A dispersive liquid-liquid microextraction combined with high-performance liquid chromatography with diode array detection (DLLME-HPLC-DAD) has been  developed for the determination of 15 sulfonamides in water samples. The effects of experimental   parameters on the performance of the method such as the type and volume of extraction and disperser solvents, pH of the solution, effect of salt, and centrifugation time were evaluated, and optimum conditions were established. Under these optimum conditions, linearity was found in the range of 2 - 1 000 μg L-1 with regression coefficients better than 0.9990. The limit of detection (LOD) and limit of quantification (LOQ) values were in the range of 0.6 - 7.8 μg L-1 and 1.5 - 8.3 µg L-1, respectively. Intra-day and inter-day precision   ranged from 1.5 - 9.7% and from 3.1 - 9.9%, respectively. The accuracy of the method were also evaluated in  samples from three different sources of water (river water, groundwater,  and tap water) were found to  range from 70.5 to 103.9% with %RSD values ranging from 0.5 to 9.8%, with the exception of sulfaguanidine in river water and groundwater where the percentage recovery ranged from 54.3 - 84.9% at the two higher concentration levels (100 and 400 µg L-1).Furthermore, the effect of sample matrix on the extraction of target compounds was evaluated using the relative recovery (%RR) of the samples from three different water sources relative to the ultrahigh purity (UHP) water sample spiked with target analytes, and found in  the range of  70.3 - 106.1%.The proposed method was successfully applied to wastewater treatment plant samples and  Sulfisoxazole (SSO) was detected in the influent samples in the range of 14.8 - 17.8 µg L-1 with corresponding %RSD, (n = 5) value of 1.6% to 1.9%.The results indicated that the proposed method is effective for the extraction and determination of the sulfonamides in water samples.

Currently,  the use of antibiotics in human and veterinary medicine has become a common practice that has led to the gradual accumulation of antibiotics in the environment, which enter through sources of contamination such as wastewater, landfills, urban sites, as well as industrial and hospital effluents.(Manzetti and Ghisi 2014) Furthermore, due to the limited knowledge of the input, fate, and effects of most antibiotics in the environment, there is no regulation on the levels of these compounds in environmental matrices, including water systems, sediment, and soil  (Kümmerer 2009)However, concerns about the occurrence, transport, and fate of veterinary drugs in the environment have been ever increasing in recent years, and several initiatives  have been launched to establish or strengthen the monitoring systems.(Seifrtová, Pena et al. 2008, Pan, Qiang et al. 2011, Iglesias, Nebot et al. 2013) 
Sulfonamides may enter natural systems following different paths, via  discharge from wastewater treatment  plants (WWTPs), leakage  from septic systems and agricultural waste storage facilities as well as the application of human and agricultural waste to land to supplement fertilisers.(García-Galán, Silvia Díaz-Cruz et al. 2008, Kümmerer 2009, Baran, Adamek et al. 2011) Sulfonamides are fairly soluble in water and their ionisation is highly dependent on the pH of the matrix (structures, KOW. and pKa values are shown in Table S1, supplemental material). Thus, after disposal in soils they may enter surface runoff or be leached into the groundwater. Furthermore, these compounds may not be completely metabolised and a high proportion of them are possibly excreted directly in faeces (digested wastes) and urine. Therefore, both the parent compounds and their metabolites can be released directly to the environment.(Białk-Bielińska, Stolte et al. 2011) It is therefore important to develop simple, efficient, and eco-friendly analytical methods for the monitoring of these drug residues. Various methods have been developed for the determination of SA residues, such as gas chromatography-mass spectrometry (GC-MS),(Chiavarino, Crestoni et al. 1998) capillary electrophoresis with ultraviolet detection (CE-UV),(Wen, Li et al. 2011)and high-performance liquid chromatography (HPLC) with ultraviolet (UV),(Chitescu, Nicolau et al. 2011) fluorescence detection (FD),(Mor, Sahindokuyucu Kocasari et al. 2012, Arroyo-Manzanares, Gámiz-Gracia et al. 2014) MS or MS/MS,(Senta, Terzić et al. 2008, García-Galán, Díaz-Cruz et al. 2010, Tölgyesi, Berky et al. 2013) and electrochemical methods.(Fotouhi, Fatollahzadeh et al. 2012). 
Since sulfonamides are found at very low (µg L-1) levels and in most cases more than one type of compound (parent and metabolites) would be determined, efficient sample preparation (extraction, enrichment, and/or clean up) techniques are mandatory prior to instrumental analysis. Liquid-liquid extraction,(Koesukwiwat, Jayanta et al. 2007) solid-phase extraction,(García-Galán, Díaz-Cruz et al. 2010)solid-phase microextraction, (Lu, Chen et al. 2007) and pressurised liquid extraction(Yu, Tao et al. 2011) have thus been widely used as sample preparation methods for the extraction of sulfonamides from environmental and food matrices. Solid-phase microextraction (SPME) offers simpler steps, compatible with GC and LC separation techniques, and it is almost solvent-free, but it has inherent drawbacks in that the fibre can break easily, the fibre coating can be damaged, and the fibre assembly is relatively expensive.(Schmidt and Podmore 2015) On the other hand, pressurised liquid extraction (PLE) requires expensive instrumentation, and it is less selective.(Yu, Tao et al. 2011) 
In recent years, liquid-phase microextraction (LPME) techniques have become widely applied techniques in analytical research areas,(Sarafraz-Yazdi and Amiri 2010, Asensio-Ramos, Ravelo-Pérez et al. 2011, Prosen 2014) due to their consumption of very small volumes (µL) of extraction solvent; one of the objectives of green chemistry is addressed by minimising the amount of solvent usage (miniaturisation) in the analytical process.
Among the above-mentioned recently developed sample preparation techniques, dispersive liquid-liquid microextraction (DLLME) is of particular interest due to the fact that the technique has several advantages, including simplicity of operation, rapidity, low cost, high recovery, and enrichment factors.(Rezaee, Assadi et al. 2006, Yan and Wang 2013). In DLLME, extraction is carried out between the sample and a cloud of fine extractant droplets formed when the mixture of extraction solvent, which is water immiscible, and the disperser solvent, which is miscible with both aqueous solvents and extraction solvents, is injected rapidly into an aqueous sample containing the analytes of interest.(Yan and Wang 2013).  A very high contact area is generated between the aqueous phase and the extraction solvent; thus, analytes are rapidly extracted into the extraction phase.(Yan and Wang 2013). Therefore, analytes have to be in their neutral form for them to efficiently partition between the aqueous phase and the hydrophobic organic phase (extraction solvent). This explains why the technique was initially applied to neutral or non-polar organic compounds such as PAH which have larger octanol-water partition coefficient (KOW) values.  However, the application of   DLLME for polar analytes poses a challenge, since they will rather remain in the aqueous phase than be extracted into the organic solvent. These analytes are soluble in water mainly in their ionised form; therefore, appropriate adjustment of sample pH is necessary to suppress ionisation of target analytes. Furthermore, by using an appropriate amount of salt to induce a “salting-out effect”, the extractability of polar analytes can be improved as their solubility in the aqueous phase is reduced. Compounds with a wider range of polarity like sulfonamides present yet another dimension of complexity. As a result of such challenges, most of the reported DLLME methods often focused on the less polar group of sulfonamide compounds(Wen, Li et al. 2011, Xu, Su et al. 2011, Li, Li et al. 2016, Wang, Li et al. 2022) with  a few exceptions  occasionally being reported for some polar group of sulfonamides such as Herrera-Herrera et al.(Herrera-Herrera, Hernández-Borges et al. 2013) in 2013 reported for the determination of 11 sulfonamides together with 14 fluoroquinolones using DLLME coupled with UHPLC-DAD. However, most of the reported methods require longer extraction times or use of advanced instrument. Furthermore, in some cases a higher volume of solvents and detection limits were reported. Therefore, it is desirable to develop a fast, cost effective and sensitive analytical method for simultaneous extraction and determination of wide number of sulfonamides with diverse polarity ranges.
Thus, the aim of the current study was to develop a fast, cost effective and sensitive DLLME method coupled with HPLC –DAD for the simultaneous extraction and determination of 15 sulfonamide compounds in diverse water samples. The effects of various parameters on the extraction efficiencies of target compounds such as the type and volume of extraction and disperser solvents, pH of the solution, effect of salt addition and centrifuge time for DLLME procedure were   evaluated and optimum conditions were established. The performance of the developed method was validated and applied to wastewater samples. Results revealed that the proposed method is effective to be used for routine monitoring of sulfonamides and related drugs in diverse water samples

2.1. Standards and chemicals
Antibiotic standards included sulfaguanidine (SGD), sulfanilamides (SAM), sulfathiazole (STZ), sulfacetamides (SAA), sulfamethizole (SMT), sulfamethoxazole (SMX), sulfasalazine (SSA), sulfamonomethoxine (SMM), sulfaquinoxaline (SQX), sulfamerazine (SMR), sulfapyridine (SPY), sulfadiazine (SDZ), sulfabenzamide (SBZ), sulfachloropyridine (SCP), and sulfisoxazole (SSO), and were purchased from Sigma-Aldrich (Steinheim, Germany). All standards had purity higher than 98%. Methanol, acetone, and formic acid were obtained from Sigma - Aldrich (Steinheim, Germany). Acetonitrile (MeCN) was purchased from ROMIL Ltd. (Waterbeach, Cambridge, UK). Sodium   hydroxide (NaOH) was supplied by Merck (Darmstadt, Germany). Trichloromethane, dichloromethane, 1,2-dichloroethane and 1,1,2,2-tetrachloroethane were of HPLC grade and were purchased from Sigma-Aldrich (Steinheim, Germany). Ultrahigh purity (UHP) water (resistivity, 18.2 MΩ∙cm at 25 ℃) was generated using the Milli-Q® system (Millipore, Billerica, MA, USA). 
2.2. Sampling of tap water, groundwater, river water and wastewaters from wastewater treatment plant (influents and effluent)
For the purposes of validating the proposed method, four different types of water samples were collected using 2.5 L sampling bottles: Tap water I sample from Pretoria, South Africa; tap water II sample from Florida, South Africa; river water sample from Suikerbosrand River (Gauteng, South Africa); and groundwater sample from Florida, South Africa.  Wastewater   treatment plant samples (influent and effluent) were collected from the Daspoort Wastewater Treatment Plant (WWTP) located at latitude 25.73450 south and longitude 28.17810   East coordinates in Pretoria, Gauteng Province, South Africa. These samples were collected from three different locations: -i) before entry into the treatment plant (Influent I); ii) before entry into the treatment plant from a different location (influent (II); and iii) after treatment at the outlet (effluent). The grab sampling method was used, and the samples were collected into amber glass bottles, which had previously been acid-washed and rinsed with UHP water and then flushed with wastewater before collection. Each sample was collected in triplicate into 2.5 L bottles. The water samples were filtered through Whatman filter paper and 0.45 µm nylon filters, respectively, and then stored in amber glass bottles in a refrigerator at 4 ℃ until analysis. 
2.3. Instrumentation
An Agilent 1260 series high-performance liquid chromatographic system (Agilent Technologies, Waldbronn, Germany) was used for all separations. The HPLC consisted of a binary pump, vacuum degasser, thermostatted column compartment, auto-sampler, diode array detector (DAD), and fluorescence detector (FLD). Data acquisition was achieved using the Agilent ChemStation (version 1.9.0) software. Chromatographic separations were carried out using ZORBAX Eclipse Plus C18 column (100 mm x 4.6 mm, 3.5 µm) from Agilent Technologies, Inc. (Santa Clara, CA, USA). A vortex mixer (VELP Scientifica, Usmate Velate (MB), Italy), centrifuge (Thermo Electron Corporation, Massachusetts, USA) and a micro balance (Mettler Toledo XP6U) were used for sample preparation. A Hamilton® 500 µL syringe (Hamilton Bonaduz AG, Switzerland) was used for withdrawing the settled phase in the DLLME procedure and nitrogen gas was used for drying the samples.
2.4. Preparation of sulfonamide standard solutions
Stock standard solutions (1 000 mg L-1) for each compound were prepared by dissolving10 mg of accurately weighed standard of each compound in 10 mL of a mixture of methanol and ultrahigh purity water (1:1; v/v). Appropriate dilution of these stock solutions with methanol-ultrahigh purity water (1:1; v/v) was used to prepare various concentrations of working solutions. All standard solutions were protected from light by covering the sample vials with aluminium foil and were kept at 4 ℃ until use.
2.5. Optimisation of chromatographic conditions for the separation of 15 sulfonamide compounds
A chromatographic method for the separation of 15 sulfonamides was first developed and optimised. To investigate the conditions for optimum peak shape and adequate resolution in the separation of target analytes, both ultrahigh purity (UHP) water and acidified ultrahigh purity water were evaluated as a solvent A and organic solvents (acetonitrile and methanol) as a solvent B. The effect of mobile phase flow rate was also evaluated in the range of 0.3 - 2 mL min-1 at 0.2 mL min-1 interval. Both isocratic and gradient elution modes were investigated Furthermore, the column oven temperature was also optimised in the range of 25 - 45 ºC and compounds were monitored using DAD at different wavelengths (260, 265, 270 and 280 nm). 
2.6. Optimisation of DLLME conditions
In DLLME, the factors that can affect the efficiency of extraction include type and volume of extraction and disperser solvent, the pH of the solution, salting-out effect and the extraction time. A disperser solvent must disperse the extraction solvent (form fine droplets) into the aqueous solution while the volume of extraction and disperser solvents would have a significant effect on the enrichment factor (EF). Therefore, it is important to optimise all the factors in order to obtain good performance.
2.6.1. Evaluation of extraction solvent
In DLLME  it is very important  to select an appropriate extraction solvent that is capable of extracting  the target compound, has a low solubility in water, good chromatographic behaviour (i.e. should not absorb  at the wavelength of target analytes), has  a higher   density than water, and is able to form two phases in the presence of a disperser solvent.(Rezaee, Assadi et al. 2006, Schmidt and Podmore 2015). Based on the above criteria, dichloromethane, CH2Cl2 (1.32 g mL-1), trichloromethane, CHCl3 (1.47 g mL-1), 1,2-dichloroethane, C2H4Cl2 (1.25 g mL-1), and 1,1,2,2-tetrachloroethane, C2H2Cl4 (1.59 g mL-1) were selected as potential solvents for this study. The extraction solvents were   evaluated at a volume of 600 µL and using 1 mL of MeCN as the disperser solvent; these are conditions adapted from previous reports in the literature.(Rezaee, Assadi et al. 2006, Wen, Li et al. 2011, Xu, Su et al. 2011, Yan and Wang 2013, Li, Li et al. 2016).
2.62. Effect of the volume of extraction solvent
The volume of the extraction solvent has an impact on the enrichment factors. If very low volumes are used, the extraction solvent may not be distributed well in the aqueous phase resulting in low enrichment factors. On the other hand, high volumes of extraction solvent may lower the optimum ratio of the disperser solvent to the extraction solvent volume. This consequently reduces the amount of droplets formed which are necessary for the extraction process thereby lowering the enrichment factors.(Saraji, Khalili Boroujeni et al. 2011) Therefore, the effect of extraction solvent volume on the enrichment factor was investigated over a range of 200 - 1 000 μL, while all other experimental parameters were maintained constant. 
2.6.3. Evaluation of disperser solvent 
Evaluation of an appropriate disperser solvent is important since it can affect the enrichment factors of target analytes. Disperser solvents must fulfil the condition of being miscible with both water and the extraction solvents.(Rezaee, Assadi et al. 2006) 
In this study, methanol, acetonitrile, ethanol and acetone were selected as potential disperser solvents and their effect on the enrichment factor was evaluated using 1 mL of each solvent and 600 μL of dichloromethane as the extraction solvent.
2.6.4. Effect of the volume of disperser solvent
To determine the optimum volume of the disperser solvent, various volumes (200 - 1 200 μL) of acetonitrile containing a fixed volume of dichloromethane (400 μL) were evaluated. 
2.6.5. Effect of the solution pH 
In principle, DLLME is most appropriate for extracting neutral or non-charged compounds from aqueous to organic phase, thus in a way, making it a limited technique. Therefore, pH control becomes essential to extend is applicability to ionic and polar compounds such as sulfonamides. Theoretically, SAs are positively charged in acidic medium at two pH units below their acidic pKa values, neutral at pH between pKa1 and   pKa2, and negatively   charged at 3.3 pH units above their basic pKa2 values. The pKa1 values of target compounds are in the range of 0.9 - 2.5 and their pKa2 values are in the range of 5.3 - 11.3. One of the challenges when dealing with analytes of diverse polarity is to find a suitable pH in which the majority of target analytes can be    extracted in their neutral form. Therefore, the selected pH must be adequate for the extraction of all analytes. Thus, the effect of the solution pH  was   evaluated by varying it in the range of 2.5 - 7.5 using HCl (0.1 M) and/or NaOH (0.1 M) and keeping the other conditions constant. 
2.6.6 Effect of salt addition
The addition of salt into the sample solution decreases the solubility of the analytes, thus resulting in the transfer of analytes into the organic phase. Therefore, the effect of salt on the enrichment factors of target compounds was tested by adding NaCl in the range of 0 - 8 % (w/v), while keeping all the other experimental conditions constant(Xu, Su et al. 2011). 
2.6.7 Effect of centrifugation time
Adequate centrifugation time is important for the formation of phase separation. A very long centrifugation should be avoided because a centrifuge generates heat as it rotates, which can destabilise phase separation. Therefore, in this study, the effect of centrifugation time was evaluated in the range of 1 - 10 min. 
2.7. Dispersive liquid-liquid microextraction procedure for water samples
Into 15 mL screw cap centrifuge tubes, 5 mL of ultrahigh purity (UHP) water, with the pH previously adjusted to 3.5 using 0.1 M HCl that contained 0.10 g (2%; w/v) NaCl, was transferred, and spiked with a mixture of SAs at a concentration of 100 μg L-1. Thereafter, the mixture of extractant and disperser (400 µL and 600 µL, respectively) was    rapidly    injected into the aqueous solution containing the sample using an aseptic disposable syringe. The mixture was then vortexed for 30 s to ensure complete dispersion and to facilitate the extraction. The dispersed fine particles of the extraction phase, which had settled at the bottom of a screw cap centrifuge tube, were transferred to a 1.5 mL HPLC vial (Agilent, Germany) through a 400-µL vial insert using a Hamilton 500-µL microsyringe and dried under a gentle stream of nitrogen gas. Thereafter, the remaining residue was   reconstituted with 100 µL of the mobile phase and subjected to chromatographic analysis.
2.8. Chromatographic conditions for the separation of 15 sulfonamide compounds
The chromatographic conditions for the separation of 15 SAs were as follows:  a binary mobile phase comprising of solvent A (0.1% formic acid at pH 2.73) and solvent B (acetonitrile) with a gradient elution of 10% B (0 - 1 min), which was gradually increased from 10% to 40

For chromatic separation of sulfonamide compounds, optimum resolution and peak shape were obtained at a flow rate of 1.8 mL min-1. The best peak shape and satisfactory resolution of the target compounds were achieved using a binary mobile phase comprising solvent A (0.1% formic acid in UHP water at pH 2.73) and solvent B (acetonitrile). Satisfactory peak shapes and resolution were achieved using a gradient elution program consisting of mobile phase B (10%) for 0 - 1 min which was gradually (at 1% intervals) increased from 10% to 40

Under optimised conditions, the developed method was validated using parameters such as linearity, limit of detection (LOD), limit of quantification (LOQ) and precision (intra-day and inter-day). Linearity was evaluated using calibration standards prepared by spiking ultrahigh purity (UHP) water at a concentration range of 0.5 - 2 000 µg L-1 and a satisfactory linearity was obtained from 2 - 1 000 µg L-1 with regression coefficients better than 0.9990. The LOD and LOQ values were calculated based on 3- and 10-times standard deviation of UHP water spiked with a minimum analyte concentration (0.3 - 0.5µg L-1) respectively. The LOD values ranged from 0.6 µg L-1 - 7.8 µg L-1 while LOQ values were obtained in the range of 1.8 - 23.4 µg L-1 (Table 1). 

Table 1: Quantitative result of linearity, and LOD and LOQ values obtained for 15 SAs

Inter-day and intra-day precision was investigated by spiking UHP water samples with a mixture of target analytes at three concentration levels (40 µg L-1, 100 µg L-1, and 400 µg L-1) for three consecutive determinations in a single day and five determinations in five days, respectively. Intra-day %RSD values at concentration levels of 40 µg L-1, 100 µg L-1, and 400 µg L-1 were in the range 4.1 - 9.5%, 3.6 - 9.7%, and 1.5 - 9.3%, respectively, while for inter-day precision at the same three concentration levels, the ranges were 3.4 - 9.9%, 5.0 - 9.6%, and 3.1 - 9.9%, respectively (Table 2).

Table 2. Intra-day and inter-day precision results (%RSD) for water samples spiked at 40 µg L-1, 100 µgL-1, and 400 µg L-1 levels of 15 SAs

The accuracy of the method was evaluated by spiking the same levels of sulfonamide standard mixtures in three different types of water samples (river water, groundwater, and tap water) and the percentage recoveries were calculated. The results shown in Table 3 revealed that satisfactory recoveries, ranging from 70.5 to 103.9% with %RSD values ranging from 0.5 to 9.8%, were obtained, with the exception of SGD in river water and groundwater where recoveries in the range of 54.3 - 84.9% at the two higher concentration levels (100 µg L-1 and 400 µg L-1) were obtained. This lower recovery of SGD may be attributed to its very low KOW value (-1.07). 

Table 3: The recovery (%R) result obtained by comparing the ratio of concentration of each analyte found in spiked river water, groundwater samples. tap water I (Pretoria, South Africa) and tap water II (Florida, South Africa) at 40 µg L-1,100 µg L-1, and 400 µg L-1 levels of a mixture of SAs with the spiked concentration and results were multiplied by 100.
Furthermore, in order to study the effect of sample matrix on the extraction process, the relative recovery (RR), which is defined as the peak area ratio of the natural water sample to the UHP water sample spiked with analytes at the same concentration levels (40 µg L-1, 100 µg L-1, and 400 µg L-1) were calculated and found in the range of 70.3 - 106.1%, 72.5 - 105.4%, and 72.6 - 104.9%, respectively (Table S2, Supplemental material). The recovery results were in agreement with those previously reported in the literature on the same sample and matrix.34 The result demonstrated that the matrices of the real water samples do not have significant effects on the proposed DLLME method for the extraction of target analytes from real water samples. The selectivity of the proposed method was also evaluated by comparing the chromatograms of the blank (unspiked) water samples with the corresponding spiked water samples. Results shown in Figures S1 – S5 (supplemental material) confirmed that no interferences were observed at the retention times of the target analytes.
5. Application of the proposed method to Wastewater samples from Wastewater Treatment Plant (WWTP) 
To demonstrate the applicability of the proposed DLLME method, influent and effluent samples of the WWTP were collected, i.e., influent I, influent II, and effluent samples; these samples were analysed to assess the presence and levels of sulfonamides. A parallel group of the same samples was also analysed after spiking with 100 µg L-1 of target compounds. As shown in Table 4, SSO was detected in the influent I sample at an average concentration of 17.8 µg L-1 with corresponding %RSD, (n = 5) value of 1.6% and in the influent II sample at 14.8 µg L-1 with corresponding %RSD (n = 5) value of 1.9%. 

Table 4: Sulfonamide levels found in wastewater samples from wastewater treatment plant (WWTP), Pretoria, South Africa and spike recovery (%R, %RSD, n = 5 and two injections for each sample)
The chromatograms of SSO in influent I and II samples are shown in Figures S6 and S7, respectively (supplemental material). Sulfisoxazole was also detected in the influent of WWTPs in the USA in the range of 3.2 to 22.1 ng L-1,(Spongberg and Witter 2008)which was lower than that detected in the current work. It has also been detected in  WWTP effluents in  Canada (19 - 34 µg L-1),(Rodríguez-Cabo, Rodríguez et al. 2011) in the USA (3.2 to 11.9 ng L-1),(Spongberg and Witter 2008) and in Japan (0.13 -1.6 ng L-1)(Díaz-Cruz, García-Galán et al. 2008) One of the  possible reasons for the presence of sulfisoxazole is its wide application in human medicine on its own or together with erythromycin for urinary tract infections.(Chung 2008) In addition,  the compound is also excreted mainly in urine in an uncharged form. On the other hand, none of the target compounds were detected in the effluent samples. 
Sulfonamides have been  detected in WWTPs in Europe, America, Asia, Australia,(Zhang and Li 2011) and in Africa.(Faleye, Adegoke et al. 2018) Among these, sulfamethoxazole was the most  frequently detected, followed by sulfamethazine, sulfapyridine, and sulfadiazine.(Faleye, Adegoke et al. 2018, Zhang, Bai et al. 2023). The highest reported concentration of sulfamethoxazole in WWTP influent was 5 597 ng L-1 and in WWTP effluent it was 6 000 ng L-1 in Europe, America, Asia and Australia,(Díaz-Cruz, García-Galán et al. 2008). United nation reported minimum and maximum concentrations for Africa were in the range of 0.1 - 29.0 μg L-1respectively(Faleye, Adegoke et al. 2018). Sulfamethoxazole has been detected wastewater effluents ranging between 0.15 and 10 μg L−1 in African wastewater effluents.(Ncube, Nuapia et al. 2021) Considering South  Africa, 
the average concentration of sulfamethoxazole in surface water was  reported as 7.3 μg L-1(Faleye, Adegoke et al. 2018) .  Due to the combination of trimethoprim and sulfamethoxazole used for the treatment of Pneumocystis pneumonia in people with HIV/AIDS, sulfamethoxazole is widely used and hence the most commonly detected sulfonamide in surface water systems in South Africa. Neube etal (Ncube, Nuapia et al. 2021) also reported the presence of sulfonamides (sulfamethiozole, sulfamethazine and sulfamethoxaxole) in their study for the presence of antibiotics in a South African stream.

The proposed method was also compared with the other reported DLLME methods based on the number of analysed analytes, volume of solvent consumption, extraction time, linearity range, recovery values, and LOD values. The results shown in Table 5 confirmed that the proposed method has a wider linear range, short extraction time (less than 5 min), a lower consumption of organic solvents than the methods reported in the literature,(Wen, Li et al. 2011, Li, Li et al. 2016) a larger number of analytes, and a wide range of KOW values (-1.07 - 3.8). However, the percentage recovery and LOD values were similar to those reported by Wen et al.(Wen, Li et al. 2011) and  Wang et al  (Wang, Li et al. 2022) and higher than the values reported by Xu et al.(Xu, Su et al. 2011) Therefore, the results indicated that the proposed method is feasible to be used as an alternative method for the determination of Sulfonamides in water samples.

Table 5: Comparison of the proposed DLLME method for water samples with other reported similar methods

In this study, dispersive liquid-liquid microextraction combined with high performance liquid chromatography-diode array detection has been evaluated and successfully developed for the simultaneous extraction and determination of  15 sulfonamides in real water samples. The effects of various parameters on the extraction efficiency were    evaluated, and optimum conditions were established. Under optimised conditions, linearity was found in the range of 2 - 1 000 μg L-1 with regression coefficients better than 0.9990. The limit of detection (LOD) and limit of quantification (LOQ) values were in the range of 0.6 - 7.8 μg L-1 and 1.8 - 23.4 µg L-1 respectively. Intra-day (n = 6) and inter-day (n = 6) precision expressed as %RSD were in the range of 1.5 - 9.7% and 3.1 - 9.9%, respectively. 
The accuracy   of the method was also evaluated in samples from three different sources of water (river water, groundwater, and tap water) and found to range from 70.5 to 103.9% with %RSD values ranging from 0.5 to 9.8% for 14 target analytes, with the exception of SGD, where recoveries in the range of 53.8 - 84.9% were obtained in river water and groundwater, with the corresponding %RSD values in the range of 0.5 - 9.5%.
Satisfactory relative recoveries (RR) in the range of 70.3 - 106.1% obtained provided evidence that the proposed DLLME   method was not affected by the matrix effects. Therefore, the proposed method was applied for the determination of sulfonamides in wastewater   treatment plant samples. Both SMX and SSO were detected; however, SMX was not quantifiable and SSO was in the range of 14.8 -17.8 µg L-1 in WWTP influents with %RSD values in the range of 1.6 - 1.9%. In comparison with other reported methods, the proposed DLLME method is adequate to be     used as an alternative method for the determination of sulfonamides in water samples.
Data availability
The raw data required to produce these findings can be obtained from authors with reasonable request
Conflict of interests
All the authors declare that they have no conflict of interest
Acknowledgments
The authors acknowledge the University of South Africa for the financial support and providing
the laboratory facilities for the student research during his PhD studies. Moreover, the authors also
Acknowledge the University of Debre Berhan (Debre Berhan, Ethiopia) for sponsoring the PhD study.