Application of silver nanoparticles toward Co(II) and Pb(II) ions contaminant removal in groundwater

Applications of nanomaterials has recently been of keen interest to researchers. This fascinating drive ensued due to their unique properties such as optical, magnetic, ease of synthesis, large surface area, and tunable surface chemistry and functionalization that are remarkably different from that of the bulk material. In this work, silver nanoparticles were synthesized and characterized using UV–Vis spectroscopy and scanning electron microscopy. The UV–Vis spectra obtained showed absorption peak at 406 nm wavelength and scanning electron microscope reveals an image of size, 20 nm. The efficiency of our nanomaterial was first tested using standard solutions of cobalt and lead. A bathochromic shift in the absorption wavelength from 406 to 445 nm and from 406 to 458 nm for Cobalt(II) and Lead(II), respectively, indicates an adsorption has taken place. Adsorption characteristics of the nanoparticles evaluated at various incubation periods indicates a percent reduction in cobalt(II) from 33.13 to 53.34% and 79.9–92.92% for lead(II) from day 1–14. This indicates high removal efficiency for lead(II) than cobalt(II) in the proof of concept experiment. The results obtained further reveal that metal ion removal from complexes using silver nanoparticles is time-dependent. The effectiveness of our nanomaterial toward metal ion removal from groundwater was tested using inductively coupled plasma optical emission spectroscopy. The results obtained reveal removal efficiencies of 24% for cobalt and 77% for lead. It is important to note that metal ion concentrations reduced as the incubation periods increased. This is in line with the proof of concept experiment using complex solutions of cobalt and lead. Notably, the percent removal of cobalt was quite low compared to proof of concept experiment. These observations suggest a further research into understanding the adsorption characteristics of metal ions to the surface of nanomaterials in aqueous media.


Introduction
Pollution of water bodies emanating from natural and anthropogenic sources has become a matter of concern to mankind ( Giri and Qiu 2016;Li et al. 2012;Vodela et al. 1997). Among the wide diversity of pollutants affecting water resources, heavy metals have received considerable attention due to their high toxicity even at low concentrations (Gaur et al. 2014;Jameel et al. 2012;Marcovecchio et al. 2007). Frontiers of research into the adverse effects of heavy metal pollution of water bodies has been enlarged in recent times due to the contaminant effects of these metals on plants and animal species. Clean water is essential to human health, but due to population growth and extended droughts; the world is facing formidable challenges in meeting rising demands for clean water (Fu and Wang 2011;Gaur et al. 2014;Larsen et al. 2016;Savage and Diallo 2005;Tiwari et al. 2008). Many technologies have been employed in water treatment, and these include ion exchange, reverse osmosis, reduction, precipitation and membrane filtrations. The major hitch and disadvantage to these technologies emanates from their operational and maintenance cost (Jadhav et al. 2015;Mutamim et al. 2013;Weng et al. 2008). Despite being costly, they are also related to generation of sludge. Monteagudo and co-workers in their work reported sludge disposal as another problem with most of these technologies (Monteagudo and Ortiz 2000;Suksabye et al. 2008). This makes improving water quality a difficult task in areas where the risk of contamination is high (Pradeep and Anshup 2009). Recently, Chouler et al. (2018) reported on screen-printed paper microbial fuel cell biosensor for detection of toxic compounds in water. Though it has a very good detection capability, an important aspect to consider lies on removal efficiency. An alternative technique to microbial fuel cell biosensors which could both detect and remove with high efficiency is nanotechnology.
Nanotechnology is currently the solution to water and wastewater treatment (Dasgupta et al. 2017;Ghasemzadeh et al. 2014;Kunduru et al. 2017;Li et al. 2018;Madhura et al. 2018;Savage and Diallo 2005;Tiwari et al. 2008;Westerhoff et al. 2016;Zhang et al. 2016). Pradeep and coworkers in their work reported the significant progress that has been made to utilize the chemistry of nanomaterial in several ways to improve water quality (Pradeep and Anshup 2009). Qu and co-workers in their research confirmed Pradeep and Anshup's assertion that nanotechnology holds great potential in advancing water and wastewater treatment (Pradeep and Anshup 2009;Qu et al. 2013;Westerhoff et al. 2016). Tremendous attention has been focused on the application of nanostructured materials as adsorbents or catalyst in the removal of toxic and harmful substances from wastewater and air (Ding et al. 2005;Santhosh et al. 2016;Yang et al. 2006;Zhang et al. 2016). For example, Shankar and co-workers reported on using chitosan to remove pentachlorophenol pesticides from aqueous media (Shankar et al. 2020). Eltayeb and Khan (2019) designed a new and novel nanocomposite with CNTs and applied it as a sensor toward environmetal remediation of polutants especially lead(II) ions. Preparation of novel wave-like zirconium tungstate nanocomposite was used for ion-selective studies in aqueous media . This illustrates the potentials of nanoparticle in environmental remediation of pollutants. According to Elangovan et al. (2008), nanoparticle-based adsorption techniques have become versatile methods for removal of metallic pollutants from aqueous solutions. The very small particle size nature (1-100 nm) of nanoparticles makes them extremely versatile remediation tools (Awual 2016). Apart from their size, they can also be magnetized and also attached onto solid matrix such as activated carbon which allows them to be attached to heavy metals (Agarwal and Joshi 2010). The technique has been so effective that even at very low concentrations, metallic ions could be removed (Azimi et al. 2017;Kogan et al. 2006;Shim and Gupta 2007). Agarwal and Joshi (2010) in their work reported reactive nanoparticles appear to be useful in remediating groundwater and thus may prove useful in removing pesticides, herbicides and heavy metals in the environment (Agarwal and Joshi 2010). Baldovi et al. used nanomaterials to understand the origin of the photocatalytic CO 2 reduction which was earlier reported in aqueous media (Baldoví et al. 2015). The applications and removal efficiencies of nanoparticles make it a versatile technology for the present environmental remediations.
This work herein investigates and scout the practicability of the use of silver nanoparticles (AgNPs) in the removal of metal ion contaminants in groundwater samples in the Cape Coast Metropolis, Ghana. The sequence adopted in this work involves the synthesis and characterization of silver nanoparticles which was followed by a proof-of-concept experiment where we tested the efficacy of our nanoparticles using standard solutions of cobalt and lead at varied periods of incubation. Finally, we applied our nanomaterial to groundwater where we confirmed their effective capabilities toward metal ion contaminate removal.

Study areas
Cape Coast Metropolis, a coastal town, with its scarcity of running tap water during the dry seasons was chosen for this research. Assin Fosu (in-land town) was chosen as a control to compare if there could be wide difference in metal ion concentrations between coastal and in-land town. From the map, longitude 1° 11′ 41″ W and latitude 5° 07′ 20″ N defines the location of Cape Coast metropolis as shown in Fig. 1a. Assin Fosu in the Assin North district of the Central Region lies within longitude 1° 25′ 41″ W and latitude 6° 05′ 20″ N of the Central Region of Ghana (Fig. 1b). All samples were taken from the main source of water (groundwater) supply which are suspected to have elevated amount of heavy metals due to the environment these wells are found.

Materials and reagents
UV-Vis spectra were measured using a T70 UV-Vis spectrophotometer (PG Instrument Ltd). Varian 715-ES inductively coupled plasma-optical emission spectroscopy was used for the quantitative analysis of cobalt(II) and lead(II) ions. Field emission scanning electron microscope with model number S4800 FE-SEM system (Hitachi Scientific Equipment, Japan) equipped with an ultrahigh resolution and low voltage 10 kV SEM inspection with advanced sample navigation package was used to characterize the morphology. Centrifuge (model 0257-210) from Wagtech International Limited, UK, was used to separate the adsorbent and treated water. All chemicals such as AgNO 3 , NaBH 4 , HNO 3 , and cobalt and lead standards for ICP-OES were purchased from Merck (Darmstadt, Germany) and were of analytical grade and used without any further treatment unless otherwise stated. Standard solutions of lead and cobalt were prepared using modified procedures as stated. All solutions were prepared using doubly distilled deionized water.

Sample collection, treatment and storage
Groundwater samples were taken from 10 sampling sites. Selected sampling sites from Cape Coast Metropolis were Kotokuraba, Mpeasem, Amamoma, Effutu, Adisadel, OLA, Ebubonku, Ayikoayiko and Pedu; a control sample from Assin Fosu in the Assin North district of the Central Region of Ghana. Nitric acid-treated 1 L polyethylene bottles were used for sampling to prevent traces of metals leaching into the water samples. In addition, nitric acid treatment procedure was intended to preserve trace metals from microbial activity and adsorption losses to the walls of the sample container (Biziuk et al. 2009;Federation and Association 2005;Shugar et al. 2001;Sundaram et al. 2009). The temperature and pH values for the samples recorded were 29 °C and pH units 2.0. Samples were transported and stored in the refrigerator at 4 °C. This is to minimize sample losses through evaporation and any possibility of chemical conversion (Federation and Association 2005).

Synthesis of silver nanoparticles (AgNPs)
AgNPs utilized in this study were synthesized using the method proposed by Mulfinger and co-workers (Mulfinger et al. 2007). The method involved the reduction of Ag + to Ag 0 . In this synthetic process, 1.0 mM AgNO 3 was continually added to 2.0 mM NaBH 4 with continuous stirring for about 20 min on an ice bath. Formation of AgNPs was observed by a change in the color of the mixture from colorless to yellow. The UV-Vis characterization showed the absorption peak of our nanomaterial at 406 nm. Morphological characterization revealed a nanomaterial of size 20 nm which is in line with literature (Mulfinger et al. 2007). To determine the concentration of the synthesized AgNPs, Eqs. (1) and (2) were used which indicated the concentration of our nanomaterial is 4.06 × 10 −7 mol/L. Equation (1) gives the number of atoms per nanoparticle which is substituted into Eq. (2) to obtain the concentration of the prepared AgNPs. N = number of atoms per nanoparticle, ρ = density of face centered cubic (fcc) silver = 10.5 g/cm 3 (1.05 × 10 −20 g/nm 3 ), D = average diameter of nanoparticles = 20 nm, M = atomic mass of silver = 108 g/mol, N A = number of atoms per mole = 6.02 × 10 23 atoms/mol C = molar concentration of nanoparticle solution, N T = Total number of silver atoms added in AgNO 3 = 1.20 × 10 20 atoms, N = number of atoms per nanoparticle (determined in Eq. 2), V = volume of the reaction solution = 2 × 10 −3 L, N A = number of nanoparticles per mole.

Treatment of standard solutions (cobalt and lead) and groundwater samples using AgNPs
Prior to the treatment of standards with AgNPs, standard solutions of cobalt and lead were prepared according to APHA (American Public Health Association) method (Federation and Association 2005). Dilute concentrations of 5, 10, 15, 20 and 25 μg/mL were prepared from the standard stock solutions. To 10 mL portions of these dilute solutions, 0.5 mL portions of AgNPs were added. The incubation period used in this work were 1 day, 3 days, 5 days, 7 days and 14 days. All solutions were centrifuged at 3000 rpm for 10-20 min, and the supernatant portions taken for UV-Vis analysis. Absorbance measurements for both solutions were taken at 510 nm wavelength. Digested groundwater samples were treated with 0.5 mL portions of AgNPs and incubated using the same process as described for the metal standard solutions. Quantitative determinations of metal ion concentrations were carried out using inductively coupled plasmaoptical emission spectrophotometer (ICP-OES). This is to

UV-Vis spectroscopy and SEM characterization of AgNPs
Noble metal nanoparticles exhibit strong UV-Vis absorption band which is absent in the bulk metal. The major drive behind application of UV-Vis in this work is to identify wavelength changes upon surface modification of the nanoparticles. The measured UV-Vis absorption spectra of our nanoparticle fell in the range of 300-700 nm as shown in Fig. 2, (left). The material showed distinctive surface plasmon absorption peak at 406 nm, which is a characteristic of AgNPs. Morphological characterization of AgNPs (Fig. 2, right) showed particles of small sizes (20.3 ± 0.8 nm) with spherical shapes.

Metal-ions adsorption effects on surface plasmon band of AgNPs
After a successful synthesis and characterization of AgNPs, a proof of concept experiment was carried out to ascertain shift in the plasmon band upon surface modification. In this experiment, 0.5 mL of AgNPs was added to about 10 mL cobalt and lead standard solutions. The shift patterns observed upon interaction between silver nanoparticles and metal ions from cobalt and lead solutions are shown in Fig. 3. A remarkable redshift from 406 to 445 nm upon adsorption of cobalt(II) ions onto the surface of AgNPs was observed (Fig. 3b). Adsorption of lead(II) ions, on the other hand, showed an outstanding higher bathochromic shift from 406 to 458 nm as indicated in Fig. 3c. These redshifts as shown in Fig. 3b, c demonstrate changes in the optical properties of the nanoparticles following the adsorption of Co 2+ and Pb 2+ ions. Hypochromic shifts which were also observed further stipulates the surface modifications of the silver nanoparticles. A proposed mechanism was developed showing the adsorption of metal ions to the surface of the Our results presented herein suggest the importance of UV-Vis spectrophotometry in tracking optical properties of nanoparticles upon surface modification. Based on these experimental facts, UV-Vis spectrometry was used to identifiably determine the level of reduction of metal ions from various concentrations of metal standards. In other to achieve this, silver nanoparticles effects before and after days of incubation were studied using 5-25 μg/mL cobalt(II) and lead(II) ion solutions recorded at 510 nm wavelength using UV-Vis spectroscopy technique. Figures 4 and 5 show the UV-Vis spectra of 5 and 25 μg/mL cobalt(II) ion solutions, while Figs. 6 and 7 show that of 5 and 25 μg/mL lead(II) ion solutions before and after days of incubation. The spectra obtained for all the metal ions concentrations (5-25 μg/ mL) used for this experiment before and after nanoparticle incubation are summarized in Tables 1 and 2. As stated earlier in this manuscript, the adsorption of cobalt and lead ions to the surface of AgNPs could result in the decrease in the absorbance intensities an indicative of cobalt and lead ion removal (Sumesh et al. 2011;Yantasee et al. 2007). According to Ball, the decrease in absorbance intensities depicts a decrease in concentration which is consistent with Beer-Lambert's law indicating the removal of metal ion in solution (Ball 2006).
It is important to reiterate that decrease in absorbance correlates with increase in the incubation period suggesting longer incubation period enhances removal efficiencies as shown in Fig. 8. The absorbance values measured at 510 nm for cobalt and lead were converted to percentages using Eq. (3) and the values shown in Tables 3 and 4 respectively. where %M n+ = percent of metal ion, A o = absorbance before nanoparticle addition, A 1 = absorbance after nanoparticle addition.
UV-Vis measurements from Tables 1 and 2 indicated a decrease in absorbance measurements for Co 2+ and Pb 2+ solutions as the incubation periods increase an indicative of greater adsorption of metal ion to the surface of the nanoparticles. This observation gave indication that our AgNPs could possibly remove Co 2+ and Pb 2+ in an aqueous environment. The data obtained from the spectra were converted to   Table 3 showed apart from 5 μg/mL, the removal efficiencies for 10, 15, 20 and 25 μg/mL were below 50%. This could mean a higher molar concentration for the AgNPs will be the most preferred for concentrations higher than 5 μg/ mL than the one used whose concentration is 0.0438 ppm (4.06 × 10 −7 mol/L) as determined from Eq. 2. Pb 2+ complex solutions of 5, 10, 15, 20 and 25 μg/mL when incubated with 0.5 mL of AgNPs from days 1, 3, 5, 7 and 14 showed a trend indicating that the absorbances reduced as the number of days increased (Alissawi et al. 2013;Alqudami et al. 2013). These observations as shown in Table 2 also confirm the ability of AgNPs to remove lead. As presented in Table 4, the maximum percent reduction was recorded for all the concentrations on day 14. Thus 5, 10, 15, 20 and 25 μg/mL had the concentrations of Pb 2+ in the complex reduced by as low as 88.56, 92.92, 84.10, 80.11 and 79.92%, respectively, which are much higher than the percentages of Co 2+ removed on the same day with the same amount of AgNPs applied. This high level of adsorption illustrates our nanomaterial is highly efficient and best choice for Pb 2+ removal (Shiri-Yekta et al. 2013). The time dependency of nanoparticles as indicated by Alqudami et al. (2013) and Alissawi et al. (2013) was once again proven by our system. The time dependence of nanoparticle adsorption effects in Pb 2+ solution is illustrated in Fig. 8 which illustrates AgNPs is a potential candidate for metal ion contaminants removal in groundwater samples.

Quantitative determination of Co 2+ and Pb 2+ ions in groundwater samples
Since the key objective of this study was to remove Co 2+ and Pb 2+ ions from selected groundwater/wells, there was the need to apply a quantitative methodology to help determine amounts of metal ions removed within the samples. ICP-OES technique was used to determine the amounts of metal ions removed. Firstly, ten (10) samples from our 10 sampling sites were analyzed using ICP-OES to provide a baseline for comparison as shown in    Odoh et al. 2011). It is important to note that Assin Fosu recording the highest Co 2+ concentration debunk the notion that sample sites closer to the sea might have higher concentrations. Another confirmation of the above statement is shown in Pb 2+ concentrations where apart from Amamoma, (coastal zone) the areas away from the sea (Effutu, Mpeasem, and Assin Fosu) still exhibited higher levels of lead meaning sea water intrusion may have very little or no effect on detectable metal ions from groundwater. Table 6 shows the mean values of three different ICP-OES analysis results of groundwater samples after AgNPs incubation from 24 h to 14 days. The choice of these incubation periods is to mimic the procedure employed in the analysis of standard solutions of Co 2+ and Pb 2+ ions presented earlier in this paper. Comparing the results in Tables 5 and 6, it can be concluded that the addition of AgNPs to the samples reduced the metal ion concentrations from the samples as indicated by a decrease in metal ion concentrations. The percent reductions of the various metals in each respective sample were computed relative to Tables 5 and 6 using Eq.
(3) and shown in Table 7. The computation from kotokuraba samples shows a reduction percent of cobalt from 1.20, 4.16, 6.02, 19.16 and 20.58% on days 1, 3, 5, 7 and 14, respectively. Pb 2+ ion percent reduction from same Kotokuraba samples were 8.83, 24.70, 32.30, 47.66 and 48.32% on days 1, 3, 5, 7 and 14, respectively. These results show a pattern indicative of reduced concentrations from day 1 to day 14. Although the percent reductions were minimal, we can say that the AgNPs were able to reduce metal ions that were initially contained in the samples. Further observation of Table 7 reveals that Mpeasem samples showed maximum removal efficiencies at 15.45% and 72.55% for Co 2+ and Pb 2+ ions, respectively. Assin Fosu, the controlled sample site also showed Co 2+ ion concentrations reduced from 8.14% in day 1 to 24.02% on   (Table 4) removed were higher than Co 2+ (Table 3) on day 14, respectively. This could mean there are other substances in the Co 2+ solution that the nanoparticle has more affinity for and hence adsorbed them leaving a less surface area for the target ions adsorption implying that AgNPs may work effectively as Pb 2+ ion removal in aqueous media than Co 2+ . It could also suggest that a higher AgNPs concentration than the one prepared might be more effective toward the removal of the Co 2+ ions in solution. It is important to note that AgNPs removal efficiencies as observed in the proof of concept experiment were higher than those recorded for groundwater samples. The result though not as expected is still considerable owing to the complex nature of groundwater analysis. We also want to reiterate that metal ion adsorption begins immediately after AgNPs were added

Conclusion
Silver nanoparticle was synthesized and characterized using UV-Vis spectrophotometer and scanning electron microscope. The UV-Vis spectra showed the characteristics surface plasmon absorption peak of 406 nm for the nanoparticle. The morphology of the nanoparticle was analyzed using scanning electron microscope which showed a mean diameter of 20 nm. The choice of silver nanoparticle in this research was due to its ease of synthesis, inexpensive starting materials, produces no sludge when used in water treatment and high metal adsorption capacity as discussed in earlier. The research presented herein showed that nanoparticles efficiencies toward metal ions removal in metal complex standard solutions are high as compared to that of groundwater. This observation we believe might be due to the complex nature of underground water samples. Nevertheless, the results indicate metal ion removal in both metal ion complex and aqueous media is time-dependent. These observations suggest a further research into understanding the adsorption kinetics of metal ions to the surface of nanomaterials is needed.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.