH and ionic strength (Benoit et al., 2013; El Badawy et al., 2010; Liang et al., 2013b; Wang et al., 2014b), and soil moisture and water flow velocity (Kumahor et al., 2015; Yecheskel et al., 2016). Indeed, the retention of surfactant-stabilized AgNP in a loamy sand soil was found with decreasing AgNP input concentration and flow velocity, and increasing ionic strength (Braun et al., 2015; Liang et al., 2013b). Furthermore, the surface coating and particle size of AgNP also determine their fate in soil (Adrian et al., 2018; El Badawy et al.,

https://doi.org/10.1016/j.envpol.2018.02.025 0269-7491/© 2018 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

2

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

2010; Whitley et al., 2013). Liang et al. (2013a) reported that free surfactants enhance surfactant-stabilized AgNP transport in sand due to competition between AgNP and free surfactants for the retention site, but in soil the availability of a high number of retention sites and the interacting surface area tends to reduce this competition (Liang et al., 2013b). The fate of AgNP in soil is, in addition to all aforementioned factors, controlled by AgNP transformation processes, such as homoaggregation (El Badawy et al., 2010; Metreveli et al., 2015), heteroaggregation (Liu et al., 2015), sulfidation (Baalousha et al., 2015), oxidation (Lui and Hurt, 2010), and dissolution (Li et al., 2012). Importantly, the interaction and association of nanoparticles with soil colloids are known to influence the fate of nanoparticles (NP) in soil. Soil column remobilization studies showed that released NP, including AgNP and carbon nanotubes (CNT), were associated to soil colloids (Liang et al., 2013b; Zhang et al., 2017). Recently, Zhang et al. (2017) quantified CNT concentrations in different particle-size fractions of the soil and the results emphasized that the soil colloids were particularly important for CNT retention and remobilization. Water-dispersible colloids (WDC), i.e. the fraction of soil colloids that disperse in water, are considered as the fraction of potentially most mobile colloids in soil (Kjaergaard et al., 2004). They are released during the infiltration of rain water (Kjaergaard et al., 2004) and can therefore be expected to be important for the transport and retention of AgNP in soil. Thus, further knowledge about AgNPesoil colloid associations is required to understand the environmental fate of AgNP in soil systems. Outdoor lysimeters and mesocosms are suitable tools for investigating the long-term behavior of NP under environmentally relevant conditions. Lowry et al. (2012) applied PVP-AgNP to freshwater mesocosms and demonstrated that 70% of the injected PVP-AgNP accumulated in the upper layer of the soil and sediments after 18 months. The outdoor lysimeter investigations of the entry pathways of AgNP to soil by sewage sludge application during crop cultivation indicated that almost all AgNP remained in the sewage sludge after 15 months, whereas some Ag uptake was observed in the roots of canola and wheat (Schlich et al., 2017). Emerson et al. (2014) investigated the transport of different coated AgNP in a packed sandy soil intermediate-scale field lysimeter, thus revealing that more than 99% of the applied AgNP were retained in soil after one year due to heteroaggregation e present mostly at the depth where they were applied. There is currently a lack of information about the long-term transport behavior of surfactant-stabilized AgNP in undisturbed soils. The influence of the intensity and type of rain events on NP transport and retention in soils has e as far as we are aware e not yet been studied in great detail. This is significant because under natural conditions, rain events and dry periods alternate and thus influence the mobility of AgNP in soil. In a previous column study (Makselon et al., 2017), we showed that in comparison to continuous flow conditions, flow interruption resulted in an overall reduced mobility of AgNP in soil columns. The enhanced AgNP retention was attributed to the increased ionic strength of the soil solution due to evaporation and an increased irreversible attachment of AgNP to the airewater interface during flow interruption. However, rain events differ in their intensity and frequency, for example heavy rain or steady rain. These differences may, in turn, influence AgNP mobility in soil due to their subsequent effects on infiltration rate, water content, ionic strength, and soil colloid mobilization (Bradford and Kim, 2010; de Jonge et al., 2004; Zhuang et al., 2009). Mobile soil colloids can act as a carrier for AgNP. Colloid-facilitated AgNP transport in soil was indicated by the simultaneous release of silver (Ag), iron (Fe), and aluminum (Al) effluent concentrations from soil columns (Hoppe et al., 2015; Liang et al., 2013b; Makselon et al., 2017) and by the TEM-EDX

measurements (Liang et al., 2013b) of the effluent samples. The objectives of this study were to investigate the transport and long-term behavior of AgNP in undisturbed outdoor lysimeters through: i) comparing environmentally relevant precipitation scenarios such as simulated heavy rain, steady rain and natural rain conditions, and ii) investigating the role of soil colloids and larger soil size fractions with respect to the fate of AgNP in soil. Therefore, the aged AgNP within the upper 30 cm of the Ap horizon were analyzed in different particle-size fractions of the soil. Furthermore, AF4-UV/Vis-ICP-MS and TEM-EDX measurements were conducted to identify and characterize AgNPesoil colloid associations within the WDC fraction. 2. Materials and methods 2.1. AgNP and artificial rainwater The AgNP (Ag, 10.16% w/w) were purchased from AgPURE, rent a scientist GmbH, Germany. They were modified by the manufacturer using a mixture of two stabilizers, 4% w/w each of polyoxyethylene glycerol trioleate and polyoxyethylene (20) sorbitan monolaurate (Tween 20). The amount of free surfactants in the stock suspension was around 5% (Liang et al., 2013a). The AgNP were spherical in shape with a diameter size range of 15e20 nm, as determined by transmission electron microscopy (TEM) (Liang et al., 2013a). Further information on the AgNP studied are given in Liang et al. (2013a) and our previous study (Makselon et al., 2017). For each lysimeter experiment, the AgNP suspension was prepared by diluting the AgNP stock suspension in artificial rainwater (for preparation of the artificial rain water, see Supplementary Information (SI)) to obtain concentrations of 1 g L
1 AgNP. The ionic strength of the artificial rain water was 0.2 mM. The suspension was then sonicated for 15 min in a sonication bath (Sonorex, Bandelin, Germany). Measurements of the hydrodynamic diameter of a 10 mg L
1 concentrated AgNP suspension taken by means of dynamic light scattering (DLS, NanoZetaSizer, Malvern, UK) over a period of 24 h indicated a constant hydrodynamic diameter of 51.1 ± 1.8 nm and ensured that the AgNP were stable in artificial rain water. It should be noted that Batch-DLS is prone to overestimating the hydrodynamic diameter of a nanoparticulate sample, particularly when the sample exhibits high polydispersity. This is due to the fact that even a small number of aggregates or larger particles contribute significantly to the overall scattering intensity, thus masking the presence of smaller sample components as scattering scales with 1/D6 (D: diameter). 2.2. Lysimeter transport experiments under different rain events Stainless steel lysimeters (0.5 m2 surface area, 1.1 m depth) were filled with undisturbed soil monoliths from an agricultural field (Kaldenkirchen- Hülst, North Rhine-Westphalia, Germany). A classification of the soil horizons as well as the physical and chemical properties of the gleyic Cambisol soil are given in the SI, €rster et al., 2008; Pütz and Klimsa, 1991). The voluTable S1 (Fo metric water content inside the soil monoliths was measured using two time-domain reflectrometry (TDR) probes installed at depths of 5 cm and 30 cm, and was obtained by analyzing the TDR waves (Topp et al., 1980). Firstly, the ploughed A (Ap) horizon in the lysimeters was wetted by adding at least 100 L artificial rainwater to avoid a dry top layer. A suspension of 3 L containing a AgNP concentration of 1.11e1.16 g L
1 in artificial rainwater was then applied uniformly by a hand-held sprayer on the lysimeter surface followed by different rain events. Indeed, the resulting target AgNP concentration of 4 mg kg
1 soil was higher than estimates from exposure models (Gottschalk et al., 2015; Sun et al., 2014), but was

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

comparably lower than that used in prior long-term soil studies (Emerson et al., 2014; Lowry et al., 2012; Schlich et al., 2017). The AgNP concentration of 4 mg kg
1 was chosen to ensure that AgNP could be detected by ICP-MS and analyzed using AF4-UV/Vis-ICPMS both in the effluent and the soil after a test duration of around 1 year. The AgNP long-term transport studies were carried out under three different rain events: heavy (Lysheavy rain), steady (Lyssteady rain) and natural rain (Lysnatural rain). Heavy rain events were defined as short rain events of high frequency of high rain intensity, which was simulated in the first lysimeter using an automatic irrigation device with a 1-min irrigation pulse of 3.03 L every 8 h for 29 days. For the simulation of a steady rain event in the second lysimeter, which is defined as a long-lasting continuous rain event of lowintensity rainfall, an additional automatic irrigation device, that forms small rain droplets was used with a continuous irrigation of 6.9 mL min
1 for 23 days. After the heavy and steady phases in both lyimeters variants, the two lysimeters were treated similarly using a hand-held irrigation device every 5 days with 3.5 L artificial rainwater. Both lysimeters were covered with a stainless steel plate to prevent natural rainfall and evaporation. The third lysimeter was open at the top and exposed to natural weather conditions. Information about precipitation was provided by the meteorological station of Forschungszentrum Jülich, Germany, and is shown in Fig. 1. The aim was to reach an irrigation volume comparable to that of the simulated lysimeters, which resulted in a longer experiment duration. After 319 days, a daily irrigation rate of 1.0 L d
1 was achieved and we decided to stop the experiment to prevent large

3

duration differences between the different lysimeter transport experiments. Detailed experimental conditions, such as duration, irrigation and leachate volume, and average water contents, are given in Table 1. The bottom boundary condition of the lysimeters was a seepage face and drainage was by gravity. The leachate outflow occurred when the lower end of the soil profile was satu et al., 2010). rated (Garre The leachate collected in a stainless steel pan at the bottom of the lysimeter was pumped out. At first, a sample of around 100 mL was taken and stored at 4  C before ICP-MS measurements were taken. Finally, the total leachate was pumped out and weighed. At the end of the transport studies, soil samples were taken by a stainless-steel soil corer (2.2 cm diameter, 1 m length) at three different points. The upper 10 cm soil layer was divided into 2 cm sections, the 10e30 cm layer into 5 cm sections, and the 30e100 cm layer into 10 cm sections. All sample sections were weighed to determine the water content at the end of the experiments. The soil samples were freeze-dried, homogenized, and three aliquots of 0.5 g from each section used for microwave-assisted digestion (turboWAVEINERT, mls, Germany) with 5 mL of 65% HNO3. Subsamples of 1 mL leachate were treated with 1 mL of 65% HNO3. Leachate and soil concentrations of AgNP were determined based on Ag concentrations that were measured by ICP-MS (Agilent 7900, Agilent, USA). A fourth lysimeter was used to determine the Ag background concentration per g soil in the soil monoliths as 0.20 ± 0.01 mg g
1 in the 0e30 cm horizon and 0.07 ± 0.01 mg g
1 in the 30e60 cm horizon. At horizon depth of >60 cm, the Ag concentration was measured as < 0.03 ± 0.01 mg g
1.

Fig. 1. Varying precipitation during experiment time of Lysnatural rain from 17th Sep. 2015 to 6th July 2016. Measurements provided by the meteorological station of Forschungszentrum Jülich, Germany. The diamonds represent precipitation per 10 min. Classification of rain intensities at 10 min interval according to Deutscher Wetterdienst (2017) (German weather service): low rain intensity < 0.5 mm; medium rain intensity  0.5e1.7 mm; heavy rain intensity  1.7 mm.

Table 1 Experimental lysimeter conditions. lysimeter

Lysheavy

rain event irrigation design

heavy rain phase 1: pulse (every 8 h), 3.03 L min
1 phase 2: every 5 days 3.5 L 29a/235b/264c 263.5a/158.3b/421.8c

steady rain phase 1: continuous, 0.007 L min
1 phase 2: every 5 days 3.5 L 22.9d/201e/223.9c 227.9d/140e/367.9c

natural natural rainfall

352.2a/120.7b/472.9c 0.29 ± 0.07a/0.19 ± 0.05b

189.6d/142.8e/332.4c 0.29 ± 0.07d/0.22 ± 0.01e

190.2 0.15 ± 0.03

0.14 ± 0.01

0.13 ± 0.02

0.12 ± 0.04

duration [d] irrigation [L 0.5 m-2] leachate [L] average volumetric water contentf weighted average water contentg a b c d e f g

rain

Lyssteady

rain

Lysnatural

rain

294 302.7

Phase 1 of heavy rain. Phase 2 of heavy rain. Total duration, irrigation and leachate volume, respectively. Phase 1 of steady rain. Phase 2 of steady rain. Average volumetric water content in 5 cm and 30 cm depth during experiment obtained from TDR measurements. Average gravimetric water content within 0e100 cm at the end of the experiments determined by weighing the soil samples taken.

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

4

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

2.3. Particle-size fractionation Particle-size fractionation of the upper 30 cm of the Ap of the lysimeter after heavy rain events was conducted according to the method of Sequaris and Lewandowski (2003) to gain a better understanding of AgNP retention in soil. The soil was fractionated in the following particle sizes: 2000e20 mm; 20e2 mm; 2e0.45 mm WDC; and 0.45 mm WDC fraction þ electrolyte phase (EP). The latter was further fractionated into 0.45e0.1 mm WDC and 0.1 mm nanoparticles (NP) þ EP. The AgNP concentrations associated with the different particle-size fractions were determined by ICP-MS measurements and divided by the total AgNP concentration to obtain the percentage of AgNP proportions in particle-size fractions. The particle-size fractionation was replicated three times and exhibited similar results. The detailed procedure of particle-size fractionation is described in the SI. 2.4. Characterization of AgNPesoil colloid associations using AF4UV/Vis-ICP-MS and TEM-EDX measurements In order to characterize and identify AgNPesoil colloid associates, AF4-UV/Vis-ICP-MS measurements of the 0.45 mm WDC þ EP and 0.1 mm NP þ EP fractions of Lysheavy rain were conducted and the concentrations of Ag, Fe, Al, and Silicon (Si) were determined as well as the particle size. A detailed description of AF4-UV/Vis-ICP-MS measurements and the applied AF4-UV/Vis parameters as well as the ICP-MS measurement conditions are given in the SI and Table S2. Particle size determination was performed using a size calibration method with polystyrene latex particles of defined sizes (NanoSphere size standards from Fisher Scientific (USA) with diameters of 21 nm ± 1.5 nm, 60 nm ± 4 nm, 125 nm ± 3 nm, 350 nm ± 6 nm). Sizing AgNP by using polystyrene latex size standards had previously applied by Loeschner et al. (2013). The obtained fractogram and the respective calibration curve, which was used to translate retention times into hydrodynamic diameter (Dh), are available in the SI, Fig. S1. Furthermore, the 0.45 mm WDC þ EP fraction was analyzed by means of TEMEDX measurements. A description of the TEM-EDX measurements is given in the SI. 3. Results and discussion 3.1. Effect of heavy, steady, and natural rain events on AgNP transport in soil AgNP transport increased over time for all rain events (Fig. S1), however the breakthrough of AgNP was very small for all rain events, resulting in a total recovery of AgNP in the lysimeter leachate of 0.06% for Lysheavy rain, 0.02% for Lyssteady rain, and 0.03% for Lysnatural rain, whilst the main portion was retained by the soil. The breakthrough of AgNP under heavy, steady, and natural rain conditions is shown in Fig. S2 as cumulative AgNP mass detected in the leachates plotted against the time after AgNP application on the lysimeter surface. The retention profiles (RPs) are plotted as mg AgNP g
1 soil against the distance from the soil surface (Fig. 2aec). It should be noted that the AgNP concentrations in leachates and soil were based on Ag concentrations determined by ICP-MS and they therefore did not allow a differentiation between AgNP, Ag ions, and AgNP transformation products. The RPs (Fig. 2) indicate that the highest AgNP concentrations were in the first soil layers independent of the type of rain events. This finding is consistent with the transport results of AgNP in unsaturated undisturbed soil columns at lab scale using the same soil and AgNP (Liang et al., 2013b; Makselon et al., 2017), thus demonstrating the strong filtering capacity of the soil. The RP of

Fig. 2. Retention profiles (RPs) of AgNP transport experiments in outdoor lysimeters under (a) heavy rain, (b) steady rain, and (c) natural rain conditions. RPs plotted as mg AgNP g
1 soil against the distance from the soil surface. AgNP concentrations based on Ag concentrations measured by ICP-MS. Triangles, squares, and diamonds represent three different positions for soil core sampling. Error bars represent the standard deviation of three aliquots of each soil section used for acid digestion.

Lysheavy rain indicated some AgNP transport within the upper 35 cm (Fig. 2a), which was not observed under steady rain conditions (Fig. 2b). Indeed, the heavy and steady rain conditions differed in terms of experiment duration and total irrigation volume (Table 1). However, based on the daily irrigation volume, both lysimeter

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

treatments exhibited an irrigation rate of 1.6 L d
1 and an average volumetric water content of 0.29 ± 0.07, indicating that the enhanced AgNP transport might be a result of the high frequency of the high rain intensity treatment causing higher infiltration rates. The leachate volume for Lysheavy rain was higher by a factor of 1.9 (Table 1) than for Lyssteady rain, indicating that under heavy rain conditions, a higher pore water velocity was present than in the Lyssteady rain. Higher pore water flow velocities enhance AgNP transport in soil (Braun et al., 2015; Liang et al., 2013b; Sagee et al., 2012). In addition, the 15% higher total irrigation volume for Lysheavy rain compared to Lyssteady rain could further enhance AgNP transport. The influence of higher infiltration rates and pore water velocities under heavy rain events was especially noticeable in the early increase of the AgNP concentration in the leachate of Lysheavy rain compared to Lyssteady rain (Fig. S2). Although there was no direct experimental evidence of colloid-facilitated AgNP transport in this study, our previous works have indicated this transport mechanism (Makselon et al., 2017). We assume that the transport of soil colloids and thus that of AgNP associated to soil colloids was enhanced by higher pore water flow velocities. Nevertheless, the release of colloid-associated AgNP as a function of flow velocity requires further investigation. Furthermore, we assume that no significant  AgNP transport is caused by preferential pathways, since Garre et al. (2010) reported on a homogenous convective-dispersive transport process and a non-preferential behavior for water movement in the same soil. At natural rain conditions the lysimeter was open at the top, which led to wet and dry cycles and evaporation occurring, and resulted in a factor of 1.9 to 1.3 lower average volumetric water content in the upper 30 cm of the lysimeter (Table 1) compared to the covered Lysheavy rain and Lyssteady rain. In our previous study (Makselon et al., 2017), it was shown that interrupted irrigation results in an increased AgNP attachment in soil due to enhanced evaporation, decreased water content, higher ionic strength of the soil solution, and increased attachment to the airewater interface. Therefore, the enhanced evaporation under natural rain conditions may have led to a reduced AgNP breakthrough. However, the total AgNP mass in the leachate was clearly higher under natural rain conditions than at steady rain conditions (Fig. S2), although the daily irrigation rate was lower (1.0 L d
1) at natural rain conditions and the total leachate volume was lower by a factor of 1.7 (Table 1) compared to Lyssteady rain. Since the rain precipitation during the experiment duration of Lysnatural rain included low, medium, and heavy rain intensities (Fig. 1), the influence of the latter two might possibly facilitate AgNP transport (Fig. 2c). This assumption is supported by the comparable course of RP of Lysnatural rain to that of Lysheavy rain, indicating transport within the upper 35 cm of the soil monolith (Fig. 2c). In Lysheavy rain and Lysnatural rain, the enhanced AgNP transport could be attributed to the remobilization of soil colloids due to high infiltration rates (Zhuang et al., 2009) and due to the mechanical breakdown of soil aggregates by rain drop impact at the soil surface layer (Kjaergaard et al., 2004; Le Bissonnais, 1996). The remobilized soil colloids might contribute to the cotransport of AgNP. Moreover, at natural rain conditions, variations of the ionic strength after wet and dry periods could mobilize soil colloids and soil-associated AgNP, and thus enhance AgNP transport in deeper soil layers (Liang et al., 2013b; Makselon et al., 2017). It is worth mentioning that the effect of surfactants on soil colloid mobilization was not studied in this study and requires further investigation. The AgNP concentrations in the topmost soil layer at natural rain conditions were twofold lower than those of Lysheavy rain and Lysstaedy rain. Heavy rain events lead to ponding (Chu, 1978), which was observed in the case of Lysheavy rain. Furthermore, at a high frequency of high rain intensity the erosivity of the rain drops could be

5

greater than the resistance of the soil surface, resulting in a silting of the topmost soil layer (Sauerborn, 1993). Both, ponding and silting of the soil surface could lead to higher AgNP concentrations in the topmost soil layer of Lysheavy rain compared to Lysnatural rain. The reduced AgNP transport at a depth >35 cm (Fig. 2) is a result of soil management. The Ap horizon was ploughed and thus exhibited a lower bulk density of the Ap (1.44e1.5 g cm
3) compared to the underlying Bw1 horizon (1.68 g cm
3). At the soil depth >35e100 cm, the average AgNP concentrations of the three sampling positions were <0.28 ± 0.34 mg g
1 for Lysheavy rain, <0.30 ± 0.36 mg g
1 for Lyssteady rain and <0.27 ± 0.38 mg g
1 for Lysnatural rain, thus indicating strongly reduced AgNP transport in the deeper soil layers. The high AgNP retention and very low transport of AgNP in soil can be attributed to several retention mechanisms. Liang et al. (2013b) described the time and depthdependent filling of retention sites of the solid phase by AgNP contributing to AgNP retention. AgNP retention might also be attributed to an attachment of AgNP at the airewater interface (Makselon et al., 2017), especially in the topmost soil layer, as indicated by the RP. Furthermore, the AgNP might be retained by heteroaggregation with soil colloids (Hoppe et al., 2015). Hoppe et al. (2015) incubated a sandy Cambisol farmland soil with the same AgNP product used in our study and found a high retention of AgNP in saturated soil columns after subsequent percolation with artificial rainwater. The retention increased with the contact time between AgNP and the soil from 87% after 3 days to 96% after 92 days of incubation (Hoppe et al., 2015). This supports the high AgNP retention in our lysimeter studies with a two to threefold longer contact time. High AgNP retention was also shown in artificially filled lysimeters exposed to natural weather conditions (Emerson et al., 2014; Schlich et al., 2017). Our study highlights the high overall AgNP retention in an undisturbed arable loamy sandy soil under varying rain frequencies and intensities. 3.2. AgNP in different particle-size fractions of the soil treated with heavy rain Particle-size fractionation of the upper 30 cm of Lysheavy rain revealed that about 94.1% of the particles were in the range of 2000e20 mm, 4.2% in the 20e2 mm range, 1.2% WDC in the 2e0.45 mm range, 0.2% WDC in the 0.45e0.1 mm range, and 0.3% NP þ EP below 0.1 mm (Fig. 3a). A standard deviation of 0.1% for each particle-size fraction indicated good reproducibility for the soil fractionation method used. Fig. 3b presents a plot of percentages of AgNP proportions in size fractions with a total recovery of AgNP of 97.4 ± 4.5%. AgNP were present in each soil fraction, with the highest percentages of AgNP proportion amounting to 55.9% for the 2000e20 mm fraction, followed by 28.5% for the 20e2 mm fraction, 9.1% for the WDC fraction in the range of 2e0.45 mm, 3.4% in the 0.45e0.1 mm WDC fraction, and 0.5% in the 0.1 mm NP þ EP fraction. The very low AgNP proportion in the 0.1 mm NP þ EP fraction even after a 6 h shaking procedure at 150 rpm, thus indicated strong AgNP association to the different particle-size fractions. Strong AgNPesoil associations were also observed in the remobilization study of Hoppe et al. (2015), where, under favorable transport conditions such as a high water saturation level and low IS (Liang et al., 2013b; Yecheskel et al., 2016), the remobilization of aged AgNP was very low. Fig. 3c shows the AgNP concentrations in g
1 of size fraction and indicates that with the increase of the specific surface area of the decreasing particle-size fractions, more retention sites for AgNP were available. The highest AgNP concentration of 750 mg g
1 was found in the smallest WDC particle-size fraction of 0.45e0.1 mm, which accounted for only 0.2% of the soil mass (Fig. 3a and c) and

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

6

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

Fig. 4. AF4-UV/Vis-ICP-MS fractogram of (a) the WDC  0.45 mm þ EP fraction and (b) the 0.1 mm þ EP fraction. ICP-MS data of Ag (black line), Al (blue line), Fe (red line), and Si (purple line) as well as Vis detector signal (green line) are shown on the y-axes. Vis detector signal recorded at 420 nm was scaled up by a factor of (a) 100 and (b) 10, respectively. X-axis shows the hydrodynamic diameter (Dh) at logarithmic scale. Dashed lines enclose the calibrated size area in the range of 21 nme350 nm. WDC: water-dispersible colloids. EP: electrolyte phase. NP: nanoparticles. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3. (a) Mass percentage of size fractions of Lysheavy rain: 2000e20 mm, 20e2 mm, 2e0.45 mm WDC and 0.45 mm WDC þ EP.  0.45 mm WDC þ EP further fractionated in 0.45e0.1 mm WDC and 0.1 mm NP þ EP represented by brackets. (b) Proportion percentage of AgNP in particle-size fraction. (c) mg AgNP g
1 size fraction. AgNP concentrations based on Ag concentrations measured by ICP-MS. Error bars represent the standard deviation of three replicates of particle-size fractionation. WDC: water dispersible colloids. EP: electrolyte phase. NP: nanoparticles.

revealed the potential of WDC for AgNP attachment. The results of the particle-size fractionation for Lysheavy rain highlight that AgNP were associated with soil particles and that these associations were formed even under a high frequency of high rain intensity. The strong association of AgNP with soil particles plays an important role in the transport and deposition behavior of AgNP. In particular, the AgNP associated with fine WDC could be transported in soil due to the infiltration of rain water. 3.3. Characterization of the AgNPesoil colloid associations of the soil treated with heavy rain The AgNPesoil colloid associations were characterized by AF4UV/Vis-ICP-MS within the WDC 0.45 mm þ EP fraction (Fig. 4a) and 0.1 mm NP þ EP fraction (Fig. 4b). The WDC 0.45 mm þ EP fraction was further analyzed by TEM-EDX (Fig. 5). In both particlesize fractions of the soil, three eluted fractions including Ag were identified.

The first fraction with a size range of 4e12 nm and a fraction peak of 8 nm (Fig. 4a and b) may contain small AgNP as a result of AgNP dissolution (Cornelis et al., 2012; Whitley et al., 2013) or Ag ions associated with natural soil NP due to the capability of natural NP to adsorb trace metals (Hartland et al., 2013; Jacobson et al., 2005). Ag ions released from AgNP are readily adsorbed to soil compounds and, therefore, the transport of Ag ions was assumed to be rather low (Hoppe et al., 2014; VandeVoort and Arai, 2012). The Al and Fe concentrations detected within the first fraction indicate soil nanoparticles such as Fe- and Al-(hydr)oxides (Jiang et al., 2015a, 2015b), with which AgNP and/or Ag ions were associated. The second size fraction of 12e35 nm (Fig. 4a and b) exhibited elevated Ag concentrations compared to Al and Fe, and thus could represent single AgNP and/or homo-aggregated AgNP. The first and second Ag-related fractions showed no Vis signal. This might be due to the AgNP concentration being too low for Vis detection. The Vis signal at 3 nme6 nm suggests nanoparticulate natural organic matter (NOM) (Missong et al., 2017) consisting of humic substances which are detectable in the Vis range (Tan, 1997). The third fraction was in the size range of 35 nme300 nm with a fraction peak of 140 nm for the WDC 0.45 mm þ EP sample (Fig. 4a) and in the size range of 35e200 nm with a fraction peak of 85 nm for the 0.1 mm NP þ EP sample (Fig. 4b). In both particle size fractions, the ICP-MS signals of Ag, Al, Fe and Si, as well as the Vis signal recorded at the same elution time suggest that AgNP were associated with fine-sized organic matter containing colloids including soil nanoparticles (<100 nm) (Jiang et al., 2015a; Missong et al., 2017). The elementary composition of the third fraction points to soil components such as Fe- and Al-(hydr)oxides and phyllosilicates (Missong et al., 2017; Tsao et al., 2011).

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

7

shape or their interactions with the membrane. A direct TEM-EDX identification of AgNPesoil colloid associations in the 0.45 mm WDC þ EP fraction (Fig. 5) also indicated the persistence of the surfactant-stabilized AgNP in the soil studied. The elemental composition revealed by EDX spectroscopy confirmed the black dots on soil colloids to be AgNP (Fig. 5b). The particle sizes of the associated AgNP were around 7e23 nm. AgNP <15 nm could be indicative of a dissolution. It is worth noting that AgNP were attached as single AgNP to the soil colloids and that attached AgNP homoaggregates were not observed. The elemental composition of the colloidal fraction revealed by EDX spectroscopy was mainly Si, Al, and Fe (Fig. 5b), which is in agreement with the AF4-UV/Vis-ICP-MS results (Fig. 4). Furthermore, the large signals of Si and Al imply the presence of phyllosilicates acting as an AgNP collector, thus confirming the results of Liang et al. (2013b) and Makselon et al. (2017) based on release experiments. The potential of Fe and clay minerals to retain AgNP was also demonstrated in the studies of Hoppe et al. (2016) by batch experiments of surfactantand citrate-stabilized AgNP with various soil minerals. Batch experiments with PVP-AgNP and different soils revealed a positive correlation between the retention of PVP-AgNP and clay-sized minerals, suggesting AgNP attachment to Fe- and Al- oxides, and the edges of clay minerals (Cornelis et al., 2012). The results of AF4UV/Vis-ICP-MS measurements and TEM-EDX spectroscopy confirmed the formation of AgNPesoil colloid associations within the lysimeter under heavy rain events and confirm the particularly important role of soil colloids in the retention and remobilization of AgNP in soil. 4. Conclusion

Fig. 5. Identification of AgNPesoil colloid associations within 0.45 mm WDC þEP fraction of Lysheavy rain by TEM (a, c) and EDX (b). Black dots were confirmed as AgNP by EDX measurements (b) in the selected area of the circle (a). Arrows indicate attached AgNP on soil colloid (c).

Phyllosilicates such as illite, montmorillonite and kaolinite occur in the used soil (Liang et al., 2013b). Furthermore, the concentrations of Ag, Al, Fe, and Si as well as the Vis signal of the third fraction were higher compared to the first and second fractions, thus indicating that these AgNPesoil associations were dominant in the WDC 0.45 mm þ EP and 0.1 mm NP þ EP, respectively. The concentration of Ag, Al, Fe, and Si in the third fraction were higher for WDC 0.45 mm þ EP (Fig. 4a) than for  0.1 mm NP þ EP (Fig. 4b), which is in agreement with the mass percentage of particle-size fraction of the soil and the percentage of AgNP in the size fractions in Fig. 3a and b. The larger size particles (>100 nm) in Fig. 4b indicated aggregates or particles which eluated later due to the non-spherical

Overall, our outdoor lysimeter studies indicated a high retention of surfactant-stabilized AgNP in the arable loamy sand soil studied, thus confirming the potential of AgNP to enrich in the soil, particularly in the upper soil layers. Rain events, differing in terms of the intensity and frequency of rainfall affected the fate of AgNP in soil. Rain events of high frequency of high rain intensity enhanced AgNP transport in the Ap horizon due to high pore water flow velocities and/or the mobilization of AgNPesoil colloid associations. Our results elucidate the impact of rain events on the behavior of AgNP in soil, which is of importance for exposure modeling approaches to predict AgNP concentrations in soil. Aged AgNP associated to soil colloids such as Fe- and Al-(hydr)oxides and phyllosilicates contributed to AgNP retention in soil, indicating the importance of AgNPesoil colloid associations for AgNP behavior in soil. AgNP appeared to be persistent in soil and thus long-term effects of AgNP on soil fauna and flora can be expected. The combined use of AF4ICP-MS and TEM-EDX enables the characterization of AgNPesoil colloid associations and the dissolution of aged AgNP. Acknowledgements The study was funded by the NanoMobil project, which is supported by the German Federal Ministry of Education and Research (BMBF) (03X0151B). We gratefully acknowledge Dr. Axel Knaps for providing the precipitation data. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.envpol.2018.02.025. References Adrian, Y.F., Schneidewind, U., Bradford, S.A., Simunek, J., Fernandez-Steeger, T.M.,

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025

8

J. Makselon et al. / Environmental Pollution xxx (2018) 1e8

Azzam, R., 2018. Transport and retention of surfactant- and polymer-stabilized engineered silver nanoparticles in silicate-dominated aquifer material. Environ. Pollut. 236C, 195e207. Anjum, N.A., Gill, S.S., Duarte, A.C., Pereira, E., Ahmad, I., 2013. Silver nanoparticles in soileplant systems. J. Nano Res. 15, 1e26. Baalousha, M., Arkill, K.P., Romer, I., Palmer, R.E., Lead, J.R., 2015. Transformations of citrate and Tween coated silver nanoparticles reacted with Na(2)S. Sci. Total Environ. 502, 344e353. Batley, G.E., Kirby, J.K., McLaughlin, M.J., 2013. Fate and risks of nanomaterials in aquatic and terrestrial environments. Acc. Chem. Res. 46, 854e862. , S., 2013. Partitioning of silver and chemical Benoit, R., Wilkinson, K.J., Sauve speciation of free Ag in soils amended with nanoparticles. Chem. Cent. J. 7, 1e7. Bradford, S.A., Kim, H., 2010. Implications of cation exchange on clay release and colloid-facilitated transport in porous media. J. Environ. Qual. 39, 2040e2046. Braun, A., Klumpp, E., Azzam, R., Neukum, C., 2015. Transport and deposition of stabilized engineered silver nanoparticles in water saturated loamy sand and silty loam. Sci. Total Environ. 535, 102e112. Chu, S.T., 1978. Infiltration during an unsteady rain. Water Resour. Res. 14, 461e466. Cornelis, G., Thomas, C.D.M., McLaughlin, M.J., Kirby, J.K., Beak, D.G., Chittleborough, D., 2012. Retention and dissolution of engineered silver nanoparticles in natural soils. Soil Sci. Soc. Am. J. 76, 891e902. de Jonge, L.W., Kjaergaard, C., Moldrup, P., 2004. Colloids and colloid-facilitated transport of contaminants in soils: an introduction. Vadose Zone J. 3, 321e325. Deutscher Wetterdienst, 2017. https://www.dwd.de/DE/service/lexikon/Functions/ glossar.html?lv2¼101812&lv3¼101906 acesss date 15/08/2017. El Badawy, A.M., Luxton, T.P., Silva, R.G., Scheckel, K.G., Suidan, M.T., Tolaymat, T.M., 2010. Impact of environmental conditions (pH, Ionic Strength, and Electrolyte Type) on the surface charge and aggragation of silver nanoparticles suspensions. Environ. Sci. Technol. 44, 1260e1266. Emerson, H.P., Hart, A.E., Baldwin, J.A., Waterhouse, T.C., Kitchens, C.L., Mefford, O.T., Powell, B.A., 2014. Physical transformations of iron oxide and silver nanoparticles from an intermediate scale field transport study. J. Nano Res. 16, 1e14. €rster, M., Laabs, V., Lamsho € ft, M., Pütz, T., Amelung, W., 2008. Analysis of aged Fo sulfadiazine residues in soils using microwave extraction and liquid chromatography tandem mass spectrometry. Anal. Bioanal. Chem. 391, 1029e1038. , S., Koestel, J., Günther, T., Javaux, M., Vanderborght, J., Vereecken, H., 2010. Garre Comparison of heterogeneous transport processes observed with electrical resistivity tomography in two soils. Vadose Zone J. 9, 336e349. Gottschalk, F., Lassen, C., Kjoelholt, J., Christensen, F., Nowack, B., 2015. Modeling flows and concentrations of nine engineered nanomaterials in the Danish environment. Int. J. Environ. Res. Publ. Health 12, 5581e5602. Hartland, A., Lead, J.R., Slaveykova, V.I., O'Carroll, D., Valsami-Jones, E., 2013. The environmental significance of natural nanoparticles. Nat. Edu. Knowl. 4, 1e7. Hoppe, M., Mikutta, R., Kaufhold, S., Utermann, J., Duijnisveld, W., Wargenau, E., Fries, E., Guggenberger, G., 2016. Retention of sterically and electrosterically stabilized silver nanoparticles by soil minerals. Eur. J. Soil Sci. 67, 573e582. Hoppe, M., Mikutta, R., Utermann, J., Duijnisveld, W., Guggenberger, G., 2014. Retention of sterically and electrosterically stabilized silver nanoparticles in soils. Environ. Sci. Technol. 48, 12628e12635. Hoppe, M., Mikutta, R., Utermann, J., Duijnisveld, W., Kaufhold, S., Stange, C.F., Guggenberger, G., 2015. Remobilization of sterically stabilized silver nanoparticles from farmland soils determined by column leaching. Eur. J. Soil Sci. 66, 898e909. Jacobson, A.R., McBride, M.B., Baveye, P., Steenhuis, T.S., 2005. Environmental factors determining the trace-level sorption of silver and thallium to soils. Sci. Total Environ. 345, 191e205. Jiang, X., Bol, R., Nischwitz, V., Siebers, N., Willbold, S., Vereecken, H., Amelung, W., Klumpp, E., 2015a. Phosphorus containing water dispersible nanoparticles in Arable soil. J. Environ. Qual. 44, 1772e1781. Jiang, X., Bol, R., Willbold, S., Vereecken, H., Klumpp, E., 2015b. Speciation and distribution of P associated with Fe and Al oxides in aggregate-sized fraction of an arable soil. Biogeosciences 12, 6443e6452. Kjaergaard, C., De Jonge, L.W., Moldrup, P., Schjønning, P., 2004. Water-dispersiblecolloids: effects of measurements method, clay content, initial soil matric potential, and wetting rate. Vadose Zone J. 3, 403e412. Klitzke, S., Metreveli, G., Peters, A., Schaumann, G.E., Lang, F., 2015. The fate of silver nanoparticles in soil solution d sorption of solutes and aggregation. Sci. Total Environ. 535, 54e60. Kumahor, S.K., Hron, P., Metreveli, G., Schaumann, G.E., Vogel, H.J., 2015. Transport of citrate-coated silver nanoparticles in unsaturated sand. Sci. Total Environ. 535, 113e121. Langdon, K.A., McLaughlin, M.J., Kirby, J.K., Merrington, G., 2015. Influence of soil properties and soil leaching on the toxicity of ionic silver to plants. Environ. Toxicol. Chem. 34, 2503e2512. Le Bissonnais, Y., 1996. Aggregate stability and assessment of soil crustability and erodibility: I. theory and methodology. Eur. J. Soil Sci. 47, 425e437. Li, X., Lenhart, J.J., Walker, H.W., 2012. Aggregation kinetics and dissolution of coated silver nanoparticles. Langmuir 28, 1095e1104. Liang, Y., Bradford, S.A., Simunek, J., Heggen, M., Vereecken, H., Klumpp, E., 2013b. Retention and remobilization of stabilized silver nanoparticles in an undisturbed loamy sand soil. Environ. Sci. Technol. 47, 12229e12237. Liang, Y., Bradford, S.A., Simunek, J., Vereecken, H., Klumpp, E., 2013a. Sensitivity of the transport and retention of stabilized silver nanoparticles to physicochemical factors. Water Res. 47, 2572e2582. Lin, S., Cheng, Y., Bobcombe, Y., Jones, K.L., Liu, J., Wiesner, M.R., 2011. Deposition of

silver nanoparticles in geochemically heterogeneous porous media: predicting affinity from surface composition analysis. Environ. Sci. Technol. 45, 5209e5215. Liu, J., Hwang, Y.S., Lenhart, J.J., 2015. Heteroaggregation of bare silver nanoparticles with clay minerals. Environ. Sci. Nano. 2, 528e540. Loeschner, K., Navratilova, J., Legros, S., Wagner, S., Grombe, R., Snell, J., Von der Kammer, F., Larsen, E.H., 2013. Optimization and evaluation of asymmetric flow field-flow fractionation of silver nanoparticles. J. Chrom. A. 1272, 116e125. Lowry, G.V., Espinasse, B.P., Badireddy, A.R., Richardson, C.J., Reinsch, B.C., Bryant, L.D., Bone, A.J., Deonarine, A., Chae, S., Therezien, M., Colman, B.P., HsuKim, H., Bernhardt, E.S., Matson, C.W., Wiesner, M.R., 2012. Long-term transformation and fate of manufactured Ag nanoparticles in a simulated large scale freshwater emergent wetland. Environ. Sci. Technol. 46, 7027e7036. Lui, J., Hurt, R.H., 2010. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environ. Sci. Technol. 44, 2169e2175. Makama, S., Piella, J., Undas, A., Dimmers, W.J., Peters, R., Puntes, V.F., van den Brink, N.W., 2016. Properties of silver nanoparticles influencing their uptake in and toxicity to the earthworm Lumbricus rubellus following exposure in soil. Environ. Pollut. 218, 870e878. Makselon, J., Zhou, D., Engelhardt, I., Jacques, D., Klumpp, E., 2017. Experimental and numerical investigations of silver nanoparticle transport under variable flow and ionic strength in soil. Environ. Sci. Technol. 51, 2096e2104. McGillicuddy, E., Murray, I., Kavanagh, S., Morrison, L., Fogarty, A., Cormican, M., Dockery, P., Prendergast, M., Rowan, N., Morris, D., 2017. Silver nanoparticles in the environment: sources, detection and ecotoxicology. Sci. Total Environ. 575, 231e246. Metreveli, G., Philippe, A., Schaumann, G.E., 2015. Disaggregation of silver nanoparticle homoaggregates in a river water matrix. Sci. Total Environ. 535, 35e44. Missong, A., Nischwitz, V., Bol, R., Krüger, J., Lang, F., Siemens, J., Klumpp, E., 2017. Phosphorus in water dispersible-colloids of forest soil profiles. Plant Soil 1e16. https://doi.org/10.1007/s11104-017-3430-7. Pappas, S.A., Turaga, U., Kumar, N., Ramkumar, S., Kendall, R.J., 2017. Uptake of silver from polyvinylpyrrolidine coated silver nanoparticles in a terrestrial system. Int. J. Environ. Agric. Res. 3, 104e114. Pütz, T., Klimsa, K., 1991. Profilbeschreibung und Analysenergebnisse des Standortes Kaldenkirchen-Hülst. Internal Report. Forschungszentrum Jülich. Institut für Radioagronomie, IRA, 6/91. Sagee, O., Dror, I., Berkowitz, B., 2012. Transport of silver nanoparticles (AgNPs) in soil. Chemosphere 88, 670e675. €t der Niederschl€ Sauerborn, P., 1993. Die Erosivita age in Deutschland e Ein Beitrag zur quantitativen Prognose der Bodenerosion durch Wasser in Mitteleuropa. Bonner Bodenkundliche Abhandlungen, 13. Schlich, K., Hoppe, M., Kraas, M., Fries, E., Hund-Rinke, K., 2017. Ecotoxicity and fate of a silver nanomaterial in an outdoor lysimeter study. Ecotoxicology 26, 738e751. Sequaris, J.M., Lewandowski, H., 2003. Physicochemical characterization of potential colloids from agricultural topsoils. Colloids Surf., A. 217, 93e99. Sun, T.Y., Gottschalk, F., Hungerbühler, K., Nowack, B., 2014. Comprehensive probabilistic modelling of environmental emissions of engineered nanomaterials. Environ. Pollut. 185, 69e76. Tan, K.H., 1997. Humic Matter in Soil and the Environment. Marcel Dekker, Inc. New York, Basel, University of Georgia, Athens, Georgia U.S.A. Topp, G.C., Davis, J.L., Annan, A.P., 1980. Electromagnetic determination of soil water content: measurements in coaxial transmission lines. Water Resour. Res. 16, 574e582. Tsao, T.M., Chen, Y.M., Wang, M.K., 2011. Origin, separation and identification of environmental nanoparticles: a review. J. Environ. Monit. 13, 1156e1163. VandeVoort, A.R., Arai, Y., 2012. Environmental chemistry of silver in soils: current and historic perspective. Adv. Agron. 114, 59e90. Wang, D., Ge, L., He, J., Zhang, W., Jaisi, D.P., Zhou, D., 2014a. Hyperexponential and nonmonotonic retention of polyvinylpyrrolidone-coated silver nanoparticles in an Ultisol. J. Contam. Hydrol. 164, 35e48. Wang, D., Jaisi, D.P., Yan, J., Jin, Y., Zhou, D., 2015. Transport and retention of polyvinylpyrrolidone-coated silver nanoparticles in natural soils. Vadose Zone J. 14, 1e13. Wang, M., Gao, B., Tang, D., 2016. Review of key factors controlling engineered nanoparticle transport in porous media. J. Hazard Mater. 318, 233e246.   Wang, Y., Bradford, S.A., Sim unek, J., 2014b. Physicochemical factors influencing the preferential transport of in soils. Vadose Zone J. 13, 1e10. Whitley, A.R., Levard, C., Oostveen, E., Bertsch, P.M., Matocha, C.J., von der Kammer, F., Unrine, J.M., 2013. Behavior of Ag nanoparticles in soil: effects of particle surface coating, aging and sewage sludge amendment. Environ. Pollut. 182, 141e149. Yang, X., Lin, S., Wiesner, M.R., 2014. Influence of natural organic matter on transport and retention of polymer coated silver nanoparticles in porous media. J. Hazard Mater. 264, 161e168. Yecheskel, Y., Dror, I., Berkowitz, B., 2016. Transport of engineered nanoparticles in partially saturated sand columns. J. Hazard Mater. 311, 254e262. Zhang, M., Bradford, S.A., Simunek, J., Vereecken, H., Klumpp, E., 2017. Roles of cation valance and exchange on the retention and colloid-facilitated transport of functionalized multi-walled carbon nanotubes in a natural soil. Water Res. 109, 358e366. Zhuang, J., Tyner, J.S., Perfect, E., 2009. Colloid transport and remobilization in porous media during infiltration and drainage. J. Hydrol 377, 112e119.

Please cite this article in press as: Makselon, J., et al., Role of rain intensity and soil colloids in the retention of surfactant-stabilized silver nanoparticles in soil, Environmental Pollution (2018), https://doi.org/10.1016/j.envpol.2018.02.025