Simultaneously determination of copper and zinc in human serum and urine samples based on amoxicillin drug by dispersive ionic liquid- liquid microextraction coupled to flame atomic absorption spectrometry

Received 11 Jun 2020 Revised form 5 Aug 2020 Accepted 28 Aug 2020 Available online 29 Sep 2020 *Corresponding Author: Kian Azami Email: Kianazami@yahoo.com https://doi.org/10.24200/amecj.v3.i03.111 -----------------------Research Article, Issue 3 Analytical Methods in Environmental Chemistry Journal Journal home page: www.amecj.com/ir


2]
. Total zinc is about 40 mg per day for adults over 20 years. Copper helps to transportation of iron, energy production, the pigmentation of skin, hair, and eyes. Copper acts as an antioxidant for defending of cell damage which was caused by free radicals. Copper deficiency cause to anemia, low white blood cell count, loss of myelin, multiple sclerosis (MS), the loss of pigmentation, the impaired growth and osteoporosis [3]. The copper intake from food/water or supplements has 10 mg per day and over limit accumulates in the liver. Based on previous studies, a ratio of 8-15 mg of zinc for every 1 mg of copper reported. However, this ratio seems to be more important for human bodies mechanism [1][2][3][4]. Zinc is absorbed in the small intestine by a carrier-mediated mechanism. Under normal physiologic conditions, transport processes of uptake are not saturated. The mean value for serum copper was normal range (17.47 ± 3.31 μmol L -1 ; 111.32 μg dL -1 ), and the mean value for serum zinc was at the lower edge of the normal value (12.24 ± 1.04 μmol L -1 ; 80.01 μg dL -1 ), while the mean value for serum copper/zinc ratio was 1.44 ± 0.31 ranging from 0.65 to 2.67 [5][6][7]. Copper and zinc concentration in liquid phase can be determined directly by inductively-coupled plasma atomic emission spectrometry (ICP-AES) [8] or electrothermal atomic absorption spectrometry (ET-AAS) [9] with low detection limit (LOD). The conventional flame atomic absorption spectrometry (F-AAS) [10] was used in many laboratories and had low interferences ions as compared to ICP-AES or ETAAS. As difficulty matrix in human blood or serum patients a sample preparation is require for preconcentration/separation/extraction of ions from samples before determination. Many procedures for metal determination in water and human matrix was used with different analytical techniques and reagents [11,12]. The Liquidliquid extraction by using salophen as an complex reagents [13], the sandwich supported liquid membrane [14], the modified carbon based solid phase extraction [15], the solid-phase extraction on MWCNTs -D2EHPA-TOPO [16], dispersive liquid-liquid microextraction of copper (II) by oxinate chelate [17] and dispersive liquid-liquid microextraction-slotted quartz tube-flame atomic absorption spectrometry [18] are well-known procedures for preconcentration and separation of trace copper or zinc from different matrix.
In this work, the pure amoxicillin antibiotic drug was used as chelating agent for copper and zinc (Cu/Zn) extraction in human serum and urine samples by DIL-CPE procedure at optimized pH. The Cu and Zn deficiency was evaluated by determining its concentration in human serum/ urine samples by F-AAS. The Cu and Zn ions can be separated from liquid phase by hydrophobic ionic liquid as green solvent.

Apparatus and Reagents
A flame atomic absorption spectrometer model, with an air-acetylene flame, was used for copper (II) and zinc (II) determination in human serum and urine samples (Shimadzu, F-AAS, model 680, Tokyo, Japan). Copper based on wavelength 324.7 nm, slit 0.5 nm and lamp current 3.0 mA (1-5 mg L -1 ) were selected. Zinc lamp with wavelength 213.9 nm, slit 0.5 nm and current 5.0 mA (0.4-1.5 mg L -1 ) were used. A pH meter with glass electrode was adjusted the pH of human samples (Metrohm, E-632). For validation, the electrothermal atomic absorption spectrophotometer (ET-AAS, GBC 932) and ICP-MS in real samples as certified reference material (CRM) was used for determination of copper (II) and zinc (II) in serum and urine samples. The calibration curve of copper (II) and zinc (II) with injecting 20 µL of standard solution to graphite tube were used. All the reagents with analytical grade were used. Deionized-distilled water (DW, Millipore, USA) was prepared for experimental run. The copper (II) and zinc (II) solutions were prepared by appropriate diluting a 1000 µg L -1 of Cu and Zn solution (Merck) with DW. The pure amoxicillin (Fig.1

Characterizations
Amoxicillin, as organic compounds with the penicillin core structure was used as ligand in this study. Amoxicillin is structurally characterized by a penam ring (C 5 H 7 NOS) bearing two methyl groups and an amide group. Fourier transform infrared (FT-IR) spectra were recorded from KBr pellets using a spectrophotometer FTIR Shimadzu (Kyoto, Japan). For amoxicillin, the C-O stretching vibrations show intense IR absorptions, due to the considerable change in the molecular dipole moment produced by this vibration mode (Fig. 2). Powder X-ray diffraction (XRD) was conducted on a X-ray diffractometer. X-ray diffraction analysis of pure amoxicillin and the optimized formulation was done by X-ray powder diffractometer (PW 3040/ 60 Xpert PRO, Panlytical, Netherlands). The X-ray diffraction patterns were recorded using Cu Kα radiations (λ=1.5405980Ả), a current of 30 ma, and a voltage of 50 Kv. The samples were analyzed over 5-35 (2θ range) with a scan step size of 0.02s and 0.5s per step (Fig. 3). Scanning electron microscopy (SEM) images were obtained using a Tescan Mira-3 Field Emission Scanning Electron Microscope (FE-SEM). The external and internal morphology of the amoxicillin was studied by scanning electron microscopy (Fig. 4).

Sample Preparation
Serum potentially contains elements and proteins which were produced in the human body. Sample preparation helps to reduce time, errors and interferences in analytical chemistry. The sample collection and handling have improved the sensitivity, selectivity, and reproducibility of Cu/Zn analysis in serum samples. For increasing accuracy and precision of results, the human serum samples were prepared. First, the glass laboratories placed in mixture of sulfuric acid and nitric acid (ultra pure grade; 0.5 molar: 0.5 molar) for 24 hours and washed for ten times with DW. The Cu and Zn concentrations in serum have an important limit concentration (~0.8-1.5 mg L -1 ) and So, the human serum sampling, storage and analysis must be carefully done. In addition, 10 mL of the human serum were prepared from personnel of multiple Sclerosis patients in Iran (MS; 25 Men, 25 women, 25-55 age), based on the world medical association declaration of Helsinki (WMADH). Clean and sterilized syringes with plastic needles were purchased for Merck, Germany for serum blood sampling. The human biological samples were maintained frozen in refrigerator (below -4°C). For long-term storage of serum samples, we placed samples at −20, −80 °C or using liquid nitrogen. Urine samples were prepared and storage based on standard method in human samples.

Extraction Copper and Zinc Procedure
As shown in Figure 5, the copper and zinc were simultaneously extracted based on pure amoxicillin by DIL-CPE procedure. Firstly, 0.1 g of hydrophobic ionic liquid [MOIM][PF 6 ] and 0.2 mL acetone was mixed together and then 0.03 g of pure amoxicillin as a antibiotic drug was added. The mixture was injected to diluted serum or urine sample (2:10) by 2 mL of syringe with PVC needle. The cloudy solution was obtained and shacked for 7 min by shaker accessory. Then, the Cu or Zn ions was extracted from serum or urine samples based on sulfur bonding of amoxicillin at pH of 7 by DIL-CPE procedure (Cu─:S:─Zn). After centrifuging for 3 minute (3500 rpm), ionic liquid separated from serum sample in end of conical tube. The Cu or Zn ions were simply back extracted with 0.5 mL of nitric acid (0.2 M) and diluted with DW up to 1 mL. Finally, the remained solution was determined by flame atomic absorption spectrometry (F-AAS). The recovery of proposed method based on pure amoxicillin antibiotic drug was achieved for Cu and Zn extraction by the recovery equation (mean of RSD% < 2.2; more than 95%). The C p is the primary concentrations of Cu or Zn in sample and C F is the final concentration of Cu or Zn by DIL-CPE/F-AAS procedure (n=10, Eq. A).

Results and Discussion
The DIL-CPE procedure provides novel and interesting approach based on amoxicillin drug for extraction of copper and zinc from human serum and urine samples. In order to obtain favorite separation and quantitative extraction of Cu and Zn ions with high sensitivity and precision, the analytical parameters of proposed DIL-CPE method must be optimized.

Optimization of pH
The pH of human serum or urine is main factor for efficient extraction of Cu and Zn ions by DIL-CPE procedure. The retention of Cu and Zn ions by amoxicillin ligand has been investigated at different pH from 2 to 11 with buffer solutions containing (100 µg L −1 , 500 µg L −1 ) and ( 50 µg L −1 , 150 µg L −1 ) for copper and zinc as LLOQ and ULOQ concentrations. It showed that amoxicillin ligand (pH PZC = 5.6), the extraction efficiencies of Cu (II) and Zn(II) were improved with the increase of pH values more than 6 and the quantitative extraction were obtained at pH 6-8 and then the recoveries were reduced at pH more than 9. Consequently, the Cu and Zn ions quantitative extracted at pH 7 (Fig.6). Due to serum pH, the extraction mechanism of Cu (II) and Zn(II) ions based on amoxicillin ligand is mainly depended on the electrostatic attractions of deprotonated sulfur groups of amoxicillin ligand with the positively charged Cu 2+ and Zn 2+ cations. At low pH (pH< pH PZC ), the amoxicillin ligand have positively charged (SH 2 + ) as a protonation system. So, the extraction efficiency can be decreased due to the electrostatic repulsion between the Cu 2+ and Zn 2+ cations and positively charge of ligand. In addition, by increasing pH, the sulfur groups in amoxicillin ligand becomes negatively charged (R-S: 2-….Cu/Zn) and so the electrostatic attraction between negatively charged sulfur groups and positively charged Cu 2+ and Zn 2+ cations increased. In more pH (pH>9) the extraction were decreased due to the

Effect of amoxicillin dose
Different amounts of amoxicillin ligand (AMOX-L) in the range of 5 to 50 mg were tested on the recoveries of Cu (II) and Zn(II) ions for the presented DIL-CPE procedure. The results were shown in Figure 7. It was found that 30 mg and 25 mg of AMOX-L was sufficient for quantitative recoveries of copper and zinc ions in serum and urine samples, respectively. So, 30 mg of AMOX-L was used as optimum amount of AMOX-L for further works. Higher amount of AMOX-L had no significant effect on the extraction of Cu (II) and Zn(II) ions. Due to capture of Cu (II) and Zn(II) on surface of AMOX-L and the metal concentration in solution, it came to equilibrium with each other. Eventually, 30 mg of AMOX-L was used as ligand for further work.

Effect of the IL amount
In the presented DIL-CPE method, 1-Methyl-3octylimidazolium hexafluorophosphate (C 12 H 23 F 6 6 ]. Therefore, in order to achieve a suitable preconcentration, 0.1 g of IL was chosen as optimum leading to a final IL for serum and urine samples (Fig.8). Moreover, the effect of IL for Cu and Zn extraction with the same experiments were evaluated without and AMOX-L. Based on results, the efficient extraction for Cu and Zn ions were obtained less than 24% and 19% for urine and serum as complexation Cu and Zn with amino acids such as cysteine (Cys) and proteins (Cu/ Zn….Pr /Cys). Therefore the IL had low effect on Cu and Zn extraction at optimum conditions. In addition, by increasing the ultrasonication time up to 60 min, the almost 27% and 31% of Cu and Zn ions extracted by 0.1 g of IL without any AMOX-L. These results confirm the critical role of AMOX-L as complex agent for Cu and Zn extraction.

Effect of eluents
The elution solutions were optimized in order to obtain the maximum back-extraction Cu and Zn ions from IL with the minimum concentration and volume of the elution solution. By DIL-CPE method, the different elution solutions were selected with high recovery. The coordination of Cu and Zn cations with AMOX-L was dissociated and ions released into the aqueous phase in acidic pH. For evaluation of the type, the concentration and the volume of acid solutions for back extraction ions from ligand, 1000 µL of different mineral acids solutions such as HCl, H 3 PO 4 , HNO 3 and H 2 SO 4 (0.1-0.5 mol L -1 ) were examined by DIL-CPE procedure. The results showed that 0.2 mol L -1 HNO 3 (0.5 mL) quantitatively back-extracted Cu and Zn from ligand/IL (Fig. 9).

Effect of sample volume
As shown in Figure 10, the effect of sample volume on the extraction of copper and zinc in serum and urine samples were evaluated and optimized by different volumes from 1-25 mL containing copper values (100 µg L −1 , 500 µg L −1 ) and zinc values ( 50 µg L −1 , 150 µg L −1 ) as LLOQ and ULOQ concentrations by DIL-CPE procedure. As shown in Fig. 10, satisfactory recoveries were obtained between 2-10 mL for urine and serum samples. In addition, the high sample volume caused to significantly decrease the extraction Cu and Zn in human samples. Therefore, 10 mL of human samples was used for further investigation.   Therefore, different times for centrifuging were examined between 1-5 min at 3500 rpm. The result showed that 2.0 minutes is sufficient to perfect separation phase.

Interference study
For evaluating of the analytical application in real samples, the important interference of coexisting ions effected on copper and zinc extraction in serum and urine samples were studied by the DIL-CPE procedure. For this proposed, the different amounts of the interfering ions added to 10 mL of liquid solution containing 500 µg L −1 of copper and 150 µg L −1 of zinc. The results showed, the most of the coexisting cations and anions had no effect on the extraction of Cu and Zn ions under optimum conditions (SD of recovery < ±5%). In fact, the tolerable concentration ratio of interference of coexisting ions (M) per Cu and Zn (M/Zn 2+ or M/ Cu 2+ ) for Hg 2+ , Ag + and Au 3+was less than 50 and 30 for zinc and copper, respectively. This ratio was almost 100-200 for Ni 2+ , Pb 2+ and Co 2+ ions. The results showed that the AMOX-L have favorite ligand for Cu and Zn extraction despite the high concentrations of the coexisting ions (Table 1). .

Analytical figures of merit
The analytical characteristics for Cu and Zn extraction in human serum and urine samples were  studied by the purposed DIL-CPE procedure. The intra-day analytical performance was shown in Table 2 for the multiple sclerosis patients (50) and healthy peoples (50). Under the optimal conditions, the linearity for the Cu (II) and Zn(II) concentration ranges between 100 -505 µg L −1 and 41-153 µg L −1 , respectively as a lower limit of quantification (LLOQ) and upper limit of quantification (ULOQ) with mean correlation coefficient of R 2 =0.9997 for Intra-day analysis. The LOD and LOQ are as an analytical signal three times higher than the background noise and three times higher than LOD, respectively. The precision of the AMOX-L/ DIL-CPE procedure showed by the relative standard deviation (RSD %) for ten replicate determination containing 100 µg L −1 of Cu and Zn concentration which was obtained lower than 2.2%. The enrichment factor was calculated based on calibration curve and curve fitting rule (tga=m 1 / m 2 ). According to proposed method, a favorite linear ranges and satisfactory EF were achieved for determination of Cu(II) and Zn (II) ions in human samples. The mean value for serum copper/zinc ratio in MS patients was obtained 1.11 ± 0.28 which was lower than normal range as 1.44 ± 0.31.

Analysis of real and certified samples
The DIL-CPE procedure was used for determination of copper and zinc in serum and urine samples by AMOX-L ligand. The validation of method was obtained based on spiking samples by known concentration of Cu (II), and Zn(II).
The efficient recovery was achieved by spiking samples, which confirms the accuracy of the Correlations are based on Pearson coefficients (r). Statistical significance will be observed if P < 0.001 Table 3. DIL-CPE procedure ( Table 3). The recoveries of spiked samples for Cu and Zn were ranged from 96 to 105%, which demonstrated that the DIL-CPE procedure was satisfactory for determination copper and zinc in urine and serum samples. On the other hand, the certified standard reference materials (CRM) were prepared in serum and urine sample by ICP-MS and used for the validation methodology (Table 4).

Conclusions
A simple, rapid, reliable and sensitive method was developed for separation and extraction of Cu (II) and Zn(II) in serum and urine samples based on AMOX-L by the DIL-CPE procedure. The [MOIM][PF 6 ] ionic liquid as a trapping phase was used for separating of Cu/Zn-loaded AMOX-L from liquid phase. IL helps to reducing the sample preparation and separation time for extraction process. Using a small amount of AMOX-L with high extraction recovery, good precision, minimal acid elution (500 µL) and green solvent caused to make the efficient extraction based on environmentally friendly for determining of Cu and Zn in urine and serum samples. Also, the low LOD and RSD% values as well as the quantitative recoveries (more than 95%) were obtained in optimized conditions. Therefore, the developed method based on AMOX-L can be used as favorite sample preparation in human biological samples in short time. As obtained results, the amoxicillin can be affected on copper and zinc deficiency in human body at human pH when the patients used it by over dosage for many times.

Acknowledgment
Department of Pharmacology and Toxicology, Pharmaceutical Sciences Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran.