Simultaneous determination of five opium alkaloids in underground waters using molecularly imprinted polymer-modified magnetic nanoparticle based dispersive micro-solid phase extraction followed by high performance liquid chromatography

ABSTRACT In this work, we present a fast, simple and selective method for the extraction/preconcentration and determination of five opium alkaloids. This method is based on the combination of dispersive solid phase extraction with magnetic molecularly imprinted polymers (MMIP) and determination via high performance liquid chromatography. The surface modified Fe3O4 nanoparticles were synthesised in-situ, from a co-precipitation method by addition of a mixture of the methacrylic acid and ammonia to a mixture of Fe2+/Fe3+ under nitrogen atmosphere. MMIP was prepared from surface modified nanoparticles in the presence of ethylene glycol dimethyl acrylate as cross-linker and azobisisobutyronitrile as initiator and template molecule under reflux in acetonitrile. MMIPs for five natural occurring opium alkaloids, including morphine (MO), codeine (CO), thebaine (TE), noscapine (NO) and papaverine (PA) were prepared in the same manner. Structural characterisation and elucidation of both synthesised nanoprticles were performed by Fourier transform-infrared spectroscopy, scanning electron microscopy. X-ray diffraction and vibrating sample magnetometry. A precise chromatographic method was developed and coupled with extraction with MMIP nanoparticles for simultaneous preconcentration and determination of above-mentioned alkaloids in underground water samples. The limit of detection obtained 0.007, 0.007, 0.004, 0.003, 0.003 mg L−1 for MO, CO, TE, NO and PA, respectively and recoveries were about 97–102%. Linear dynamic range and relative standard deviation were 0.03–100 mg L−1 and less than 1.5% for each analyte.


Introduction
Opium has been known as pain relief material from predates written history.It is exuded from surface incisions in the unripe seed heads capsule of papaver plant.Opiates are very potent analgesics commonly used as therapeutic agents [1].Some of these compounds are also frequently abused as illicit drugs.The most significant alkaloids of opium, in terms of their quantity and pharmaceutical importance, are morphine (MO), codeine (CO), thebaine (TE), noscapine (NO) and papaverine (PA), also called poppy alkaloids (Fig. S1).The determination of these drugs in aqueous media is a major concern in environmental and clinical toxicology [2,3].
Several analytical methods were developed for both simultaneous and individual detection and determination of opium alkaloids in biological, environmental and industrial samples.The most common methods of determination of opium alkaloids are based on chromatographic techniques such as gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-diode array detection (HPLC-DAD) [4][5][6].The major disadvantage of GC methods is the complicated, difficult and costly sample preparation needed for the derivatisation of the analytes prior to GC analysis.The analysis of opium alkaloids by HPLC-DAD is hindered by the similar chromatographic behaviour of the minor alkaloids papaverine and noscapine and the likely presence of substantial matrix interferences, thus requiring purification steps.Recently, LC-MS/MS is carried up for determination of six opium alkaloids from bakery products and poppy seed after a solvent extraction with a ternary solvent mixture [7].LC-MS/MS is extensively used in forensic research on opium alkaloids, at the same time it is an expensive and costly technique [8,9].There is also a report on simultaneous hydrodynamic amperometric and differential pulse voltammetric determination of morphine and codeine in aqueous media [10].Chemiluminescence detection of opium poppy alkaloids in combination with HPLC, capillary electrophoresis and microfluidic instrumentations was also reported [11].Quantitative analysis of trace levels of opium alkaloids is an important challenge requiring an effective and fast sample preparation procedure prior to analysis.Analytical procedures like liquid-liquid extraction (LLE) and solid-phase extraction (SPE) have been developed for this purpose.Electromembrane extraction and magnetic SPE with multiwall carbon nanotubes-magnetite-nanoparticles (CNT-Fe 3 O 4 -NPs) prior to HPLC were done as sample preparation method in determination of thebaine and morphine, respectively [12].Dispersive liquid-liquid microextraction was also developed for determination of some opium alkaloids [13].In the recent years, the application of SPE procedures involving the use of synthetic antibody mimics, such as molecularly imprinted polymers (MIPs), has received more and more attentions as an attractive alternative for the analysis of complex samples [14,15].The advantages of MIPs are short synthesis time, long term stability in laboratory conditions, regeneration and reusability [16,17].Several analytical methods using MIPs were developed for the preconcentration of trace amounts of opium alkaloids, as adsorbent in SPE, prior to various chromatoghraphic, electrochemical, luminescence and spectrometric detection techniques [18][19][20][21].However, MIPs suffer from drawbacks in some applications, for example, slow mass transfer, incomplete template removal, heterogeneous distribution of the binding sites, poor site accessibility to the target molecules [19].To solve these problems, one of the promising alternatives is surface imprinting on the magnetic solid support.The combining magnetic separation and molecular imprinting has many advantages, improving the binding capacity and site accessibility of imprinted materials, showing high selectivity (in terms of shape, size, and functionality) for target molecules, and rapid isolation of analytes from samples under an external magnet.At the same time, polymeric coating prevents the oxidation of magnetic support.
Dispersive SPE is an attractive approach based on the utilisation of the solid phase dispersion in a sample solution, in comparison to conventional SPE.The dispersion phenomenon accelerates the interaction of the analyte with the adsorbent particles, resulting in shorter extraction times.The most common adsorbent in the D-micro-SPE method is nano-Fe 3 O 4 , which is a magnetic nanoparticle (MNP) [22,23].Regarding the high chemical activity of bare MNPs, their oxidisation in the air may lead to the loss of dispersibility as well as magnetism.Thus, their surface should be appropriately coated to maintain their stability.In this context, coatings with MIPs can be helpful [24].The obvious advantages of MNP adsorbents which appeared to be suitable carriers for MIPs in D-micro-SPE are short diffusion route, providing improved extraction dynamics, large surface area and, therefore high extraction efficiency, simple and fast isolation of MNPs from the sample solution and ease of surface functionalisation/modification [25,26].
Considering the combined impacts of these materials, the present work addresses the synthesis and characterisation of MIP-modified Fe 3 O 4 NPs and evaluates their uses in D-micro-SPE.To assess the viability of the developed method, the appropriateness of the MIP modified-Fe 3 O 4 NPs was simultaneously evaluated as the adsorbent for D-micro-SPE of five opium alkaloids from various water samples.The factors affecting the efficiency of the extraction were optimised by the one factor at a time (OFAT) method.Furthermore, high-performance liquid chromatography was employed to separate and detect alkaloids.The proposed method was used to analyse the selected alkaloids in real water samples from some pharmaceutical industrial zones.

Exprimental
Chemicals and materials, apparatus and instruments, preparation of alkaloid molecules in base form and chromatographic condition are described in Supplementary Online Material.

Preparation of MAA@Fe 3 O 4 NPs
Solution (A): MAA@Fe 3 O 4 nanoparticles were prepared using conventional chemical coprecipitation method with some modifications [22,23].Formation and surface modification of iron oxide nanoparticles were simultaneously performed using alkaline MAA solution in one step.To briefly describe, 5.50 g (20 mmol) of FeCl 3. 6H 2 O and 2.10 g (10 mmol) of FeCl 2. 4H 2 O were transferred to the dissolution flask that contains a mixture of 50 mL distilled water (conductivity less than 0.1 µs/cm), 1.00 g (11.5 mmol) of methacrylic acid and resulting solution immediately was treated in an ultrasonic bath by purging nitrogen for 5 min.The degassed solution was transferred to the reactor and purged by nitrogen all the time, while mixing the mixture (vigorously at 1000 rpm).Temperature of the reaction mass was raised slowly to 70°C in an oil bath.
Solution (B): 2.00 g (23 mmol) of methacrylic acid was dissolved in a 100 mL 15% ammonia and degassed at room temperature by nitrogen for 10 min.
Procedure: Solution (B) was transferred to a dropping funnel and added drop-wise to the solution (A) by flowrate of 3 mL min −1 at controlled temperature of 70°C.After the reaction was subsided, the reaction mixture stirred for 15 min more at the same temperature, then allowed to cool to room temperature slowly and the produced MAA@Fe 3 O 4 NPs were collected by an external magnet.The product was washed 5 times by 50 mL portions of distilled water till the supernatant solution became neutral.The obtained MAA@Fe 3 O 4 NPs was dried at 60°C for 6 h and was stored in the laboratory for the next uses.

Preparation of MIP@MAA@Fe 3 O 4 NPs
0.100 g (0.35 mmol) of morphine base (MO) was dissolved in 50 mL acetonitrile, degassed by ultrasonic and nitrogen bubbling for 5 min at room temperature.Then 0.35 g (4.06 mmol) of MAA was added and the degassing procedure continued for 15 min and mix to form template-monomer complex [25].Then 0.22 g of MAA@Fe 3 O 4 was suspended in acetonitrile, degassed and added to the reaction mixture.5.00 g (25.2 mmol) of EGDMA and then 0.05 g (0.3 mmol) AIBN added and the reaction mixture cold to near zero, added.Then, the reaction mixture was warmed up to 60 °C in reflux condition and maintained at this temperature for 6 h.After completing the MIP@MAA@Fe 3 O 4 NPs, solvent was evaporated and the product washed three successive times with 20 mL portions of methanol: acetic acid (9:1, v:v).The leached solutions were collected and determined by HPLC to ensure complete removal of template.This same procedure was performed sepatrately for each of the other selected alkaloids as template.Procedure of the synthesis and analysis is schematically shown in Figure 1.As control, magnetic non-imprinted polymers (MNIP) nanoparticles were synthesised under similar conditions but omitting the template in the reaction system.

Dispersive micro-SPE procedure
Twenty-five microgram of each of five MMIPs were taken and mixed together and then added to the 100 mL sample solution in a glass beaker.pH of solution was adjusted to 9 by adding Britton-Robinson buffer and the suspension was sonicated for 5 min in an ultrasonic bath at 20°C.A permanent NdFeB magnet was put at the bottom of the beaker to settle the NPs.After discarding the supernatant solution, MMIPs was dried under nitrogen flow.By adding 500 μL of desorption solvent (a mixture of methanol: acetic acid 9:1(v/v)), desorption of analytes was done under sonication for 2 min.Finally, MMIP-NPs were isolated from the solvent, using a permanent magnet, and after the filtration with a 0.45 µm PTFE filter, 10 μL of the solution was injected into the HPLC system.MMIP-NPs were regenerated by washing with 2 × 5 mL of methanol: acetic acid 9:10 and conditioned with 2 × 5 mL deionised water for further uses.

Preparation and characterisation of magnetic imprinted nanoparticles
The schematic diagram of the synthesis procedure of MIP @ MAA@Fe 3 O 4 nanoparticles (MIP@MNPs) is shown in Figure 1.Fe 3 O 4 nanoparticles (MNPs) were prepared to produce ferromagnetic properties that could be collected by the external magnet and enhance the surface to volume ratio to produce nano-sized polymers, resulted in having more cavities to adsorb template molecules.Methacrylic acid was selected as the monomer to prepare the imprinted polymers.Molar ratio of template molecule/functional monomer/crosslinker agent is a critical factor which affects the adsorption properties of imprinted polymers.We used molar ratios of the reagents in accordance with our previous studies [27].After the completion of surface polymerisation, MMIPs were collected using an external magnet and then rinsed several times with methanol: acetic acid solution to remove unreacted reagents and extract template molecules till no template molecule could be seen in supernatant by HPLC.
The morphological structure of synthesised NPs can be observed in scanning electron microscopic (SEM) images of them.As shown in Figure 2(a) Figure 2(b), MAA@Fe 3 O 4 and MIP@MAA@Fe 3 O 4 nanoparticles both appear as a spherical form.The images confirm nanostructures with average diameter of 60 nm for MIP@MAA@Fe 3 O 4 , which is larger than that of the MAA@Fe 3 O 4 confirming that MIP layers were coated on the MAA@Fe 3 O 4 surface successfully.Porous aggregates of spherodial-shaped particles are also clearly visible in microscopic image of MIP@MNPs.
To further confirm polymerisation, energy dispersive X-ray (EDX) analysis was performed.EDX spectrum (Figure 2(c)) indicates the presence of Fe, O and C, confirming that polymerisation occurs on the surface of Fe 3 O 4 MNPs.The magnetic properties of bare MNPs and MIP@MNPs were investigated by using vibrating-sample magnetometry (VSM) at room temperature.In Figure 2(d), no hysteresis was observed, indicating the NPs' characteristic superparamagnetic behaviour at room temperature.Indeed, no magnetism remained in the NPs after removing the magnetic field [28].According to Figure 2(d), for the non-coated MNPs, the saturation magnetisation value was 65 emu g −1 .After the MIP coating process, a decreasing over 29 emu g −1 was seen in saturating magnetisation.Nevertheless, as a result of a change in the chemical structure of Fe 3 O 4 NPs in this procedure, no considerable impact was found on the magnetic properties of MIP@MNPs.Furthermore, the measured saturation magnetisation value of MIP@MNPs was strong enough for quick separation from aqueous solution, which allows them an appropriate selection for magnetic SPE procedure, so dispersed MIP@MNPs in a solution can be isolated by an external magnetic field.
Figure 3(a) presents the X-ray diffraction (XRD) patterns of bare Fe 3 O 4 , MAA@Fe 3 O 4 and MIP@MAA@Fe 3 O 4 nanoparticles.The diffraction peaks were observed at 2Ɵ = 30.1°,33.6°, 43.0°, 53.4°, 56.9°, 62.4° which can be allocated to the (220), (311), (400), (422), (511), (440) planes, respectively.These XRD patterns correspond to an inverse cubic spinel phase of Fe 3 O 4 NPs according to the standard data in the literature for magnetite (with JCPDS card no.85-1436) [28].Amorphous layers can be also observed in the range of 2Ɵ = 15-25, proving the existence of a polymer layer on the modified Fe 3 O 4 NP surface.The decrease in the intensity of the peaks and their broadening is the result of the polymer coating of the surface of nanoparticles [22].
As shown in Figure 3

Optimisation of extraction condition
The extraction conditions were independently optimised using OFAT method.pH of sample, time of extraction, amount of sorbent and temperature were investigated as major effective factors on extraction efficiencies (Figure 4).

Effect of pH
pH value of sample solution is an important variable affecting adsorption of analytes on the surface of sorbent.The effect of pH on the extraction at constant concentrations of selected analytes, was investigated in the range of 2.5-11.MIP@MNPs were collected from the solution by applying an external magnet and supernatant solution was injected to HPLC after the filtration with a 0.45 µm PTFE filter.All experiments were performed in triplicates.Results show that the recovery of opium alkaloids increases with increasing to pH of 9. Further increase in pH leads to a clear decrease in extraction efficiency of morhine while it remains almost constant for the other analytes.Increasing the pH increases the fraction of molecular species, decreases the fraction of protonated form of the alkaloids and consequently leads to more efficient extractions.Different behaviour in extraction of morhine, is due to dissociation of phenolic proton at higher pHs.Extraction of morphine is maximised in pH 9, which is compatible with acidic constants (pK a1 = 8.32, pK a2 = 9.65) and isoelectric pH (pH isoelectric = 8.98) of morphine.Therefore, pH 9.0 was selected as optimal in simultaneous extraction of five selected alkaloids (Figure 4(a)).

Effect of temperature
Extraction procedure was repeated at 10, 20, 25, 30 and 35°C in adjusted pH 9.0.As shown in Figure 4(b), the results showed that temperature variation in range of 25-35°C has insignificant effect on extraction recoveries, but a clear decrease was happened at lower temperature.Thus, 25°C was used as optimal temperature in further experiments due to accuracy and simplicity of control.

Effect of extraction time
Effect of time on the extraction efficiency was also investigated in optimal pH and controlled temperatue 25°C.Time intervals of 1, 2, 3 and 5 min were selected as extraction times.The results showed that extraction was completed in early minutes and applying more time does not improve the extraction efficiency.As shown in Figure 4(c), the optimum response was observed applying 5 min, thus magnetic separation of MMIP NPs from the solution was done after 5 min.

Effect of amount of adsorbent
Definite weight of MIP@MNPs was added to 20.0 mL of 20 ppm solution of each analyte at optimised pH (9.0) and temperatue (25°C), and magnetic separation of adsorbent was done after 5 min.Supernatant solution was filtered by 0.45 µm PTFE filter and injected to the HPLC.Effect of adsorbant amount was studied for 2, 8, 16, 25 and 30 mg of MMIP NPs.This study was performed for all five alkaloids in a similar way and as shown in the Figure 4 (d), extraction efficiency of the alkaloids is similarly increased with raising weight of the sorbent up to 25 mg and then remained constant.Therefore, the weight of adsorbent used to extract each of the alkaloids was selected to be 25 mg.

Type and volume of desorption solvent
The type and volume of the desorbing solution used for the desorption of the analytes from the sorbent are another important parameters affecting the extraction efficiency of the procedure.The effect of the nature of the eluent on the desorption of selected alkaloids from the MIP@MNPs was studied using 1.0 mL of various eluents such as methanol, ethanol, acetone, acetonitrile and the solutions of methanol containing 5%, 10% and 15% acetic acid (by volume).The extraction recovery of each alkaloid from the related sorbent was found to be more than 97% with solutions of methanol containing 10% or 15% acetic acid (Figure 5).The average recoveries with the other eluents were as follows: methanol (51.0%), ethanol (28.6%), acetone (30.9%) and acetonitrile (22.6%), methanol containing 5% acetic acid (88.7%).Protonation of the monomer, which weakens interaction with analyte molecules, increases the recovery by methanol acidic solutions.So, 9:1 mixture of methanol/acetic acid was selected as the desorbing solution for future studies (Figure 5).The volume of the eluent is an important factor that affects the extraction recovery and preconcentration factor (PF) of the system.A decrease in the volume of the eluent will increase the PF, but it may reduce the recovery of the analyte from the sorbent.To find the optimum volume of the desorbing solution, various experiments were carried out by varying the volumes of in the range of 0.4 − 2.0 mL.The results implied that 0.5 mL of the eluent is sufficient for complete desorption of selected alkaloids.Thus, in order to get high PFs and, consequently, improve detection limits, 0.5 mL of methanol:acetic acid (9:1) was used for efficient recovery of retained alkaloids from the sorbent for the subsequent experiments.

Selectivity of adsorbents
In order to study selectivity of each adsorbent for the target molecule and evaluate potential interference of other alkaloids, a series of experiments were performed to measure adsorption of target alkaloid in presence of equal concentration of other four analytes from aqueous sample.The experiment was repeated for all five alkaloids repeatedly as same.As shown in Figure 6, each MMIP shows very good adsorption for its template molecule (79.9-88.3%),while the adsorption of other species was negligible (0.1-5.3%).Therefore, it is possible and logical to use a mixture of five MMIP adsorbents for the simultaneous quantitative extraction and preconcentration of the five analytes from water samples prior to HPLC determination.

Reusability of adsorbents
Reusability is a key feature of sorbent materials from the perspective of stability, cost effectiveness, and also environmental protection.Stability of MMIP adsorbents was evaluated by investigating their possibility of regeneration and reusability, which is one of the most important factors for their application.On the basis, a batch of MMIPs is evaluated by 10 consecutive adsorption-desorption (regeneration) cycles.MMIPs could be regenerated by washing with 2 × 5 mL of methanol: acetic acid 9:10 and conditioned with 2 × 5 ml deionised water for further uses.The results revealed that, the extraction recovery of MMIPs still maintains more than 92-95% of the first recovery for the selected alkaloids after ten cycles.In other word, MMIPs can be employed again to adsorb the interested analytes after regeneration and retained their adsorption capacity at an approximately fixed value.Figure 7 illustrates that average loss in adsorption of selected alkaloids between first and 10th cycle was about 6%.The decreased adsorption capacity indicates that a small number of imprinted sites have been destroyed after the 10 successive regeneration process.Meanwhile adsorption capacity of MNIPs remains almost unchanged, because their affinities towards targets are non-specific and the effect of washing can be negligible.Long-term stability (shelf life) of MMIPs was also checked in 3, 6, 9 and 12 month periods and extraction properties of adsorbents remained unchanged in closed vessel after 12 months in room temperature, ambient humidity and away from light.The results demonstrate that the MMIPs are reusable, stable and selective for five selected alkaloids, so that they have potential values in practical applications.

Analytical performance
Under optimum conditions, a series of experiments were performed to consider the figures of merit of the proposed method including linear range, correlation coefficient (R 2 ), precision, limit of detection (LOD) and limit of quantification (LOQ).The LOD and LOQ of the developed method for the selected alkaloids, defined as the ratio of 3 and 10 times of the standard deviation of the blank signals to the slope of the calibration curve, were found to be 0.003-0.007and 0.010-0.024mg L −1 , respectively (Table 1).The calibration graphs were linear over the range of 0.03-100 mg L −1 of the alkaloids.
The result also indicates good linearity for the preconcentrated alkaloids from 100 mL of the aqueous solution.Thus, when the concentration of alkaloids in the sample is higher, the extraction procedure can be applied to a lower sample volume so that the concentration of the analyts in the final extract is in the linear range of the calibration graph.The preconcentration factor, which is defined as the ratio of volume of initial sample to the volume of final extracted solution, is equal to 200.Enrichment Factor (EF) that is defined as the concentration ratio of final extract to initial solution (EF = C f /C i ), gives another expression of extraction.EF values for the selected alkaloids were contained in the range of 158.2-176.4 at optimum conditions.Extraction recovery (ER) was utilised to assess the optimum conditions.ER is defined as the percentage of the total amount of analyte (n 0 ), which is extracted into the extracted phase (n ex ).The ER was calculated correctly as an analytical response as follows: where C 0 and C ex represent the analyte's initial concentration in the aqueous sample and the analyte's concentration in the extracted phase, respectively.C 0 is calculated using a calibration curve obtained via direct injection of the standard solutions into the HPLC system.V sam and V ex denote the volumes of the aqueous sample and extracted phase, respectively [21].Extraction recovery can also be defined as ratio of enrichment factor to preconcentration factor (Table 1).

Analysis of real samples
To evaluate performance of the proposed method, the developed procedure was applied to determine target alkaloids in the water samples.Underground water samples from some pharmaceutical industrial zones, were selected as real samples.Recovery tests were used to evaluate the matrix effects, accuracy and reliability of the method.When no aqueous standard reference materials with a certified content of target analyte are available, an added-found method can be considered as an alternative for validation studies.The different standard amounts of selected alkaloids were spiked into the samples and their concentration was determined at optimum conditions, applying the extraction procedure by using the related calibration curves, for each sample.The results of study are summarised in Table 2, as can be seen indicate the potential of this method for efficient and sensitive determination of target alkaloids in real water samples.The results show similar recoveries (reported as relative recoveries) and are in good agreement with those obtained by aqueous standards, demonstrating the practical analytical utility of the method.Relative recovery (RR) is defined as the ratio of the found concentration of the analyte (C found ) to its added concentration (C added ), in percent (Equation 2).C found is calculated using the related calibration curve, which was obtained from results of analysis of calibration standard solutions by the proposed procedure.In other word, RR values which compare ER of real samples with standard solutions, confirm the agreement of analytical results of them.Linearity was investigated until concentration of 100 mg mL −1 of each alkaloid.b Percent relative standard deviations were calculated for three-replicate measurements of analytes with concentration of 0.1 mg L −1 .
The intra-day and inter-day precisions of the proposed method were used to evaluate the precision of the method.For this purpose, three similar experiments were carried out for the analysis of samples on the same day and on three consecutive days, respectively.The resulted RSDs for the analysis of intra-day and inter-day spiked samples were in the range of 0.9-3.1% and 1.4-4.4%,respectively.The summarised results are presented in Table 2.

Comparison with other methodologies
The comparison between the proposed method and other similar reported sample preconcentration methods for extraction and determination of opioid alkaloids are presented in Table 3.As can be seen the presented procedure shows good detection limits and wide linear dynamic ranges comparable with several reported methods.Moreover, it can be applied for simultaneous determination of five alkaloids and at the same time, the present method provides good results with minimum absorbent and consumption of   organic solvents.In addition, the developed method can be one of the simplest and fast methods for extraction and preconcentration of these analytes.

Conclusion
The present work provided a faithful and valid adsorbent constituted of magnetic Fe 3 O 4 NPs functionalised with MIPs.A series of efficient adsorbents were successfully synthesised through coating MNPs with appropriate MIPs.Various techniques systematically characterised the nanomaterials.A mixture of five prepared nanostructures was utilised as a magnetic D-micro-SPE adsorbent for the extraction of five selected alkaloids from water samples before their analysis by HPLC.The validation experiment results showed acceptable precisions in the linear ranges.The synthesised nano-sorbents can be considered as attractive and innovative candidates to substitute traditional DSPE adsorbents utilised for extracting target compounds from a complex matrix.The proposed method combined the excellent regeneration ability and selectivity of the magnetic MIP with the resolution and sensitivity of HPLC and appeared to be suitable for the determination and preconcentration of trace levels of these alkaloids.Through the proposed method, much lower quantities of the sorbent are required to achieve adequate extraction efficiency and, subsequently, a minimal amount of the organic solvents is needed which is one of the major principles of green chemistry.
The validation experiment results showed acceptable precisions in the linear range.According to the data obtained from this method, LOD values were 0.003-0.007µg/mL for the selected alkaloids.The developed method was applied to underground water and quantitative extraction recoveries were obtained in the range of 79-87% with a preconcentration factor of 200.

Figure 1 .
Figure 1.A schematic representation of the proposed method.
(b), the infrared spectra were acquired for Fe 3 O 4 , MAA@Fe 3 O 4 and MIP @ MAA@Fe 3 O 4 nanoparticles.The main functional groups of the predicted structures can be identified with corresponding infrared absorption bands.The main absorption band around 569 cm −1 corresponds to stretching vibration of Fe-O bond at Fe 3 O 4 NPs.An important absorption band around 1650 cm −1 corresponds to stretching vibration of the C = C bond in methacrylic acid which trapped in Fe 3 O 4 (FTIR spectrum of MAA@Fe 3 O 4 ).Also, Figure 3(c) shows absorption bands at 1735, 1257, 1155, 1456 and 1391 cm −1 which were assigned to C = O stretching vibration of carboxyl, symmetric and asymmetric stretching vibration of esteric C-O bond, olefinic C-H and aliphatic C-H bending vibrations in MMIP, respectively, confirm formation of MIP [29,30].

Figure 4 .
Figure 4. Effect of (a) pH of solution, (b) temperature, (c) contact time, and (d) dosage of MMIP nanoparticles on extraction of selected alkaloids.

Figure 5 .
Figure 5.Effect of type and volume of desorption solvent on recovery.(a) Effect of solvent type, V = 1 mL.(b) Effect of volume of selected solvent (methanol/acetic acid (9:1)).

Figure 6 .
Figure 6.Selectivity of MMIPs in extraction of target analytes from their mixtures with equal concentrations.MMIP weight = 25 mg, V sample = 100 mL, Sample containing 5 mgL −1 of each analyte.

Figure 7 .
Figure 7. Reusabilty of MMIPs for extraction of selected alkaloids up to 10 successive extraction cycles.
film modified-palladized aluminium electrode, b Multiwall carbon nanotubes, c Electromembrane extraction, d Dispersive liquid liquid microextraction-Solidified floating organic drop, e Chemiluminescence, f Magnetic molecularly imprinted solid phase extraction, g Surface plasmon resonance, f

Table 1 .
Analytical performance of MMIPs for determination of selected alkaloids in water samples.

Table 2 .
Determination selected alkaloids in underground waters (n = 3).Calculated from the related calibration curve which was obtained from results of analysis of extracted solutions from the standard solutions using the proposed extraction procedure.b Relative recovery is defined as ratio found/spiked concentration values, in percent.
c Not detected.

Table 3 .
Comparison of some procedures used in the determination of selected alkaloids in aqueous samples.