Influence of Humic Acid on the Transport of Two Types of Synthesized Zinc Oxide Nanoparticles in Quartz Sand

In this study, transport and retention behaviors of the two types of nZnO prepared with separate manufacturing methods were compared/analyzed according to the presence/absence of Suwannee River humic acid (SRHA) adsorbed into the sand surface and the SRHA suspended in bulk solution, and to changes in the solution ionic strength (0.1–10 mM) in sand-repacked water-saturated columns. In the absence of suspended SRHA, nZnO-1 breakthrough was observed only in SRHA-coated soil, and the breakthrough amount decreased with the increase in the ionic strength (23.8% to 17.2% at 0.1 mM to 10 mM, respectively). In contrast, nZnO-2 breakthrough was not observed over the entire ionic strength range, regardless of the SRHA sand coating. With the presence of suspended SRHA, neither nZnO-1 nor nZnO-2 showed a significant difference in the breakthrough amount regardless of sand coating or ionic strength. However, the breakthrough amount of nZnO-1 was higher than that of nZnO-2 (51.5% versus 37.7% at 10 mM with 1 mg/L SRHA). From confirming the difference in transport between the two types of nZnO, the amount of SRHA adsorbed into nZnO-1 was less than the amount adsorbed into nZnO-2 (0.29 mg/g versus 0.64 mg/g at 10 mM with 1 mg/L SRHA). This result was considered to be due to the larger nZnO-1 breakthrough amount than the nZnO-2 breakthrough amount, which was caused by the larger amount of suspended SRHA that could occupy the deposition sites in the nZnO-1 suspension. Because the rate of SRHA deposition on the sand surface was higher than on nZnO, nZnO transport improved in the presence of suspended SRHA, and changed according to the amount of suspended SRHA.


Introduction
Metal oxide nanomaterials are widely used in industrial settings [1][2][3][4]. In particular, zinc oxide nanoparticles (ZnO-NPs) are used widely in the fields of cosmetics, optics, and the chemical industry [5,6] due to their multifunctional properties, and also for nanocomposite development [7]. Adverse effects of ZnO-NPs on living organisms have been reported frequently, e.g., [8][9][10][11][12][13][14][15]. For example, the toxicity of ZnO-NPs in plants [10,11], bacteria [8,16], and rodents [17] has been reported. The environment has been exposed more and more to ZnO-NPs with the increase in the demand for and production of ZnO-NPs. For detailed and accurate analyses on the risk of ZnO-NPs, their exposure probability must be evaluated, and studies on their fate and transport must be conducted [1].
Studies on the fate and transport of ZnO-NPs have been conducted in various soil environment conditions, such as flow rate, ionic strength, pH, ion valence, and type of porous media, which have been shown to be important factors in nanoparticle transport behavior, e.g., [4,[18][19][20][21][22][23][24]. Humic acid has been reported to improve the ZnO-NPs transport level in the soil environment [14,16], and one of the main mechanisms causing such trend was confirmed as deposition site competition between natural organic matter (NOM) and nanoparticles [25]. These studies on ZnO-NPs transport have targeted single particles. Although variously synthesized particles were used in previous studies on ZnO-NPs fate and transport, comparisons among their outcomes may not be appropriate due to the differences in the soil environment conditions, such as flow rate, ionic strength, pH, ion valence, and porous media. Although ZnO-NPs are known to be synthesized by various methods [26,27], no comparative studies on the fate and transport behavior of ZnO-NPs synthesized with diverse methods have been reported so far. In our recent comparative study [28], ZnO-NPs synthesis methods were confirmed to affect the physicochemical properties (particle size, surface functional groups, specific surface area, and particle shape) of ZnO-NPs. Significant differences in their properties, which affected the aggregation and dissolution tendencies of the two types of ZnO-NPs, were also confirmed [28].
It should be noted that the physicochemical properties of nanoparticles have been reported to be critical factors affecting their fate and transport [4,29,30]. Thus, it is reasonable to expect that the nanoparticles synthesized with different approaches could show different transport behavior, but no comparative studies on this issue have been reported; thus, the relevant study is required.
The first objective of this study was to compare/analyze the transport and retention behaviors of the two types of commercial ZnO-NPs, which were prepared with different methods, in saturated porous media as a function of the presence/absence of the SRHA that is present in the soil media surface and the SRHA that is present in the bulk solution, and to changes in the ionic strength. The second objective of this study was to reverify the deposition site competition that was reported as the mechanism of the coexistence of recently-reported suspended SRHA and NPs in the soil environment using two types of the ZnO-NPs that were synthesized by different methods, and eventually to enhance the reliability of the suggested mechanisms of ZnO-NPs.

ZnO-NPs Suspension Preparation
ZnO-NPs can be synthesized via various approaches (e.g., precipitation, spray pyrolysis, hydrothermal synthesis, solvothermal synthesis, electrochemical methods, sol-gel, microwave synthesis, and green synthesis) [31][32][33]. Among those, a chemical precipitation method with different synthesis conditions, which was modified from published procedures [26,27,34], was used to prepare ZnO-NPs in the present study. The synthesis processing of ZnO-NPs is described in the Supplementary Information. The synthesized ZnO-NP samples from synthesis 1 and synthesis 2 were designated as nZnO-1 and nZnO-2, respectively. The nZnO stock suspensions (100 mg L −1 ) were prepared by dispersing the nZnO in Milli-Q water (Q-Gard, Millipore Inc., Billerica, MA, USA). To enhance dispersibility, ultrasound (ultrasonic homogenizer, KNSN-RAB, KOEN, South Korea) was applied for 30 min. A stock suspension was prepared each time prior to an experiment and was diluted in the desired salt solution (NaCl, 0.1, 1, and 10 mM) to prepare the nZnO input suspension (20 mg L −1 ) that was necessary for column tests. Then, the nZnO suspension was stirred for 2 h at 200 rpm to adjust its pH to 9.0 using 0.1 M NaOH.

ZnO-NPs Characterization
X-ray diffraction (XRD) patterns were recorded with a Bruker D8 HRXRD X-ray diffractometer using Ni-filtered Cu Kα radiation (λ = 0.154606 nm, 40 kV, 40 mV). XRD patterns of the samples were collected in the range of 20-65 • (2θ) with a step size of 0.02 • and step time of 2 s. Transmission electron microscope (TEM) observations were conducted with JEOL-2010F microscope operated at 200 kV. The samples for TEM measurements were suspended in ethanol and dropped onto holey carbon films that were supported on Cu grids for imaging. The specific surface areas of the samples were determined via gas physisorption at −196 • C using a Micromeritics ASAP 2020M analyzer and assessed from nitrogen adsorption data in the partial relative pressure from 0.05 to 0.25 using the Brunauer-Emmett-Teller (BET) method [34,35]. The zeta potential and particle size of nZnO with the presence/absence of SRHA (1 and 5 mg L −1 as TOC concentration) in the NaCl (0.1, 1, and 10 mM) solution were measured using ELS-Z (Otsuka, Osaka, Japan); the measurements were conducted 10-15 times repeatedly and immediately prior to the column test at room temperature (25 • C). The zeta potential was converted from measured electrophoretic mobility values using the Smoluchowski equation, and the particle size was determined using a dynamic light scattering (DLS) method [36,37]. The zeta potential and particle size of the two types of nZnO, which were measured in NaCl in the presence/absence of SRHA and according to the ionic strength, are shown in Figure 1. nZnO with the presence/absence of SRHA (1 and 5 mg L −1 as TOC concentration) in the NaCl (0.1, 1, and 10 mM) solution were measured using ELS-Z (Otsuka, Osaka, Japan); the measurements were conducted 10-15 times repeatedly and immediately prior to the column test at room temperature (25 °C). The zeta potential was converted from measured electrophoretic mobility values using the Smoluchowski equation, and the particle size was determined using a dynamic light scattering (DLS) method [36,37]. The zeta potential and particle size of the two types of nZnO, which were measured in NaCl in the presence/absence of SRHA and according to the ionic strength, are shown in Figure 1.

Suwannee River Humic Acid (SHRA) Solution Preparation
In this study, humic substances, which are abundant in the freshwater environment, were used as an NOM. SRHA (Cat. #2S101H, International Humic Substances Society), which is used frequently in colloid transport studies, was used as a humic substance. The SRHA stock solution was prepared according to a method reported previously [15,19]. First, 25 mg of dried SRHA was placed into 50 mL Milli-Q water (Q-Gard) and was stirred for 24 h. The solution was filtered using a 0.22 μm filter, and its pH was adjusted to 8.0 using 0.1 M NaOH. This filtered and pH-adjusted solution was kept in a darkroom at 4 °C to be used as the SRHA stock solution. The concentration of the SRHA stock solution was analyzed using a TOC meter (TOC-VCPH, Shimadzu Co., Kyoto, Japan), the median diameter of which was ~475 μm, and the uniformity factor 1.5. The porous media were cleansed to remove chemical heterogeneity using the method suggested in previous studies [38,39]; the soil media were placed into Milli-Q water and shaken to remove impurities, then soaked in 37% HCl at 1:1 ratio for 24 h. The media were cleaned again with Milli-Q water, and the above process was repeated one more time before baking them at 800 °C for 8 h. To rehydrate the baked soil media prior to the column test, they were placed into Milli-Q water and boiled for at least 1 h. The zeta potential was measured at the desired NaCl solution ionic strength after the soil media were ground.
Column experiments were conducted with NaCl as a salt at various ionic strengths, ranging from 0.1 mM to 10 mM. The transport and retention behaviors of the two types of nZnO were compared and analyzed. Cleansed sand was wet-packed in the cylindrical glass chromatography column (10 cm in length × 2.5 cm in inner diameter (Omnifit, UK)).

Suwannee River Humic Acid (SHRA) Solution Preparation
In this study, humic substances, which are abundant in the freshwater environment, were used as an NOM. SRHA (Cat. #2S101H, International Humic Substances Society), which is used frequently in colloid transport studies, was used as a humic substance. The SRHA stock solution was prepared according to a method reported previously [15,19]. First, 25 mg of dried SRHA was placed into 50 mL Milli-Q water (Q-Gard) and was stirred for 24 h. The solution was filtered using a 0.22 µm filter, and its pH was adjusted to 8.0 using 0.1 M NaOH. This filtered and pH-adjusted solution was kept in a darkroom at 4 • C to be used as the SRHA stock solution. The concentration of the SRHA stock solution was analyzed using a TOC meter (TOC-VCPH, Shimadzu Co., Kyoto, Japan), the median diameter of which was~475 µm, and the uniformity factor 1.5. The porous media were cleansed to remove chemical heterogeneity using the method suggested in previous studies [38,39]; the soil media were placed into Milli-Q water and shaken to remove impurities, then soaked in 37% HCl at 1:1 ratio for 24 h. The media were cleaned again with Milli-Q water, and the above process was repeated one more time before baking them at 800 • C for 8 h. To rehydrate the baked soil media prior to the column test, they were placed into Milli-Q water and boiled for at least 1 h. The zeta potential was measured at the desired NaCl solution ionic strength after the soil media were ground.
Column experiments were conducted with NaCl as a salt at various ionic strengths, ranging from 0.1 mM to 10 mM. The transport and retention behaviors of the two types of nZnO were compared and analyzed. Cleansed sand was wet-packed in the cylindrical glass chromatography column (10 cm in length × 2.5 cm in inner diameter (Omnifit, UK)). The wet sand was packed little by little (at an interval of~1 cm) with a slight wobble and layering or minimal air inflow. The porosity of the packed column was measured at 0.36.
After packing, deionized water (DI) was supplied for~10 pore volumes (PV, defined as t/VQ; here, t, V, and Q represent time, column void volume, and volumetric flow rate) to remove impurities in the column, and the NaCl set at the desired ionic strength and pH was injected for at least 10 PV to reach an equilibrium prior to the experiment. Upon completion of equilibration, the nZnO suspension with or without SRHA was injected for 5 PV. Then, particle-free NaCl with the same ionic strength and pH was injected for 5 PV. The nZnO and salt solution were injected in up-flow mode using a peristaltic pump (Cole-Parmer, Thermo Fisher Scientific, Waltham, MA, USA). The concentration of the injected nZnO suspension was 20 mg L −1 . All the flow rates in the experiments were set at 1.144 mL min −1 . The nZnO released from the column were continuously (~8 mL) sampled in a 10 mL test tube. To investigate the distribution of the retained NPs in the column, the column was dissected at approximately every 1 cm after the transport experiment, and soil media were placed into 0.5 M HNO 3 to dissolve nZnO. The concentrations of the nZnO released from the column and the nZnO retained in the column were analyzed using ICP. The areas below the breakthrough curves represent the relative concentration of the nZnO released from the column against the concentration of the nZnO that was injected initially. The obtained retention profiles represent the fraction (i.e., retained nZnO amount/total injected nZnO amount) of the nZnO retained in the unit mass of the soil media. The whole collected nZnO fraction obtained from the breakthrough curve and retention profile are shown in Table S2 (mass balance).

SRHA Adsorption Experiments for ZnO-NPs
Previous studies reported that SRHA can adsorb onto the surface of ZnO-NPs and alter the surface properties [15,28]. Based on the procedure described in the previous studies, the amounts of SRHA adsorbed into the two types of nZnO as a function of the ionic strengths (0.1 mM and 10 mM) were evaluated at pH 9. The initial SRHA concentrations were set at 1 mg L −1 and 5 mg L −1 , and the added nZnO suspension concentration was set at 20 mg L −1 . nZnO and SRHA were stirred for 30 min to facilitate adsorption. Upon completion of adsorption, the nZnO suspension was placed into a 50 mL tube to centrifuge (20 min, 15,000× g; 1580MGR, Gyrozen, South Korea). Then, 2 mL of supernatant was carefully collected from each tube. The collected samples were put in a 15 mL tube to centrifuge again (20 min, 15,000× g). nZnO and SRHA were separated completely. Initial and final SRHA concentrations were measured at 254 nm using a spectrophotometer [28] (HS-3300, Humas, South Korea). The results of the experiments in each condition were expressed as mg SRHA/g nZnO.

Derjaguin-Landau-Verwey-Overbeek (DLVO) Interaction Energy Calculation
To confirm the details of nZnO transport behaviors in the presence/absence of SRHA, through the column experiments, the total interaction energy of NPs-NPs and quartz-NPs according to the distance was calculated using the classic DLVO theory. As shown in Equation (1), the DLVO interaction energy was calculated as the sum of retarded van der Waals energies (Φ VDW ) and electrostatic double layer energies (Φ EDL ).
In Φ VDW , a sphere-sphere equation was used for NPs-NPs, and a sphere-plate equation for NPs-sand. Each equation is expressed as Equations (2) and (3) [40].
where a p1 and a p2 in Equation (2) refer to the radii of the two interacting spherical NPs, whereas a p1 in Equation (3) refers to the radius of the NPs, h represents the distance between the two NPs (Equation (2)) or the distance between NPs and sand surface (Equation (3)), and A 131 is the Hamaker constant. In the NPs-water-NPs system, 1.23 × 10 −20 J was used in the absence of SRHA, and 6.09 × 10 −22 J in the presence of SRHA. A 132 is the Hamaker constant in the NPs-water-collector system; it was calculated at 6.96 × 10 −21 J in the absence of SRHA, and 1.54 × 10 −21 J in the presence of SRHA. The combined Hamaker constant NPs-water-NPs system (A 131 ) and NPs-water-sand system (A 132 ) were calculated through Equations (4) and (5), respectively.
A 11 is the Hamaker constant for nZnO, and 9.21 × 10 −20 J was used [41]. In the presence of SRHA, 4.71 × 10 −20 J was used for the Hamaker constant of nZnO with adsorbed SRHA [42]. A 22 was the Hamaker constant for silica, and 6.50 × 10 −20 J was used, from [43]. A 33 was the Hamaker constant for water, and 3.70 × 10 −20 J was used same as A 22 , considering [43]. λ was the characteristic wavelength adsorbed in the interaction, and 100 nm was typically used. Initial particle sizes measured prior to transport experiments were used in Equations (2) and (3). Adsorption of SRHA on the surface of nanoparticles could stabilize the nanoparticles in suspension and alter particles size of nanoparticles, obtaining a different a p used in Equation (3).
To calculate Φ EDL , NPs-NPs was assumed to have had a sphere-sphere form, and NPs-sand, a sphere-plate form. Each equation is shown as Equations (6) and (7) [44].
n ∞ denotes the bulk number density of ions (N m −3 , where N is the number), k is the Boltzmann constant (J K −1 ), T the absolute temperature of the system (K), ϕ p1 and ϕ p2 are the zeta potentials of NPs 1 and 2 (ϕ = zeψ/kT), and ψ p and ψ c are the zeta potentials of the nanoparticle and sand (V), respectively. ε 0 represents the permittivity of a vacuum (C V −1 m −1 ), ε r the dielectric constant or relative permittivity of water, κ represents the Debye-Hückel reciprocal length (m −1 ). z i is the ion valence, e is the electron charge (C), and n i0 represents the number concentration of the i ions in the bulk solution. The zeta potential and particle size values that were used for calculating DLVO interaction energies are shown in Table S1.

Characterization of Prepared Two Types of ZnO-NPs
The XRD and TEM analysis results for the nZnO prepared by two different synthesis methods are presented in Figure 1, and the key properties are also summarized in Table 1. Overall, two nZnO was observed to exhibit different physical properties, and the detailed information on all properties for two nZnO is provided in the Supplementary Information. Figure 2 shows the results of the zeta potential and DLS particle size of the two types of nZnO as a function of the change in SRHA concentration and ionic strength at pH 9. Note that zeta potential represents the electrical potential of a particle measured at slipping plane, and the value is affected by many variables (e.g., salt type, salt concentration, polymer sorption, pH) [45]. In the absence of SRHA (Figure 2a,b), the zeta potential of nZnO-1 was less negative than that of nZnO-2 over the whole ionic strength range (0.1-10 mM), showing no difference as a function of the change in the ionic strength (0.1-10 mM); and the zeta potential of nZnO-2 was formed at approximately the isoelectric point (IEP) in the whole ionic strength range (0.1-10 mM). In the presence of SRHA, the zeta potential for two types of nZnO was more negative than that in the absence of SRHA, showing a consistent tendency over the whole ionic strength range (0.1-10 mM), regardless of the SRHA concentration changes (1 and 5 mg L −1 ). With the presence of SRHA, no difference in the zeta potential was observed between the two types of nZnO. The reason for the more negative charge in the zeta potential in the presence of SRHA was that the small amount (1 and 5 mg L −1 ) of SHRA added in the solution was adsorbed into the nZnO to accelerate the change. This has been reported previously; SRHA was adsorbed onto metal oxide NPs to negatively increase the zeta potential [19,[46][47][48][49].
i. 2021, 11, x FOR PEER REVIEW 6 of 17 mM); and the zeta potential of nZnO-2 was formed at approximately the isoelectric point (IEP) in the whole ionic strength range (0.1-10 mM). In the presence of SRHA, the zeta potential for two types of nZnO was more negative than that in the absence of SRHA, showing a consistent tendency over the whole ionic strength range (0.1-10 mM), regardless of the SRHA concentration changes (1 and 5 mg L −1 ). With the presence of SRHA, no difference in the zeta potential was observed between the two types of nZnO. The reason for the more negative charge in the zeta potential in the presence of SRHA was that the small amount (1 and 5 mg L −1 ) of SHRA added in the solution was adsorbed into the nZnO to accelerate the change. This has been reported previously; SRHA was adsorbed onto metal oxide NPs to negatively increase the zeta potential [19,[46][47][48][49].  In terms of the two types of nZnO size (Figure 2c,d), the nZnO-1 size in the absence of SRHA ranges from 190 to 220 nm over the whole ionic strength range, showing little difference as a function of change in ionic strength. This trend is consistent with the DLVO  a Crystallite size calculated according to Scherrer equation from XRD data; b Primary particle size was determined by averaging more than 400 particles from TEM micrographs; c Specific surface area was measured using BET model based on N 2 adsorption of isotherm.
In terms of the two types of nZnO size (Figure 2c,d), the nZnO-1 size in the absence of SRHA ranges from 190 to 220 nm over the whole ionic strength range, showing little difference as a function of change in ionic strength. This trend is consistent with the DLVO interaction energy calculation between particles, which shows the favorable interaction condition even at the lowest ionic strength tested (0.1 mM), and the same is expected at higher ionic strengths; this will be discussed in detail in the Section 3.2. In the presence of SRHA, the size was~180 nm or smaller, showing little difference as a function of the change in ionic strength or SRHA concentration. In the absence of SRHA, the nZnO-2 size ranges from 220 to 330 nm over the whole ionic strength range, showing little difference as a function of the change in ionic strength. In the presence of SRHA, the size was 260 nm or smaller, showing little difference as a function of the change in ionic strength or SRHA concentration. Eventually, the size of the two types of nZnO decreased slightly in the presence of SRHA. The reason was that SRHA further negatively changed the zeta potential of nZnO to increase the repulsive electrostatic force among the particles, and to decrease the nZnO aggregation level, resulting in a decrease in the nZnO size. These outcomes corresponded to those of the zeta potential. In previous studies, SRHA was reported to increase the stability of colloidal particles such as C 60 , TiO 2 , NPs, hematite, latex, and clay particles [50][51][52][53][54][55]. The size of nZnO-1 tended to be smaller than that of nZnO-2, regardless of the presence/absence of SRHA and the ionic strength.

Transport and Retention Behavior of ZnO-NPs without SRHA
To investigate the transport behavior of nZnO in the packed column, experiments were conducted in the general soil environment conditions of NaCl (0.1-10 mM) and at pH 9. Figure 3a,b, as well as Figure 4a,b, show the breakthrough curves and retention profiles of the two types of nZnO in the absence of SRHA. In the breakthrough curves of both type of nZnO (Figure 3a,b), no breakthrough was observed over the whole ionic strength range (0.1-10 mM). Because no breakthrough was observed with either type of nZnO, they were considered to have been deposited in the soil media, which corresponded to the results of mass balance (Table S2).
Because no breakthrough was observed with either type of nZnO from the column, it was difficult to compare the apparent transport behaviors. Accordingly, the distribution of the nZnO retained in quartz sand was investigated by confirming the nZnO retention profiles. As described, all the nZnO were retained on the column according to the mass balance confirmation (Table S2). According to the retention profiles (Figure 4) of the two types of nZnO, no difference in the deposition amount was observed as a function of change in ionic strength (0.1-10 mM) between the two types of nZnO, and the deposition of both type of nZnO was confirmed mainly near the column inlet. The similar graph pattern (i.e., large amount of deposition at the column inlet) was also observed from many previous studies [20,25,[56][57][58][59][60], and they attributed this trend to the straining caused by the concurrent aggregation among nZnO. Appl. Sci. 2021, 11, x FOR PEER REVIEW 8 of 17 . The inset of (a,b) is the enlarged image of the BTC profiles for clarity; note that the BTC percent is ≤2% for all cases, which is considered negligible. C/C0 represents normalized concentration; here, C and C0 represent the concentration of NPs passing through the column at certain pore volume and of injected NPs suspension, respectively. The pore volume is defined as t/VQ; here, t, V, and Q represent time, column void volume, and volumetric flow rate. The inset of (a,b) is the enlarged image of the BTC profiles for clarity; note that the BTC percent is ≤2% for all cases, which is considered negligible. C/C 0 represents normalized concentration; here, C and C 0 represent the concentration of NPs passing through the column at certain pore volume and of injected NPs suspension, respectively. The pore volume is defined as t/VQ; here, t, V, and Q represent time, column void volume, and volumetric flow rate.  Here, the dimensionless depth represents the normalized column depth (i.e., column depth at certain position divided by total length of the column). The x-axis represents the normalized concentration of NPs deposited at a certain depth per unit sand mass; Nc, Ntc, and gsand represent the number of the deposited NPs, the number of total NPs injected to the column, and the sand weight at a certain depth, respectively. Here, the dimensionless depth represents the normalized column depth (i.e., column depth at certain position divided by total length of the column). The x-axis represents the normalized concentration of NPs deposited at a certain depth per unit sand mass; N c , N tc , and g sand represent the number of the deposited NPs, the number of total NPs injected to the column, and the sand weight at a certain depth, respectively.

Transport and Retention Behavior of ZnO-NPs with SRHA
To compare and analyze the transport behaviors of both types of the nZnO in the packed column in the presence of SRHA (1 and 5 mg L −1 ), experiments were conducted using NaCl (0.1-10 mM) and at pH 9. The breakthrough curves and retention profiles of both type of nZnO affected by SRHA (1 and 5 mg L −1 ) are shown in Figure 3a-f as well as Figure 4c was seen either. The reasons for the increase in the transport of both types of nZnO in the presence of SRHA (1 and 5 mg L −1 ) include first that the suspended SRHA is adsorbed into the surface of nZnO, resulting in the nZnO zeta potential being more negative, and eventually, the repulsive electrostatic force between nZnO and the sand surface increases. Second, there is deposition site competition between NPs and suspended SRHA. These two reasons were verified as follows.
To verify the first point, DLVO interaction energy profiles were calculated; the results are shown in Figure 5a. Because the difference in transport as a function of the change in the ionic strength and SRHA content was not significant between the two types of nZnO, the results of representative conditions, such as 0.1 mM ionic strength and 5 mg L −1 SRHA, are shown. The interaction energy profiles for other conditions (1 mg L −1 SRHA in 1 and 10 mM NaCl) can also be found in the Supplementary Information (Figures S4 and S5). In Figure 5a, the energy barrier was observed in all conditions; the level was significantly low in the absence of SRHA, whereas the level was higher in the presence of 5 mg L −1 SRHA. In the absence of SRHA, the NPs-sand energy barrier was low, NPs could probably overcome the energy barrier, all the nZnO were deposited on the column, and consequently, no breakthrough occurred. In the presence of 5 mg L −1 SRHA, the NPs-sand energy barrier level was significantly higher, so nZnO could not deposit on the sand, due to the stronger repulsive force, and consequently, breakthrough occurred [19,21]. To verify the second point, SRHA was coated on the sand for the column test. If deposition site competition was the main mechanism, the column packed with SRHA-coated sand might show improved nZnO transport versus the column packed with bare sand. NaCl with a desired ionic strength was supplied to the column packed with sand to reach equilibrium; then, 5 mg L −1 SRHA was supplied in as much as 5 PV to coat the quartz sand. To remove the SRHA remaining in the column pores, NaCl with the same ionic strength was supplied at 2 PV. All the tests were conducted at pH 9 and two levels of ionic strength (0.1 and 10 mM NaCl). The transport and retention behaviors of both type of nZnO with the presence/absence of suspended SRHA (5 mg L −1 ) are shown in Figure 6. In the absence of suspended SRHA, nZnO-1 showed breakthrough with SRHA-coated quartz sand, and the breakthrough amount decreased with the increase in ionic strength (0.1 and 10 mM). The reason was that with the increase in ionic strength, the electrostatic repulsive force between NPs and SHRA-coated sand decreased, and more and more NPs were deposited, resulting in a low level of breakthrough. This corresponded to the results of previous studies that confirmed an increase in the deposition of various nanoparticles (C60, TiO2, and Fe 0 ) with decreasing ionic strength [25,61,62]. The nZnO-1 breakthrough curves showed a ripening pattern due to the aggregation among NPs in the porous media. However, nZnO-2 showed no breakthrough at any ionic strength regardless of the SRHA coating. The reason for the difference in the transport behavior between the two NPs was due mainly to the difference in the zeta potential between nZnO-1 and nZnO-2. In the presence of suspended SRHA, no difference in the breakthrough amount of nZnO-1 was observed regardless of sand coating or ionic strength. In nZnO-2, no difference in the breakthrough amount was observed regardless of sand coating or change in the ionic strength either. The reason for the lack of difference regardless of SRHA coating was that the 5 mg L −1 suspended SRHA could be sufficiently adsorbed into the heterogeneous sites that were present in sand in a patch form. Based on these results, deposition site competition between nZnO and suspended SRHA was confirmed as the main retention mechanism of nZnO. In addition, because the rate of SRHA depositing to sand was higher than that of nZnO, the nZnO transport improved in the presence of suspended SRHA. To verify the second point, SRHA was coated on the sand for the column test. If deposition site competition was the main mechanism, the column packed with SRHAcoated sand might show improved nZnO transport versus the column packed with bare sand. NaCl with a desired ionic strength was supplied to the column packed with sand to reach equilibrium; then, 5 mg L −1 SRHA was supplied in as much as 5 PV to coat the quartz sand. To remove the SRHA remaining in the column pores, NaCl with the same ionic strength was supplied at 2 PV. All the tests were conducted at pH 9 and two levels of ionic strength (0.1 and 10 mM NaCl). The transport and retention behaviors of both type of nZnO with the presence/absence of suspended SRHA (5 mg L −1 ) are shown in Figure 6. In the absence of suspended SRHA, nZnO-1 showed breakthrough with SRHA-coated quartz sand, and the breakthrough amount decreased with the increase in ionic strength (0.1 and 10 mM). The reason was that with the increase in ionic strength, the electrostatic repulsive force between NPs and SHRA-coated sand decreased, and more and more NPs were deposited, resulting in a low level of breakthrough. This corresponded to the results of previous studies that confirmed an increase in the deposition of various nanoparticles (C 60 , TiO 2 , and Fe 0 ) with decreasing ionic strength [25,61,62]. The nZnO-1 breakthrough curves showed a ripening pattern due to the aggregation among NPs in the porous media. However, nZnO-2 showed no breakthrough at any ionic strength regardless of the SRHA coating. The reason for the difference in the transport behavior between the two NPs was due mainly to the difference in the zeta potential between nZnO-1 and nZnO-2. In the presence of suspended SRHA, no difference in the breakthrough amount of nZnO-1 was observed regardless of sand coating or ionic strength. In nZnO-2, no difference in the breakthrough amount was observed regardless of sand coating or change in the ionic strength either. The reason for the lack of difference regardless of SRHA coating was that the 5 mg L −1 suspended SRHA could be sufficiently adsorbed into the heterogeneous sites that were present in sand in a patch form. Based on these results, deposition site competition between nZnO and suspended SRHA was confirmed as the main retention mechanism of nZnO. In addition, because the rate of SRHA depositing to sand was higher than that of nZnO, the nZnO transport improved in the presence of suspended SRHA. In the absence of SRHA, the nZnO was largely retained in the column inlet. To investigate the reason, the NPs-NPs DLVO interaction energy profiles were calculated, and the results are shown in Figure 5b. As a result of confirming the NPs-NPs interaction energy profiles, an active NP aggregation developed because the significantly low energy barrier in the absence of SHRA was overcome by nZnO Brownian motion or collision force [63]. The massive nZnO deposition in the column inlet area could be explained by NP aggregation. In the presence of SRHA, a high level of energy barrier was confirmed due to the NPs-NPs interaction energy (unfavorable conditions). The significantly strong repulsive force reduced NPs aggregation, resulting in no mass deposition in the column inlet area.
While no difference in breakthrough curves and retention profiles was observed between the two types of nZnO according to the change in ionic strength in the absence of SRHA, differences in the transport behavior were observed between nZnO-1 and nZnO-2 in the presence of 1 mg L −1 SRHA (Figure 7a,b). More specifically, the nZnO-1 breakthrough amount was confirmed at 47-52%, and nZnO-2, 34-40%. In the presence of 5 mg L −1 SRHA, the breakthrough amount of nZnO-1 was 48-53%, and that of nZnO-2, 41-48%. The transportability of nZnO-1 was confirmed to be higher than that of nZnO-2. To explain the difference in the transport of the two types of nZnO, the zeta potential of nZnO was investigated. No difference in the zeta potential was observed between the two types of nZnO according to the SRHA concentration and ionic strength in the presence of SRHA In the absence of SRHA, the nZnO was largely retained in the column inlet. To investigate the reason, the NPs-NPs DLVO interaction energy profiles were calculated, and the results are shown in Figure 5b. As a result of confirming the NPs-NPs interaction energy profiles, an active NP aggregation developed because the significantly low energy barrier in the absence of SHRA was overcome by nZnO Brownian motion or collision force [63]. The massive nZnO deposition in the column inlet area could be explained by NP aggregation. In the presence of SRHA, a high level of energy barrier was confirmed due to the NPs-NPs interaction energy (unfavorable conditions). The significantly strong repulsive force reduced NPs aggregation, resulting in no mass deposition in the column inlet area.
While no difference in breakthrough curves and retention profiles was observed between the two types of nZnO according to the change in ionic strength in the absence of SRHA, differences in the transport behavior were observed between nZnO-1 and nZnO-2 in the presence of 1 mg L −1 SRHA (Figure 7a,b). More specifically, the nZnO-1 breakthrough amount was confirmed at 47-52%, and nZnO-2, 34-40%. In the presence of 5 mg L −1 SRHA, the breakthrough amount of nZnO-1 was 48-53%, and that of nZnO-2, 41-48%. The transportability of nZnO-1 was confirmed to be higher than that of nZnO-2. To explain the difference in the transport of the two types of nZnO, the zeta potential of nZnO was investigated. No difference in the zeta potential was observed between the two types of nZnO according to the SRHA concentration and ionic strength in the presence of SRHA (1 and 5 mg L −1 ). As described, the zeta potential was not sufficient to explain the difference in the transport, so the NPs-sand DLVO interaction energy profiles were calculated to compare the NPs. As a result, the energy barrier of nZnO-1 was confirmed to be~200 kT, and that of nZnO-2,~400 kT. The NPs-sand DLVO interaction energy profiles were not enough to explain the difference because the energy barrier of both particles was quite high, and nZnO-2 showed a higher energy barrier but lower transportability than nZnO-1. Finally, the main mechanism of NP transport was confirmed in Section 3.3 to be deposition site competition between suspended SRHA and NPs in the presence of SRHA, so the amount of adsorption to each nZnO, according to the SRHA concentration level, was confirmed, and the results are shown in Figure 7c. nZnO-1 showed less adsorption of SRHA than nZnO-2 regardless of the SRHA concentration. This indicates that the amount of remaining SRHA (i.e., suspended SRHA) was larger than that adsorbed in the nZnO-1 suspension. Accordingly, in the case of nZnO-1, which had more suspended SRHA, more deposition sites on the sand surface were occupied by SRHA, and, consequently, the number of NPs deposition sites decreased to improve transport. These phenomena were supported by the absence of a difference in the adsorption amount between nZnO-1 and nZnO-2 regardless of the change in ionic strength, and by the lack of difference in transport.
In previous studies, more suspended SRHA amount was confirmed to have resulted in more TiO 2 transport amount [25,64].
(1 and 5 mg L −1 ). As described, the zeta potential was not sufficient to explain the difference in the transport, so the NPs-sand DLVO interaction energy profiles were calculated to compare the NPs. As a result, the energy barrier of nZnO-1 was confirmed to be ~200 kT, and that of nZnO-2, ~400 kT. The NPs-sand DLVO interaction energy profiles were not enough to explain the difference because the energy barrier of both particles was quite high, and nZnO-2 showed a higher energy barrier but lower transportability than nZnO-1. Finally, the main mechanism of NP transport was confirmed in Section 3.3 to be deposition site competition between suspended SRHA and NPs in the presence of SRHA, so the amount of adsorption to each nZnO, according to the SRHA concentration level, was confirmed, and the results are shown in Figure 7c. nZnO-1 showed less adsorption of SRHA than nZnO-2 regardless of the SRHA concentration. This indicates that the amount of remaining SRHA (i.e., suspended SRHA) was larger than that adsorbed in the nZnO-1 suspension. Accordingly, in the case of nZnO-1, which had more suspended SRHA, more deposition sites on the sand surface were occupied by SRHA, and, consequently, the number of NPs deposition sites decreased to improve transport. These phenomena were supported by the absence of a difference in the adsorption amount between nZnO-1 and nZnO-2 regardless of the change in ionic strength, and by the lack of difference in transport. In previous studies, more suspended SRHA amount was confirmed to have resulted in more TiO2 transport amount [25,64].

Conclusions
Regarding the transport and retention behavior of nZnO, no breakthrough was observed in either type in the absence of suspended SRHA. The NPs-NPs concurrent aggregation resulted in a massive deposition of nZnO on the column inlet. In the presence of suspended SHRA, however, no massive deposition was observed. Specifically, 47.0-53.2% of injected nZnO-1 passed through the column, while 33.4-47.5% passed through for nZnO-2. These outcomes were based on the increase in the nZnO transport that was facilitated by the increase in the electrostatic repulsive force among nZnO, which was caused by the SHRA adsorbed into the nZnO surface.
The transportability of nZnO-1 was observed to be higher than that of nZnO-2, and in the absence of SRHA in the coated soil column, breakthrough was observed only with nZnO-1 (17.2-23.8%). To further investigate the nZnO transport behavior, tests determining the amount of SRHA adsorbed to nZnO were conducted. As a result, nZnO-1 showed less SRHA adsorption than nZnO-2 regardless of ionic strength. Eventually, with more remaining SHRA not adsorbed to nZnO, more nZnO transport was confirmed in the study. Based on the results above, deposition site competition between nZnO and suspended SHRA was confirmed as the major retention mechanism of nZnO transport. Furthermore, nZnO transport increased in the presence of abundant suspended SRHA because the deposition rate of SRHA to soil media was higher than that of nZnO.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/app11198957/s1, Additional Details on Materials and Methods, Additional Results and Discussion, Figure S1: TGA profiles for the prepared two types of nZnO by difference synthesis conditions, Figure S2: ATR-FTIR spectra of two types of nZnO obtained from synthesis conditions, Figure S3: Specific surface areas of two types of nZnO from Brunauer-Emmet-Teller (BET) method, Figure S4: DLVO interaction energy profiles of nZnO-1 for NPs-NPs (solid line) and NPs-sand (dashed line) surfaces in NaCl (1-10 mM) solutions in the absence/presence SRHA, Figure S5: DLVO interaction energy profiles of nZnO-2 for NPs-NPs (solid line) and NPs-sand (dashed line) surfaces in NaCl (1-10 mM) solutions in the absence/presence SRHA, Table S1: Parameters used to calculate DLVO interaction energy profiles, Table S2: Mass balance for two types of nZnO in column experiments.
Author Contributions: Writing-original draft preparation, methodology, investigation: D.K.; conceptualization, writing-original draft preparation, formal analysis, review and editing: G.H. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.