Removal of Total Organic Carbon from Sewage Wastewater Using

Jan 15, 2014 - Centro de Investigacion en Materiales Avanzados, S.C. (CIMAV), Miguel de Cervantes 120, Complejo Industrial Chihuahua, Chihuahua, Apdo...
2 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/Langmuir

Removal of Total Organic Carbon from Sewage Wastewater Using Poly(ethylenimine)-Functionalized Magnetic Nanoparticles Ramnath Lakshmanan,† Margarita Sanchez-Dominguez,‡ Jose A. Matutes-Aquino,∥ Stefan Wennmalm,§ and Gunaratna Kuttuva Rajarao*,† †

School of Biotechnology, Royal Institute of Technology (KTH), Stockholm, Sweden Centro de Investigacion en Materiales Avanzados, S. C. (CIMAV), Unidad Monterrey, Alianza Norte 202, 66600 Apodaca, Nuevo Leon, Mexico ∥ Centro de Investigacion en Materiales Avanzados, S.C. (CIMAV), Miguel de Cervantes 120, Complejo Industrial Chihuahua, Chihuahua, Apdo. Postal 31109, Mexico § Science for Life Laboratory, Department of Applied Physics, KTH-Royal Institute of Technology, 171 65 Solna, Sweden ‡

ABSTRACT: The increased levels of organic carbon in sewage wastewater during recent years impose a great challenge to the existing wastewater treatment process (WWTP). Technological innovations are therefore sought that can reduce the release of organic carbon into lakes and seas. In the present study, magnetic nanoparticles (NPs) were synthesized, functionalized with poly(ethylenimine) (PEI), and characterized using TEM (transmission electron microscopy), X-ray diffraction (XRD), FTIR (Fourier transform infrared spectroscopy), CCS (confocal correlation spectroscopy), SICS (scattering interference correlation spectroscopy), magnetism studies, and thermogravimetric analysis (TGA). The removal of total organic carbon (TOC) and other contaminants using PEI-coated magnetic nanoparticles (PEI-NPs) was tested in wastewater obtained from the Hammarby Sjöstadsverk sewage plant, Sweden. The synthesized NPs were about 12 nm in diameter and showed a homogeneous particle size distribution in dispersion by TEM and CCS analyses, respectively. The magnetization curve reveals superparamagnetic behavior, and the NPs do not reach saturation because of surface anisotropy effects. A 50% reduction in TOC was obtained in 60 min when using 20 mg/L PEI-NPs in 0.5 L of wastewater. Along with TOC, other contaminants such as turbidity (89%), color (86%), total nitrogen (24%), and microbial content (90%) were also removed without significant changes in the mineral ion composition of wastewater. We conclude that the application of PEI-NPs has the potential to reduce the processing time, complexity, sludge production, and use of additional chemicals in the WWTP.

1. INTRODUCTION

WWTPs are designed to reduce the load of nutrients such as carbon, nitrogen, and phosphorus in incoming wastewater. Generally, colloidal and settleable organic carbons are removed in primary sedimentation. Bacteria play an important role in breaking down complex organic chemicals into fragments, and they also use carbon as a source of growth. However, this process depends upon the load, wastewater composition, and time taken for the biological process.5 Previous studies indicated that 40−50% of the total organic content in secondary effluents consists of humic substances, and other dissolved sources of organic carbon found are carbohydrates (∼11.5%), fatty acids (∼8.3%), proteins (∼22.4%), tannins (∼1.7%), and anionic detergents (∼13.9%).6 The existing treatment process including biological steps is not sufficient, which is the reason that certain treatment facilities are equipped

Organic matter from wastewater is a heterogeneous mixture of molecules with a varied structure and molecular weight, ranging from simple to complex substances. Wastewater has a mixed population of organic compounds and aggregates, with an average size range from 1 nm to 1 mm, consisting of dissolved organic carbon (∼42%), settleable organic carbon (∼27%), supracolloidal organic carbon (∼20%), and colloidal organic carbon (∼11%).1 Organic compounds not properly eliminated during the wastewater treatment process (WWTP) are discharged into lakes/seas, which has a negative impact on the quality of drinking water and also affects aquatic life.2 Recently, it has been recognized that persistent organic pollutants (POPs) such as endocrine-disrupting pharmaceuticals and other organic molecules are difficult to degrade and can be detected in wastewater effluents.3 Therefore, a proper treatment of wastewater is essential to maintaining the effluent composition within the standard limits.4 © 2014 American Chemical Society

Received: October 25, 2013 Revised: December 27, 2013 Published: January 15, 2014 1036

dx.doi.org/10.1021/la404076n | Langmuir 2014, 30, 1036−1044

Langmuir

Article

Observation was carried out using a field emission transmission electron microscope (JEM-2200FS, 200 kV, 0.19 nm resolution in TEM mode, 0.1 nm resolution in scanning transmission electron microscopy (STEM) mode, and spherical aberration correction in STEM mode). 2.2.2. X-ray Diffraction. Dried powders were characterized by X-ray diffraction (XRD) using a Panalytical Empyrean diffractometer. The crystallite size was estimated using the Debye−Scherrer equation. 2.2.3. Infrared Analysis. Infrared spectroscopy was performed in order to identify organic functional groups and to find the molecular interactions and bonds between the groups. The powders of TSC-NPs (as core NPs), PEI-NPs, and treated PEI-NPs were directly run in a Thermo Nicolet 6700 FT-IR spectrophotometer, and attenuated total reflectance (ATR) measurements were carried out. 2.2.4. CCS AND SICS Analyses. Confocal correlation spectroscopy (CCS)28 and scattering interference correlation spectroscopy (SICS)29 are novel techniques that closely resemble the more established technique of fluorescence correlation spectroscopy (FCS). As in FCS, the diffusion coefficient of particles is estimated from the diffusing particles’ transit time through the confocal detection volume; however, in contrast to FCS, which is based on fluorescence, SICS is based on light scattering plus interference and CCS is based on light scattering only. CCS measurements were performed on a home-built microscope as described earlier.29 The untreated and treated PEI-NPs were diluted in 80% water and 20% ethanol. 2.2.5. Magnetization Studies. Magnetization curves were determined using a physical properties measurement system from Quantum Design with a maximum applied magnetic field of 20 000 Oe. The masses of the TSC-NP and PEI-NP samples were 11.1 and 12.1 mg, respectively. 2.2.6. TGA Analysis. Thermogravimetric analysis (TGA) was carried out by heating a few milligrams of sample from 25 to 800 °C in a TGA-50 from Shimadzu at a heating rate of 10 °C/min. 2.3. Removal of TOC Using PEI-Coated Magnetic NPs. Wastewater was obtained from Hammarby Sjöstadsverk, Sweden, and experiments were performed on the same day of sample retrieval in two stages. In the first stage, PEI-NPs with different concentrations (10, 50, 100, 200, 300, 400, and 500 μg/mL) were mixed to a final volume of 10 mL of wastewater in a 50 mL falcon tube. Samples for analysis were collected at different time intervals (20, 40, 60, 90, and 120 min), and the removal efficiency was calculated as shown in eq 1. The stirring speed was kept constant in all experiments as mentioned in our previous studies.48 In the second part of the experiments, 20 μg/mL of PEI-NPs were incubated with 0.5 L of wastewater for 60 min in order to assess the removal efficiency on a large scale. These experiments were performed in triplicate, and treated samples were separated using an external magnet. The total organic carbon was analyzed using a TOC-5000 from Shimadzu Corporation, Japan, and the reduction percentage was calculated from eq 1.

with advanced methods such as activated carbon, photooxidation, ozonation, and UV radiation.7,8 However, these methods are expensive to use long term and pose some drawbacks: the use of activated carbon requires high temperature during its preparation and regeneration, and additional tertiary filtration is also often needed. Treatment processes with ozonation and UV light demand technical expertise and are energy-intensive.9−11 To remove nutrients such as total organic carbon (TOC) from wastewater efficiently, new technologies are sought and valued while keeping in mind their affordability, time, and ease of use.12,13 Nanoparticles are an attractive possibility for water treatment because their large surface area and specific interactions with organic contaminants may allow a more efficient process to be developed.14 The adsorption of pollutants is economically favorable and easy to use in large-scale applications. Hence, magnetic nanosorbents involving an external magnetic field would be a new generation tool appropriate for the WWTP.15,16 The functionalization of NPs with polymeric materials results in effective adsorption and helps to minimize agglomeration.17,18 Poly(ethylenimine) (PEI) is a cationic organic polymer with branched amino groups on the surface with a wide range of buffering capacity in aqueous solution and thus works in a broad pH range.19−21 Under laboratory conditions, PEI-coated magnetic NPs (PEI-NPs) can specifically remove polycyclic aromatic hydrocarbons22,23 and heavy metals such as zinc, lead, copper, cadmium, nickel, and uranium(IV) from aqueous solutions.14,24,25 However, the application of PEI-NPs to the removal of organic carbon from sewage wastewater remains to be explored. The goal in the present study is to establish the optimum concentration of PEI-NPs for the effective removal of organic contaminants such as TOC from sewage wastewater. The PEINPs were synthesized and characterized with respect to particle size and crystallinity (transmission electron microscopy, TEM) for their interaction with organic matter (Fourier transform infrared spectroscopy, FT-IR), aggregate size, and polydispersity (confocal correlation spectroscopy, CCS, and scattering interference correlation spectroscopy, SICS), magnetic properties (vibrating sample magnetometry, VSM), and thermal stability (thermogravimetric analysis, TGA). Furthermore, analyses of concentration and time kinetics were performed for the removal of total organic carbon in wastewater. In addition, parameters such as total nitrogen, turbidity, color, and microbial reduction were also analyzed and are reported.

2. EXPERIMENTAL SECTION

reduction(%) =

2.1. Synthesis of PEI-Coated Magnetic Nanoparticles. The synthesis of PEI-iron oxide NPs was carried out using the chemical coprecipitation method with a slight modification.26 The iron oxide NPs stabilized with trisodium citrate (TSC) were prepared as described earlier.27 A volume of 2 mL of PEI solution (30%) was added drop-by-drop to the TSC-NP dispersion and stirred continuously at room temperature for 6 h. Thereafter, the NPs were washed to remove unbound PEI from the solution and stored at 4 °C until further use. The concentration of PEI-NPs was expressed in terms of the dry weight of particles per unit volume of suspension medium. 2.2. Characterization of PEI Magnetic NPs. 2.2.1. TEM Analysis. High-resolution transmission electron microscopy (HRTEM) was performed for the analysis of particle size, morphology, and crystallinity. The sample was prepared as follows: 0.5 mg of magnetic NPs was dispersed in water (4 mL) and sonicated; the large agglomerates were removed with a magnet. For analysis, a drop of this dispersion was deposited onto a Formvar/carbon copper grid.

⎛ initial − final ⎞ ⎜ ⎟ × 100 ⎝ ⎠ initial

(1)

2.4. Other Parameters. For large-scale experiments, the total nitrogen content was analyzed using a HR total nitrogen reagent set (persulfate digestion method, 10 to 150 mg/L N) from VWR International, Sweden. The color was estimated by measuring the absorbance at 420 nm using a UV−vis spectrophotometer from Aquamate Thermospectronic, England. The turbidity was measured using a Hach portable turbidimeter (2100Q ISO standard). A nutrient agar plate was prepared and used to determine the colony-forming units (CFU/mL) in order to determine the total number of microorganisms before and after treatment. 2.5. Ion Analysis. Chemical ion analysis was carried out as described earlier.30 Briefly, nitrate, phosphorus, potassium, magnesium, calcium, sulfur, chloride, manganese, boron, copper, zinc, iron, molybdenum, silicon, and aluminum were measured using ICP-AES (inductively coupled plasma atomic emission spectroscopy). The untreated wastewater was used as a control, and water samples treated 1037

dx.doi.org/10.1021/la404076n | Langmuir 2014, 30, 1036−1044

Langmuir

Article

with PEI-NP were sent to LMI (Lennart Månsson Internation) AB, Helsingborg, Sweden.

3. RESULTS AND DISCUSSION Magnetic NPs were synthesized by the coprecipitation method, and trisodium citrate (TSC) was used for primary functionalization as published previously.31 In a second step, PEI was adsorbed onto the TSC-functionalized NPs. The resultant PEIcoated NPs had a long chain of amino groups (NH2) that impart a positively charged surface (Figure 1).

Figure 1. Schematic illustration of PEI-coated magnetic nanoparticles (PEI-NPs).

The long-chain amino groups (NH2) interact with organic matter present in sewage wastewater. Complexes formed on the PEI-NPs were further separated using an external magnet, and the effluent was analyzed. 3.1. Characterization Studies. 3.1.1. TEM and XRD Analysis. According to transmission electron microscopy analysis (Figure 2a−f), both TSC-coated (TSC-NPs) and (TSC + PEI)-coated magnetic nanoparticles (PEI-NPs) were globular in shape, with very similar sizes (average particle sizes of 11.5 nm for TSC-NPs and 11.8 nm for PEI-NPs). The highresolution TEM images (Figure 2d,f) as well as the selected area electron diffraction (SAED) pattern (Figure 2b) demonstrate the crystallinity of the NPs. D spacings of 2.95, 2.52, 1.61, 1.48, 1.32, and 1.29 Å were observed, which may correspond to either the 220, 311, 511, 440, 620, and 540 lattice planes of maghemite (γ-Fe2O3) or the 022, 113, 115, 044, 026, and 335 lattice planes of magnetite (Fe3O4). Both of these possible iron oxides have a cubic spinel structure. X-ray diffraction analysis (Figure 2i,j) confirmed that the correct structure was maghemite (γ-Fe2O3) because all of the reflections were more consistent with this structure (ICDD card no. 00-039-1346). The crystallite size was assessed from the full width at half-maximum (fwhm) of the reflection with the highest intensity (2θ at 35.75°) using the Debye−Scherrer formula, resulting in 12.7 nm for TSC-NPs and 12.0 nm for PEI-NPs, respectively, in very good agreement with the particle size determined by TEM, which confirms that each nanoparticle comprises one nanocrystal, as observed in the HRTEM images. 3.1.2. Magnetization Studies. The reversible behavior of the magnetization curves is a fingerprint of superparamagnetism (Figure 3). It can be observed that none of the samples reach saturation under a maximum applied magnetic field of 20 000 Oe, which is related to surface anisotropy effects in small magnetic nanoparticles. Additionally, the TSC-NPs sample

Figure 2. Transmission electron microscopy and X-ray diffraction analyses of TSC-NPs and PEI-NPs: (a, c, d) HRTEM images of TSCNPs. (b) SAED pattern of image a. (e, f) HRTEM images of PEI-NPs. (g, h) Particle size histograms of TSC and PEI-NPs, respectively. (i, j) X-ray diffractograms of PEI and TSC-NPs, respectively. hkl Miller indexes of maghemite (γ-Fe2O3) are indicated in the diffractogram of PEI-NPs.

shows specific magnetization values above the values of the PEI-NPs sample, which is related to the additional nonmagnetic mass of the PEI coating. Under a maximum applied magnetic field of 20 000 Oe, the TSC-NP sample had a maximum specific magnetization of σmax = 66.4 emu/g and the PEI-NP 1038

dx.doi.org/10.1021/la404076n | Langmuir 2014, 30, 1036−1044

Langmuir

Article

bending,34 and the band at 1556 cm−1 was assigned to C−N stretching. The band at 1470 cm−1 can be assigned to νsym(COO−) and CH2 deformation, and the small bands at 790, 1000, and 1100 cm−1 correspond to N−H wagging.33,35 However, the IR spectrum of the treated PEI-NPs sample presented even stronger bands, which confirms the strong presence of organic matter. Bands at 3250, 2952, 2915, 2850, 1700, 1630, 1550, 1540, 1450, 1400, 1210, and 1010 cm−1 are characteristic of the following functional groups: hydroxyl, amine, methylene, carbonyl, and carboxylate (Figure 4c). Thus, IR spectroscopy could confirm the successful functionalization of iron oxide NPs with TSC and PEI as well as the adsorption of organic contaminants from wastewater. 3.1.4. CCS and SICS Analysis. PEI-NPs and treated PEI-NPs were analyzed using the recently developed techniques of confocal correlation spectroscopy (CCS)28 and scattering interference correlation spectroscopy (SICS).29 The measurements on PEI-NPs showed a good fit when using a model containing one diffusion component and one exponential part. In the intensity trace, the spikes are of very similar heights, and thus neither the trace nor the autocorrelation function (ACF) curve indicates the presence of any aggregates. The diffusion time in these curves was on average 1.3 ms. In contrast, the measurements on treated PEINPs show an intensity trace with spikes of varying heights, indicating the presence of particles of very different sizes. As expected from the intensity trace, the diffusion time of treated PEI-NPs was also longer, on average 24 ms (ms) (i.e., about 20 times longer than the mean diffusion time of the untreated PEINPs, which would indicate a much larger diameter). To analyze and compare the size distribution of the treated and untreated PEI-NPs further, a histogram of the estimated diffusion time in different measurements was obtained (Figure 5). A histogram from 21 measurements of 10 s duration on the PEI-NPs shows that all diffusion times were below 5 ms, indicating a homogeneous population. In contrast, a histogram from 21 measurements of 10 s duration on the treated PEI-NPs shows that the diffusion time in each measurement varies from less than 5 ms up to several hundred milliseconds, indicating a large distribution of particle sizes. It is important to note that in cases when a large aggregate transits the detection focus during the time course of the measurement the estimated diffusion time of that measurement will be dominated by the diffusion time of the aggregate. Therefore, large aggregates may constitute a minority of all diffusing units, as they do in this case, and still dominate the estimated diffusion time in many of the measurements. However, because the diffusion time scales with the radius of the analyzed particles, a 100-fold increase in diffusion time, which was observed in a few cases for the treated PEI-NPs, corresponds to a particle volume that is 1003 = 106 times larger than the particles yielding the shortest diffusion times. Thus, even though the large aggregates constitute a minority of all diffusing units, a majority of the monomers are most likely bound up in aggregates. 3.1.5. TGA Analysis. To estimate the content of organic matter in each sample, thermogravimetric analysis was carried out. Figure 6 shows the thermograms of TSC-NPs, PEI-NPs, and treated PEI-NPs. It was clear that the treated PEI-NPs had the greatest amount of organic matter because the weight loss reached nearly 71 wt %. The PEI-NPs had a total weight loss of nearly 27%, and TSC-NPs had a total weight loss of around 17%. The weight loss attributed to water (below 200 °C) was 11.5 wt % for TSC-NPs, 5.7 wt % for PEI-NPs, and 7.5 wt % for

Figure 3. Magnetization curves taken at room temperature for TSCNPs and PEI-NPs samples.

sample had a maximum specific magnetization of σmax = 58.8 emu/g. Using the law of approach to saturation,36 specific saturation magnetizations of σsat = 71.2 and 62.2 emu/g were calculated for TSC-NP and PEI-NP samples, respectively. These values are below the value of 76.0 emu/g reported for bulk maghemite at 20 °C.37 This reduction is again related to surface anisotropy effects in small magnetic nanoparticles. 3.1.3. FTIR Analysis. To verify the functionalization of citrate (TSC) and PEI on the surface of magnetic NPs, Fourier transform infrared (FT-IR) measurements were carried out. For TSC-NPs and PEI-NPs, OH-stretching bonds were detected in the FTIR spectra as a wide peak in the region from 3000 to 3500 cm−1 (Figure 4a,b). This can be assigned to water

Figure 4. IR spectra of (a) TSC-NPs, (b) PEI-NPs, and (c) treated PEI-NPs.

adsorbed on the oxide surface as well as the OH functional group of the citrate molecule. The band at 535 cm−1 present in the three samples is attributed to the Fe−O bond vibration. The TSC-NPs sample exhibited bands at 1595 and 1392 cm−1 attributed to the asymmetric and symmetric stretching bonds of the carboxylate group COO − (ν asym (COO −) and ν sym (COO−), respectively). The PEI-NPs sample presented several other bands, confirming the successful functionalization with the polymer. The small bands at 2920 and 2820 cm−1 were attributed to the asymmetric and symmetric CH2 stretching of the PEI chain, respectively. The band at 1302 cm−1 was assigned to the C−N stretching mode of PEI,32,33 the band at 1590 cm−1 can be attributed to the asymmetric stretching bonds of the carboxylate group νasym (COO−) as well as NH2 1039

dx.doi.org/10.1021/la404076n | Langmuir 2014, 30, 1036−1044

Langmuir

Article

Figure 5. Confocal correlation spectroscopy of the PEI-NPs before and after treatment with wastewater. (A) Example autocorrelation function (ACF) curves for treated and untreated PEI-NPs. In this particular example, the diffusion time is 3 times longer, indicating a 3 times larger hydrodynamic radius. (B) Two superimposed histograms, one displaying the estimated diffusion time of the untreated PEI-NPs (single black bar) and the other displaying the estimated diffusion time of the treated PEI-NPs (red bars), indicating a broad size distribution of the treated PEI-NPs.

important weight loss up to 800 °C, equivalent to 68.5 wt % of the total organic matter. These results are in good agreement with the efficient adsorption of organic contaminants from wastewater using PEI-NPs. 3.2. Removal of TOC Using PEI-Coated Magnetic NPs. The PEI-NPs were used first in an exploratory experiment intended to establish the optimum concentration and time required for the adsorption of total organic carbon; the concentration and time were screened in the ranges of 10−500 μg/mL and 20−120 min, respectively. The initial concentration of total organic carbon (TOC) present in wastewater was 140 ± 12 mg/L from the day of sample collection. As seen in Figure 7a, the removal of TOC was found to be evident up to a concentration of 100 μg/mL PEI-NPs in 60 min. However, above 200 μg/mL the reduction was decreased and a further increase in TOC was found in the treated samples. This could be due to the large aggregates that are formed and precipitate out of solution and therefore inhibit the interaction with most of the contaminants present in sewage wastewater. Additionally, more collisions between the PEI-NPs resulting from higher concentrations might also be a factor that influences the negative effect. To understand the reduction, similar analyses were performed at low and high concentration of PEI-NPs in Milli-Q water. This test was performed as a blank in order to

Figure 6. TGA thermograms of TSC-NPs, PEI-NPs, and treated PEINPs.

treated PEI-NPs. From the thermogram of TSC-NPs, it was observed that trisodium citrate is lost between 200 and 500 °C, in agreement with previous studies;38 thus the content of trisodium citrate in TSC-NPs sample was 5.5 wt %. The PEI-NPs sample presented its main weight loss of 21.3 wt % between 200 and 680 °C, which was attributed to organic matter (TSC + PEI). The treated PEI-NPs sample had an

Figure 7. Study of the removal of total organic carbon (TOC) from sewage wastewater. (a) Concentration and time kinetics. (b) Low-concentration kinetics at 60 min. 1040

dx.doi.org/10.1021/la404076n | Langmuir 2014, 30, 1036−1044

Langmuir

Article

Figure 8. Contaminants removed using PEI-coated magnetic NPs under optimum conditions. (a) Removal of other contaminants in large-scale treatment (TOT N, total nitrogen). (b) Microbial reduction with control and treated wastewater with PEI-NPs.

The wastewater pH was in the range of 6.8 to 8.4, and earlier studies show that PEI-NPs have a broad range of pH susceptibility between 2 and 12.43 In the present study, the zeta potential for PEI-NPs was found to be 33.14 mV at pH 7, which correlates to an earlier report.44 Several studies have reported that amino-functionalized magnetic nanoparticles can effectively remove heavy metals such as copper, cadmium, nickel, zinc, arsenic, and lead from an aqueous solution and that the efficiency of removal of heavy metals is strongly dependent upon the pH of the solution, in particular, under acidic conditions.45−48 In contrast, another study reported that both adsorption and complex formation occur with amino-functionalized magnetic nanoparticles at higher pH, thus leading to an almost 100% removal of heavy metals except for arsenic.49 Earlier studies have also demonstrated that the use of polyelectrolytes has an average of 30% coagulation in the removal of humic substances from polluted water.50 The degree of removal depends on both the operating conditions and the characteristics of organic matter present in sewage wastewater. However, this is the first study to present the use of PEI-NPs in real wastewater collected from a treatment plant. The pH and other parameters were unaltered in order to study the ability of PEI-NPs to remove contaminants under natural conditions so that their potential applications in wastewater treatment might be highlighted. Removing color, turbidity, and microbes is important in the WWTP where chemicals are used. The results obtained from higher volumes of wastewater treated with the optimized concentration (20 mg/L for 60 min) of PEI-NPs are shown in Figure 8a. PEI-NPs reduced 86% of the color and 89% of the turbidity, which will be an added benefit for the removal of contaminants. Around a 90% reduction of the microbial content was observed (Figure 8b), which could help to reduce the use of disinfectants or further treatment steps. The reduction of microbes may be explained by the binding of the cationic surface charge of the PEI-NPs to the negatively charged cell surface.51,52 The formation of larger flocs was observed after 20 min of incubation. CCS results indicate a 20fold-longer diffusion time for the treated PEI-NPs, leading to a 20-fold-larger diameter than for the nontreated PEI-NPs. This is in accordance with earlier studies14 that highlight the larger

determine the influence of PEI in the water environment. The concentration of TOC increased when using >100 μg/mL PEINPs. Subsequently, total nitrogen was evaluated in addition to TOC measurements in order to determine whether the increase in TOC was due to the release of PEI from the NPs. The total nitrogen was increased as the concentration of PEI-NPs was increased above 100 μg/mL (data not shown). The increase in total nitrogen or carbon concentration is neither practical nor desirable in wastewater treatment application.3 Therefore, an experiment with a concentration range of PEI-NPs between 10 and 50 μg/mL and an interaction time of 60 min was performed (Figure 7b). The reduction of TOC when using concentrations from 20 to 40 μg/mL PEINPs was achieved. However, when experiments were carried out with Milli-Q water there was no significant effect of the increase in TOC or total nitrogen in the treated samples. Other recent developments in the removal of organic carbon in wastewater include photocatalysis, ozonation, and Fenton processes. An earlier study revealed that 20% of the total organic carbon was removed when using 5 mg/L ferrate-Fe (VI) in municipal secondary effluents. Additionally, when compared to the removal of a specific organic compound (i.e., methylene blue), removal using a photocatalytic reaction showed that a 95% reduction in total organic carbon was achieved.39,40 In the present study, the concentration of PEI-NPs from 20 to 40 μg/mL showed the maximum removal efficiency (∼40%) of TOC as compared to the efficiency for higher concentrations (up to 50 μg/mL). The initial concentration of total nitrogen present in wastewater was 42 ± 4 mg/L. No additional nitrogen compounds were found in the treated samples using 20 μg/mL PEI-NPs, which indicates that PEI did not detach from the NPs. Normally, anionic polymers are less toxic to aquatic organisms than cationic polymers. The limits set for cationic polymers are generally