Introduction

In recent decades, contamination by toxic cations has become a global concern for the environment and food safety. This issue has attracted a lot of attention because it is a serious threat to public health worldwide. Therefore, the assessment of the levels of trace cations in various environmental and food samples has become critical for occupational and environmental exposure assessment. Unlike other elements, cadmium has no structural role in the human body. Even in small amounts, cadmium and its soluble compounds are very toxic and can accumulate in organs and harm the body (Marahel et al. 2011; Zounr et al. 2018). The Joint FAO/WHO Committee on Food Additives recommended a temporary maximum tolerable weekly intake of cadmium, which is set at 7 μg per kg of body weight (World Health Organization 1996). Copper plays an important role in the body because it is a component of several important enzymes. It also helps the metabolism of brain cells that are not working properly (Wang et al. 2019; Ellingsen et al. 2007). The recommended daily intake of copper is about 0.6 mg for women and 0.7 mg for men, which are equivalent to approximately 11 pg kg−1 of body weight for both sexes (World Health Organization 1996). However, consuming excessive amounts of copper can lead to various health problems such as diarrhea, nausea, vomiting, cirrhosis, anemia, and bronchitis (Moraes et al. 2009; Khaghani et al. 2010). It is important to note that exceeding the safe limit of copper and other heavy metals can be toxic. Considering the possible harm of heavy metal residues in the environment and food products, the pollution caused by these metals has attracted considerable attention (Wu et al. 2010). As environmental pollution continues to increase, it is necessary to monitor and determine the levels of toxic metals in water and food samples, which can have a significant impact on human health. As a result, the detection of heavy metals in water and food samples has become an area of ​​interest due to their rapid accumulation in the human body.

The most commonly used methods for the analysis of heavy metal ions are flame atomic absorption spectroscopy (FAAS) (Kazantzi et al. 2019; Sodan et al. 2020), graphite furnace atomic absorption spectroscopy (Liu et al. 2020; Manjusha et al. 2019), inductively coupled plasma (ICP)—mass spectrometry (Xu et al. 2021; Chen et al. 2020), ICP-atomic emission spectroscopy (Rehan et al. 2021; Paktsevanidou et al. 2021), and electrochemical techniques (Wu et al. 2019; Vajedi and Dehghani 2019). However, direct measurement of heavy metal ions in water and food samples is often a challenging due to the presence of interfering matrix components and low concentrations of the metal ions of interest. Consequently, it is necessary to use a preconcentration/extraction techniques as a preliminary step. The most commonly used methods for preconcentration and extraction of heavy metal ions from different samples include liquid–liquid extraction (Lee et al. 2020; Mane et al. 2016), solid phase extraction (SPE) (Shahryari et al. 2022; Suo et al. 2019), co-precipitation (Deng et al. 2019; Li et al. 2020), cloud point extraction (Racheva et al. 2021; Akl et al. 2021), dispersive solid phase extraction (DSPE) (Karlıdağ et al. 2023), and solid phase microextraction (Rohanifar et al. 2018). In DSPE, unlike conventional SPE there is no need to sorbent packing into a cartridge, and the extractive solid particles are dispersed into solution by vortexing or sonication (Chisvert et al. 2019; Khezeli and Daneshfar 2017; Ghorbani et al. 2019), thereby resulting in high extraction recoveries (ERs). DSPE has been used as a pretreatment technique for the analysis of several compounds by using various adsorbents (Chisvert et al. 2019; Ghorbani et al. 2019; Sajid et al. 2021). In recent years, in situ formation of extractive phase has been the focus of most researchers. In these conditions an unlimited contact area between the extractive phase and analytes is achieved during extraction step (Farajzadeh and Sattari Dabbagh 2020; Abed Altuwaijari et al. 2023). Dispersive liquid liquid microextraction (DLLME) is based on a three-component solvent system in which a mixture of a water-immiscible solvent (non-polar, extraction solvent) and a water-miscible solvent (polar, dispersing solvent) is rapidly injected into an aqueous sample solution, and a cloudy solution is formed (Sorouraddin et al. 2020; Saleem et al. 2023; Altunay et al. 2019). After centrifugation, the droplets of the water-immiscible analyte extraction solvent containing the desired analytes are separated from the aqueous phase.

This study aims to develop a new approach for the extraction and preconcentration of Cd(II) and Cu(II) ions from water and food samples using a DSPE procedure based on in situ formation of extractive phase. In the present study a homogeneous mixture, using ibuprofen and choline chloride (ChCl), is formed at high temperature. In the following, an aliquot of this mixture is injected into the sample solution. A cloudy solution was formed because of decreasing solubility of the mixture in water at room temperature. After a few seconds, fine particles of the mixture were formed and dispersed into the solution by vortexing. In the following, the particles containing the analytes were collected and dissolved in methanol. In order to achieve further preconcentration, the proposed method is combined with a DLLME step. The developed DSPE method needs no desorption step, which is usually a time-consuming process. Easy operation, inexpensiveness, high extraction efficiency, and low extraction time can be the other advantages of the proposed procedure.

Materials and methods

Reagents and solutions

A stock solution containing Cd(II) and Cu(II) ions at a concentration of 100 mg L−1 of each was made by dissolving Cd(NO3)2·4H2O and Cu(NO3)2·6H2O (Merck, Germany) in deionized water with 18.2 MΩ cm resistivity that was purchased from Ghazi company (Tabriz, Iran). This stock solution was diluted with deionized water to prepare the daily used working standard solution containing 25 μg L−1 of each cation. Additionally, a mixture solution with a concentration of 2 mg L−1 of each cation in deionized water was prepared daily and directly injected into FAAS to ensure the quality of the instrumental system. The ChCl and ibuprofen used for the synthesis of the desired homogeneous mixture were obtained from Merck and Daana pharmaceutical company (Tabriz, Iran), respectively. Sodium diethyldithiocarbamate (SDDTC), methanol, acetonitrile, 1,2-dibromoethane (1,2-DBE), 1,1,2,2-tetrachloroethane (1,1,2,2-TCE), 1,1,2-trichloroethylene (1,1,2-TCE), chloroform, acetone, and carbon tetrachloride were supplied from Sigma (St. Louis, MO, USA) or Janssen (Beerse, Belgium).

Real samples

One surface water was collected from the suburb of Tabriz city in East Azerbaijan province (Iran). An urban water was collected from our laboratory. Also one well water was collected from the city of Tabriz (Iran). Mango, orange, cherry, and apple juice samples packed in tetra pack boxes were purchased from local supermarkets in Tabriz. The samples were stored in a refrigerator at 4 °C. All samples were centrifuged at 8000 rpm for 6 min before their using in the proposed method.

Instruments

Absorbance measurements were taken using a Shimadzu AA-6300 FAAS (Kyoto, Japan). The instrument had a 100-mm burner head and a deuterium lamp for background correction. The flame was produced by a combination of acetylene and air, with the flow rates of 2.3 and 15 L min−1, respectively. For radiation sources, cadmium and copper hollow cathode lamps (Shimadzu, Japan) were used. These lamps worked in the currents of 10 and 12 mA at the wavelengths of 228.8 and 307.6 nm, respectively. The spectral resolutions were 0.5 and 0.7 nm for cadmium and copper, respectively. To measure pH, a Metrohm pH meter model 654 from Herisau, Switzerland, was employed. Phase separation was accelerated using a Hettich centrifuge, model ROTOFIX 32A from Kirchlengern, Germany. Throughout the experiments, a vortex mixer (Labinco L46, Breda, Netherlands) was used. To prepare a mixture solution, a laboratory hot plate from Gerhardt in Konigswinter, Germany, was utilized.

Study of effect of important parameters on the extraction recoveries (ERs) of the analytes

To achieve suitable mole ratio of the constituents for synthesis of the homogeneous mixture, different molar ratios of ibuprofen to ChCl (1:0, 1:1, 1:2, 1:3, 2:1, and 3:1) were tested in the preparation of the mixture solution.

The effect of SDDTC amount on the ERs of Cd(II) and Cu(II) ions was studied by adding different volumes (50–300 µL) of SDDTC solution (with a fixed concentration of 0.1 mol L−1 in water) into working solutions.

To investigate the effect of volume of the synthesized solution, different volumes of the synthesized solution from 20 to 100 μL were investigated.

The duration of the examined vortexing varied between 1 and 5 min.

In the proposed method, to transfer the extracted analytes into solution, acetone, acetonitrile, and methanol were tested as the potential dissolving solvents for the sorbent and complexes of the studied ions. For this purpose, after extraction and removing the supernatant, 1.5 mL of the selected solvents was added onto the collected solid particles, vortex for 2 min, and the obtained solution was mixed with 255 µL of 1,2-DBE. In the following, the mixture was injected into 5 mL deionized water by a 2-mL syringe. Two 100 µL of the settled phase (208 ± 4 µL) was removed and injected into FAAS by a home-made microsample introduction system (Sorouraddin et al. 2016).

In the following to study the volume of methanol, different volumes in the range of 0.25–1.75 mL of methanol were added to the separated solid particles.

Preparation of ibuprofen-Choline Chloride homogenous mixture Ibuprofen-ChCh homogeneous mixture was prepared by mixing ibuprofen and ChCl at the ratio of 3:1 in a 10-mL glass test tube with a screw cap. The tube was sealed and placed in an oil bath at a temperature of 120 \(^\circ{\rm C}\) for 6 min. Then the tube was vortexed for 1 min, and this process of heating and vortexing was repeated two more times. Finally, a uniform liquid was produced.

Extraction procedure

Seven milliliters of the working/sample solution was poured into a 10-mL glass test tube. Then 150 μL of SDDTC solution (0.1 mol L−1, as a complexing agent) and 50 μL of the prepared ibuprofen-ChCl homogeneous solution were added to it. The cloudy solution was formed, and after a few seconds, very fine particles were formed. After vortexing for 2 min, the solution was centrifuged at a speed of 7000 rpm for 6 min. After removing the supernatant, 1.25 mL of methanol was added onto the collected particles. Two-minutes vortexing was done, and a clear solution obtained. In the next step, 275 μL of 1,1,2,2-TCE (water-immiscible analytes extraction solvent) was added and the resulting mixture was dispersed into 5 mL of deionized water by a 2-mL glass syringe. This resulted in the formation of a cloudy solution that was subsequently separated into two phases by centrifuging for 6 min at 7000 rpm. The water-immiscible phase containing the analyte of interest was subsequently collected and injected into the FAAS for analysis. In all experiments, the differences of absorbances between the working or sample solution and the blank solution have been used in the calculations.

Extraction recovery and enrichment factor calculation

To evaluate the effectiveness of the proposed method, ER and enrichment factor (EF) were investigated. The EF described in Eq. (1) is the ratio of concentration of the analyte in the collected water-immiscible phase (Ccoll) to the initial concentration of the analyte (C0) in the aqueous solution. On the other hand, the ER calculated by Eq. (2) is the percentage of the total amount of analyte (n0) transferred to the collected water-immiscible phase (ncoll). Vaq refers to the volume of the aqueous phase, while Vcoll refers to the volume of the collected water-immiscible phase.

$${\text{EF}}=\frac{{C}_{{\text{Coll}}}}{{C}_{0}}$$
(1)
$${\text{ER}}\%=\frac{{n}_{{\text{coll}}}}{{n}_{0}}\times 100=\frac{{C}_{{\text{coll}}}}{{C}_{0}}\times \frac{{V}_{{\text{coll}}}}{{V}_{{\text{aq}}}}\times 100={\text{EF}}\times \frac{{V}_{{\text{coll}}}}{{V}_{{\text{aq}}}}\times 100$$
(2)

Method validation tests

Calibration curve of each analyte was plotted by performing the optimized procedure on ten working solutions containing the intended heavy metal ions in the concentration range of 0.1–150 µg L−1. Limit of detection (LOD) was calculated based on 3Sb/m (Sb: representing standard deviation of blank and m: representing slope of the calibration graph). Limit of quantification (LOQ) is the lowest amount of analyte in a sample which can be quantitatively determined with suitable precision and accuracy. It was calculated as 10Sb/m. The method's repeatability, represented as relative standard deviation (RSD%), was determined by analyzing the working standard solutions at the concentrations of 10 and 25 µg L−1 for each analyte. To assess the accuracy of the developed procedure, concentrations of the analytes in a certified reference material (CRM); SPS-WW2 Batch 108, were determined by the proposed method.

Results and discussion

Investigation mole ratio of the constituents involved in synthesis of the homogeneous mixture

In the proposed microextraction method, the formed homogeneous solution should have certain characteristics such as high efficiency in extracting the studied ions and minimal solubility in water. For this purpose, different mole ratios according to “materials and methods” section were investigated. The outcome of the experiment indicated that clear and homogenous solutions were obtained only at the molar ratios of 1:0 and 3:1 for ibuprofen to choline chloride. Injecting these solutions into water created cloudy solutions, which after a few seconds, solid particles were formed inside the solution. The results (Fig. 1) showed that the amount of the analytes extraction by the second solution is about three times the first solution. Therefore this ratio (3:1 for ibuprofen to ChCl) was used in the next steps.

Fig. 1
figure 1

Effect of components of the synthesized extraction solution in DSPE step. Experimental conditions: aqueous solution, 7 mL of deionized water spiked with 25 μg L−1 of each Cd (II) and Cu (II) ions; mixture volume, 80 μL; extraction time, 4 min; and desorption solvent (volume), acetonitrile (1.5 mL). DLLME step: aqueous phase volume, 5 mL deionized water; water-immiscible analyte extraction solvent volume, 250 μL of 1,2-DBE; and centrifugation rate (time), 7000 rpm (6 min). Error bars represent the standard deviations of three repeated determinations

Full size image

Optimization complexing agent

SDDTC is a suitable chelating agent that can form stable complexes with Cd(II) and Cu(II) ions.. To study the effect of this parameter, different amounts of SDDTC according to “materials and methods” section were added into the solution. Based on the findings (Fig. 2), when the volume of SDDTC solution increases from 50 to 150 μL, the ERs increase for the target analytes. However, further increase in the volume of the chelating agent does not lead to a significant change in ER values. It is assumed that at the volume of 150 μL, the target analytes are completely chelated, and beyond this point, increasing the volume of SDDTC solution has no effect on ERs.

Fig. 2
figure 2

Effect of SDDTC concentration on the ERs of the analytes. Conditions: the same as those used in Fig. 1, except ibuprofen:ChCl (3:1) was used in the synthesize of the homogeneous solution

Full size image

The synthesized solution volume

In the proposed method, the volume of the formed solution has a significant effect on the extraction of the formed complexes. To investigate the effect of this parameter, different volumes (according to “materials and methods” section) were tested. The results in Fig. 3 indicate that extraction efficiency of the target ions is the highest in the volume of 50 µL. The ERs of the analytes in the volumes less and higher than 50 µL show a decreasing trend.

Fig. 3
figure 3

Effect of volume of the mixture solution on the efficiency of the method. Conditions: the same as those used in Fig. 2, except volume of SDDTC solution (0.1 mol L−1) which was fixed at 150 µL

Full size image

Study of vortexing time

In the proposed method, the main purpose of vortexing is complete dispersion of the formed solid particles of into aqueous sample to increase efficiency of the extraction process. The results in Fig. 4 show that the ERs of the analytes increase with increasing vortexing time from 1 to 2 min and after that, they remain constant. It shows that extraction of the analytes is completed within 2 min and high vortexing times cannot affect the extraction process. Therefore 2 min was selected as the suitable vortexing time in the next steps as the extraction time.

Fig. 4
figure 4

Vortexing time optimization (extraction time). Conditions: the same as those used in Fig. 3, except volume of the mixture solution which was fixed at 50 µL

Full size image

pH effect

One of the effective parameters that may play an important role in extraction procedures is pH of the sample solution. In the proposed DSPE procedure, pH of the sample solution can be effect on the existing forms of the analytes and the functional groups of the prepared homogeneous solution. To study the effect of this parameter, pH of working solutions was changed in the range of 2–12. According to the results in Fig. 5, the maximum extraction efficiency is obtained in the pH range of 6–8. It is noticeable that at highly acidic or alkaline pHs, the ERs are decreased. It seems that at acidic solution, hydronium ions disturb complex formation between SDDTC and the analytes. Also, instability of the formed solution can be another reason of decreasing ERs at lower pHs. At high pHs, the formation of hydroxide precipitates or complexes of the cations can be reason for the ERs decreasing. Due to the fact that pH of the used samples was between 6 and 8, thus there was no need to adjust pH.

Fig. 5
figure 5

Study effect of sample pH on the ERs of the analytes. Conditions: the same as those used in Fig. 4, except vortexing time was fixed at 2 min

Full size image

Selection of solvent type and volume in dissolving sorbent and complexes

To study the effect of type of solvent, different solvents (according to “materials and methods” section) were tested. The results in Fig. 6 show that methanol is the most effective solvent. In addition, the volume of methanol was also investigated and the results are shown in Fig. 7. When the methanol volume was 0.25 mL, the sorbent particles were not completely dissolved, resulting in low extraction efficiency. On the other hand, when the volume of methanol exceeded 1.25 mL, the polarity of the aqueous solution used in DLLME step was reduced and partition coefficients of the analytes were decreased accordingly, which led to a decrease in the extraction efficiency. Based on these findings, 1.25 mL of methanol was selected as the optimal volume for the dissolving dispersion solvent in the next steps.

Fig. 6
figure 6

Effect of type of dissolving solvent to recover the ions. Conditions: the same as those used in Fig. 5

Full size image
Fig. 7
figure 7

Optimization of methanol volume. Conditions: the same as those used in Fig. 6, except type of desorption solvent which was selected methanol

Full size image

Investigation of water-immiscible analyte extraction solvent type and volume in DLLME

Physicochemical properties of the water-immiscible analyte extraction solvent used in DLLME step play important roles in enhancing efficiency of the extraction method. The solvent must be able to effectively extract the analytes and at the same time be compatible with the analysis instrument. Factors such as low viscosity, miscible with methanol, and low solubility in water should also be considered. In this study, several organic solvents including 1,2-DBE, 1,1,2,2-TCE, 1,1,2-TCE, chloroform, and carbon tetrachloride were investigated as potential water-immiscible analyte extraction solvents. Based on the results in Fig. 8, solvent 1, 1, 2, 2-TCE shows the highest efficiency for the target analytes. To optimize the volume of 1,1,2,2-TCE, analytical signals ​​of the analytes were measured using different volumes of 1,1,2,2-TCE, including 250, 275, 300, 325, and 350 μL. According to the results (Fig. 9), analytical signals ​​for both cations are the highest in the volume of 275 µL. At higher volumes, by keeping the other conditions constant, concentrations of the concentrated analytes decrease. As a result, analytical signals of the analytes decrease. Therefore, 275 μL of 1,1,2,2-TCE was selected for the next steps.

Fig. 8
figure 8

Selection of water-immiscible analyte extraction solvent type in DLLME. Conditions: the same as those used in Fig. 7, except methanol volume was fixed at 1.25 mL

Full size image
Fig. 9
figure 9

Optimization of water-immiscible analyte extraction solvent volume

Full size image

Investigation of coexisting ions

The impact of the presence of coexisting ions in water or fruit juice samples on the amount of the analytical signals of Cd(II) and Cu(II) was investigated. In these experiments, 7 mL of standard solution along with different concentrations of the coexisting ions was treated using the established method. If a coexisting ion leads to a ± 5% change in absorbance of the analytes, it was considered interfering ion (Jabbari et al. 2023; Pourmohammad et al. 2020). Table 1 shows the acceptable concentration ratios of the investigated ions to the analytes. The data show selectivity of the method toward the analytes owing to obtaining high tolerance interferent/analyte ratios in the range of 400–2500.

Table 1 Tolerance limit of interferent/analyte ratioa in the analysis of Cu(II) and Cd(II) based on the studied method
Full size table

Method validation

The optimized experimental conditions were used to evaluate quantitative characteristics of the developed method. The validation included studying coefficient of determination (r2), linear range (LR), LOD of the calibration curve, LOQ, EF, repeatability, and ER. The results are presented in Table 2 which indicate that the calibration curves for the selected heavy metal ions exhibit good linearity with r2 values greater than 0.99. The LODs were determined to be 0.87 µg L−1 for Cd(II) and 0.30 µg L−1 for Cu(II). The LOQs were found to be 2.5 µg L−1 for Cd(II) and 0.50 µg L−1 for Cu(II). The method's repeatability, represented as RSD%, was determined by analyzing the working standard solutions at the concentrations of 10 and 25 µg L−1 for each analyte. The calculated values were found to be between 2.5 and 3.1% for intra-day precision (n = 6) and 3.2 and 4.2% for inter-day precision (n = 6). The ERs for Cd(II) and Cu(II) were found to be 97% and 94%, respectively. Additionally, the EFs for Cd(II) and Cu(II) were determined to be 33 and 32, respectively. To investigate the accuracy, concentrations of the analytes in SPS-WW2 Batch 108 (CRM) were determined by the proposed method. The certified concentrations of Cd(II) and Cu(II) in the CRM were 100 ± 0.5 and 2000 ± 10 µg L−1, respectively. Therefore, before analysis, it was diluted 40 times with deionized water. The obtained concentrations were 102.7 ± 3.5 and 2085.7 ± 81.3 µg L−1 (n = 3) for Cd(II) and Cu(II), respectively. The student t test showed that a good agreement between the determined and the certified values was obtained (texperimental = 1.3 and 1.82 < t0.05,2 = 4.3).

Table 2 Quantitative characteristics of the proposed method for the analysis of Cu(II) and Cd(II) ions
Full size table

Real samples analysis

In order to show the reliability and applicability of the proposed approach, it was used to determine concentrations of Cd(II) and Cu(II) ions in different water and fruit juice samples. Treatment and determination were done in triplicate for all samples based on the developed method. The concentrations of the target ions in the selected samples are listed in Table 3. It is worth noting that one or both analytes were found only in three samples, while in other samples their concentrations were lower than the LODs. In order to investigate the effect of samples matrices in different water and fruit juice samples, they were spiked with the analytes at the concentrations of 20 and 50 μg L−1 of each cation. Then the proposed method was applied to those samples. The results of the samples compared to the results obtained from the spiked deionized water with the same concentrations, and the results are presented as relative recovery in Table 3. Based on these results, it can be concluded that different sample matrices do not have a significant effect on the effectiveness of the proposed method.

Table 3 Results of assays to check the matrices effect of various samples in the determination of Cu(II) and Cd(II) and their concentrations in the selected samples
Full size table

Compatibility of the developed method with the principles of green chemistry

The evaluation of greenness of analytical methods is difficult because of the diversity of analytes and analytical methods and the complexity of sample matrices. Therefore, the presence of instruments and procedures should be considered and provide an answer to whether a proposed method can be regarded as green or not is necessary. Tools such as Green Analytical Procedure Index (Plotka–Wasylka 2018), Analytical Eco-Scale (AES) (Gałuszka et al. 2012), and National Environmental Methods Index (Tobiszewski 2016) are used to study the greenness of analytical methods. In this study, AES tool, based on assigning penalty points (PPs) to parameters of the analytical protocol, was used for the estimation and assessment of the environmental impact or “greenness” of the proposed method. The PPs concern the amounts of the needed reagents, their hazards, generated waste, and consumed energy. Sum of these PPs was calculated and subtracted from 100. According to AES guideline an ideal green method should have a score of 100. The values higher than 75 are considered the method as a green procedure. All of the parameters of the proposed method are listed in Table 4, and the related PP was mentioned for each factor. Considering the score of 81, it can be concluded that the proposed method is greenness and can be used for the routine analysis of the studied heavy metal ions with a minimal detrimental impact on human health and environment.

Table 4 Evaluation of greenness of the proposed method using AES
Full size table

Comparison of the developed method with other methods

The efficiency of the proposed method was evaluated in comparison with the other analytical methods used for the determination of Cd(II) and Cu(II) in different samples by considering parameters such as LOD, LOQ, LR, ER, and RSD (Table 5). This method has wider ranges compared to the other methods. The RSDs and ERs obtained through this technique are satisfactory and comparable to the other methods. The LODs and LOQs of the proposed method are lower than the other mentioned methods.

Table 5 Comparison of the present method with the various approaches employed for the preconcentration and analysis of Cu(II) and Cd(II) in samples of water and fruit juice
Full size table

Conclusions

In this study, a simple, efficient, and environmentally friendly DSPE method based on a new hydrophobic mixture composed of ibuprofen and choline chloride was proposed to extract trace amounts of Cd (II) and Cu (II) ions from water and fruit juice samples. The proposed DSPE method needs no desorption step, which is usually a time-consuming process. In order to further preconcentrate the extracted analytes, the proposed method was combined with a DLLME. The proposed method achieved excellent ER, satisfactory linearity, high precision, and low LODs and LOQs for the analytes. In addition, in the proposed DSPE, a small amount of a green mixture was used. In conclusion, the proposed DSPE-DLLME in combination with FAAS improved the efficiency and accuracy of measuring low levels of Cu(II) and Cd(II) ions in water and fruit juice samples.