A Highly Selective Analytical Method Based on Salt-Assisted Liquid-Liquid Extraction for Trace-Level Enrichment of Multiclass Pesticide Residues in Cow Milk for Quantitative Liquid Chromatographic Analysis

In this study, a simple, inexpensive, selective, and fast salting-out assisted liquid-liquid extraction (SALLE) technique coupled with high-pressure liquid chromatography-diode array detection (HPLC-DAD) was developed for the extraction, preconcentration, and analysis of trace level seven multiclass pesticide residues in pasteurized and raw cow milk samples. The significant factors that affect the extent to which the target analytes are extracted, such as the type of extraction solvent and its volume, the type and concentration of salting-out salts, the pH of the solution, and the extraction time, have been investigated. Under optimum conditions, the correlation coefficient (r2) was obtained within a range of 0.9982–0.9997 for a broad linear range concentration of 2–1500 ng·mL−1. Reliable sensitivity was achieved with limits of detection (LODs) and limits of quantification (LOQs) ranging from 0.58–2.56 ng·mL−1 and 1.95–8.51 ng·mL−1, respectively. While precision with interday and intraday in terms of relative standard deviations (RSDs) was observed in the range of 1.97 – 7.88% and 4.52 – 8.04%, respectively. The results of the precision studies reveal that good repeatability and reproducibility (RSDs <9) were achieved, thus showing a low variability extraction of the developed method. Finally, the proposed and validated approach was effectively used to extract and determine pesticide residues in real milk matrices; however, the target analytes were not detected in all samples.


Introduction
Pesticides are chemical compounds used all over the world to control, prevent, or eliminate pests that threaten plants, animals, and human environments.It is well documented and known that their use increases agricultural productivity, though their residues greatly contaminate environmental components [1].Nowadays, pesticide use has signifcantly increased throughout the world and similarly in Ethiopia [2] mainly as a result of the country's continuous agricultural reform.Out of the enormous quantities of pesticides applied, only less than 0.1% actually reach the intended pests; the remainder may end up on other environmental surfaces and accumulate in grasslands and feed additives given to cattle and other animals [3][4][5].Pesticide residues at trace levels can be hazardous to unanticipated targets, posing a serious threat to human health and the ecosystem [2,6,7].Humans come into contact with these chemicals through unsafe use, food, or the environment [5,8].Food security is a condition in which everyone, at all times, have physical, social, and economic access to sufcient quantities of wholesome foods to meet their dietary needs and food preferences for an active and healthy life [9].Due to these facts, monitoring for pesticides in food matrices on a regular basis is crucial and has become one of the hot research topics these days [10,11].
Pesticides typically exist in low concentrations in the environment and food matrices, and determination of these trace quantities requires various analytical instruments, including gas chromatography-mass spectrometry (GC-MS) [7,10,12], gas chromatography-tandem mass spectrometry (GC-MS/MS) [13], high-performance liquid chromatography (HPLC) combined with a diode array detector (DAD) [14][15][16], tandem mass spectrometry (MS/MS) [1,5,[17][18][19], and ultraviolet detector (UVD) [20][21][22].Most of these instruments provide good selectivity, sensitivity, low detection capacity, and so on; however, there are fnancial limitations to acquire them at laboratories that are not well equipped to meet the demanded requirements.However, relatively less expensive techniques, such as HPLC-DAD, are routinely used for monitoring of pesticide and other pollutant residues.In addition, when combined with a DAD detection system, HPLC procedures are typically favored over GC ones since HPLC is used without derivatization and is a sufciently selective and sensitive analytical method [23].Terefore, HPLC-DAD was chosen for monitoring of multiclass pesticides in milk samples for the designed sample preparation methods in the current study.
Dairy farming is one of the most proftable businesses in Ethiopia, particularly in the central Oromiya regional state.Furthermore, Ethiopia has one of the highest populations of cattle in Africa, estimated at 60 million heads, and around 90% of milk products are obtained from cows [24].Milk is one of the required food item for mankind, but the question of its contamination with trace-level pesticides must be given attention, particularly when its handling personnel are untrained farmers and agricultural extension workers who lack knowledge of pesticide management, how to use for agronomy, and veterinary care are involved [8,25].Studies have revealed that despite the fact that most pesticides are often present in low concentrations, their persistence causes them to accumulate in animal tissues where they enter the food chain [5,10,11,19].Contamination of milk and milk products is extremely concerning because these foods hold a very special place in the diets of infants, young children, and the elderly for whom milk is a complete diet enriched with proteins, vitamins, fats, and essential minerals [26,27].
Te health concerns posed by trace pesticide residues in food can be signifcant, especially for young children whose enzymatic and metabolic systems are still developing [28][29][30].Research on pesticide residues in the environment and various foods that have detrimental efects on human health is receiving special attention [30].Because milk has dissolved proteins, carbohydrates, and minerals, it is difcult to recover trace-level multiclass pesticide residues with diferent physicochemical properties, and thus developing an amenable sample extraction technique and cleanup step is very crucial before chromatographic analysis [26].
Te majority of these methods are labor intensive, timeconsuming, and environmentally unfriendly, despite the fact that they ofer clear advantages for extraction of pesticides from milk.Besides, as stated explicitly in published literature, industrially produced QuEChERs kits, SPME needles, and SPE cartilage materials are quite expensive [5,6,37].
Preconcentration of multiclass pesticide residues in food samples nowadays needs the development of analytical techniques that are miniaturized, efcient, simple, fast, and afordable.Te most popular method, dispersive liquidliquid microextraction (DLLME), is limited to the use of nonpolar, water-immiscible solvents with low dielectric constants and poor extraction efciency of polar organic and inorganic compounds [38].As a result, the introduction of salt-assisted liquid-liquid extraction (SALLE), an efcient extraction method for polar to moderately polar organic compounds, was made feasible by using more polar and water-miscible organic extraction solvents like acetonitrile, isopropanol, acetone, ethanol, and methanol, among others.In the SALLE method, organic solvent is separated from the mixture, and a two-phase system is created as a result of addition of inorganic salt [39].When using inorganic or organic salts, the salting out efect increases the ionic strength of the solution and decreases the solubility of the weak electrolyte in water, which causes the target analyte to be extracted into the organic solvent, resulting in high extraction efciency of polar or slightly polar target analyte in an aqueous sample [39].Te SALLE method produces extracts with solutes in organic solvent that may be evaporated and reconstituted with an appropriate solvent for preconcentration and analysis by HPLC or GC [40,41].On the other hand, in the SALLE methods, extraction solvents are compatible with the majority of analytical instruments, particularly chromatographic ones, making it possible to directly inject the extract into these methods of analysis [36,42,43].
SALLE has been used successfully to analyze pesticides in foods [20,40,44], biological matrices [36], and environmental water [45][46][47].Researchers put a lot of work into making the method automated and high throughput during the development step to reduce processing time and chemicals required [35,37].Tough various pretreatment technologies have been developed, the method of SALLE has still been widely used, since it integrates sample cleanup, preconcentration, and extraction in one single step and shares the advantages of the sample pretreatment technique gained from QuEChERS [48,49].Even though numerous advantages were reported, the application of SALLE for enrichment of multiclass pesticides in milk samples is scarce in the reported literature.
To the best of our knowledge, there are no reports in the literature on the use of the SALLE technique coupled with HPLC-DAD for simultaneous extraction and determination of multiclass pesticide residues including carbamate (carbrayl), organophosphate (methidathion), triazines (cyanazine, atrazine, and propazine), neonicotinoid (thiamethoxam), and strobilurin (azoxystrobin) in cow milk samples.Terefore, the present study was designed to develop, optimize, and validate a simple, fast, inexpensive, 2 International Journal of Analytical Chemistry and an environment friendly (green) analytical technique based on SALLE, as an alternative for preconcentration and extraction of seven multiclass pesticide residues in cow milk samples.

Milk Samples.
A total of 7 milk samples (one fresh raw milk collected from a dairy cattle farm and three pasteurized milk processed and packed by two dairy product suppliers) were taken.Pasteurized milk samples were bought from randomly selected local supermarkets in Addis Ababa, and raw milk samples were donated from a randomly selected dairy cattle farm in Sheger city (in sululta subcity) in April 2023 for the multiclass pesticide residue analysis.After arrival at the laboratory, the pasteurized milk samples in their original packing and raw milk in a brown bottle were kept in a refrigerator at 4 °C until the time of analysis, when not in use.Note that the names of the producers have been kept confdential to protect their business and reputation.
2.6.Procedure of SALLE.Aliquots of 0.5 mL of milk sample were placed in a 15 mL falcon centrifuge conical bottom tube and then diluted to 5.0 mL with ultrapure water (pH 8.0) to reduce the matrix efect of the sample.Te sample solution pH was adjusted using 0.1 M HCl or 0.1 M NaOH solution and spiked with appropriate amount of mixed standard solutions of the pesticides.Te sample solution was then kept to stand for 20 min to equilibrate, and 1.0 mL ACN was added and vortexed for 0.5 min.Tis was followed by the addition of 2.0 g MgSO 4 to the mixture and vortexed for an additional 2 min to dissolve the salt to be used as a salting out agent.After centrifuging the resulting content at 4000 rpm for 5 min, 500 μL of the supernatant was carefully withdrawn with a micropipette and transferred to a vial fltering through a 0.22 μL flter membrane.Ten, 15.0 μL was injected into the HPLC-DAD system for extract analysis.
International Journal of Analytical Chemistry 2.7.Statistical Analysis.Descriptive statistical analysis of means, standard deviations, and relative standard deviations for data obtained during parameter optimizations and validations of the method was performed using Microsoft Ofce Excel 2010 software, and fgures were drawn using Origin 2019b software.Te organic solvents with the desired characteristics such as high capability to dissolve the analyte, miscibility with water, ease of inducing phase separation upon addition of the appropriate salt and having good chromatographic behavior were tested as extraction solvent.

Results and Discussion
Moreover, the solvent peak should not interfere with the analyte peak under the selected HPLC conditions.In this work, solvents such as MeOH, ACN, IPA, acetone, diethyl ether, and ethyl acetate were tested.A series of experiments were performed using a 5 mL ultrapure water sample containing 30% NaCl (m/v) and 2 mL of each organic solvent with the exception of methanol and acetone, in which the two phase systems were not observed.Similar observations were also noted for methanol and acetone and reported in literature by other workers [20,22,50].Te reason for the absence of phase separation in methanol could be due to the high polarity of methanol caused by its hydroxyl group and the hydrogen bond formed between this solvent and water which as a result increases its solubility [51].Figure 2 depicts the observed maximum peak area when ACN was used as the extraction solvent.Tis might be attributed to its closer polarity with water and its promising protein precipitation reagent for milk [52].Additional advantages of using ACN as an extraction solvent are its ability to extract a wide range of compounds [53] caused by its higher polarity and its less toxic and less harmful nature compared to other common extraction solvents.Tese characteristics also make it more suitable from the viewpoint of green chemistry.Tus, ACN is selected to be used as extraction solvent in this study.

Volume of the Extraction Solvent Efect. Te volume of
the extraction solvent is a very crucial parameter that infuences the extraction performance of the SALLE technique since it afects the amount of analyte solubility in the sample solution [51].Generally, the volume of extraction solvent used should be as low as possible to achieve the highest possible enrichment and the least toxicity hazards for environment.In this context, the infuence of the ACN volume on the extraction efciency was investigated between 700 and 1800 μL.As shown in Figure 3, peak areas of all the analytes increased with the volume of ACN from 700 to 1000 μL and then decreased upon further increase in the volume of the ACN.With low volumes, i.e., lower than 700 μL, the interface between the extraction solvent and the aqueous phases was not clear, and collection of the organic layer was found to be difcult.A decrease in extraction efciency above 1000 μL may be due to the dilution efect resulting from the higher volume of the organic phase obtained after extraction, and hence further higher volumes were not performed [22].Hence, based on the observed experimental results, 1000 μL ACN was selected as the optimum volume in all the subsequent experiments.

Efects of the Salt Type.
Te solubility of both the analytes as well as the extraction solvent in the aqueous phase could be decreased by salt addition, and this in turn enhances the analytes transfer into the organic phase [44,48].As diferent salts have the capacity to cause diferent degrees of phase separation [52], in this study, the efect was evaluated by addition of the salts such as NaCl, (Na) 2 SO 4 , MgSO 4 , (NH 4 ) 2 SO 4 , and NH 4 CH 3 CO , using 30% (m/v) of each salt, as potential salting out agent.It was observed that all salts could induce phase separation, but as it can be seen from Figure 4, the highest instrumental response for all analytes was obtained when MgSO 4 was used as the salting out agent.Tis could be due to its high ionic strength per unit concentration in the aqueous phase [43].

Efect of Salt Concentration.
Varying salt concentrations may cause the degrees of phase separation to vary [43,54].A salting-out study was carried out by adding diferent amounts of MgSO 4 in the range of 0.75 g-2.5 g (or 15-50%, m/v) in the aqueous sample solution.It was shown that in Figure 5, the peak area of the target analytes was slightly increased as the concentration of the salt increases from 1 g to 2 g.However, at higher concentrations, the peaks were observed to slightly decrease for all the target analytes, and thus 40% m/v (2 g) was chosen to be the optimum for the following experiments.Similar quantities of this salt were found to cause a signifcant salting-out efect in the SALLE analytical method, reported in the literature and employed for fruit juice, yogurt, and carbonated drink matrices [41,44].

Efect of Sample pH.
In SALLE, the sample solution pH also has a signifcant role on the extraction efciency of the multiclass pesticides, as it afects the extent of their ionization as well as the solubility in aqueous media [35,51,55].Te efect of this parameter was evaluated by carrying out a series of experiments varying the pH values from 3.0 to 9.0 in the aqueous solution.Tese pH values were adjusted using HCl and NaOH.Te experimental results International Journal of Analytical Chemistry obtained revealed that pH 8 was the optimum value, as shown in Figure 6.Tis could mainly be associated with the enhanced stability of the target analytes in the weakly alkaline solution, while they were easily degraded in acidic and strongly alkaline environments [4].Terefore, pH 8 was selected as the optimum value for the subsequent studies.

Efect of Centrifugation Time.
In SALLE procedures, optimizing the time required for phase separation is also an important analytical step, in order to obtain a clear extract [44].In order to establish the optimum conditions: centrifugation time was varied between 1 and 7 min, with a 2 min interval, at a constant speed of 4000 rpm.Based on the peak areas representing the target analytes, the highest results were obtained at the centrifugation time of 5 min (shown in supplementary material Figure S1).Terefore, centrifugation time of 5 min was selected as the optimum time for the subsequent studies.

Efect of Vortex Agitation Time.
Mass transfer is a time-dependent process and one of the most important factors in most of the extraction procedures [44].Vortex was performed to strengthen the contact between acetonitrile and the aqueous sample solution (i.e., infuence the kinetics of the extraction), thus facilitating the formation of the twophase system.Besides, in the present study, vortex agitation was also employed to enhance the dissolution of the saltingout salt.Terefore, a vortex time was evaluated in the range of 0.25-4 min, at the maximum vortex speed, and thus a slight increase of peaks was obtained when the vortex time increased from 15 sec to 30 sec.Tis indicates that the diffusion of pesticides from the sample to the acetonitrile medium was found to require a short time.A decrease in extraction efciency after 30 sec (Figure 7) may be associated to the back extraction.Tus, extraction time of 30 sec was chosen in the present study.

Calibration Curves and Precision Study.
Te proposed analytical method was validated through linearity and analytical fgures of merit under optimal conditions.Linearity validation of the method was performed with the establishment of the linear calibration using external standard,

6
International Journal of Analytical Chemistry and the corresponding curves (supplementary information, Figure S2) were generated by plotting the area of the analyte peak against the standard concentration (ng/mL).Good linearity, with a correlation coefcient >0.998 was obtained for all the target analytes considered in this study over the studied concentration range (Table 1).Individual chromatograms of the target analyte considered in this study are given in Figure S3.Te precision of the proposed method was also evaluated in terms of repeatability (intraday precision) and reproducibility (interday precision).To study repeatability of the method, pasteurized milk sample one (PSM) was spiked with the mixture of seven pesticides at two concentration levels (μg L −1 ): Level 1 : 10 for CAR; 15 for THE, CYZ, ATZ, and PRZ; 30 for AZO; and 45 for MET, and Level 2 : 100 for CAR; 130 for THE, CYZ, ATZ, and PRZ; 260 for AZO; and 390 for MET.Te sample was extracted in triplicate and injected in duplicate on the same day under the optimized experimental conditions.Te reproducibility of the method was also validated using the same milk sample at concentration values used above to evaluate reproducibility for four consecutive days, following single extraction and injection.As shown in Table 1, the RSD % of the method was in the range of 1.97-7.88for intraday and 4.52-8.04for interday.Te results of the precision studies reveal that good repeatability and reproducibility (RSD <9) were achieved, thus showing a low variability extraction technique [35,56].

Sensitivity.
Te sensitivity of the method guaranteed the detection and confrmation of pesticide residues in milk found at levels below or above the limits of detection (LODs).Te calculations for LODs and limits of quantitation (LOQs) were based on the standard deviation (σ) of the seven extraction responses of blank milk for each type of milk samples and the slope of the calibration curve (S) using equations 3 × σ/S and 10 × σ/S, respectively [55].Te results are given in Table 1, showing that the LODs ranged from 0.58 to 2.56 ng•mL −1 while LOQs from 1.95 to 8.51 ng•mL −1 .

Applications of the SALLE Method to Real Milk Samples.
Te suggested method's accuracy was validated using three real milk samples including pasteurized milk sample one (PMM), pasteurized milk sample two (PSM), and raw sululta milk sample (RSM).None of the tested milk samples produced signals corresponding to values above the LODs when the unspiked sample was evaluated to determine whether the seven selected target analytes were identifed or not.Te observed results may indicate that the samples tested were either free of pesticide residues or contained amounts below the detected limits.Te average relative recovery (RR%) of each sample spiked at two concentration levels and extracted in triplicate was used to determine the accuracy of the proposed SALLE technique (Table 2).Relative recovery was intended using the standard addition on the blank real samples to evaluate the methods accuracy [21,22,35].Mean relative recoveries (RR %) at two concentration levels were in the range of 85.9-108.8%,with %RSD <11.5 for the studied milk samples.Te results obtained for recovery were in the acceptable range [56], indicating that the matrices of milk samples have no intense efect on the performance of the proposed method.Similar results were also reported by other workers both for accuracy and precision for the analysis of pesticides in the studied milk [35].
Te chromatograms of the target multiclass pesticide residues in the PSM milk sample before and after spiking at concentration (level 2) used for precision study using the developed SALLE methods are shown in Figure 8. Te separation of target analytes in the chromatogram obtained using reversed-phase high-performance liquid chromatography is in the order of their polarity in which the more polar elute frst and the less polar one retained more (Figure 1, logK ow value).Te chromatograms selectivity was assessed by comparing blank and fortifed sample peaks.It is evident from these chromatograms that absence of the chromatographic peaks from coextracted components and is well resolved for all analytes, demonstrating a high level of selectivity at the same retention time as the target pesticides.Terefore, the reported chromatogram endorses the selectivity of the developed SALLE technique.Te other milk samples evaluated by this study also had the same profles (supplementary information, Figures S4 and S5).

Comparison of the Proposed Method to Other Previously
Reported Methods for the Analysis of Milk Samples.Te presented analytical method, i.e., SALLE-HPLC-DAD for preconcentration and determination of multiclass pesticide residues was compared with other methods reported in the literature, such as dispersive liquid-liquid microextraction with gas chromatography mass spectrometry (DLLME-GC-MS) [17], solid-phase extraction with high performance liquid chromatography coupled with ultraviolet detector (SPE-HPLC-UV) [16], cloud point extraction with HPLC-UV (CPE-HPLC-UV) [3], head space solid-   International Journal of Analytical Chemistry phasemicroextraction with GC-MS (HS-SPME-GC-MS) [33], quick, easy, cheap, efective, rugged, and safe (QuEChERS) coupled with HPLC and diode array detector (QuEChERS-HPLC-DAD) [14], and dispersive solid-phase extraction combined DLLME with HPLC-DAD (DSPE-DLLME-HPLC-DAD) [57], and the results are shown in Table 3.As can be seen, in terms of the LODs, precisions and accuracy of the present method were better than or comparable to those of the other methods applied for extraction of pesticides from the same type of matrices, i.e., milk sample.For HS-SPME [33] and SPE [16] methods, there may be the problem of facing sample carryover efects which leads to false-positive results.Te proposed SALLE is simple, and unlike the SPE method, it does not require multisteps conditioning, washing, loading, and elution [5,6].In addition, the proposed method is found to use simpler equipment and exhibits a wider linear range, integrated  International Journal of Analytical Chemistry pretreatment and preconcentration in a single step, which would make the procedure simpler, cost-efective, time saving, and eco-friendly.

. Conclusions
In this study, the SALLE-HPLC-DAD analytical technique, that is, simple, fast, and green, was developed and optimized for routine monitoring and quantitative determination of seven multiclass pesticide residues with a wide range of physicochemical properties including methidathion, atrazine, azoxystrobin, cyanazine, carbrayl, thiamethoxam, and propazine in samples of raw and pasteurized milk.Compared with more traditional extraction techniques like LLE and SPE, this method uses a signifcantly smaller extraction solvent and sample volume.For all of the experimental factors taken into account during the investigation, the approach was optimized utilizing univariate methods.Te optimized method ofers sufcient accuracy, precision, linearity, and sensitivity under optimum extraction conditions in a short extraction time.No matrix interferences were coextracted or seen in the analysis at their respective retention times while this method was being used to extract trace-level pesticides from milk samples.Comparatively to other reported research that used hazardous halogenated organic solvents as extraction solvents, the extraction solvents used in the current extraction approach are more environmentally benign.As a result, the trace level enrichment of multiclass pesticides using the SALLE analytical technique could be thought of as a good alternative for selective and sensitive extraction and practical assessment of multiclass pesticide residues in milk as well as enrichment of other trace compounds in complex samples in routine laboratory analysis.
HPLC column was equilibrated with the mobile phase for 10 min.Analysis was performed with a fow rate of 1 mL/min, column temperature of 35 °C, injection volume of 15 μL, and UV detection at 224 nm for all the target analytes.Peak area was used as instrumental response and comparison of the responses.Under these chromatographic conditions, baseline separation was maintained for all the target analytes.2.4.Standard Solution Preparation.Te stock standard solution of each target analyte, with the concentration of 0.1 mg/mL, was prepared by weighing the appropriate amount and dissolving it in methanol.Intermediate standard solutions of 10 μg/mL were obtained by diluting the stock solution with ultrapure water.Other working solutions of lower concentrations were also prepared by diluting the intermediate solution in the ultrapure water.
Industries (Mumbai, India, 99%).Ammonium acetate (BDH Chemical Ltd, England, 96%) was obtained from VWR International (Radnor, PA, USA).Sodium sulphate and sodium acetate anhydrous were from (BDH Chemical Ltd, England, 96%).Common chemicals such as NaCl were obtained from Sigma-Aldrich (Steinheim, Germany), hydrochloric acid (HCl) was purchased from Sigma-Aldrich (St. Louis, MO, USA), and sodium hydroxide (NaOH) was the product of Merck chemicals (Darmstadt, Germany).Ultrapure water was prepared by purifying with a double distiller, a 8000 Aquatron water Still (Bibby Scientifc, Stafordshire, UK), and a deionizer (EASY Pure LF, Dubuque).2.2.Instruments and Equipment.Chromatographic analyses were carried out using the Agilent 1200 series HPLC system (Agilent Technologies, Waldbronn, Germany) outftted with a quaternary pump, vacuum degasser, standard and preparative autosampler, thermostated column compartment, autosampler thermostat, and a diode array multiple wavelength detector.LC Chemstation software (B.02, 01-SR1) was used for sample processing and data acquisition.Chromatographic separation was performed using a ZOR-BAX ODS-C 18 (150 × 3 mm, i.d., 3.5 μm particle size) analytical column from Agilent Technologies.Te sample solution pH was measured using an Adwa pH meter, model 1020, made by Adwa Hungary Kft. in Szeged, Hungary.For sample preparation, a centrifuge, Model 800, Jiangsu Zhenji insturuments Co., Ltd.(Jiangsu, China), a 15 mL centrifuge tube, Corning integrated (Corning, NY, Mexico), and an ultrasonic heater, Dacon ® , were utilized.2.3.Chromatographic Conditions.Chromatographic separations were achieved using the isocratic condition of a binary mobile phase, consisting of solvent A (40% ultrapure water) and solvent B (60% methanol).Prior to the sample injection, the All standard solutions were stored in the refrigerator below 4 °C, when not in use.Te chemical structures, common names, abbreviations, the octanol-water partition coefcient (logP; at pH 7 and 20 °C), and other relevant physicochemical properties of the target pesticides considered are shown in Figure1.

Table 2 :
Relative recovery (RR) values of the proposed method in the milk samples.