Magnetic Nanoparticles for Toxic Wastewater Cleaning:Experimental Study on Phenol ^†

Abstract

This study focuses on the possibility of cleaning of industrial wastewater with catalytically active magnetic nanoparticles. Cobalt ferrite synthesized by the co-precipitation method was used, as prepared or after surface modification with a silica precursor. Electronic absorption spectra were recorded and analyzed to obtain the phenol degrading rate for various experimental design variants. Treating with pristine magnetic nanoparticles under simultaneous exposure to ultraviolet radiation resulted in similar degrading rates for 4 g/L and 8 g/L pristine nanoparticles, while, for silanized nanoparticles, the degrading rates were slightly increased. Along with ultraviolet irradiation and magnetic nanoparticles, hydrogen peroxide was also added, which led to significant enhancement of phenol degradation, for both pristine and silanized nanoparticles. It is proposed that photo-Fenton processes, triggered by metal ions at the nanoparticle surface and water photolysis and sustained by hydrogen peroxide decomposition, occurred to gradually decompose phenol to simpler compounds.

1. Introduction

Industrial wastewater issues related to biosphere safety and human health preoccupy researchers today more than ever. Since nanotechnology development revealed new methods for water cleaning, based mainly on nanoparticle interaction with the chemicals polluting used waters, magnetic nanoparticles (MNPs) have been much studied with regard to the catalyst properties of the metal ions in their composition [1,2,3,4]. The photo-catalyst properties of cobalt ferrite result from its crystalline structure, which has many surface active sites that ensure high Fenton catalytic activity and attract scientists’ attention for their utilization in removing heavy metals [5,6] and in various anionic and cationic dyes [7,8,9,10,11,12]. The supplementary addition of hydrogen peroxide (H₂O₂) or ultraviolet radiation (UV) exposure has also been studied [13,14,15,16], being successfully applied to enlarge the practical procedure array for pollutant degradation, such as in the case of phenol. We present the results of our experiments with a phenolic wastewater model treated with cobalt ferrite pristine nanoparticles and silanized ones.

2. Experimental

2.1. Reagents

Ferric chloride hexahydrate (FeCl₃×6H₂O), cobalt sulfate heptahydrate (CoSO₄×7H₂O), sodium hydroxide (NaOH), citric acid (C₆H₈O₇), TEOS (tetraethylorthosilicate, Si(OC₂H₅)₄), phenol (C₆H₆O), and hydrogen peroxide (H₂O₂) were acquired from Sigma-Aldrich, part of Merck KGaA, Darmstadt, Germany and used without additional purification.

2.2. Synthesis

According to the adapted Massart method [17], the precursor metal salts were dissolved in deionized water with respect to their stoichiometry, i.e., ferric chloride 3.62 g in 100 mL deionized water (134 mM), and cobalt sulfate 1.88 g in 50 mLaqueous solution (67 mM). Magnetic stirring was carried out on a heating plate (DLAB Scientific CO., LTD., Beijing, China), at approximately 80 °C, and then hot 1.7 M sodium hydroxide (3.4 g in 50 mL deionized water) was carefully poured, to trigger co-precipitation of magnetic nanoparticles.

The brown-black sludge was separated from the reaction medium using a permanent laboratory magnet, and was then washed three times consecutively with 200 Mlof deionized water to eliminate unwanted reaction products. After drying in a low-vacuum oven, the sample of CoFe₂O₄ magnetic nanoparticles was obtained. To prepare the coated cobalt ferrite sample, the same procedure was carried out and, first, 1.7 g of citric acid in 3.4 mL of deionized water was added to the wet powder under mechanical stirring (1200 rpm) for one hour in a water bath at a temperature of 90 °C.

2.3. Stabilization and Silanization

The colloidal particles thus stabilized with citrate ions were washed with deionized water repeatedly to eliminate excess citrate until the pH reached ~5.5, which was tested with pH paper since applications in life sciences favor this method [18]. Secondly, particle silanization was carried out by applying an adapted variant of the sol–gel method [19]. According to this, 0.25 g of magnetic nanoparticles withdrawn from citrate-stabilized suspensions and dried were dispersed in water of an amount up to 7 mL and mixed with 35 mL of 2-propanol under continuous mechanical stirring, then with 0.07 g of sodium hydroxide and 1 mL tetraethylorthosilicate, the vigorous stirring continuing for three hours.

Then, the silica-coated particles were separated from the reaction medium by centrifugation at 3500 rpm and washed repeatedly with deionized water until the pH reached ~6. Finally, the waxy (gelatinous) magnetic material was dried under vacuum at 90 °C for 6 h according to [19] and then heat-treated for three hours at 165 °C to complete the synthesis of the composites.

2.4. Phenol Wastewater Model

Phenol was used, which is the most widely used pollutant model for wastewater treatment studies because it was found to be one of the most toxic of the recalcitrant water pollutants [20].

For the phenol solution, the concentration of 5.0 μM proved to be the most appropriate for evidencing the degrading rate of magnetic nanoparticles, as our preliminary studies showed.

2.5. Ultraviolet Radiation Exposure

UV radiation exposure of the phenolic water samples treated with magnetic nanoparticles (in the C subdomain of UV range) was performed using a low-pressure mercury vapor discharge tube (from Philips, Signify Romania, București, Romania), with a length of L = 0.87 m, which emits UV-C radiation with a maximum at 253.7 nm (corresponding to the mercury spectrum)—while the glass walls of the tube filter the ozone line (from 185 nm). The total emission power was 30 W, of which, according to the manufacturer’s specifications, the power emitted in the UV-C range was P = 12 W. The vessel with 10 mL of phenolic water sample, with an exposed surface area of 12.57 cm² (diameter of 4 cm), was placed at D = 20 cm below the center of the discharge tube.

2.6. Microstructural and Spectral Investigations

Considering that the total surface area of interaction with water is the determining factor of the efficiency of this process, favored precisely by the high surface area-to-volume ratio, the dimensional and morphological characteristics were analyzed using the TEM (Transmission Electron Microscopy) visualization and measurement technique (Hitachi High-Tech HT7700 model device from Hitachi High-Tech Corporation, Tokyo, Japan).

A Shimadzu PharmaSpec 1700 spectrophotometer, from Shimadzu Corporation, Tokyo, Japan (equipped with quartz cuvettes and specialized software for data recording and processing) was used to record the electronic absorption spectra of the wastewater model samples at different time points, between 0 and 120 min. The nanoparticles were removed from the reaction medium by magnetic decantation, before the sample absorbance was recorded. The determinations were made in triplicate, with double-distilled water as reference, and the average values were represented on graphs. The standard deviation was 3–4%. The rate of decrease in phenol concentration was evaluated using the relationship below, in accordance with the Beer–Lambert law (the concentration of the absorbing molecule, phenol in this case, is directly proportional to the light absorbance in the sample in the spectral maximum at a wavelength of 270 nm [21] relative to a control sample):

Phenol degrading rate (%) = ((A₀ − A_f) × 100)/A₀

(1)

where A₀ is absorbance corresponding to the initial concentration of phenol and A_f is absorbance corresponding to the concentration after degradation.

3. Results and Discussion

TEM images are presented in Figure 1, where particles with rather symmetrical geometrical shapes can be seen—quasi-spherical but also polyhedral with apparently rare association tendency.

The mean values (Dm) of particle size were estimated by the histogram approximation with statistical function. As a result of stabilization by surface modification with silica, the size of the most frequent nanocomposites increases significantly, from about 20 nm in the case of pristine nanoparticles (Figure 1a) to about 28nmfor the silanized particles (Figure 1b). The latter appear to be found not only in a single layer, but can also be overlapped within the same silica matrix volume, locally generating images with intense blackening.

According to [3], because of nanometric size and narrow energy band gaps, the cobalt ferrite nanoparticles tend to agglomerate and, during the catalytic reactions they are involved in, the photogenerated carriers can thus rapidly recombine.

In previous studies [22], it has been highlighted that some silanized nanoparticles are grouped in the same silica matrix, structures that may appear as aggregates or overlapped particles when investigated by the TEM method. In Figure 2, the influence of UV-C radiation exposure on the phenol wastewater model is given in the presence of magnetic nanoparticles as well as in their absence.

It is known that UV rays can penetrate only a very thin molecular layer, but, due to the continuous stirring of the sample, the volume molecules were able to reach the surface of the fluid sample, while, at the same time, the free water radicals released following surface water photolysis were able to pass into the sample volume.

The efficacy of the process strongly depends on the sample homogenization, which seems to be quite a sensitive operation, especially because control of it seems to be a difficult task. We observed that irradiation alone is able to lead to a degradation rate of about 28% for 120 min (Figure 2). The literature mentions that degradation under exposure to UV radiation alone can continue up to 100%, but over a long period of time—24 h [20]. Shortening this time interval can be achieved precisely by the concomitant use of nanoparticles. According to Figure 3a, supplementation with hydrogen peroxide significantly improved the phenol degrading rate. The influence of hydrogen peroxide of 20 mM for 8 g/L MNP under UV-C exposure led to a degrading rate of about 70% over 120 min with continuous careful stirring (Figure 3a), which is better than for a MNP concentration of 4 g/L (Figure 3b), when the maximal phenol degradation level was about 64%.

The Fenton reactions involving iron ions at the cobalt ferrite MNP surface can be written as:

Fe²⁺ + H₂O₂ → Fe³⁺ + HO⁻ + HO^•

(2)

Fe³⁺ + H₂O + (hν) → Fe²⁺ + H⁺ + HO^•

(3)

HO^• + RH → H₂O + R^•

(4)

where hν is an UV photon, and R^• is a molecular radical—for example, in a RH molecule as the phenol. Reactions (3) and (4) are responsible for the production of the hydroxyl radical which, according to (4), can oxidize organic compounds (RH). We note that the dot ^• marks the unpaired electron that characterizes electrically neutral free radicals. The efficiency of the photo-Fenton process has been attributed to the reactivity of the HO^• group, which can attack various chemical species present in the liquid phase. In the literature, photo-Fenton-type catalytic activity of cobalt has also been highlighted [23], even on phenol and in the absence of iron ions (for example, using zeolite structures carrying Co²⁺ ions).

However, over a period of time of approximately 200 min, the amounts of 2 g/L of zeolite carrying cobalt ions caused, through oxidation reactions, a decrease in the phenol concentration of only 12%, the main oxidation product being the hydroquinone. The MNP interaction with phenol and phenol degradation products could alsoconsists of the adsorption of some molecules from the reaction medium, which could also be presumed to occur in the case of silanized MNPs. According to Figure 4, in the absence of supplemental hydrogen peroxide but under exposure to ultraviolet radiation, the phenol degradation rate was up to 32%, similar to the case of pristine MNPs, suggesting that the release of hydroxyl radicals through water photolysis represents the main cause of the process. Higher amounts of hydroxyl free radicals, like those originating in the decomposition of the instable hydrogen peroxide, resulted in phenol degrading by up to 42% for 8 g/L silanized MNPs—for 120 min of UV-C exposure (Figure 5a)—and up to 31% for 4 g/L silanized MNPs (Figure 5b). In Figure 6, the decrease in the electronic absorption spectrum, at 270 nm, can be seen.

The fact that the silanized nanoparticles can actually be nanoparticle groups in the same silica matrix [19] may be a cause of the decrease in the total interface area with the phenolic environment.

Thus, the lower rate of phenol degrading was evidenced after cobalt ferrite nanoparticle silanization compared to pristine ones, but the photocatalytic processes were still considered to be the main cause overcoming the putative surface adsorption. To demonstrate that phenol can be adsorbed without decomposition on silica structures [24], we refer to the specialized studies [14] that demonstrate that hydroxyl radicals produced by photo-Fenton processes can generate hydroxylated by products. During irradiation, the absorption spectrum we recorded for the phenol sample diminished in intensity as result of degrading through photo-Fenton processes triggered by hydrogen peroxide supply and UV-C exposure, with new unstable species (catechol and benzoquinone) being generated, as presented in Figure 6. The attack of HO^• radicals leads to the addition of another HO^• group in the ortho or para positions, which leads to the formation of catechol, then of ortho-benzoquinone, hydroxyhydroquinone, and hydroxybenzoquinone, which can be further decomposed into smaller hydrocarbon fragments and finally mineralized (carbon dioxide and water). It should be noted that, when the degradation process is not completed for all phenol molecules, it is possible that intermediate catechol or hydroquinones still absorb in the ultraviolet domain [25,26], at 260–280 nm; thus, the spectra of phenolic wastewater treated with MNPs, H₂O₂, and UV-C are still present with non-zeroed intensity after treatment.

4. Conclusions

The experiments we carried out, designed to study phenolic wastewater cleaning with magnetic nanoparticles, revealed some factors influencing the efficacy of the process. While under ultraviolet radiation alone, the degradation rate was practically the same in the presence or absence of magnetic nanoparticles (of 27–28%); the supplemental treatment with 20 mM hydrogen peroxide and 8 g/l pristine nanoparticles led to a degrading rate of up to 70%. In such circumstances, the photo-Fenton reactions involving direct interaction of metal ions with the surrounding dispersion medium were supposed to catalytically transform the phenol in byproducts with gradually diminished absorption in ultraviolet range, close to the phenol absorption band, at 270 nm. Nevertheless, the adsorption of pollutant molecules to the nanoparticle surface could have a significant role in the decrease in the intensity of the phenol absorption band, especially for silanized particles where the direct contact of metal ions with the aqueous medium is lower.

This study could be further developed by considering partial doping with cobalt of magnetite nanoparticles and testing different pollutant molecules.