Photo-catalytic degradation of sulfamethoxazole from aqueous solutions using Cu-TiO2/ CQDs hybrid composite, optimisation, performance and reaction mechanism studies

ABSTRACT Cu doped TiO2 decorated with carbon quantum dots (CQDs), called Cu-doped TiO2/CQD, was synthesised as a photocatalyst, and then used in the degradation of sulfamethoxazole (SMX) antibiotic. The properties of the synthesised catalyst were determined by FT-IR, XRD, FESEM, TEM, EDX, BET and UV–Vis DRS analysis. These analyses showed successful doping of TiO2 with Cu coated by CQDs. Variable parameters such as different CQDs ratios, photocatalyst dosage, initial concentration of SMX, pH and intensity of visible light were investigated. The best removal efficiency of SMX was obtained at initial concentration = 20 mg/L SMX, photocatalytic dosage = 0.8 g/L, pH = 6, visible light intensity = 75 W/m2 and CQDs ratios in the composite = 4% wt during 60 min reaction time. The photocatalytic degradation kinetic of SMX fitted well with the pseudo-first-order kinetic model. Experiments showed that HO• and O2• – were active species in the photocatalytic destruction of SMX and O2• – had more inhibiting effects in the SMX degradation process. The synthesised photocatalyst was used for six cycles with a slight reduction in the SMX degradation efficiency. TOC removal efficiency was 81% at optimal condition.


Introduction
Antibiotics and their derivatives enter the environment through wastewater from the pharmaceutical industries, hospitals, disposal of human and animal wastes and finally found in both drinking water and treated water [1,2].Although residues of antibiotics are present in trace levels in the environment, the continuous entry of antibiotics into the environment resulted in toxic effects on living species in the ecosystem and also bacterial resistance [3,4].Sulfamethoxazole (SMX) is a synthetic antibiotic with antimicrobial properties that belongs to the class of sulphonamides.This antibiotic is used as antimicrobial agents to remedy infectious diseases in humans such as urinary tract infections, meningitis, bronchitis, and a number of other types of infections [2,5].Also, a large amount of this antibiotic is used in aquaculture and animal husbandry to remedy bacterial infections and usually added to animal feed as a growth stimulant [6].According to the past reports, about 2,000 tons of sulphonamide antibiotics were discharged into the aquatic environment through sewage and the pharmaceutical industries per year, and the antibiotic SMX is one of the well-known sulphonamides that detected in µg/L level in water sources [6].The high resistance and stability of antibiotics and their non-biodegradability and this fact that they can pose problems for public health have raised many concerns in recent years [4].Therefore, there is a need to use efficient methods to treat wastewater containing this type of contaminant.Many methods used to remove antibiotics such as absorption, biological degradation and many current treatment methods are not able to completely detoxify antibiotics, and these reasons have attracted the attention of researchers to study new treatment methods [7].
Advanced oxidation processes (AOPs) based on the production of free and highly active radicals such as the hydroxyl ( °OH) and super oxide ( °O) species are recommended for the degradation of organic pollutants in aqueous environments [1,8].Photocatalytic processes based on nanocatalysts are one of the most promising methods for degradation of pollutants in aqueous media.In this process, using photon energy, the electron layers in the semiconductor material are excited and electron-hole pairs (e − / h + ) are produced, which eventually leads to the production of many active radicals for degradation and detoxification of resistant contaminants [6].There are many semiconductors such as WO 3 [9], V 2 O 5 [9], TiO 2 [10], CdS [11], and ZnO [12], which have photocatalytic properties.TiO 2 is one of the common semiconductors for photocatalytic applications due to its non-toxicity, chemical stability, and high photocatalytic properties.Pure TiO 2 has drawbacks including less effective use of photons compared to TiO 2 doped with other elements, fast recombination of electron-hole pairs, and activity in the wavelength range less than 400 nm and therefore have a large gap band (3.2 ev) [13,14].Therefore, in order to improve the photocatalytic efficiency, prevent electron-hole recombination, shift its effective light absorption to the visible light region, and decrease the gap band, modifications have been made on the pure TiO 2 , which can be referred to doping TiO 2 with non-metal (such as graphene, carbon, carbon nanotubes, graphitic carbon nitride and CQD) and metal ions [13,14].Doping TiO 2 by transition metals such as Cr 3+ , Mn 2+ , Fe 3+ , Co 2+ , Ni 2+ and Cu 2+ is one of the proposed solutions [6].Among these dopants, copper ions have been used extensively due to their ionic radius close to the Ti 4+ ionic radius (Ti = 0.68 Å, Cu = 0.72 Å) which can be easily replaced in the TiO 2 lattice.Also, the proximity of the redox potential of Cu +2 /Cu + and Ti 3+ /Ti 4+ leads to an increase in the light absorption by TiO 2 in the visible region.Additionally, Cu ions have a narrow gap band that reduces the gap band of TiO 2 down to 1.8-2 eV.On the other hand, Cu ions trap electrons generated and increase photocatalytic activity by reducing e − /h + pairs recombination [15][16][17].In recent years, to increase the photocatalytic activity of TiO 2 in the visible region, various types of carbon nanomaterials such as graphene fluorescence, carbon nanotubes and so on have been used to produce carbon/TiO 2 compounds [18].Recently, carbon quantum dots (CQDs) have been considered as a new member of the carbon family due to their unique properties such as low toxicity, low solubility, and high stability against light and photoluminescence properties.The photoluminescence emission of CQDs is at 430-550 nm [18,19].CQDs stimulate TiO 2 under light irradiation to improve the production of the e − /h + pairs.On the other hand, the CQDs decorated on the Cu-TiO 2 photocatalyst act as an electron reservoir, and the photogenerated electrons in the TiO 2 conduction band are transferred to the CQDs, and CQDs causing the separation of the e − -h + pairs and prevention of recombination of e − / h + pairs.Finally, long-life holes are produced at the TiO 2 surface and the photocatalytic properties increase [20].In recent years, much research has been reported on the synthesis of TiO 2 /CQDs and usage of those to increase photocatalytic activity in the range of visible light and the degradation of pollutants [18,20].Therefore, according to the above, the use of pure TiO 2 in photocatalytic degradation systems compared to TiO 2 doped with metal elements and combined with CQDs (due to their unique properties) in real scale and industrial systems is undesirable because of the long process of degradation of pollutants.For these purposes, in the present study, Cu-TiO 2 /CQDs were synthesised and used as a photocatalyst in SMX degradation process.

Chemicals
SMX (C10H11N3O3S, MW: 253.27 g/mol) was obtained from Sigma-Aldrich Co.The rest of the chemicals used in this study were purchased from Merck Co. and used without any purification.All the required stock solutions were prepared using double distilled water (DDW).

Synthesis of CQDs
This step was performed using hydrothermal method [21].Briefly, 3 g of pure citric acid powder and 3 g of urea were mixed in 25 ml of DDW at room temperature and then stirred for 30 min under ambient conditions to homogenise the mixture.The resulting mixture was placed in a Teflon-lined steel autoclave and heated at 180°C for 5 h.Next, autoclave was placed at room temperature to cool, and finally a dark brown solution was obtained.Then, to remove impurities, the solution was centrifuged at 4500 rpm for 5 minutes and a uniform solution was obtained.

Synthesis of Cudoped TiO 2
Cu doped TiO 2 was prepared by the sol-gel method according to previous reports [16,22].In summary, 11.44 ml of titanium tetraisopropoxide (TTIP) was added to 9 ml of anhydrous ethanol and stirred for 1 h using a mechanical stirrer (solution A).Next, 0.28 g of copper nitrate (Cu (NO 3 ) 2 .3H 2 O) (3% wt.) was dissolved in DDW and stirred with a magnetic stirrer for 15 min (solution B).Then, solution B was added drop wise to stirring solution A, and mixed for 2 h to obtain a gel-like mixture.To prevent the deposition of hydroxide species during the synthesis process, the pH of the solution was adjusted at 2 by nitric acid.The gel obtained in the oven was dried for 5 h at 103°C.

Syntheses of Cu-TiO 2 /CQDs composite
A stoichiometric amount of 3% Cu-TiO 2 was added to a solution containing 40 ml of ethanol and 20 ml of DDW and dispersed by ultrasonic for 0.5 h.Then, different amounts of pre-prepared CQDs suspension (2, 5 and 8 mL) were added to the Cu-TiO 2 mixture and the mixing process was continued with a mechanical stirrer at 60°C for 2 h.Finally, the product was dried at 60°C for 8 h and dehydrated at 300°C for 3 h.This composite was named %2 CQDs/Cu-TiO 2 , %4 CQDs/Cu-TiO 2 and %8 CQDs/Cu-TiO 2 , respectively.

Characterisation tests
To evaluate the physical, chemical, morphological, structural and optical properties of the catalysts and various advanced techniques such as FT-IR, XRD, FESEM, TEM, EDX, BET and UV-vis DRS were used, which are given in Table S1.

Analytical techniques and apparatus
The concentration of SMX was analysed using a high performance liquid chromatography (HPLC) device (Waters 2010, USA) equipped with an ultraviolet doctor.The mobile phase consisted of acetonitrile/water containing 0.2% formic acid (V/V 35:65).The injection volume was 20 µL, and the flow rate was 1 mL/min.The samples were injected into a C 18 column (250 mm × 4.6 mm Waters, USA), and SMX was detected at a wavelength of 270 nm.Before measuring the antibiotic residue, the calibration curve of the device was plotted using different concentrations of SMX in the range of concentrations tested.A TOC analyser was used to measure the total organic carbon (TOC).

Adsorption equilibrium studies and photocatalytic experiment procedure
Photocatalytic degradation of SMX antibiotic was performed in a 500 mL Pyrex batch photochemical reactor containing 200 mL of sample.For a uniform distribution of light, a xenon lamp (55 w) with 75 mW/cm 2 radiation intensity and 472 nm maximum light output was placed in the centre and above the photoreactor, and the photoreactor was placed on a magnetic stirrer to mix thoroughly.Before irradiation, in the dark and to achieve adsorption/desorption equilibrium, a certain amount of catalyst was added to the solution containing a certain concentration of antibiotic at a certain pH and the mixture was stirred at 200 rpm, and adsorption tests were performed in the first 30 min before the lamp was turned on.The photocatalytic process was started by turning on the xenon lamp, and the sample was stirred again using a magnetic stirrer.At specified time intervals, 1 ml was removed from the sample and centrifuged for 5 min.The sample was filtered using a 0.22 μm syringe filter and injected into the HPLC device to determine the residual concentration of antibiotics.The effect of various variables including solution pH, initial SMX concentrations, photocatalytic doses, visible light intensity and SMX degradation time were investigated by the Cu-TiO 2 /CQDs process under visible light.
After optimisation of variables, degradation kinetics, the effect of co-existing ions and scavenger agents, recyclability and stability of the photocatalyst, mineralisation degree and the ability of the process to treat real pharmaceutical wastewater was evaluated.Finally, an experimental degradation pathway was proposed.The decontamination rate of the system is based on the equation below: Where, C t and C 0 indicate the SMX concentration at time t and reaction start time, respectively.
Where, TOC 0 and TOC t represent amounts of TOC at time t and at time t 0 , respectively.

XRD
XRD is one of the methods to determine the crystalline properties and structure of nanoparticles. Figure 1(a) shows the XRD spectra for the Cu-TiO 2 catalyst, Cu-TiO 2 /CQDs 4wt% , and TiO 2 nanoparticles.In all samples, peaks in 2ϴ = 25.30,37.9, 47.9, 54, 55, 62.64, 69.2 and 75.6 that are indexed 101, 004,200, 105, 211, 204, 116 and 215 are visible.These peaks are in accordance with the JCPDS, 21-1272 of TiO 2 anatase phase [6].Peaks attributed to the brookite and rutile phase of TiO 2 were not observed.This indicates that the samples were grown in a crystalline phase corresponding to the anatase phase of TiO 2 .In the Cu-TiO 2 /CQDs spectrum, despite the presence of CQDs in the Cu-TiO 2 /CQDs catalyst, the lattice phase is completely preserved and no specific peak related to CQDs was observed, which can be due to the overlapping the main peak of CQDs at 2ϴ = 25 indexed in 002 with the main peak of the TiO 2 (101) in the anatase phase.Also, no specific peak presenting copper was observed in both catalysts, which could be due to the low amounts of dopant copper in the TiO 2 structure and the uniform distribution of copper in the titanium lattice [23].

FTIR analysis
FTIR analysis of TiO 2 , Cu-TiO 2 /CQDs and CQDs is shown in Figure 1(b).In the curve of A and B, peaks below 1000 are related to metal oxide bands [24].In the A and B curves, the two strong peaks at 640 cm −1 and 670 cm −1 are represented by the Ti-O-Ti vibration bands and the 726 cm −1 and 750 cm −1 peak are attributed to the Ti-O stretching bands of titanium, and the peak Cu-O is located in 516 cm −1 [6].In curves A and B, peaks 3420 cm −1 and 3452 cm −1 correspond to H-O-H stretching vibrations and vibration bands at ~1619 cm −1 and 1640 cm −1 correspond to the bending vibrations of O-H groups, which represent the water adsorbed on the catalyst surface and hydroxyl groups on the catalyst surface [6].H-O-H and O-H vibrational bands are weaker in TiO 2 samples than in Cu-TiO 2 /CQDs samples.For the curve C, the vibrations for C-O at ~1023 cm −1 , the bond C-N at ~1173, the vibration for COCH 2 at ~1390 cm −1 and N-H stretching vibration at ~3515 cm −1 and 3202 cm −1 are observed [25,26].

UV-Vis DRS
DRS analysis was used to determine the optical absorption properties of the synthesised samples.
The light absorption spectrum for the TiO 2 anatase phase is in wavelengths shorter than 400 nm (see Figure 1c), according to previous reports [27].
The curve of TiO 2 , shows an absorption edge below 380 nm, which is typically attributed to electron transfer from the valence energy level to the conduction energy level and the high TiO 2 gap band [6].Compared to pure TiO 2 , Cu-TiO 2 nanoparticles and Cu-TiO 2 /CQDs composites with different ratios of CQDs show the ability to absorb high light in the visible region.For Cu-TiO 2 nanoparticle, the absorption peak was at 430 nm and the absorption range was better at 380 to 500 nm.Doping of copper metal in the TiO 2 structure leads to increased light absorption in the visible spectrum by the catalyst and absorption at 390 nm, which can result in the creation of a new energy level near the TiO 2 conduction band and the Ti-O-Cu stretching bands [6].Cu-TiO 2 /CQDs with increasing the CQDs content show better light absorption higher than 400 nm and the light absorption intensity has increased.The light absorption peaks for Cu-TiO 2 /CQDs 2 wt% , Cu-TiO 2 /CQDs 4 wt% and Cu-TiO 2 /CQDs 8 wt% are in 400, 420 and 430 nm in the visible light region, respectively.CQDs s can improve the optical activity of Cu-TiO 2 and increase photocatalytic activity and further reduce pollutants with charge-transfer transition from the conduction band TiO 2 to CQDs and producing more e − /h + pairs [20].CQDs also show strong absorption in the range of 300 to 700 nm.The band gap energy of the samples was calculated using the well-known modified Kubelka-Munk function according to Equation (3) [28]: The band gaps of TiO 2 , Cu-TiO 2 , and Cu-TiO 2 /CQDs 2%, Cu-TiO 2 /CQDs 4% and Cu-TiO 2 /CQDs 8% are found 3.12, 2.97, 2.90 eV.The obtained results show a reduction of band gap of titanium after surface modifications.

BET
To determine the porosity of the samples and its surface characteristics, N 2 adsorption and desorption methods were used.
As shown in Figure 1(d), TiO 2 particles with hysteresis loop of H 3 and Cu-TiO 2 particles and Cu-TiO 2 /CQDs 4wt% nanocomposite with hysteresis loop of H 2 are classified as type IV isotherms according to IUPAC classification, which proves the existence of a porous structure in the synthesised samples.
According to the results of BET analysis, the specific surface areas (S BET ) for TiO 2 , Cu-TiO 2 and Cu-TiO 2 /CQDs 4 wt% were found to be 65, 92.3 and 78.2 m 2 g -1 , respectively.
The results show that the S a of Cu-TiO 2 have increased compared to TiO 2 , because copper inhibits the well growth of crystallite and leads to the formation of smaller grains, which is consistent with the results of previous studies [6].On the other hand, with the addition of CQDs into the structure of Cu-TiO 2 nanoparticles, it has led to a decrease in the S BET of the nanocomposite.Pore size distribution pattern of each sample using BJH analysis shows that the average pore diameters of TiO 2 , Cu-TiO 2 and Cu-TiO 2 /CQDs 4 wt% were 6.40, 5.3 and 6.20 cm 3 g -1 , respectively.Also, larger pore volume of Cu-TiO 2 and Cu-TiO 2 /CQDs 4 wt% compared to TiO 2 shows that Cu and CQDs create more number of pores in catalyst (Table 1).Increasing the S BET and pore volume of the catalyst can increase photocatalytic activity by increasing the reaction sites.

FE-SEM and EDX analysis
To evaluate the morphological characteristics of the synthesised samples, FE-SEM analysis was used.In Figure 2(a), the Cu-TiO 2 /CQDs 4wt% nanocomposite is spherical and well crystallised.FE-SEM analysis of TiO 2 and Cu-TiO 2 is shown in Figure S1.TiO 2 nanoparticles  with spherical morphology are observed.The morphological properties of TiO 2 have not changed with the entering copper into the TiO 2 lattice.In Figure 2(b), EDX analysis for Cu-TiO 2 /CQDs 4wt% shows high purity for the sample.EDX analyses for Cu-TiO 2 , Cu-TiO 2 /CQDs 2 wt% and Cu-TiO 2 /CQDs 8wt% are shown in Figure S2.The results of this analysis show that CQDs were well placed in the titanium structure and also with increasing the weight percentage of CQDs, the concentration increases.

Kinetic and isotherms of adsorption
The adsorption isotherms and kinetics of SMX in the Cu-TiO 2 /CQDs 4wt% system were investigated.The experimental data with the pseudo-first-order and pseudo-secondorder models were analysed under a concentration of 20 mg/L SMX, catalyst dose of 0.8 g/L at pH = 6 at 60 min time period.The absorption rate was high in the first 30 min and no significant changes in absorption were observed after that time, therefore this time was considered as the absorption equilibrium time.The results are shown in Table 2.The results showed the coefficient of determination (R 2 ) in the pseudo-second-order model was higher than the pseudo-first-order model and the q e, cal value (24.32 mg/g) in the pseudo-second-order model was higher than the q e, cal value (30.35 mg g −1 ) in the pseudo-first-order model.Examination of adsorption kinetics models showed that in the adsorption process, the chemical adsorption mechanism is predominant and electron exchange between SMX molecules and the catalyst occurs at the junction (see Figure 3a).The Langmuir and Freundlich adsorption isotherms were used to investigate the distribution of SMX molecules on Cu-TiO 2 /CQDs 4 wt% at concentrations of 5-50 mg/L SMX at equilibrium time, and the results are presented in Table 2.The Langmuir isotherm has a larger R 2 than the Freundlich isotherm (as shown in Figure S3).For the Langmuir isotherm, the R 2 = 0.991 and the q 0 and k L values are 54.94 mg/g and 0.0608 L/mg, respectively.For Freundlich isotherm model, the value of R 2 = 0.978 was obtained and value of n 1.69, that value n > 1 indicates that the adsorption follows a physical process and the adsorption system is desirable.Therefore, according to the results, the adsorption system follows the Langmuir isotherm model.

Effect of CQDs content
The effect of CQDs loading in the structure of the photocatalyst on the degradation efficiency of SMX was investigated under optimum conditions, initial concentration of SMX = 20 mg/l, light intensity = 75 mW/cm 2 , photocatalyst dosage = 0.8 gr/L and pH = 6.
Increasing the CQDs loadings from 2 to 4 wt% increased the SMX removal efficiency from %81 to %100 (see Figure 3b).This increase in efficiency can be due to the increase in the active sites on the surface of the catalyst to absorb more pollutants.Also, increasing the amount of CQDs prevents the recombination of the e − /h + .However, with increasing the amount of CQDs from 4 to 8 wt%, the degradation efficiency of SMX reduced down to 89%, which can be due to the reduction of the active surface of the catalyst.Also, CQDs cover surface of catalyst and reduce direct contact between light irradiation and the active sites on the surface of TiO 2 [29].Therefore, CQDs 4 wt% was selected as the optimum value for photocatalytic experiments.

Effect of initial solution pH
The pH of aqueous media is one of the most important parameters in the detoxification efficiency of pollutants in water and wastewater treatment.That affects the capacity of adsorption and decomposition of pollutants, distribution of electrical charge on the photocatalyst surface and oxidation potential of the valence band [4].As shown in Figure 3(c), with increasing the solution pH from 3 to 6, the SMX degradation efficiency by the photocatalyst under visible light increased from 64% to 100% in concentration SMX 20 mg/L, light intensity of 75 mW/cm 2 and photocatalyst dose 0.8gr/L.Also, with an increase in pH of 6 to 9, destruction efficiency of SMX reduced to 76%.The effect of pH can be attributed to the electrostatic forces between the SMX molecule and the photocatalyst, the ionic form of the SMX molecule and the surface charge of the photocatalyst at different pH [6].The zerocharge point was 6.03 for Cu-TiO 2 /CQDs 4 wt% photocatalyst.That means at this point the surface of the photocatalyst is neutral.In acidic conditions, the surface of the photocatalyst has a negative electric charge, and in alkaline conditions, species with positive electric charge such as H + are adsorbed on the surface of the photocatalyst and the photocatalyst becomes positively charged.On the other hand, the SMX molecule has two pK a (pK a, 1 = 1.85 and pK a, 2 = 5.6).At pH < 5.6, the SMX molecule is neutral, and the adsorption between the SMX molecule and the Cu-TiO 2 / CQDs 4 wt% photocatalyst can be attributed to hydrogen bond interactions [6].At pH > 6.03, strong repulsive force between the SMX molecule and the Cu-TiO 2 /CQDs 4 wt% photocatalyst reduced the efficiency of the SMX degradation process.Therefore, the optimum pH was found to be 6.Because the surface of the photocatalyst has a positive charge and the surface of the SMX molecule has a negative charge, the electrostatic absorption force between the SMX and the photocatalyst increased the process efficiency [6].There are many similar studies regarding the effect of pH on photo-destruction of SMX.

Effect of photocatalyst loading
With increasing the photocatalyst dosage from 0.2 to 0.8 g/L, the degradation efficiency increased from 68% to 100% (Figure 3d), when concentration of SMX = 20 mg/L, light intensity = 75 mW/cm, 2 and pH = 6.Increasing the photocatalyst dosage leads to the production of more active sites, increasing the level available for SMX adsorption and increasing the photon absorption by the photocatalyst, which leads to the production of more oxidising radicals and increases the photocatalytic activity [4].However, with increasing the photocatalyst dose up to 0.8 g/L, the SMX degradation rate reduced to 83%.This phenomenon can occur due to reduction of visible light penetration as a result of solution turbidity increase, accumulation and deposition of photocatalyst particles and thus reducing the number of active sites for SMX adsorption, and the phenomenon of self-consumption of active radicals at high dose photocatalyst (most hydroxyl radicals are formed for this catalyst) [4,30].Therefore, due to the maximum efficiency at dosage of 0.8 g/L, it was selected as the optimum dose.

Effect of SMX initial concentration
According to the results, with increasing the concentration of SMX from 10 to 50 mg/L at pH = 6, light intensity of 75 mW/cm 2 and photocatalyst dosage 0.8 gr/L, the efficiency of the photocatalytic process decreased (see Figure 4a).It could occur due to this fact with increasing the concentration of SMX, the active sites of the photocatalyst surface are occupied and consequently production of active radicals reduce [4,6].
Also, high concentration of SMX result in the absorption of light photons by SMX molecules, thus reduced light capture on the active sites of the photocatalyst surface and consequently lead to reduce photocatalyst excitation, resulted in the SMX degradation removal efficiency by the photocatalyst process [6].As can be seen in Figure 4(b), the increase in the intensity of light irradiation from 15 to 75 mW/cm 2 increased the SMX degradation efficiency from 74% to 100%, which can be due to the increase in light absorption by the photocatalyst when light irradiation intensity increased.Consequently, the number of electron-holes produced by light photons increased and led to an increase in the production of active radicals to destroy the SMX antibiotic [6,31].The optimal dose of light intensity in this study was chosen 75 mW/cm 2 for photocatalytic decontamination of SMX Cu-TiO 2 / CQDs 4 wt% composite.

Kinetics
The kinetics of the SMX removal process was investigated using the Langmuir-Hinshelwood (L-H) kinetic model.When the substrate concentration is less than 1, a pseudo-first-order model can be used, according to the literature.The pseudo-firstorder kinetics in the L-H model is based on Equation (4) [4]: Where, C t and C 0 are contaminant concentrations at t = t and t = 0, respectively.The k and t parameters are also reaction rate constant (min -1 ) and reaction time, respectively.According to Equation ( 4), the ln(C t /C 0 ) graph against t presents a linear slope equivalent to k.
The plots of ln (C t /C 0 ) versus reaction time are shown in Figure 4(c) for Cu-TiO 2 /CQDs 4wt%.
The k parameter and the coefficient of determination (R 2 ) obtained from Equation (4) for different concentrations of SMX are prepared in Table 3.As can be seen in Table 3, the data well fitted with an R 2 higher than 0.98 and indicated the process mechanism pseudofirst-order kinetic model, which agrees with previous studies.
Also, with increasing the concentration of SMX from 10 to 50 mg/L, the k value decreased from 0.0853 to 0.0246 min -1 .Therefore, low concentration of SMX showed a higher rate of degradation than higher concentration, which can be due to increase light absorption by SMX molecules at low concentrations.Comparison between previous studies with the present study on SMX degradation is prepared in Table 3.This comparison shows that other processes were less efficient in removing SMX under UV or visible light and at higher reaction times.Therefore, the use of Cu-TiO 2 /CQDs photocatalyst in the destruction of SMX can be appropriate.

Trapping study
To determine the role of radicals and oxidising active species in the photocatalytic degradation of SMX by Cu-TiO 2 /CQDs 4wt% under visible light, a series of experiments were performed under optimal conditions (concentration SMX = 20 mg/L, light intensity = 75 W/m 2 , photocatalyst dose = 0.8 gr/L, pH = 6, and 60 min visible light irradiation) using different scavengers.In these experiments, 1 mM Na 2 -EDTA (disodium ethylenediaminetetraacetic acid), BQ (1, 4-benzoquinone) and IPA (isopropanol) were used as a special scavenger for h + , super oxide (O 2 •-) and hydroxyl radicals, respectively [6].As shown in Figure 4(d), the reduction of SMX degradation efficiency is observed in the presence of all trapping agents.The addition of Na 2 -EDTA reduced the SMX degradation efficiency to % 78.16 indicating the importance of the h + in SMX degradation process.With addition of BQ to the reaction medium, the SMX degradation efficiency reduced to 45.18%, indicating the major role of the O 2 •radicals in the SMX degradation process.
Also, the addition of IPA reduced to 59.2% SMX degradation efficiency indicating the importance of the HO• radicals in the destruction process.According to the results, O 2 •and HO • radicals played an important role as oxidant species during the SMX degradation process, and O 2 •species had major role in SMX degradation in the Cu-TiO 2 /CQDs photocatalyst process.The results agree with previous studies.

Effects of anions
Mineral ions in aqueous media can have an effect on the properties of the solution and affect the fate of the target pollutant and the catalytic activity and photocatalytic decomposition of the treatment processes.The effect of minerals such as SO 4 2-, Cl − , NO 3 − , and CO 3-in the destruction of SMX antibiotics by the Cu-TiO 2 /CQDs system under optimal operating conditions showed that the photocatalytic degradation efficiency of SMX reduced in the presence of each anion after 60 min (Figure S5).The lowest inhibitory effect was related to HCO 3 − anion, and the highest inhibitory effect was related to SO , respectively.The main reasons for this decrease in the process efficiency in the presence of anions include 1) reaction of anions with active radicals and their consumption and finally production of radicals with lower oxidation potential (Cl • , Cl 2 •-, ClOH •-, CO 3 •-, NO 3 •-, and NO 2 •-), as described in Equations ( 22)-( 27), 2) adsorption at the catalyst surface and reduction of active sites, 3) consumption of reactive species in reactions to produce non-radical species, and 4) Creating competition between polluting molecules and anions to react with active radical species [4].SO 4 2-anions can be adsorbed on the catalyst surface by Vander Waals force and hydrogen bond and reduce the active sites on the photocatalyst surface and then react with hydroxyl radicals and positive holes in the photocatalyst capacity strip (h + VB ) to produce SO 4 •-radicals, Equation ( 22) [32].Because sulphate radicals can act selectively, the probability of their reaction with SMX molecules is reduced.SO 4 2-anions can also compete with organic pollutants to react with photocatalytic oxidising species and reduce the degradation efficiency of SMX [33].In the case of anion Cl −, • OH oxidative radicals and holes are consumed and Cl • radical species are produced, which due to the lower oxidation potential of these species compared to • OH radicals, reduces the efficiency of the photocatalytic process, Equations ( 23)-( 24) [4].For NO 3 − ions, the process efficiency is reduced by adding NO 3 − ions to the reaction solution.NO 3 − anion reacts with active radicals and holes to produce NO 3 • (E0 = 2.30 V) and NO 2 • (E0 = 1.03 V) radicals, which have less oxidation potential than active species such as • OH and reduced process efficiency, Equations ( 25)-( 26) [4].
With adding CO 3 − to the reaction solution, the removal rate decreases slightly, this can be due to the adsorption of CO 3 − anions on the photocatalyst surface and the reduction of the active surface.Also, CO 3 − anions produce CO 3 −• radicals with lower oxidation potential (E0 = 1.78 V) with consuming hydroxyl radicals [4].On the other hand, CO 3 − anions react with h + VB to produce CO 3 −• radicals and prevent the production of • OH radicals, Equation ( 27) [34].

Stability and reusability of Cu-TiO 2 /CQDs
Reusability of photocatalysts is one of the important items in determining operational and economical profits.Additionally, photocatalyst stability and maintenance of photocatalytic activity are important.To do that, the photocatalyst was washed using dilute HCl after each cycle to remove SMX molecules and intermediates from the catalyst surface.The photocatalyst was washed 3 times using ethanol and dried at 85°C for 2 h.The results showed that after 6 times reuse of the photocatalyst during 60 min of reaction under optimal conditions, the degradation efficiency decreased from 100% to 94% (Figure S5).Blockage of the pores on the surface of the photocatalyst by SMX molecules and intermediates produced by SMX decomposition may be the reason for that result.
Considering the results, no significant reduction in the removal efficiency was observed, which indicated high activity of the synthesised catalyst after six reuses.The results of XRD analysis did not show significant changes in peak intensities after six consecutive uses, revealed no change in the structure of the Cu-TiO 2 /CQDs catalyst.

Mineralisation of SMX
To evaluate the Cu-TiO 2 /CQDs photocatalyst process in SMX mineralisation, the removal rate of TOC under optimal conditions was investigated.The removal rate of SMX was 100% after 60 min of reaction, when the removal rate of the TOC was about 81% (Figure S5).The reasons for the reduced efficiency of TOC removal compared to SMX molecules attributed to this fact that some by-products of SMX degradation are resistant to free radicals, and there is competition between by-products with the main compounds for reacting with free radicals [6].

Comparative study
To compare the photocatalyst effectiveness and evaluate the capability of the Cu-TiO 2 /CQDs 4wt% /Vis photocatalytic process with each of the under Vis, TiO 2 /Vis, Cu-TiO 2 /Vis, and Cu-TiO 2 /CQDs 4wt% , comparative experiments were performed under the operating conditions.The most efficient method for SMX degradation was the Cu-TiO 2 /CQDs 4wt %/Vis process, which had a removal efficiency of 100% in comparison to the other studied methods (Figure S5).The removal efficiency of SMX by Vis only, TiO 2 /Vis, Cu-TiO 2 /CQDs 4 wt% and Cu-TiO 2 /Vis was 8%, 31%, 38% and 64%, respectively.The Vis only process had less efficient than other processes, which was due to the lack of active radicals such HO˙ in the process, resulting in reduction of the Vis only process efficiency.In the photocatalytic process of TiO 2 /Vis, the removal efficiency is higher than the Vis only process.That is due to the stimulation of TiO 2 nanoparticles by light irradiation and the production of e − /h + pairs that produce active radicals such as HO˙ and O 2 ˙− and eventually leads to oxidation and decomposition reactions of SMX.In the Cu-TiO 2 /Vis process, the performance of the process is significantly improved after the addition of copper to the structure of TiO 2 nanoparticles.In this process, separation efficiency of e − /h + pairs increased and in consequence producing the desired amount of oxidising radicals in SMX decomposition improved.In the Cu-TiO 2 /CQDs 4wt% process, the removal efficiency was 38% in 60 min, which is due to the adsorption mechanism during the SMX removal process, indicating the synthesised catalyst has the adsorption capacity to adsorb SMX molecules.The higher removal efficiency in the Cu-TiO 2 /CQDs 4wt% /Vis process than other investigated processes is because of significant synergistic effect between the processes applied in Cu-TiO 2 /CQDs 4wt% /Vis.Catalyst particles produce more electron-cavity pairs under Vis, which ultimately lead to the production of more active oxidising species and increased the removal efficiency of SMX during oxidation and decomposition reactions [29].

Photocatalytic mechanisms
The mechanism of the Cu-TiO 2 /CQDs photocatalyst process for SMX degradation was investigated.The main reactions in the electron transfer process for Cu-TiO 2 /CQDs photocatalysts under visible light are described in Equations ( 28) and (29).
Initially, when the photocatalyst is exposed to visible light, e − /h + pairs are produced under photocatalytic reactions (Equation ( 28)).The copper in the TiO 2 lattice increases the migration of electrons from the valence band to the conduction band and reduces the recombination of the photogenerated e − /h + pairs and causes better performance of photocatalytic in the presence of visible light.On the other hand, the CQDs decorated on the Cu-TiO 2 photocatalyst acts as an electron reservoir, and the photogenerated electrons in the TiO 2 conduction band are transferred to the CQDs, and CQDs causing the separation of the e − /h + pairs Equation ( 29) and increase the performance of the photocatalyst in the degradation of pollutants.In the next step, HO˙ radicals are formed in the valence bond through oxidation of water molecules adsorbed on the catalyst surface, Equations ( 30)- (31).The electrons produced by light in the conduction band and electrons transferred from the conduction band to CQDs react with dissolved oxygen, and species of HO 2 ˙, O 2 ˙−, HO˙ and H 2 O 2 produce, Equations ( 32)-(40).In addition, the H 2 O 2 molecules formed react with O 2 ˙− radicals to form HO˙ radicals.Finally, all radical species produced in photochemical reactions are involved in the degradation and oxidation of the SMZ molecules and the formation of intermediate products, Equation (41).

Conclusion
In this study, Cu-TiO 2 and Cu-TiO 2 /CQDs nanocomposite were successfully synthesised and investigated for photocatalytic degradation of SMX under visible light irradiation of aqueous solution.The main concluded points are listed below: • Adding CQDs to Cu-TiO 2 in a suitable weight ratio improved light absorption in the visible light range, which DRS analysis also confirms this.• In the XRD analysis, no spectrum related to the element copper was observed, indicating low porosity and replacement in the titanium lattice.
• O 2 •and HO • radicals played an important role as oxidant species during the SMX degradation process, and O 2 •species had the greatest effect in SMX degradation.
• The lowest inhibitory effect on the SMX degradation was for CO 3 − anion and the highest inhibitory effect was for SO 4 2-anion.
• The results of kinetics fit well with the first-order model and also Langmuir model described better adsorption isotherm.
Scheme 1.The possible mechanism for the degradation of SMX by Cu-TiO 2 /CQD 4 wt% system under visible light.
• The catalyst showed acceptable recycling properties.
-However, further studies are still needed to utilise this nanocomposite in real water and wastewater treatment plants.

Figure 1 .
Figure 1.(a) XRD spectra, (b) the FTIR spectrum of the samples, (c) UV -Vis DRS spectra of the samples, (d) N2 adsorption/desorption pattern of the samples.

Figure 3 .
Figure 3. (a) equilibrium adsorption capacity and pseudo-second-order model (inset), (b) effect of different amount of the CQD, (c) effect of pH, and (d) Effect of photocatalyst loading.

Figure 4 (
Figure 4(b) shows the effect of changes in the light intensity in the range of 15 to 75 mW/ cm 2 in the SMX degradation efficiency by the photocatalytic process under visible light.As can be seen in Figure4(b), the increase in the intensity of light irradiation from 15 to 75 mW/cm 2 increased the SMX degradation efficiency from 74% to 100%, which can be due to the increase in light absorption by the photocatalyst when light irradiation intensity increased.Consequently, the number of electron-holes produced by light photons increased and led to an increase in the production of active radicals to destroy the SMX antibiotic[6,31].The optimal dose of light intensity in this study was chosen 75 mW/cm 2 for photocatalytic decontamination of SMX Cu-TiO 2 / CQDs 4 wt% composite.

Figure 4 .
Figure 4. (a) Effect of SMX initial concentration, (b) Effect of light intensity, (c) pseudo first order kinetics of SMX degradation by Cu-TiO 2 /CQD 4 wt% , (d) Identification of active species on SMX decontamination process.
comparison of the intervening role of these ions in the photocatalytic removal efficiency of SMX is as following: ions.SMX degradation efficiency in the absence of anion was 100%, but with adding 10 mg/L anion to the reaction solution, degradation efficiency reduced to 72%, 85%, 89% and 94% in the presence of

Table 1 .
The specific surface area, pore diameter, and pore volume of the samples.

Table 2 .
The values of kinetics and isotherms of SMX adsorption on Cu-TiO 2 /CQD 4 wt%.