Introduction

The prevalence and detection of residual pharmaceuticals and antibiotic-resistant bacteria (ARB) in effluents of wastewater treatment plants (WWTPs) are a global concern (Wang et al. 2020; Faleye et al. 2019). Most of the pharmaceuticals get excreted in their unmetabolized form either through urine or feces (Gašo-Sokač et al. 2017). Despite the low levels of pharmaceuticals detected in the environment, continued exposure poses substantial risks as their long-term effects in humans and animals remain unclear. Seasonal fluctuations also affect the frequency and concentrations of pharmaceuticals detected; the highest concentrations of pharmaceuticals are often detected during the winter season, and this is related to the increased utilization of medicine during winter when more people are prone to various illnesses (Wu et al. 2016).

The current study reports on a widely used class of pharmaceuticals known as sulfonamides which account for 16–21% of all antibiotic consumption annually (Göbel et al. 2005). Due to their extensive use, sulfonamides have been detected in surface water, secondary effluents and sludge (Paumelle et al. 2021; Ngigi et al. 2020; Raich-Montiu et al. 2007; Batt et al. 2007). For example, concentrations of sulfonamides varying from 7 to 88 µg/L have been reported (Wang et al. 2020). On the other hand, concentrations between 1.8 and 7.2 µM of sulfasalazine (SSZ) were reported to constitute significant threats to antimicrobial resistance. Additionally, the presence of sulfonamides can also increase antimicrobial resistance (Neafsey et al. 2010). Wang et al. (2020) reviewed the removal efficiencies of sulfonamides in full-scale treatment plants, and the finding showed that sulfonamides have a high propensity to resist biodegradation and generally exhibit a low removal efficiency of 52.6%. Other sulfonamide removal efficiencies were reported in the ranges 38–74%, 33–75%, as well as 33% and 35%, respectively (Blair et al. 2015; Kasprzyk-Hordern et al. 2009; Snyder et al. 2007).

Most conventional processes in WWTPs were constructed more than 50 years. These WWTPs are ineffective in removing sulfonamides because they were not designed to deal with pharmaceutical contaminants. WWTPs were primarily designed to eliminate pollutants with chemical oxygen demand (COD) removal in the range 80–95% which is estimated to be 25% higher than most pharmaceuticals. As it stands, the removal efficiencies of antibiotics range between 40 and 70% and can be compared to the removal of total nitrogen in WWTPs (Qiu et al. 2010). The possibility of using disinfection (chlorination or UV) in the abatement of sulfanomides in WWTPs has been investigated previously. The reports revealed that after chlorination or UV treatment sulfamethoxazole concentrations were reduced to between 10 adn 70 ng/L and 20 and 40 ng/L, respectively (Yan et al. 2022). Although disinfection (chlorination and/or UV treatment) showed ability to treat sulfamethoxazole in the WWTPs, concerns relating to the generation of toxic disinfection by-products (DBPs) and the costly treatments cannot be ignored (Mazhar et al. 2020; Achour and Chabbi 2014). In addressing some of the limitations such as DBPs, researchers are suggesting the degradation of SSZ using advanced oxidative processes (AOPs). AOPs can be employed as an alternative standalone method or can be a hybrid complementing in existing water treatment procedures (Fan et al. 2011; Ji et al. 2018; Pelalak et al. 2020).

In AOPs, the highly reactive species generated via chemical and photochemical reactions are the main oxidants responsible for the degradation of contaminants. The most used peroxides in AOPs as sources of reactive radicals include persoxydisulfate (PDS), hydrogen peroxide (H2O2) and peroxymonosulfate (PMS). These peroxides were previously activated either by using transition metals, ultraviolet (UV) or by applying thermal treatment (Keyikoglu et al. 2020; Jazić et al. 2020; Chen et al. 2016). Peracetic acid (PAA; CH3 C(O)OOH) is an organic peroxide that has great potential of producing similar highly reactive species (Chen et al. 2019). PAA has also been considered as a possible replacement for chlorine-based oxidants in the wastewater treatment largely because it does not generate toxic by-products during the disinfection process (Kitis 2004; Rossi et al. 2007).

In the current work, the use of a low-impact process that can simultaneously degrade pharmaceuticals and inactivate microorganisms using peracetic acid (PAA) activated by metal-free graphene oxide quantum dots (GQDs) is proposed. Unlike, in previously reported articles on the activation of PAA (Keyikoglu et al. 2020; Jazić et al. 2020; Chen et al. 2016), this study employs carbon-based nanomaterials and very little work has been openly published in this regard. The activation of PAA by GQDs typically generates free radicals [i.e., acetlyoxyl (CH3C(O)O·) and/or acetylperoxy CH3C(O)OO·] that are not prone to scavenging in complex water matrices by other possible ions (Tshangana et al. 2022; Cai et al. 2017; Rossi et al. 2007). Unlike chlorine, PAA exhibits disinfectant capabilities that are not dependent on pH, and it has been shown that PAA does not generate DBPs in the treated effluent (Kitis 2004). One of the core objectives of this work was therefore to determine and investigate the mutagenic potential of the reaction by-products of sulfasalazine degradation.

The main objectives of this study were to (a) investigate the degradation efficiency and kinetics of GQDs/PAA (keeping the concentration of GQDs constant while varying the concentrations of PAA) system against SSZ, (b) identify the main reactive radicals responsible for the degradation of SZZ, (c) investigate the mutagenic potential of the reaction by-products, (d) evaluate the effect of water matrices, initial concentration of SSZ and PAA dosage on SSZ degradation, (e) propose a photodegradation mechanistic pathway based on the detected reaction by-products determined by LC-Q-TOF-MS and finally (f) evaluate the inactivation of microbes in real wastewater and synthetic water with or without spiking the sample with S. aureus by GQDs/PAA. Further this study proposes the inactivation mechanism of GQDs/PAA by observing morphological changes on SEM and confirms the viability of the bacterial cells using the LIVE/DEAD cell viability test kit.

Materials and methods

Chemicals

Graphene oxide quantum dots suspension (1000 mg/L in H2O), peracetic acid solution (36–40 wt% in acetic acid) and the chemicals used in determining reactive radical species (p-benzoquinone, 2,4-hexadiene, ethylenediaminetetraacetic acid disodium salt (EDTA-2Na), silver nitrate and methanol) were obtained from Sigma-Aldrich, South Africa. Sulfasalazine (Fig. 1), acetic acid, hydrogen peroxide, physiological saline solution, iodonitrotetrazolium chloride (INT), Salmonella typhimurium tester strains TA98 and TA100, nutrient broth, sodium thiosulfate, the 2,2 diphenyl-1-picrylhyrazyl (DPPH) glutaraldehyde, ascorbic acid and catalase were obtained from Merck, South Africa. The pH values of the solutions were adjusted using HCl or NaOH (0.1 and 0.01 M). All chemicals were used as received without further purification. Throughout the study, Milli-Q ultra-pure water was used, unless otherwise noted.

Fig. 1
figure 1

Molecular structure of sulfasalazine (C18H14N14O5S)

Full size image

Sampling site and sample collection

The wastewater (WW) effluent sample was collected from a wastewater treatment plant in the Gauteng province (South Africa) and the water quality parameters (WQPs) are summarized in Table 1. Wastewater in this treatment plant passes through the following stages: (a) primary clarifier, (b) an aeration tank, (c) secondary clarifier and (d) through the disinfection process. For this study, the chosen sampling points were before the primary clarifier and after the secondary clarifier (labeled sampling point 1 and 2, respectively) as shown in the schematically in Scheme 1.

Table 1 Water quality parameters of the collected wastewater
Full size table
Scheme 1
scheme 1

Representation of the wastewater treatment plant where the samples were collected

Full size image

LC-QTOF-MS analysis

An Agilent 1200 binary pump system was used to perform LC-QTOF-MS analysis on an Acquity BEH C18 column (2.1 × 100 mm, 1.7 µm): the mobile phase (solvent A) (water containing 40 mM ammonium acetate with 2.5% acetonitrile at pH 7.8 adjusted with ammonia solution 2.5%) and solvent B (acetonitrile); after injection, isocratic conditions (100%) mobile phase A for 0.1 min and then linear gradient to 95% mobile phase over 2 min; followed by a gradient up to 55% mobile phase A in 8 min and isocratic conditions of 55% mobile phase A for a further 0.4 min before returning to 100% in 0.1 min; and followed by a reconditioning step at 100% mobile phase A for 5 min. The flow rate was 0.40 mL/min at 50 °C.

Detection conditions

Mass spectrometry (MS) tuning was conducted in negative electrospray ionization (ESI) mode. This was accomplished by employing a T-connector to infuse each analyte solution separately (concentration = 10 µg/mL), flow rate of 10 µL/min mixed with HPLC flow composed of solvent A and B (50: 50, v/v; 0.40 mL/min). Quantitative analysis was carried out using tandem MS in selected reaction monitoring (SRM) mode, alternating two or three transition reactions with varying dwell periods for each molecule.

Photodegradation of sulfasalazine (SSZ)

The degradation of SSZ experiments was performed at room temperature (20–22 °C) using a custom-made photoreactor from Lelesil Innovative Systems (India). A 250 W medium-pressure mercury vapor lamp of quartz bulb was used to irradiate light each aqueous sample solution where it was needed (Fig. 2). To determine the effect of initial SSZ concentration on the photodegradation process, three different concentrations of SSZ were used (200, 300 and 500 µM). The pH optimization studies were carried out, and the optimal pH value was 5. Other researchers (Omrania et al. 2019; Wu et al. 2019) have also reported pH 5 as ideal in the photodegradation of sulfasalazine and acetaminophen (Ghanbari et al. 2021). To this end, the pH of each of the SSZ solutions was maintained at pH 5 by adding (1:1) of 1 M HCl/NaOH. A set of experiments were conducted with real wastewater samples collected at the plant and with synthetic water from the laboratory. Both real wastewater and synthetic water samples (40 mL) were spiked with SSZ (10 mL; 10 ppm) to ensure detection with analytical instruments. For the photodegradation experiments, 150 mg/L of GQDs was added to the SSZ solution (10 ppm). The solutions (mixture of GQDs and SSZ) were initially mixed in the dark under stirring for 30 min, after which the UV–Vis light was turned on. The start of the reaction was timed shortly after adding the PAA at various concentrations (0.5, 1.0 and 1.5 mM). Aliquots (4 mL) were collected at 30-min intervals for 150 min and were immediately injected into a 1 mL phosphate buffer solution (0.20 mol/L, pH 8) to quench the reaction. The aliquots were further filtered through a 0.45 µm PVDF filter and analyzed using a UV–Vis spectrophotometer.

Fig. 2
figure 2

Schematic representation of the photoreactor by Lelesil Innovative Systems used for the photodegradation of SSZ. Visible light mode of the reactor system was used in this study while the UV controller was turned off

Full size image

The absorbance of the filtrate was measured using a spectrophotometer to assess variations in SSZ concentrations under visible light irradiation. The percentage degradation of SSZ was calculated using Eq. 1.

$${\text{Sulfasalizine}}\;{\text{degradation}}\,\% = \frac{{\left( {C_{0 } - C_{t} } \right)}}{{C_{0} }} \times 100\%$$
(1)

where C0 and Ct represent the sulfasalazine concentration in the solution before and after irradiation, respectively.

The kinetic data (kinetic rate constants (k) and half-live (\(t_{\frac{1}{2}}\))) of the photodegradation of SSZ were analyzed and calculated employing the integrated Langmuir–Hinshelwood (L–H) model Eq. 2 (Muleja and Mamba 2018):

$$\ln \left( {\frac{C}{{C_{{\text{o}}} }}} \right) = - k_{1} t$$
(2)

where \(C\) and \(C_{{\text{o}}}\) are the concentration of SSZ after and before photodegradation experiments and k1 is the pseudo-first-order rate constant. The determination of coefficient (R2) values was used to determine goodnessof model fitness.

The half lifetimes were obtained using Eq. 3

$$t_{\frac{1}{2}} = \frac{0.693}{{k^{{\prime }} }}$$
(3)

Radical species identification

The radical scavenging experiments were conducted similarly as detailed for the photodegradation experiments. The sole difference with previous experiments consisted of adding four radical scavengers at the beginning of each separate experiment. Silver nitrate, ethylenediaminetetra acetic acid diasoduim (EDTA-2Na), methanol and benzoquinone were used as scavengers at a concentration of 60 mM (Shafaee et al. 2018). In the case of acetyloxyl radicals, 4.0 mM of 2,4-hexadiene was used as a quencher (Li et al. 2021). Thereafter, the aliquots were filtered and analyzed using a UV–Vis spectrophotometer.

Determining mutagenic and genotoxic potential of possible reaction by-products from SSZ photodegradation using Ames test

The mutagenic and genotoxic potential of the reaction by-products formed during the SSZ degradation was tested using a Salmonella microsome assay as detailed by Mortelmans and Zeiger (2000). In the Ames assay, two Salmonella typhirium tester strains TA98 and TA100 without metabolic activation were used. 100 µL of the bacterial strains was incubated in Oxoid No.2 broth (20 mL) at 37 °C on a rotary shaker for 12 h. The cultured Salmonella typhirium tester strains (100 µL) were added to 100 µL of the aliquots collected at 0, 30, 60, 90, 120 and 150 min with 500 µL of phosphate buffer and top agar (2 mL) made up of biotin histidine (0.5 mM). From this, the top agar mixture was added to cover the surface of agar plate and incubated for 2 days at 37 °C. The positive and negative controls in the work were 2-nitroflurene for the TA98 and nitrofurantoin for T100.

Antimicrobial activity

Antimicrobial activity of a representative bacteria (S. aureus)

The model microbe used in this work was S. aureus (ATCC 25923); the efficiency of the antimicrobial activity of the GQDs/PAA system was tested against S. aureus. The experiments were conducted using the serial dilution method as reported by Elisha et al. (2017). Stock solutions of the GQDs, PAA and GQDs/PAA (1 mL) were used to serially dilute 96-well plates containing an overnight culture of S. aureus. The plates were incubated overnight at 37 °C and stained with iodonitrotetrazolium chloride (INT). All experiments were carried out in triplicates.

Antimicrobial activity in raw water

The experiments were similarly carried out as described for the photodegradation of SSZ with the difference that in this case SSZ solution was spiked with S. aureus. Another difference was that after quenching the PAA, 20 µL of the aliquots was collected at various intervals of time (0, 30, 60, 90, 120 and 150 min) and subsequently plated on nutrient agar (NA) plates. The nutrient agar plates were incubated at 37 °C, and the number of colonies were counted.

Scanning electron microscopy (SEM) analysis of S. aureus

S. aureus cells (logarithmic phase) were exposed to GQDs/PAA and irradiated using UV–Vis light for 10 min. The S. aureus cells were collected every 30 s and centrifuged at 5000 rpm for 3 min before being washed three times with sterile saline solution. The S. aureus cells were fixed overnight at 4 °C with glutaraldehyde and dehydrated with sequential treatment of 50%, 70%, 85%, 90% and 100% ethanol for 10 min each and gold sputter coated and observed using SEM instrument.

Cell viability using LIVE/DEAD Baclight staining kit

LIVE/DEAD Baclight staining kit was used to determine the viability of the S. aureus cells. The kit consisted of dyes, nucleic acid-binding SYTO® 9 and propidium iodide. SYTO® 9 can penetrate all bacterial membranes and stain green all S. aureus cells with intact cell membranes, while propidium iodide can only penetrate S. aureus cells with damaged membranes which are considered dead or to be dying. The viability tests were conducted as per manufacturer’s guidelines, where 50 µL of SYTO® 9 and propidium iodide were mixed in a microfuge tube. 3 µL of this mixture was added to the GQDs/PAA treated with S. aureus suspension. The microfuge tubes were incubated for 30 min in the dark at room temperature. Following that, 5 µL of the stained S. aureus cells was pipetted onto a microscope slide, covered with a coverslip and imaged using a Zeiss laser scanning microscope LSM 780.

Results and discussion

Photodegradation of SSZ

Influence of PAA dosage on the photodegradation of SSZ

The dose of the oxidant was an important parameter in establishing whether GQDs/PAA system was an effective method for degrading SSZ. The effect of the dosage of PAA on the photodegradation of SSZ was evaluated at pH 5 by varying the PAA concentration (0.05, 0.10 and 0.15 mM) while maintaining the GQDs concentration constant (50 mg/L). When 0.05 mM of PAA was added to the GQDs/PAA system, 80% of the drug was degraded after 150 min, while 99% and 99.8% of SZZ were degraded when using 0.10 and 0.15 mM of PAA, respectively (Fig. 3a). Increasing the PAA concentration resulted in almost complete degradation of the SSZ, implying that the trend observed (Fig. 3a) was dose-dependent. Interestingly, there was a negligible increment in the kobs value from 0.036 to 0.038 min−1 at higher PAA concentrations. While the kobs value increased at higher PAA concentrations, the growth rate remained almost the same which was ascribed to the PAA scavenging effect on the ·OH. Moreover, the phenomena were attributed to the limited reactive sites on GQDs available to accelerate the PAA activation in agreement with the literature (Chen et al. 2016). Another possibility is that the excessive PAA molecules can possibly absorb more photons which in turn inhibit the photodegradation of SSZ or the high loading of GQDs may produce too many reactive sites for PAA activation, quenching the reactive species produced by PAA activation (Zhang et al. 2020). Although the literature on the degradation of SZZ using PAA or GQDs is not readily available, Cai et al. (2017) investigated the UV/PAA on the degradation of seven pharmaceuticals and the results showed no significant degradation of the pharmaceuticals when using 1 mg/L (13.1 µM) of PAA at pH 7. When increasing the PAA dosage to 1 g/L and adding 0.11 g/L H2O2, Cai et al. (2017) reported 62.50% destruction for naproxen and 29.70% for diclofenac and the removal rate for the other pharmaceuticals was less than 11%. The degradation efficiencies of the GQDs/PAA in this study were all above 80% indicating improved removal efficiencies.

Fig. 3
figure 3

a Influence of PAA dosage from 0.05 to 0.15 mM on the photodegradation of SSZ while keeping the concentration of the GQDs constant (50 mg/L: 0.10 mM) and pH 5; b influence of initial concentration of SSZ from 10 to 50 mg/L on the photodegradation process while keeping the concentration of the GQDs constant (0.05 g/L: 0.10 mM) and pH 5

Full size image

Effect of initial SSZ concentration

The initial concentration of SSZ was varied (10–50 mg/L) while maintaining the concentration of GQDs/PAA constant (50 mg/L: 0.10 mM) at pH 5 (Fig. 3b). From Fig. 3b, the lowest concentration of SSZ (10 mg/L) achieved the highest photodegradation efficiency of 100%, while the highest concentration of SSZ (50 mg/L) recorded the lowest removal efficiency of 20%. The degradation efficiency of SSZ was expected to decline with the increasing of the initial concentration of SSZ. The trend was observed in this study less catalytic sites when the concentration of SSZ was increased. Additionally, the decrease in the removal efficiency was also ascribed to less reactive species being available to degrade SSZ in agreement with previous finding (Santhosh et al. 2018). Gopinath and Krishna (2019) observed the same phenomenon when the initial concentration of 2, 4 dichlorophenol was increased during their experiment. In contrast, in this study at lower SSZ concentrations (10 and 20 mg/L) the number of catalytic sites was not a hindrance, and the rate of degradation was proportional to substrate concentration as per apparent first-order kinetics.

Effect of water matrix

Real wastewater [collected at two sampling points at a South African wastewater treatment plant (Scheme 1)] and synthetic water were used as target matrices. The SSZ degradation profile in both water matrices is presented in Fig. 4a. In synthetic water using GQDs/PAA (50 mg/L: 0.10 mM), the obtained degradation efficiency was almost 100%, while 35% and 80% removal efficiency were recorded for wastewater collected at sampling points 1 and 2, respectively. The lower degradation efficiency of SSZ in wastewater compared to synthetic water was expected and it was largely ascribed to the complex nature of real wastewater. Numerous substances in real wastewater including humic acid, bicarbonates, chlorides and carbonates ions can scavenge radicals needed for the degradation of SSZ which in turn diminishes the degradation potential (Luo et al. 2015). The difference in degradation efficiencies of SSZ at the two sampling points was therefore due to the different composition of the water samples. In this WWTP, raw wastewater influent was pre-treated prior to reaching sampling point 1. The only purpose of preliminary treatment was to screen coarse and large materials, remove the grit as well as to prevent organic solids from settling. Other researchers reported similar trends (Kirk et al. 2002). It has been demonstrated that approximately 50% of the incoming biochemical oxygen demand (BOD), 70% of the total suspended solids (TSS) and some organic nitrogen, phosphorous and heavy metals are only removed after the primary clarifier (Sonune and Ghate 2004). Sampling before the primary clarifier (sampling point 1) meant there were a lot of competition ions during the photodegradation process in the real wastewater. Luo et al. (2015) showed that carbonate and bicarbonate ions can affect the degradation process as they tend to react with ·OH and form CO3·. In another instance, humic acid was shown to affect the degradation process either by acting as a scavenger or through the introduction of an inner filter effect arising from the large absorbance of humic acid (Shi et al. 2018). Cl− ions can also affect the degradation process as they interact with ·OH and SO4·− resulting in less reactive radicals such as ClOH·−, Cl·, Cl2·− (Kläning and Wolf 1985). At sampling point 2 (after secondary clarifier), the effluent from the primary clarifier was treated using activated sludge to remove nitrogen, phosphorous, dissolved minerals as well as non-biodegradable organics similarly as reported in the literature (Sonune and Ghate 2004). The primary clarifier treatment implied that the water matrix was less complex and that most ions were removed and were not able to scavenge the radicals (responsible for photodegradation), hence the better degradation efficiency of 81% compared to the 35% at sampling point 1. In addition to efficiently degrading SSZ, the synergy of disinfection capabilities of GQDs/PAA was investigated using S. aureus as a representative bacterium in real wastewater (refer to "Antimicrobial activity of GQDs/PAA" section).

Fig. 4
figure 4

a Photodegradation profile of SSZ in various water matrices while keeping the concentration of the GQDs constant (50 mg/L: 0.10 mM) at pH 5, b absorption spectra of SSZ during photodegradation process with increasing irradiation time from 0 to 150 min while keeping the GQDs concentration constant (50 mg/L: 0.10 mM) at pH 5 and c photodegradation kinetics of different water matrix when using PAA/GQDs system with constant GQDs parameters (50 mg/L: 0.10 mM) and pH 5

Full size image

Photodegradation kinetics of SSZ

Variations in the absorption spectra of SZZ during the photodegradation are shown in Fig. 4b. Over time, the SSZ peak (wavelength 300–400 nm) decreased and eventually flattened out indicating complete degradation of SSZ. The data fit a pseudo-first-order model and a linear plot were obtained by plotting \(\ln \left( {\frac{{C_{0} }}{C}} \right)\) versus irradiation time. The kinetic data of SSZ photodegradation in synthetic water and wastewater samples (sampling point 1 and 2) are presented in Fig. 4c. The linear relationship observed between the \(\ln \left( {\frac{{C_{0} }}{C}} \right)\) over a period implied that first-order kinetics was followed in all samples. The synthetic water matrix had a higher rate constant (0.032 min−1) compared to the wastewater samples (0.011 min−1 and 0.030 min−1) for sampling point 1 and 2, respectively, and this can be explained by the complexity of the water matrix. Photodegradation efficiency and kinetic data for the degradation of SSZ are presented in Table 2. The degradation of SSZ in synthetic water was higher and showed shorter half-lives.

Table 2 Photodegradation efficiency of SSZ degradation
Full size table

Identifying radicals responsible for the degradation of SSZ

The combination of GQDs and PAA degrades SSZ using a synergistic combination of different radicals to oxidize SSZ. The contribution of each radical was investigated by radical scavenging experiments, to identify the radicals responsible of the degradation of SSZ. Methanol, EDTA-2Na, benzoquinone and silver nitrate were used as scavengers for ·OH, h+, ·O2− and e−, respectively. To differentiate between the contribution of ·OH and the acetlyperoxyl radicals, 2,4-hexadiene was used as a radical scavenger. The C=C double bonds of the 2,4-hexadiene can be readily attacked by the acetylperoxyl radicals, and it has been reported that 2,4-hexadiene can also react with ·OH at a reaction rate constant of 9.2 × 10 M−1 s −1, meaning it quenches both ·OH and acetylperoxyl radicals (Zhang et al. 2020). SSZ was not affected by the addition of EDTA-2Na and silver nitrate, suggesting that h+ and e− had no active role in the photodegradation of SSZ (Fig. 5) and concurs with other previous studies (Wang et al. 2020; Cai et al. 2017). However, ·O2− was responsible for less than 8% of the active species. Methanol inhibited 43.9% of the ·OH radicals, indicating that ·OH contributed to the degradation of SSZ. The contribution of the ·OH was attributed to the presence of GQDs in the GQDs/PAA system. Recently, the contribution of ·OH in the degradation of two pharmaceuticals using the UV/PAA system has been highlighted (Cai et al. 2017). The contribution of ·OH occurs because of the molecular oxygen in the GQDs/PAA being reduced to ·O2−, and the resulting ·O2− will react with H+ to form ·HO2. The ·HO2 will react with trapped e− to produce H2O2. These H2O2 groups will interact with the generated conduction band e− resulting in the production of more radicals. The presence of 2,4-hexadiene significantly inhibited the degradation of SSZ (76.4%), suggesting that the primary reactive species responsible for the photodegradation of SZZ were the acetylperoxyl ((CH3C(O)O·) or acetylperoxy CH3C(O)OO·)) radicals. Other radicals such as ·CH3, CH3OO·, CH3C(=O)OO· that were formed in the GQDs/PAA system had no impact on the photodegradation of SSZ, likely because since CH3OO· is a weak peroxyl whose formation is dependent on the amount of O2 present in the system. Additional tests were performed to rule out the potential of either acetic acid or H2O2 being involved in the photodegradation of SSZ. These tests employed the same amounts of acetic acid and H2O2 as the PAA solution (298.0 μM acetic acid and 86.0 μM H2O2 at 200.0 μM PAA). The results are depicted in Fig. 5b, c revealed that the contribution of both acetic acid and H2O2 was negligible in the degradation of SSZ. The above finding also confirmed that GQDs were better suited to activate PAA compared to H2O2 owing to the peroxide bond energy of the PAA (159 kJ/mol) being less than H2O2 (213 kJ/mol). To support our argument, Wang et al. (2020) reported that the dissociation of the peroxide bond in PAA produces reactive species than H2O2.

Fig. 5
figure 5

a Radical scavenging experiments (For comparison purposes experiments in the dark and in the presence of light without scavengers are also included). b photodegradation of SSZ in the presence of H2O2 and c photodegradation of SSZ in acetic acid

Full size image

The proposed radicals of PAA activated by GQDs in the degradation SZZ are shown in Eqs. 4–8. The rate-determining step and the homolysis of the oxygen–oxygen bond that is triggered by GQDs is represented by Eq. 4. CH3· and CH3O2 (Eqs. 5, 6) were ruled out as contributors because CH3· (k = 2.8–4.1 × 109 M−1 s−1) tends to quickly react with O2 resulting in the formation of CH3O2. CH3O2 also has a significantly lower oxidation capacity (Cai et al. 2017); hence, the photodegradation of SSZ was likely due to ·OH, CH3C(=O)OO· and CH3C(=O)O·. The CH3C(O)OO· radicals were reported to be stronger compared to the CH3C(O)O· (Wang et al. 2020). Based on the latter, we propose various ways of the degradation of SSZ: (a) via the direct oxidation of SSZ consisting of the involvement of CH3C(O)O· occurs via the direct reaction with SSZ or (b) indirectly when CH3C(O)O· reacts with GQDs/PAA to form CH3C(O)OO· which subsequently reacts with SSZ. The participation CH3C(O)O· was informed by the cumulative effect of three reactions, namely the reaction of CH3C(O)O· with SSZ, the direct reaction of CH3C(O)O· with GQDs/PAA as well the self-decomposition of CH3C(O)O·.

$${\text{GQDs}} + {\text{CH}}_{3} {\text{C}}\left( { = {\text{O}}} \right){\text{OOH}} \leftrightarrow {\text{GQDs}} + {\text{CH}}_{3} {\text{C}}\left( { = {\text{O}}} \right){\text{O}}\cdot + \cdot{\text{OH}}$$
(4)
$${\text{GQDs}} + {\text{CH}}_{3} {\text{C}}\left( { = {\text{O}}} \right){\text{O}}\cdot \to {\text{CH}}_{3} \cdot + {\text{CO}}_{2}$$
(5)
$${\text{CH}}_{3} \cdot + {\text{O}}_{2} \to {\text{CH}}_{3} {\text{O}}_{2}$$
(6)
$${\text{CH}}_{3} {\text{C}}\left( { = {\text{O}}} \right){\text{OOH}} + \cdot{\text{OH}} \to {\text{CH}}_{3} {\text{C}}\left( { = {\text{O}}} \right){\text{OO}}\cdot + {\text{H}}_{2} {\text{O}}$$
(7)
$${\text{CH}}_{3} {\text{C}}\left( {\text{O}} \right){\text{OOH}} + {\text{CH}}_{3} {\text{C}}\left( { = {\text{O}}} \right){\text{O}}\cdot \to {\text{CH}}_{3} {\text{C}}\left( {\text{O}} \right){\text{OO}}\cdot + {\text{CH}}_{3} {\text{C}}\left( {\text{O}} \right){\text{OH}}$$
(8)

Proposed degradation pathway

The transformation and degradation pathway of SSZ with GQDs/PAA were evaluated using degradation products obtained using LC-QTOF-MS. Molecular ion masses and MS fragmentation patterns were employed to predict the molecular structure of the reaction intermediates. Typically, the chemical structure and functional moieties of SSZ allow it to be both oxidized and reduced during the photodegradation process. The oxidation process can occur as result of the –OH, –NH and –COOH of the SSZ attacking the OH, h+ or ·O2− radicals. Conversely, the –N=N– and –O=S=O bonds in SSZ facilitates its reduction (Omrania et al. 2019). From Scheme 2, ·OH and CH3C(=O)O· attacked the –N=N– bonds of the parent molecule SSZ broke the SSZ in two intermediates, namely 5-aminosalysylic acid (m/z 153.13) and sulfapyridine (m/z 249.0984) (Fig. 6a, b). Other studies documented that further attack of the 5-aminosalysylic acid by either ·OH or h+ ought to result in the intermediates 1,3,4-triol-2-carboxilic-1,3 dibutene and acetaldehyde or into maleic acid and ethandiol amine (Omrania et al. 2019); however, that was not the case in the present study. It is postulated that sulfapyridine was further attacked by the CH3C(=O)OO· and the intermediate produced aniline (m/z 93.13) and (m/z 159.0581) (Fig. 6c, d). This intermediate underwent desulfonation and was further broken into pyridine 2,3,5-triol (m/z 125.9860) and pyridine-3(4H)-one (m/z 95.9471) (Fig. 6d). Although no smaller molecular masses were picked up on the LC-QTOF-MS, the resultant intermediates were able to undergo ring-opening caused by the reactive radicals (CH3C(=O)OO·). The latter formed lighter alcohols and acids which completely mineralized into CO2 and H2O.

Scheme 2
scheme 2

Proposed transformation and photodegradation pathway of SSZ using GQDs/PAA

Full size image
Fig. 6
figure 6figure 6

LC-QTOF-MS degradation products of SSZ showing various steps of the pathway caused by the GQDs/PAA under visible light irradiation

Full size image

Total organic carbon measurement

The TOC measurement graphs are presented in Fig. 7. TOC analysis showed 83.7% TOC eliminated when 0.15 mM PAA was used, even though there was an almost 100% degradation of SZZ. The incomplete TOC elimination was corroborated by LC-QTOF-MS results depicted in Fig. 6. This TOC elimination was attributed to the residual by-products that include aromatic ring structure of SSZ which was not completely by the GQDs/PAA. TOC showed that GQDs/PAA did not completely break the aromatic ring of SZZ within the reaction time into smaller molecules with low molecular weight such as CO2 and H2O. However, the residual by-products were less toxic and appeared in much lower concentration than the parent pollutant. Additionally, PAA is an organic peroxide and has previously been reported to contribute partially to the TOC in the reaction (Zhang et al. 2020). Elsewhere, a low degree of mineralization of SSZ was observed when using Fenton-like processes, and in that case, however, the authors ascribed the 20% TOC elimination to the complexation of SSZ with Fe3+ (Fan et al. 2011). In this study, since SSZ was not completely mineralized, further tests were necessitated to evaluate the mutagenic and genotoxic potential of the reaction by-products formed during the degradation process.

Fig. 7
figure 7

TOC elimination of SSZ using GQDs/PAA

Full size image

Genotoxicity and mutagenicity test using Ames test

The TOC results (Fig. 7) indicate that some photodegradation intermediates remained in solution even after SSZ was completely degraded and that complete mineralization was not achieved. The trend corroborated to a report by Fan et al. (2011). To verify whether the reaction intermediates formed during the photodegradation of SSZ were mutagenic, Ames assay was carried out. This assay, developed by Bruce Ames in 1970 (Ames et al. 1975) uses either Salmonella or E. coli bacterial strain to establish whether a specific chemical (in the case of this study the reaction by-products) are mutagenic. Typically, the bacterial strains used have a point mutation (histidine in Salmonella typhimurium) and (tryptophan in E. coli) that make it impossible for the bacterial strain to produce the corresponding amino acid. The point mutation results in the inability of the bacteria to produce corresponding amino acids and inhibit the growth of his- or trp-organisms unless histidine or tryptophan is supplied.

Culturing His-Salmonella in a medium containing the reaction by-products formed during the photodegradation of SSZ, may result in mutation on the histidine encoding gene, which allows Salmonella to regain the ability to synthesize histidine. If the reaction by-products formed during the degradation of SSZ cause this reversion, the by-products are considered as mutagens. The mutagenicity of the degradation by-products of SSZ are proportional to the number of bacterial colonies counted on the test plate. The assay was prepared using the aliquots collected at 30, 60, 90, 120 and 150 min. The genotoxicity was expressed as the % of micronuclei per 10.000 nuclei and presented as the average per dose ± SD (Table 3).

Table 3 Mutagenicity of reaction by-products in the photodegradation of SSZ using Salmonella typhimurium TA 98 and TA 100 tester strains
Full size table

Results presented in Table 2 show that the reaction by-products are negative or weakly positive as per previous reports. The negative results give evidence that none of the reaction by-products were toxic. This agrees with previous reports on PAA forming less toxic or no by-products at all when used in the treatment of wastewater (Monarca et al. 2002a, b; Baldry 1988). Work reported by Monarca et al. (2002a, b) showed that no halogenated reaction by-products were formed after treating wastewater with PAA and mostly only carboxylic acids (which are not mutagenic) were found. While weakly positive results were observed at 90 and 120 min in the TA 100 strains, they were attributed to peroxy radicals and hydroxyl radicals formed during the synergy of GQDs/PAA. Li et al. (2012), Levin et al. (1982) and Dillon et al. (1998) found that the presence of superoxide, singlet oxygen, aldehydes gave positive results in the Ames assay. The reaction by-products from the SSZ degradation process did not induce any mutation in the two Salmonella typhimurium tester strains.

Antimicrobial activity of GQDs/PAA

Inactivation of representative bacteria S. aureus

The antimicrobial activity of the GQDs/PAA was evaluated on a representative bacterium, S. aureus. For comparative purposes the antimicrobial activity of PAA and GQDs was also evaluated. Results demonstrated that under the same experimental conditions both PAA and GQDs were able to inactivate S. aureus; however, the addition of PAA significantly enhanced the antimicrobial activity (Table 4). The minimum inhibitory concentration (MIC) is the concentration required to completely inhibit the growth of a microorganism. The lowest MIC was recorded for GQDs alone (45.1 µg/mL) and the highest recorded was for the GQDs/PAA (21.5 µg/mL). The improved antimicrobial activity of GQDs/PAA can be explained as follows: both GQDs and PAA have biocidal properties, the combined effect of GQDs/PAA led to complete cell destruction. The acetic acid of PAA is postulated to have reduced the intracellular pH as well as disrupted the chemiosmotic functions of the lipoprotein cytoplasmic membrane while the GQDs inactivate bacteria because of direct contact as well as the oxidation of cellular components (Hui et al. 2016; Baldry 1988). Chang et al. (2006) reported the MIC of S. aureus to be 0.33 mM when using 1 mM of PAA.

Table 4 Minimum inhibitory concentration (µg/mL) of GQDs, PAA and GQDs/PAA
Full size table

Bacterial inactivation mechanism of S. aureus and morphological observation by SEM

The mechanism of action of the GQDs/PAA against S. aureus was determined by morphological observation using SEM instrument. The inactivation process was tracked by monitoring changes in morphology of the cell wall and the cell membrane of the cellular materials in every SEM micrograph. The SEM micrograph of the negative control (bacteria not exposed to GQDs/PAA) is shown in Fig. 8a; from this micrograph, an intact cell wall and the lining of the cellular material can be observed. The cell wall of the negative control is devoid of any artifacts and is relatively smooth. After 1 min of exposure to GQDs/PAA, the cell wall of the S. aureus started to disintegrate. It can be postulated that the PAA initially diffused through the cell wall of S. aureus, resulting in changes in the morphology of the cell wall. The cell wall visibly looks thinner and appears to erode at the surface and has a rougher edge, the same can be said about the cellular contents after 1 min (Fig. 8b). After the diffusion of the PAA into the S. aureus cell wall, the GQDs easily attach to the phospholipid lipid bilayer due to the electrostatic interactions resulting in further roughening of the cell wall and this is accompanied by an apparent deformation of the cellular material (Fig. 8c). Once inside the cell, GQDs accumulate in the nucleus and the oxidative radicals inhibit adenosine triphosphate (ATP), the replication of cells as well as cell respiration. Further exposure to GQDs/PAA (Fig. 8d, e) resulted in the complete destruction of the cell cortex and DNA compression. After 5 min, the S. aureus cells were destroyed and reduced to microscopic debris (Fig. 8f). Based on the SEM micrographs, it can be postulated that no regrowth of bacterial cells occur due to the damage on the S. aureus being irreparable.

Fig. 8
figure 8

SEM micrographs of a S. aureus cells without GQDs/PAA (control), S. aureus cells after GQDs/PAA treatment for b 1 min, c 2 min, d 3 min, e 4 and f 5 min under irradiation

Full size image

Cell viability

To further confirm the antimicrobial activity of the GQDs/PAA on S. aureus, LIVE/DEAD BacLight staining kit was used. Live healthy S. aureus cells (negative control) before GQDs/PAA treatment were green in color (Fig. 9). A1 min after GQDs/PAA treatment, both red and green stains were seen in the photomicrograph, suggesting that some S. aureus cells were beginning to die. The observed changes in the integrity of S. aureus cell membrane integrity are due to the combined effect of GQDs/PAA in denaturing proteins and oxidizing enzymes which results in impaired intracellular cellular solute levels. After 3 min, a complete inactivation of S. aureus cells was marked by all the S. aureus staining red. These photomicrographs verified the role that GQDs/PAA in the loss of cell membrane integrity of S. aureus. Using the same staining kit, results obtained by Costa et al. (2015) found that 2% and 0.25% of PAA were able to inactivate S. aureus after 30 min contact time, while elsewhere (Lee et al. 2016) obtained 100% inactivation of S. aureus on a biofilm after 60 s.

Fig. 9
figure 9

Photomicrographs of S. aureus cells without GQDs/PAA and after GQDs/PAA treatment after 1 and 3 min

Full size image

Antimicrobial activity in real wastewater without spiking with model S. aureus

To examine the feasibility of the GQDs/PAA to be use as a disinfectant, the disinfection ability was tested in raw wastewater collected at sampling point 1 and 2 (Scheme 1). The GQDs/PAA was added to WW samples under irradiation (experiments carried out as detailed in "Photodegradation of sulfasalazine (SSZ)" section) except using raw wastewater and no addition of SSZ. (Only GQDs/PAA was used.) The aliquots were collected at 30-min intervals and immediately mixed with 200 µL of sodium thiosulfate and 500 µL catalase to stop the PAA from reacting further. A total of 20 µl of the solution was spread-plated and incubated at 37 °C for 24 h.

Figure 10 shows the microbial plate count, at time = 0 (before treatment with GQDs/PAA) and a significant number of colonies were observed on the plate. Increasing the exposure time to GQDs/PAA led to reduced number of colonies on the plate. After 150 min, only 9 colonies were present, indicating that the synergistic effect of the combination of GQDs and PAA was able to inactivate microbial species even in complex water matrices. Further studies need to be carried out to correctly identify the bacterial species that were not inactivated by GQDs/PAA. In an earlier study, Lefevre et al. (1992), Liberti and Notarnicola (1999) cited that the limitation of PAA being used as a disinfectant was in its inability of inactivating viruses and protozoa at reasonable doses. More recently, Mezzanotte et al. (2003) have showed that PAA efficacy was limited to other bacteria at lower doses. The recommendation from the study would be to vary the doses of PAA, the photocatalyst and even the contact time to yield better results.

Fig. 10
figure 10

Bacterial colonies formed after treating raw wastewater with GQDs/PAA after 150 min

Full size image

Conclusion

The combination of graphene oxide quantum and peracetic acid (GQDs/PAA) was used to photodegrade sulfasalazine in municipality wastewater. A dose-dependent trend was observed wherein increasing the PAA concentration resulted in almost complete photodegradation of the SSZ. The degradation was accompanied by an increment in the kobs value at higher PAA concentrations. Increasing the initial concentration of SSZ (10–50 mg/L) at a constant concentration of GQDs/PAA (50 mg/L: 0.10 mM) resulted in 10 mg/L of SSZ yielding the highest photodegradation efficiency of 100%; in contrast, the highest concentration of SSZ (50 mg/L) recorded the lowest removal efficiency of 20%. The primary reactive radicals in the photodegradation were hydroxy (·OH) as well as peroxy radicals CH3C(=O)OO· and CH3C(=O)O·. Furthermore, the genotoxic and mutagenic potential of the degradation products formed during the degradation of sulfasalazine was non-mutagenic. GQDs/PAA completely inactivated S. aureus and eliminated more than 90% of bacteria present in raw municipal wastewater. This contribution presents an opportunity to simultaneously degrade pharmaceuticals and their active metabolites as well as inactivate microorganisms using GQDs/PAA. The results indicate practical application in real wastewater treatment plant where PAA-based GQDs advanced oxidation processes could be utilized in the tertiary stage of a wastewater reclamation treatment facility for water reuse and environmental remediation.