Introduction of electropolymerization of pyrrole as a coating method for stir bar sorptive extraction of estradiol followed by gas chromatography

Journal of Chromatography A 1604 (2019) 460478

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Introduction of electropolymerization of pyrrole as a coating method for stir bar sorptive extraction of estradiol followed by gas chromatography Paria Asadi Atoi, Zahra Talebpour∗, Lida Fotouhi∗ Department of Chemistry, Faculty of Physics and Chemistry, Alzahra University, P.O. Box 1993891176, Tehran, Iran

a r t i c l e

i n f o

Article history: Received 2 March 2019 Revised 22 August 2019 Accepted 22 August 2019 Available online 23 August 2019 Keywords: Stir bar sorptive extraction Electropolymerization Polypyrrole Estradiol Gas chromatography

a b s t r a c t In this study, fabrication of a stir bar sorbent is presented by electropolymerization of pyrrole via cyclic voltametry for the first time. The fabricated stir bar was applied as an efficient sorbent for extraction and pre-concentration of trace amounts of estradiol in urine samples through stir bar sorptive extraction (SBSE) method followed by gas chromatography-flame ionization detector. For this purpose, first the surface of stainless steel rod was modified by hyroxide functional group. Then electropolymerization of pyrrole monomers took place on the surface of functionalized steel rod under the optimized conditions including pyrrole concentration of 0.03 mol L−1 , equal concentration ratio of pyrrole to sodium dodecyl sulfate, 10 cycles of cyclic voltammetry and potential scan rate of 10 mV s−1 . Characterization of the produced sorbent was confirmed by scanning electron microscope imaging and energy-dispersive X-ray and infrared spectroscopy. Evantually, under the optimized conditions, the stir bar sorbent was used for extraction of estradiol from human urine samples. The presented SBSE method showed a good linearity range of 50–700 ng mL−1 with coefficient of determination 0.9910, limit of detection 10 ng mL−1 and theoretical limit of quantification 33 ng mL−1 . Moreover, better enrichment factor (87) and extraction recovery (43%) were obtained using the fabricated stir bar compared with two commercial stir bars for estradiol. The intra- and inter-bar relative recoveries were obtained 92.0% and their coefficient of variations were less than 5.4%. © 2019 Elsevier B.V. All rights reserved.

1. Introduction Sample preparation is a crucial part of any chemical analysis and so is defined as the bottleneck of a whole analytical process [1]. Lately, emerging miniaturized and environmental friendly techniques have been introduced and played a significant role in sample preparation [2]. Sorbent-based microextraction techniques which represent a beneficial approbation, apply an immobilized solid/semi-solid organic or inorganic material on a substrate, such as fused silica fiber in solid phase microextraction (SPME) [3], glass-coated bar magnet in SBSE [4, 5], thin film in thin film microextraction (TFME) [6, 7], glass fabric in fabric phase sorptive extraction (FPSE) [8, 9], or silica particles in matrix solid phase dispersion (MSPD) [10]. In all of them, sorbents may vary in extraction capability, solvent resistance, or thermal, mechanical and chemical stability. There are common methods for deposition

∗

Corresponding authors. E-mail addresses: [email protected] (Z. Talebpour), [email protected] (L. Fotouhi). https://doi.org/10.1016/j.chroma.2019.460478 0021-9673/© 2019 Elsevier B.V. All rights reserved.

of sorbents onto the substrates including: dipping [11], coating adhesion [12], sol-gel technology [13], chemical bonding [14], and electrochemical methods [15]. Physically deposited coatings indicated some hitches that were prevailed by chemical bonding between the modified sorbent material and support. In this regard, sol-gel technology has been widely applied due to its adhesive advantages. An alternative method for preparation of microextraction coatings is through electrochemistry. The electrochemical methodology for the preparation of the coatings is classified into four main groups electrodeposition of metal oxides [16], electropolymerization of conductive polymers (CPs) [17], anodization of metal wires [18] and electrophoretic deposition [19]. The privileges of these methods are their higher solvent stability, low cost, simple setup, and capability to offer coatings with variable thickness. Since Wu et al. introduced electropolymerization method to fabricate SPME coatings [20], CPs such as polyaniline [21], polypyrrole (PPy) [22], polythiophene [23] and their derivatives, and composites of these CPs with other materials have been expansively applied in SPME. These CPs can be electropolymerized by anodic oxidation of the monomers in appropriate electrolytes, and adjust-

2

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

ing the electropolymerization conditions such as applied potential, monomer and counter ion concentrations, and dopant ion type, can result in the production of desirable coatings. Since the applicability of SPME is limited by its low extraction efficiency due to its small volume of coated sorbent, SBSE was presented by Sandra et al. in 1999 [4] for improvement of analyte’s enrichment factor. In SBSE, the sorbent is coated on a magnetic stirring bar, which simply stirs in the liquid samples to extract the analyte. The classic SBSE utilizes polydimethylsiloxane (PDMS) as a common sorbent, which is typically used for extraction of nonpolar to weakly polar analytes, and has low affinity to strongly polar compounds. Therefore, there has been a growing interest towards new coatings for SBSE to improve its properties, such as modified PDMS, polar polymers [24], molecularly imprinted polymers [25], and monolithic polymers [26, 27]. Alternatively, electropolymerization of CPs, has proven to be a simple, economic and fast procedure for fabrication of SPME sorbents; but there are only a few reports of its application in SBSE technique so far. Noroozian et al. used a CP composite film synthesized at constant potential coulometry technique as a coating on stir-bar sorptive extraction of polycyclic aromatic hydrocarbons [28]. In all of these reports, the prepared stir-bar, was hung in a vial without any magnetic stirring. Among a variety of CPs, PPy and its derivatives are one of the most common SPME coatings because of their unique features including: one-step polymerization in organic or aqueous media at neutral pH by electrochemical practice, stability in air and aqueous solution, biocompatibility due to their hydrophilic characteristic, various interactions with the analytes (acid-base, π -π and dipole, ion exchange and hydrogen bonding), and commercial availability [29]. The preparation route of traditional PPy based SBSE coating, is usually tedious, arduous, and difficult to control the thickness and surface smoothness of PPy film, due to PPy’s ramified structure [30]. Whereas electropolymerization technique is easily controlled (by adjustment of potential, current and time), consumes a short time, and electrode potential is the only driving force of polymerization initiative reaction without any added initiator agent [31]. Estradiol is an estrogen steroid hormone that involves in the regulation of menstrual female reproductive cycles. It is also responsible for female secondary sexual characteristic and maintains female reproductive tissues. Its significant estrogenic effects on reproduction at mRNA and proteins lead to some types of tumors, such as breast cancers. Both types of natural and synthetic estrogens exist ubiquitously in the ecosystems and their endocrine disrupting chemicals are excreted either in free form or their metabolites, primarily through urine. Accordingly, it is vital to develop an efficient sorbent for SBSE phase via electrosynthesis method to monitor estrogens in human samples such as blood or urine [32]. Herein, electropolymerization of pyrrole (Py) as a new homemade stir bar coating technique, is developed for the analysis of estradiol as the target analyte from human urine samples through SBSE technique. The determination of analyte was done by injecting the extracted estradiol to GC equipped with flame ionization detector (FID). The effect of various parameters in electropolymerization, extraction and liquid desorption steps on the extraction efficiency of estradiol were studied and optimized consecutively. The extraction efficiency of prepared stir bar was also compared with two commercial stir bars PDMS and Acrylate Twisters®. Finally, under optimal conditions, the validity of the proposed technique was evaluated and linearity range, absolute and relative recoveries, limit of detection and limit of quantification, repeatability and reproducibility of electropolymerized PPy stir bar are reported.

2. Experimental 2.1. Chemicals and reagents Methanol (MeOH, 99.9%), acetonitrile (ACN, 99.9%), acetone (Ace, 99.9%), ethanol (EtOH, 99.9%), hydrochloric acid (HCl), dihydrated oxalic acid (C2 H2 O4 .2H2 O), sodium dihydrogen phosphate, disodium hydrogen phosphate, sodium hydroxide (NaOH), sodium dodecyl sulfate (SDS), and sodium chloride (NaCl), were provided by Merck (Darmstadt, Germany). Estradiol and amitriptyline powders were supplied by Sigma-Aldrich (Steinheim, Germany). Pyrrole (Py) was purchased from Evonik Rohm (Darmstadt, Germany). Water was purified using a Direct Q3 water purification system (Millipore, Massachusetts, USA). Urine samples were obtained from healthy volunteered ladies between the ages of 25 to 30 in February 2017. 2.2. Apparatus and equipment All electrochemical experiments for the electropolymerization of Py monomers via cyclic voltammetry (CV) were carried out by a Metrohm Autolab model 204 with potential range of ±10 V and current ranges of 10 nA to 100 mA. Saturated Ag/AgCl electrode, a steel wire and a stain steel rod with 1.5 mm i.d. × 1.5 cm length were employed as reference, counter and working (stir bar) electrodes, respectively. A model YL6500 GC - FID (Young Lin, Anyang, Korea) equipped with a HB5 capillary column, 30 m × 0.53 mm I.D., with 1 μm coating of stationary phase was used for estradiol analysis. The oven temperature was programmed from 120 °C (held for 1 min) at the rate of 30 °C min−1 to the final temperature of 280 °C (held for 10 min). The morphology of coatings was inspected by using a scanning electron microscopy (SEM) instrument (Model S4160, Hitachi, Japan). FT-IR was accomplished on a Tensor 27 FTIR instrument (Bruker, Germany). Confirmation of the modification on the metal surface, was accomplished by setting the IR instrument on the attenuated total reflectance (ATR) mode. Polydimethylsiloxane (PDMS Twister®, 1.0 cm length) and acrylate (Acrylate Twister®, 1.0 cm length) stir bars manufactured by Gerstel were used for comparison (Germany). 2.3. Preparation of standard and real sample solutions The stock solution of estradiol at concentration level of 10 0 0 μg mL−1 was prepared in MeOH, and the working solutions were prepared by diluting the stock solution to the desired concentration with phosphate buffer at pH 7. For real sample analysis, urine samples of a healthy volunteer were collected, and subjected to 5 times dilution with phosphate buffer at pH 7. Finally, 100 ng mL−1 of standard estradiol was added to the extraction solutions of urine sample. In water sample, 100 ng mL−1 spiked solution of estradiol was prepared after adjusting pH on 7 using phosphate buffer. All the stock and working solutions were capped and preserved at 4 °C. 2.4. Fabrication of PPy coated stir bar by electropolymerization procedure To begin with, magnetic stainless steel rods were first washed via stirring for 5 min in consecutively deionized water and acetone. After drying in room temperature, to the aim of metal surface preparation for modification, each rod was placed in the electrochemical cell, to get anodized via CV. A three electrode system (including steel wire counter electrode, saturated Ag/AgCl reference electrode and the target steel rod as the working electrode) was

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

3

employed in 0.1 mol L−1 NaCl solution. CV occurred with a potential range of 0 to +0.4 V, scan rate of 100 mV s−1 during 10 cycles. Afterwards, for metal surface modification with hydroxide functional group, the anodized steel rods which were washed by deionized water, stirred in 1 mol L−1 NaOH solution for 3 h. They were then removed, dried, and washed for 15 min in deionized water and 0.1 mol L−1 HCl successively, to rub off the excess hydroxide groups. After modification, the electrodeposition of PPy as the sorbent was carried out through the following procedure: a 10 mL aqueous solution of 0.03 mol L−1 Py monomers with 0.03 mol L−1 SDS as the counter ion generator in 0.4 mol L−1 dihydrated oxalic acid was prepared and put in the three electrode electrochemical cell including modified steel rod as the working electrode. Electropolymerization of Py took place using 10 cycles of cyclic voltammetry with potential range of 0 to +1.8 V and scan rate of 10 mV s−1 . After electropolymerization, the stir bar was washed with deionized water and dried with a lint-free tissue at room temperature. Finally, stir bars were conditioned prior to SBSE process. Each synthesized stir bar was subjected to stirring for 30 min in each of MeOH, deionized water, and ACN solvents subsequently. Next, it was placed in 0.5 mL mixture of water/ MeOH (1:1 v/v) and stirred with stirring rate of 200 rpm for 15 h. The stir bar was dried after each step. The effect of variable factors such as: concentration of Py monomer (0.02, 0.03, 0.04, 0.05, 0.1 and 1.2 mol L−1 ), concentration ratio of Py/SDS (1:1, 5:1, and 10:1), number of cycles in CV method (10, 20, and 30), and the CV potential scan-rate (10, 20, 50 and 100 mV s−1 ) on the extraction efficiency were investigated. In these experiments, each conditioned stir bar was placed in a 10 mL solution including deionized water and 1 μg mL−1 estradiol (drawn out from the 10 0 0 μg mL−1 stock solution), and stirred for 1 h with stirring rate of 700 rpm in order to extract estradiol from the water sample. Then for desorption, the stir bar was totally immersed in a 0.5 mL vial containing 200 μL MeOH, which was sonicated for 22.5 min.

were performed for standard estradiol solutions with concentration range of 20–700 ng mL−1 under optimized condition. The calibration curve was plotted via the peak area of analyte vs. its initial concentration. The linearity of the calibration curve was evaluated by least-squares regression method, which was used to calculate the coefficient of determination (R2 ), intercept, and the line’s slope and their uncertainties. Limit of detection (LOD) and theoretical limit of quantification (LOQ) were calculated by 3 and 10 (σ /m); where σ refers to the standard deviation of 6 repetitive injections of extracted blank sample, and m refers to the slope of the analyte calibration curve. Experimental limit of quantification was also obtained according to the low concentration of estradiol on calibration line. Accuracy and precision of the proposed method were determined using quality control solutions at concentration level of 100 ng mL−1 in both human urine and water samples, which established by the relative recovery (RR%) and its coefficient of variation (CV%), respectively. Intra- and inter-bar analyses were performed for extraction of the quality control solution of estradiol using three replicate of a coated stir bar and three coated stir bars to investigate its repeatability and reproducibility, respectively. In order to calculate the extraction recovery (ER%) of the fabricated stir bar, the theoretical enrichment factor was calculated based on the ratio of extraction to desorption volumes, then it was compared to the experimental calculated enrichment factor, which was obtained based on extrapolation of the extraction peak area into the calibration curve, and comparison of the gained final concentration to the initial estradiol concentration of the extraction sample. Lastly, the applicability of the proposed method was tested through analyzing estradiol in the human urine samples of a volunteer healthy lady as well as water sample using spiking method, and estradiol final concentration in real sample was calculated via calibration curve. In order to investigate the matrix effect, peak areas of extracted estradiol solution of 100 ng mL−1 in urine and water samples were compared according to the Eq. (1):

2.5. SBSE procedure

Matrix effect (% ) =

Stirring extraction and ultrasonic liquid desorption methods were employed for the SBSE procedure. The optimization of the SBSE procedure using prepared PPy-coated stir bar stared with extraction step. The effects of variable factors on the extraction efficiency were investigated; counting: ionic strength (adding 0, 5, 10, and 15% NaCl to the extraction solution), extraction time (30, 60, 90, 120 and 180 min), stirring rate (600, 700, 800, and 1000 rpm), pH (prepared phosphate buffers of 0.1 mol L−1 with pH 3, 5, 7, and 9), extraction temperature (30, 40, 50, and 60 °C), and extraction volume (10, 50, 10 0, and 20 0 mL). Finally, the variable factors on desorption step including: desorption time (15, 22.5 and 30 min), and desorption solvent (EtOH, MeOH, and ACN) on analyte’s enrichment factor were inspected. After optimization, the extraction step took place in a 50 mL container including aqueous phosphate buffer solution of 0.1 mol L−1 with pH 7 and 1 μg mL−1 estradiol with stirring rate of 800 rpm for 90 min. After extraction, the stir bar was removed from the sample solution by clean tweezers, and gently dried using a lint-free tissue. In the desorption step, the stir bar was totally immersed in a 0.5 mL vial containing 250 μL desorption solvent, and by applying ultrasonic liquid desorption mode for 30 min, the extracted analyte was released from the polymeric coating to the desorption solvent. Finally, 2 μL of desorption solution was injected to GC system. 2.6. Method validation In order to plot the calibration curve for validation of the proposed method, successive extraction and desorption procedures

Area in Urine − Area in Water × 100 Area in Water

(1)

3. Results and discussion 3.1. Fabrication of PPy coated stir bar by electropolymerization procedure In the first step, in order to prepare the metal surface for modification, after testing different processes, anodization technique via CV resulted in the smoothest uniformed scratches on the metal surface. As shown in Fig. 1a, the anodic current increases with the increase of positive potential. According to the necessity of using an adhesion agent on the metal surface to increase the mechanical stability of the polymercoated stir bar, modification with hydroxide functional group (–OH) proved to be practical. The pattern of ATR-IR spectrum can confirm the existence of -OH functional group on the surface of steel rod. It can be interpreted from Fig. 1b that, the peak of tensile bonding of O–H, is only observed after modification. Moreover, to confirm the presence of –OH group on the metal surface, SEM imaging, and energy dispersive X-ray analysis (EDX) were employed. As shown in Fig. 1c, the surface of the anodized steel after modification with –OH group seems darker in color compared to bare anodized steel. Also, based on the EDX spectrums and charts of element percentages, (Fig. 1d), the percentage of oxygen obviously increases from 1.76% to 8.71% after modification. In this study, synthesis of PPy sorbent as a stir bar, was accomplished through electrochemical technique for the first time. In this regard, irreversible oxidation of the Py monomers via CV in aqueous solution of oxalic acid resulted in the proper polymeric coat-

4

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

Fig 1. (a) Voltammograms of steel rod anodization in NaCl 0.1 mol L−1 solution via CV with potential range of 0 to +0.4 V, scan rate of 100 mV s−1 during 10 cycles. (b) Overlayed ATR-IR spectrums of bare anodized steel (I) vs. –OH modified steel (II). (c) SEM image of the anodized (I) and –OH modified (II) steel rod at 40 0 0× magnification. (d) EDX spectrum intensity of elements with emphasize on oxygen percentage on steel surface before –OH modification (I), and after modification (II) at 40 0 0× magnification.

ing. The voltammogram of PPy electropolymerization is presented in Fig. 2a, in which the oxidation peak summit of Py is placed at the potential 0.6 V. It also shows the shape of solvent-conditioned PPy-coated stir bar. The common mechanism for electropolymerization of PPy in (as in this case) acidic environments is presented in Fig. 2b. As shown, the Py monomers are oxidized to free radicals on the electrode surface and linked to one another in the form of a dimer, tetramer and so on during a chain propagation reaction. The deposition of the PPy onto the electrode surface formed the polymer as a result [33]. Fig. 2c presents a glimpse of steel rod surface preparation and electropolymerization of PPy on it. Based on the proposed mecha-

nism, binding of Py monomer to the oxygen on the metal surface, provides a strong adhesion that can increase the mechanical stability of the prepared stir bar. In order to attain the best coating thickness and uniformity, electropolymerization was performed in different environments including water, hydrochloric acid (HCl) with pH 2, and 0.4 mol L−1 oxalic acid, and they were compared according to their appearance and SEM images (Fig. 3a–c). According to these evidence, oxalic acid (Fig. 3c) was chosen as the best environment for PPy electrodeposition, due to its highest porosity and uniformity of polymeric structure of all. After fabrication, each stir bar went through a conditioning step, with the aim of washing the extra unattached monomers

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

5

Fig 2. (a) Voltammograms of PPy electrosynthesis on the modified steel rod and the appearance of PPy-coated stir bar, (b) simplified reactions involved in electrochemical pyrrole (Py) polymerization in acidic media, and (c) proposed mechanism of metal –OH modification and its binding.

or impurities, opening up the cavities, and arousing more porosity through the polymeric structure. Since the mechanical stability of the stir bars were ruined during thermal conditioning for upper than 60 °C, solvent conditioning was tested. Subsequent conditioning in MeOH, deionized water, and ACN solvents, resulted in the finest structural readiness of stir bars for extraction, as shown in SEM images of before (Fig. 3c) and after (Fig. 3d) conditioning.

3.2. Optimization of effective analytical parameters during SBSE method Based on the parameters influencing the SBSE efficiency [34], and the operating principles in this study, the fabrication and SBSE method consist of three major steps: sorbent electrochemical fabrication, extraction, and liquid desorption. Accordingly, the optimization procedure was performed with the mentioned order. For each

6

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

Fig 3. SEM images of electropolymerization of PPy in (a) water, (b) HCl with pH 2, and aqueous oxalic acid 0.4 mol L−1 (c) before and (d) after solvent conditioning with the magnitude of respectively 10, 200, and 500 μm from left.

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

7

Fig 4. Charts of GC peak areas vs. (a) number of CV cycles, (b) pyrrole (Py) concentration, (c) concentration ratio of [Py]:[SDS]; and (d) the appearance of the fabricated stir bars with different potential scan rates.

effective parameter, the optimal choice for extraction of estradiol 1 μg mL−1 from standard aqueous samples, was chosen based on the highest extraction peak area of GC chromatogram. All of the experiments were performed based on optimization of one parameter at a time. In order to obtain the extraction chromatograms, first extraction and desorption steps for the blank sample were performed, which confirmed that chromatogram’s background was peakless in the retention time of estradiol (8.4 min in the corresponding temperature program). Thereafter, estradiol extraction and desorption steps were executed. After the first desorption, the stir bar was removed from the desorption vial placed in a second desorption vial with the same conditions and went through a second desorption step, in order to measure carryover of the stir bar. 3.2.1. Optimization of the electropolymerization procedure The most important variables of the electropolymerization are: number of CV cycles, Py concentration, concentration ratio of [Py]:[SDS], and scan rate of potential during the PPy electrochemical synthesis [35] of a solution containing Py and SDS in aqueous 0.4 mol L−1 oxalic acid. The effect of number of CV cyclic in the range of 5–30 on the extraction of estradiol was investigated. As shown in Fig. 4a, the best polymer thickness and uniformity, and highest peak area for extraction of estradiol, was gained for electropolymerization with 10 cycles of CV. At 5 cycles of CV due to thin thickness and mechanical loss of sorbent, the performing extraction procedure was impossible. Based on the peak areas of first and second desorption (carryover) (Fig. 4b), the best result for Py

concentration was 0.03 mol L−1 , due to formation of ramified dendrites for higher concentrations that could not extract estradiol. The best concentration ratio of [Py]:[SDS] was 1:1 which maintained the electric neutrality during the polymerization, as shown in Fig. 4c. Finally, among the tested potential scan rates in CV, all the scan rates above 10 mV s−1 destroyed the mechanical stability of the stir bar during the liquid desorption step. Therefore, the optimal choice was 10 mV s−1 , as observed in Fig. 4d. The proposed electrochemical method owns some advantages, it produces the stir bar through an easy one-step process and it can work at room temperature. Moreover, deposition progress, the thickness and degree of consistency of coating can be easily monitored by the amount of passed charge, and deposition time [36]. Electrochemical method is also a method with high-throughput for fabrication of an efficient SBSE. On the other hand, since the process of other techniques for synthesis of a coating on stir bar involves chemical reactions between several ingredients in solution, contains undesired compounds, in the final sorbent material. Also, the cost of raw materials, time-consuming processing and large volume shrinkage as well as cracking during drying and high sensitivity to environmental conditions are the other limitations of these techniques.

3.2.2. Optimization of the SBSE procedure Based on former reports, during absorption step, parameters such as extraction time, temperature, volume, pH, ionic strength and sample immersion stirring rate [30]; and during ultrasonic as-

8

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

Fig 5. Charts of estradiol peak areas vs. (a) extraction time, (b) stirring rate, (c) pH, (d) sample volume, (e) desorption solvent, and (f) desorption time.

sisted liquid desorption step, desorption solvent and desorption time are the considered optimizable variables [37]. The extraction procedure was sequential and based on different analytical parameters. Moreover, from this step onwards, the enrichment factors were used as the response for all the variables. We started the optimization of the SBSE with holding on to the previous optimized conditions of the electropolymerization procedure. Firstly, in order to optimize the ionic strength, different amounts of NaCl were introduced to the extraction sample. However, since the added salt even at low percentages, caused the stir bar destruction, it was omitted from the rest of experiments. To investigate the best extraction time, for reaching the equilibrium point of mass transfer between the organic and aqueous phases, different extraction times were tested. According to diagram of extraction time vs. enrichment factor of estradiol (Fig. 5a), with more time the enrichment factor of estradiol increases, until it reaches to 90 min. For times longer than that point a lag on the slope is observed. The slope would get smoother until it reaches the mass transfer equilibrium point (data not shown). Eventually, based on the considerations of the overall experiment time, 90 min was chosen as the optimum extraction time.

Next, the effect of immersion stirring rate on extraction was studied, since the increase of stirring rate, can cause more shuffling of the sample solution and result in increase of mass transfer and extraction efficiency. On the other hand, too much increase of stirring rate may jeopardize mechanical stability of the stir bar and so, decrease the extraction efficiency and stir bar’s life time. Among the tested rates, the highest enrichment factor for estradiol extraction, occurred for stirring rate of 800 rpm, as shown in Fig. 5b. Theoretically, according to the pKa of estradiol (10.71), its nature is considered as basic, so alkaline pHs are more compatible for absorbing it into the organic phase [38]. On the other hand, mechanical stability of the stir bar is reliant on pH. Therefore, pH optimization is essential to gain the highest extraction efficiency. Fig. 5c shows the diagram of estradiol enrichment factor vs. pH. Since the highest enrichment factor was achieved for the phosphate buffer solution with pH 7, it was chosen as the optimum environment for extraction. This can be due to the π -π interaction of PPy with a big molecule like estradiol, and the fact that the extraction follows adsorption mechanism. In pHs higher than 7, the mechanical stability and thus the repeatability of the stir bar was damaged, so that pHs higher than 9 destructed the stir bar structure.

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

9

Table 1 Method validation parameters, achieved from calibration curve and calculated data for accuracy and precision for estradiol extractions. Linear range (ng mL−1 ) g

50 -700 a b c d e f g

Linear equation

LODa /LOQb (ng mL−1 )

R2 c

Spiked amount (ng mL−1 )

Intra-bar (%RRd /CVe )

Inter-bar (%RR/CV)

y = 1.5x−33

10/33

0.9910

100

92.0/5.4

92.0/3.3

ERf (%) PPy

PDMS

Acrylate

43

10

12

Limit of detection. Theoretical limit of quantification (LOQ). Coefficient of determination. Relative recovery. Coefficient of variation. Extraction recovery of polypyrrole (PPy), polydimethylsiloxane (PDMS) and acrylate stir bar. Experimental limit of quantification.

The extraction volume affects the molecular motion inside the sample tourbillon. As bigger the tourbillon is (caused in bigger volumes), more molecular transfer from aqueous into organic phase happens. Conversely, since extraction volume is dependent to extraction time, huge volumes need longer extraction times for mass transfer and therefore, the extraction efficiency is diverse for different volumes in a constant time. Based on the results shown on Fig. 5d, 50 mL was reached as the most suitable extraction volume for 90 min extraction time. Extraction temperature, is another important parameter that needs optimization. As temperature rises, an increase of kinetic energy increases the mass transfer, and thus speeds up the extraction of analyte. Yet, in case of sorbent’s mechanical instability, it can reduce distribution coefficient and consequently decay extraction efficiency. In this report, the increase of temperature higher than 30 °C, damaged the stir bar’s repeatability. Therefore, 30 °C was fixed as extraction temperature for the rest of experiments. In the third step, optimization of liquid desorption parameters which play an important role in release of the extracted analyte from sorbent to the desorption solvent was investigated. The most crucial optimizable factors in ultrasonic liquid desorption step, are the type of desorption solvent and desorption time. Firstly, the best desorption solvent was chosen, relying on the compatibility with our hand-made PPy stir bar. Among the three most credible tested solvents, MeOH provided the highest enrichment factor of estradiol (Fig. 5e). Based on estradiol’s Ko/ w = 4.15, it is considered a relatively hydrophobic compound, but according to its structure [39], the two hydroxyl functional groups on the molecule sides, can possibly engage in hydrogen bonding with the small molecules of MeOH. Finally, the best desorption time based on the estradiol enrichment factors was selected according to the obtained results (Fig. 5f). Based on considerations of total experimental timing, and also the fact that times longer than 30 min, damaged the repeatability of the stir bar, 30 min was chosen as the optimal time for desorption. 3.3. Method validation After applying all the optimal conditions in SBSE procedure, the calibration curve was plotted through the peak areas of extracted estradiol from standard aqueous samples as a function of their concentration levels of 20 to 10 0 0 ng mL−1 . The method validation results including linear dynamic range, coefficient of determination (R2 ) and LOD are listed in Table 1. Linearity range of the proposed method was from 50 to 700 ng mL−1 , based on the verification of calibration curve. The value of R2 for the calibration curve was obtained 0.9910. LOD and theoretical LOQ were obtained 10 and 33 ng mL−1 , respectively, through 6 repetitive extractions of blank sample while the experimental value of LOQ as lower limit in linear range was obtained 50 ng mL−1 . ’The LOD can be expected to improve if a more sensitive detector such as mass spectrometry is used.

Fig 6. The chromatograms of (a) non-extracted standard at concentration level 10 μg mL−1 and (b) water extract standard at concentration level 100 ng mL−1 and extracted urine sample of a volunteer healthy lady (c) before and (d) after spike of 100 ng mL−1 estradiol.

In order to calculate the extraction recovery (ER%) of the fabricated stir bar, the theoretical enrichment factor was compared to our experimented enrichment factor (87), which ER% = 43% was achieved. Based on these results, it can be concluded that the proposed method possesses acceptable accuracy and precision at extraction of estradiol from urine samples. For further evaluate, the extraction efficiency of the PPy sorbent was compared with two commercial stir bars PDMS and Acrylate Twisters®. The extraction recovery values were given in Table 1, which indicates the PPy sorbent has a better efficiency than other commercial stir bars for extraction of estradiol. The affinity of PPy coating may be due to a more porous structure of coating in electrochemical technique, which affords a fast mass transfer of analyte by increasing the surface area of coating. 3.4. Real sample analysis To investigate the applicability of the electrosynthesized PPycoated stir bar in SBSE method for determination of estradiol, urine samples of a healthy volunteered lady and water samples were analyzed through spiking method, by adding 100 ng mL−1 of standard estradiol to the real samples. Fig. 6 shows the chromatograms of a urine sample before and after spike of estradiol compare to non-extracted and water extract sample. Based on the obtained results the matrix effect was calculated according to Eq. (1) and

10

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

obtained −9 (±1)%. As can be seen, the proposed sample preparation method using PPy sorptive phase could extract estradiol, selective and interferenceless from complex matrix of urine. The recoveries in water and urine sample were obtained 92% and 84%, respectively. To the aim of evaluating the intra- and inter-bar accuracy and precision, three successive extractions of the quality control solution using one coated stir bar and three individual extractions of the quality control solution using three coated stir bars were performed in human urine. Results demonstrate good recovery values for the assay with coefficient of variation (CV%) 5.4% and 3.3% for intra- and inter-bar precision, respectively (Table 1). Based on the results of intra-bar repeatability of the stir bar, it showed the ability of being repeatedly reconditioned and used for 3 times without significant loss of extraction efficiency. Moreover, due to good reproducibility (inter-bar precision) and low price, simplicity and high throughput of the electrochemical method in successive preparation of PPy-coated stir bars, one can expect the possibility to fabricate several stir bars during the same procedure without worrying about loss of quality. 4. Conclusion The proposed new electro-fabrication procedure of PPy-coated stir bar via CV was capable of quantifying estradiol as a target analyte from urine samples through SBSE-GC methodology. Fabrication through electropolymerization has significant advantages, compared to previous chemical methods, such as one step production, shorter time consumption, eliminating consumption of toxic organic solvents, simplicity, and the ability of atomization of the procedure. Thus, it is a straightforward and high-throughput method for fabrication of a new extraction phase SBSE. The low price of the steel rod and applying a given potential to selective electropolymerization of PPy in the successively preparation of PPy-coated stir bars can be economically feasible too. PPy-coated stir bar manifested excellent extraction efficiency towards estradiol compared with PDMS and acrylate (two commercial SBSE coatings). The proposed method was evaluated using state-of-theart validation criteria: linearity (50–700 ng mL−1 with R2 value 0.9910), accuracy (RR% value 92.0), repeatability (CV% 5.4) and reproducibility (CV% 3.3). The matrix effect was obtained −9 (±1)%. The validated method was applied to estradiol assessment, as a result, the amount of spiked estradiol was successfully evaluated from urine matrix (with recovery 84%). This method opens up the way to future investigations concerning the potential stir bar fabrication through co-electropolymerization of Py with hydrophilic CPs or some nanoparticles, to promote the extraction ability of a wide range of analytes with various polarities. Declaration of Competing Interest None. Acknowledgment The authors gratefully acknowledge their gratitude to the research council of Alzahra University. References [1] S. Moldoveanu, V. David, in: Modern Sample Preparation for Chromatography, Elsevier, 2015, pp. 132–154. [2] Y. Wen, L. Chen, J. Li, D. Liu, L. Chen, Recent advances in solid-phase sorbents for sample preparation prior to chromatographic analysis, Trends Anal. Chem. 59 (2014) 26–41. [3] C.L. Arthur, J. Pawliszyn, Solid phase microextraction with thermal desorption using fused silica optical fibers, Anal. Chem. 62 (1990) 2145–2148. [4] E. Baltussen, P. Sandra, F. David, C. Cramers, Stir bar sorptive extraction (SBSE), a novel extraction technique for aqueous samples: theory and principles, J. Microcolumn Sep. 11 (1999) 737–747.

[5] Z. Talebpour, M. Taraji, N. Adib, Stir bar sorptive extraction and high performance liquid chromatographic determination of carvedilol in human serum using two different polymeric phases and an ionic liquid as desorption solvent, J. Chromatogr. A 1236 (2012) 1–6. [6] H. Lan, D. Pan, Y. Sun, Y. Guo, Z. Wu, Thin metal organic frameworks coatings by cathodic electrodeposition for solid-phase microextraction and analysis of trace exogenous estrogens in milk, Anal. Chim. Acta 937 (2016) 53–60. [7] S. Karimi, Z. Talebpour, N. Adib, Sorptive thin film microextraction followed by direct solid state spectrofluorimetry: a simple, rapid and sensitive method for determination of carvedilol in human plasma, Anal. Chim. Acta 924 (2016) 45–52. [8] M.C. Alcudia-León, R. Lucena, S. Cardenas, M. Valcarcel, A. Kabir, K.G. Furton, Integrated sampling and analysis unit for the determination of sexual pheromones in environmental air using fabric phase sorptive extraction and headspace-gas chromatography–mass spectrometry, J. Chromatogr. A 1488 (2017) 17–25. [9] R. Kumar, A. Gaurav, R. Heena, A.K. Malik, A. Kabir, K.G. Furton, Efficient analysis of selected estrogens using fabric phase sorptive extraction and high performance liquid chromatography-fluorescence detection, J. Chromatogr. A 1359 (2014) 16–25. [10] H. Chen, J. Liu, M. Cui, J. Chen, X. Li, Y. Chen, Simultaneous determination of four amides in Saururus chinensis by matrix solid phase dispersive and high-performance liquid chromatography method, J. Food Drug Anal. 26 (2018) 362–368. [11] A. Piñeiro-García, G. Gonzalez-Alatorre, F. Tristan, J.C. Fierro-Gonzalez, S.M. Vega-Diaz, Simple preparation of reduced graphene oxide coatings for solid phase micro-extraction (SPME) of furfural to be detected by gas chromatography/mass spectrometry, Mater. Chem. Phys. 213 (2018) 556–561. [12] H. Lan, T. Ronkko, J. Parshintsev, K. Hartonen, N. Gam, M. Sakeye, J. Sarfraz, M.L. Riekkola, Modified zeolitic imidazolate framework-8 as solid-phase microextraction arrow coating for sampling of amines in wastewater and food samples followed by gas chromatography–mass spectrometry, J. Chromatogr. A 1486 (2017) 76–85. [13] A. Kumar, A. Gaurav, A.K. Malik, D.K. Tewary, B. Singh, A review on development of solid phase microextraction fibers by sol-gel methods and their applications, Anal. Chim. Acta 610 (2008) 1–14. [14] S. Kesani, A. Malik, Front cover, sol-gel niobia sorbent with a positively charged octadecyl ligand providing enhanced enrichment of nucleotides and organophosphorus pesticides in capillary microextraction for online HPLC analysis, J. Sep. Sci. 41 (2018) 1663–1673. [15] M.O. Aziz-Zanjani, A. Mehdinia, Electrochemically prepared solid-state microextraction coating-a review, Anal. Chim. Acta 781 (2013) 1–13. [16] L. Fotouhi, M. Naseri, Recent electroanalytical studies of metal-organic frameworks: a mini-review, Crit. Rev. Anal. Chem. 46 (2016) 323–331. [17] M. Naseri, L. Fotouhi, Ehsani A, Recent progress in the development of conducting polymer-based nanocomposites for electrochemical biosensors applications: a mini-review, Chem. Rec. 18 (2018) 599–618. [18] M.M. Abolghasemi, S. Parastari, V. Yousefi, A nanoporous anodized alumina wire with a nanosized hydroxyapatite coating for headspace solid-phase microextraction of phenol and chlorophenols, Microchim. Acta 183 (2016) 241–247. [19] J. Feng, X. Wang, Y. Tian, Y. Bu, C. Luo, M. Sun, Electrophoretic deposition of graphene oxide onto carbon fibers for in-tube solid-phase microextraction, J. Chromatogr. A 1517 (2017) 209–214. [20] J. Wu, J. Pawliszyn, Preparation and applications of polypyrrole films in solid-phase microextraction, J. Chromatogr. A 909 (2001) 37–52. [21] H. Abedi, H. Ebrahimzadeh, J.B. Ghasemi, Solid phase headspace microextraction of tricyclic antidepressants using a directly prepared nanocomposite consisting of graphene, CTAB and polyaniline, Microchim. Acta 182 (2015) 633–641. [22] H. Asiabi, Y. Yamini, M. Shamsayei, Development of electrochemically controlled packed-in-tube solid phase microextraction method for sensitive analysis of acidic drugs in biological samples, Talanta 185 (2018) 80–88. [23] C. Hu, M. He, B. Chen, B. Hu, A sol-gel polydimethylsiloxane/polythiophene coated stir bar sorptive extraction combined with gas chromatography-flame photometric detection for the determination of organophosphorus pesticides in environmental water samples, J. Chromatogr. A 1275 (2013) 25–31. [24] Z. Talebpour, S. Rostami, H. Rezadoost, Evaluation of a method for the simultaneous quantification of N-nitrosamines in water samples based on stir bar sorptive extraction combined with high-performance liquid chromatography and diode array detection, J. Sep. Sci. 38 (2015) 1601–1609. [25] Z. Xu, Z. Yang, Z. Liu, Development of dual-templates molecularly imprinted stir bar sorptive extraction and its application for the analysis of environmental estrogens in water and plastic samples, J. Chromatogr. A 1358 (2014) 52–59. [26] V. Babarahimi, Z. Talebpour, F. Haghighi, N. Adib, H. Vahidi, Validated determination of losartan and valsartan in human plasma by stir bar sorptive extraction based on acrylate monolithic polymer, liquid chromatographic analysis and experimental design methodology, J. Pharm. Biomed. Anal. 153 (2018) 204–213. [27] M. Torabizadeh, Z. Talebpour, N. Adib, H.Y. Aboul-Enenin, Preparation of a novel sorptive stir bar based on vinylpyrrolidone-ethylene glycol dimethacrylate monolithic polymer for the simultaneous extraction of diazepam and nordazepam from human plasma, J. Sep. Sci. 39 (2016) 1316–1325. [28] E. Noroozian, M. Rahimi, Application of poly(Pyrrole-co-Phenol) electrosynthesized for stir-bar sorptive extraction of polycyclic aromatic hydrocarbons and

P. Asadi Atoi, Z. Talebpour and L. Fotouhi / Journal of Chromatography A 1604 (2019) 460478

[29]

[30]

[31]

[32]

[33] [34]

liquid chromatographic analysis, J. Liq. Chromatogr. Relat. Technol. 38 (2015) 172–181. M. Ansari Dogaheh, M. Behzadi, Preparation of polypyrrole/nanosilica composite for solid-phase microextraction of bisphenol and phthalates migrated from containers to eye drops and injection solutions, J. Pharm. Anal. (2019) doi.org/, doi:10.1016/j.jpha.2019.03.006. L.P. Melo, A.M. Nogueira, F.M. Lancas, M.E.C. Queiroz, Polydimethylsiloxane-polypyrrole stir bar sorptive extraction and liquid chromatography (SBSE/LC-UV) analysis of antidepressants in plasma samples, Anal. Chim. Acta 633 (2009) 57–64. O.I. Istakova, D.V. Konev, A.T. Glazkov, T.O. Medvedeva, E.V. Zolotukhina, M.A. Vorotyntsev, Electrochemical synthesis of polypyrrole in powder form, J. Solid State Electrochem. 23 (2019) 251–258. H. Liu, L. Qiao, N. Gan, S. Lin, Y. Cao, F. Hu, J. Wang, Y. Chen, Electro-deposited poly-luminol molecularly imprinted polymer coating on carboxyl graphene for stir bar sorptive extraction of estrogens in milk, J. Chromatogr. B 1027 (2016) 50–56. M. Wang, Emerging multifunctional NIR photothermal therapy systems based on polypyrrole nanoparticles, Polymers 8 (2016) 373–399. A. Prieto, O. Basauri, R. Rodil, A. Usobiaga, L.A. Fernandez, N. Etxebarria, O. Zuloaga, Stir-bar sorptive extraction: a view on method optimisation, novel ap-

[35]

[36]

[37]

[38]

[39]

11

plications, limitations and potential solutions, J. Chromatogr. A 1217 (2010) 2642–2666. A. Nezhadali, G. Ahmadi Bonakdar, H. Nakhaei, Electrosynthesis of polypyrrole on steel fiber for solid-phase microextraction of citalopram in serum, Anal. Bioanal. Chem. 403 (2012) 593–600. S. Lin, N. Gan, J. Zhang, L. Qiao, Y. Chen, Y. Cao, Aptamer-functionalized stir bar sorptive extraction coupled with gas chromatography-mass spectrometry for selective ecrichment and determination of polychlorinated biphenyls in fish samples, Talanta 149 (2016) 266–274. M. Barriada-Pereira, P. Serodio, M.J. Gonzalez, J.M. Nogueira, Determination of organochlorine pesticides in vegetable matrices by stir bar sorptive extraction with liquid desorption and large volume injection-gas chromatography–mass spectrometry towards compliance with European Union directives, J. Chromatogr. A 1217 (2010) 119–126. M. Kawaguchi, R. Ito, N. Sakui, N. Okanouchi, K. Saito, H. Nakazawa, Dual derivatization–stir bar sorptive extraction–thermal desorption–gas chromatography–mass spectrometry for determination of 17β -estradiol in water sample, J. Chromatogr. A 1105 (2006) 140–147. J. Aufartová, C. Mahugo-Santana, Z. Sosa-Ferrera, J.J. Santana-Rodriguez, L. Novakova, P. Solich, Determination of steroid hormones in biological and environmental samples using green microextraction techniques: an overview, Anal. Chim. Acta 704 (2011) 33–46.