Accepted Manuscript

Magnetic solid phase extraction of trace paracetamol and caffeine

in synthetic urine and waste water samples by a using core shell
hybrid material consisting of graphene oxide/multiwalled carbon
nanotube/Fe3O4/SiO2

Halil İbrahim Ulusoy, Erkan Yılmaz, Mustafa Soylak

PII: S0026-265X(18)31160-3
DOI: https://doi.org/10.1016/j.microc.2018.11.056
Reference: MICROC 3501
To appear in: Microchemical Journal
Received date: 16 September 2018
Revised date: 23 November 2018
Accepted date: 28 November 2018

Please cite this article as: Halil İbrahim Ulusoy, Erkan Yılmaz, Mustafa Soylak , Magnetic
solid phase extraction of trace paracetamol and caffeine in synthetic urine and waste water
samples by a using core shell hybrid material consisting of graphene oxide/multiwalled
carbon nanotube/Fe3O4/SiO2. Microc (2018), https://doi.org/10.1016/
j.microc.2018.11.056

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Magnetic solid phase extraction of trace paracetamol and caffeine in

synthetic urine and waste water samples by a using core shell hybrid
material consisting of graphene oxide/multiwalled carbon
nanotube/Fe3O4/SiO2
Halil İbrahim Ulusoy 1,#, Erkan Yılmaz2, Mustafa Soylak3

1
Cumhuriyet University, Faculty of Pharmacy, Department of Analytical Chemistry, 58140, Sivas, Turkey

PT
2
Erciyes University, Faculty of Pharmacy, Department of Analytical Chemistry, , 38039 Kayseri, Turkey
3
Erciyes University, Faculty of Sciences, Department of Chemistry, 38039, Kayseri-Turkey

RI
A new hybrid material was synthesized including magnetic graphene oxide/multiwalled carbon

SC
nanotube core shell (GO/MWCNT/Fe3O4/SiO2) and used as magnetic solid phase extraction
adsorbent for separation and pre-concentration of paracetamol and caffeine subsequent
NU
detection by high performance liquid chromatography. The morphology, composition, and
properties of the synthesized hybrid material was characterized by Fourier transform infrared
spectrometry, Raman spectrometry, X-ray diffraction spectrometry, scanning electron
MA

microscopy, mapping photographs and BET surface area analysis. The experimental variables
on the extraction efficiency were studied and optimized. The optimum conditions for
D

quantitative extraction of paracetamol and caffeine were obtained as of adsorbent amount, 70

mg; pH, 8.0; and elution with 1 mL of methanol by vortex mixer for 60 sec. Validation
E

experiments showed that the developed method had good linear ranges 5-800 ng mL-1 for
PT

caffeine and 10-1000 ng mL-1 for paracetamol, respectively. The limits of detection were found
as 1.48 and 3.32 ng mL-1 for caffeine and paracetamol, respectively. The accuracy of the
CE

developed magnetic solid phase extraction method was proved by addition-recovery tests by
using model solutions and the developed method was successfully applied for the determination
AC

of paracetamol and caffeine in synthetically prepared urine and waste water samples.
Keywords: Hybrid magnetic material, magnetic solid phase extraction, high performance
liquid chromatography, Paracetamol, caffeine
Corresponding Author:
Halil İbrahim ULUSOY
E-mail: hiulusoy@yahoo.com
Address: Cumhuriyet University, Faculty of Pharmacy, Department of Analytical Chemistry,
58140, SIVAS, TURKEY
Tel: +90 346 219 10 10/3905 Fax:+90 346 219 16 34
 ACCEPTED MANUSCRIPT

1. Introduction

The mixtures of paracetamol and caffeine are currently used as analgesic and antipyretic agents

in many commercial formulations including tablets and capsules. Multidrug formulations

contain the different components for the therapy of pain of weaker genesis can improve the

pharmacological effects of these formulations and they sometimes play as synergists which lead

to a better effect [1–6].

PT
Up to now, different determination methods have been developed for directly analysis of

RI
paracetamol and its combinations in pharmaceuticals or in biological fluids, such as

SC
spectrofluorimetric, spectrophotometric, chromatographic, chemiluminescent and

electrochemical techniques [7–14]. Take into account the properties of the analytes
NU
investigated, such as mid polarity, as well as thermolability and low volatility, HPLC

determinations have been the most used [15–18].

MA

Pharmaceuticals contain a number of components and these components show adversely affect

the subsequent separation and identification of analytes of interest if they are not removed.
D

Thus, sample enrichment and purification techniques (sample preparation techniques) are
E

highly required prior to HPLC analysis.

PT

In recent years, magnetic materials, which prepared the combination of magnetic inorganic
CE

material and non-magnetic adsorbent material, has attracted considerable interests in material

chemistry because of their great physical and chemical properties and used as adsorbent in
AC

magnetic solid-phase extraction (MSPE) methods [19–22].

However, there are still some obstruction in the use of magnetic particles of MSPE. For

example, the magnetic particles could be easily detached from magnetic adsorbents and these

materials are not stable for a long time in solution conditions. Hence, it is difficult to completely

isolate the adsorbents from aqueous solution by magnet. In order to solve such problems, the
 ACCEPTED MANUSCRIPT

modification of magnetic particles with silica (formation of magnetic core shell) is a commonly

used method to synthesize magnetic sorbents [23–25].

When compared with the conventional solid phase extraction (SPE) techniques, MSPE has

unique advantages in separation techniques now. The using of SPE cartridge is removed in

MSPE and magnetic adsorbents universally dispersed in a sample solution for the achievement

extraction by vortex, ultrasonication or shaking. Hereby, the contact area between the sorbents

PT
and the analytes is sharply increased, which provide high extraction performance and fast mass

RI
transfer, even from large-volume samples. In MSPE methods, there is no need for extra

SC
apparatus and operations such as centrifugation or filtration require a long time, the magnetic

sorbent tagged with analytes is separated from the solution by using an external magnetic field
NU
which makes separation easier and faster. After completion of the extraction procedure, the

sorbent is regenerated and used for the further process [26–30].

MA

To the best of our knowledge, there is no any report for the simultaneous separation and pre-

concentration of paracetamol and caffeine. Therefore, this study is focused on development of

D

a simple and effective solid phase hybrid material including graphene oxide/multi-walled
E

carbon nanotube (GO/MWCNT/Fe3O4/SiO2) and application of magnetic solid phase extraction

PT

(MSPE) for paracetamol and caffeine in synthetically prepared urine and waste water samples.
CE

The most of published methods in literature for directly determination of these molecules in

pharmaceutical samples. As known, the concentration of drug molecules in this kind of samples
AC

is relatively high and this type analysis can be easily carried out by conventional methods. Most

of sensitive analysis in the complex samples (blood, urine) was carried out by expensive tools

such as HPLC-MS, LS-MS-MS, etc. So, the proposed method has important advantages

because it can determine the target molecules by conventional HPLC system and easy sample

preparation method.
 ACCEPTED MANUSCRIPT

2. Experimental Section

2.1.Reagents and standard solutions

All reagents used were of analytical grade. Ultra-pure water with a resistivity of 18.2 MΩ cm

was used in all experiments provided by Milli-Q system (Millipore, USA) water purification

system. Paracetamol was obtained from a drug company and Caffeine was purchased from

PT
Sigma (St. Louis, Steinheim, Germany), and methanol and isopropyl alcohol were from Sigma

(St. Louis, MO, USA). Graphite powder (Sigma, Co. Germany) was used for the synthesis of

RI
graphene oxide (GO). Multi-walled carbon nanotubes (MWCNTs) were purchased from Sigma

SC
Aldrich Co. (Germany).
NU
2.2.Fabrication of GO and GO/MWCNT/Fe3O4/SiO2

GO was synthesized from graphite powder by the modified Hummers method [28]. 0.25 g of
MA

the synthesized GO and 60 mg of MWCNT was dispersed into 150 mL of water and then

sonicated for 2 h in a ultrasonic bath to ensure the homogenous dispersion of GO and MWCNT
D

particles. The magnetic particles have obtained by using Fe3O4. In this synthesis procedure,
E

0.745 g of FeCl3·6H2O and 0.383 g of FeSO4·4H2O were added into the above mixture solution
PT

and these salts were dissolved into the above mixture under argon gas with vigorous stirring in
CE

a round bottom flask at 85 ℃. 20 mL of 30 % NH4OH was added dropwise under vigorous

stirring (pH ~12), a black color solution was resulted immediately and the mixture was stirred
AC

on magnetic stirrer to age the product under the argon gas. The product was magnetically

collected and washed with water/ethanol for several times and dried at 60 ℃ in an oven for 6 h.

The formation of core shell on the surface of GO/MWCNT/Fe3O4 material was conducted

according to the procedure previously reported [31]. First, 0.10 g of GO/MWCNT/Fe3O4

particles homogeneously dispersed in the mixture of ethanol (80 mL), deionized water (20 mL)

and concentrated ammonia aqueous solution (1.0 mL, 28 w/w %), followed by the addition of

tetraethyl orthosilicate (TEOS, 0.1 mL). After stirred for 6 h at room temperature, the
 ACCEPTED MANUSCRIPT

GO/MWCNT/Fe3O4/SiO2 obtained were magnetically separated and washed with

water/ethanol for several times and dried at in an oven for 6 h.

2.3. Instrumentation

A Perkin-Elmer Spectrum 400 FT-IR spectrometer (Waltham, MA, USA) was used for the FT-

IR spectra of the GO, and GO/MWCNT/Fe3O4/SiO2 materials. For Raman analyses of

MWCNT, GO and GO/MWCNT/Fe3O4/SiO2 materials, a Renishaw 1000 micro spectrometer

PT
using an excitation wavelength of 514.5 nm was used. The morphology and structures of GO,

RI
CNT and GO/ND/Fe3O4/SiO2 materials were investigated by scanning electron microscopy

SC
(LEO 440 SEM). X-ray diffraction (XRD) spectrums for GO, pristine MWCNT and

GO/ND/Fe3O4/SiO2 materials were recorded by using a Bruker AXS D8 advanced

NU
diffractometer. The chromatographic system used is equipped with a pump model LC20-AD

(Shimadzu, Tokyo, Japan), a thermostatic oven, CTO-10 AS (Shimadzu), auto sampler, SIL-
MA

20Ac (Shimadzu) and detectors: a DAD detector model SPD-M20A (Shimadzu). An LC

solution software (Shimadzu) was used to transfer data to the computer. A Inertsil C18 (150
D

mm×4,6×5 µm) column were used for chromatographic separation. A pH meter with a glass-
E
PT

calomel electrode (Selecta, Spain) was used to measure the pH values. An ultrasonic water bath

(Kudos, China) was used for sample preparation.

CE

Previous determinations, all solvents used in the chromatographic system were filtered through
AC

a 0.45 µm PDFA membrane filter (HNWP, Millipore) using a vacuum pump (Buchi,

Switzerland) and degassed for 10 min in an ultrasonic bath (JP Selecta, Barcelona, Spain).

2.4. Chromatographic analysis

The mobile phase composition was 0.01 M KH2PO4−methanol-acetonitrile-isopropyl alcohol

(840: 40: 60: 60) (v/v/v/v) at gradient mode throughout analysis. The flow rate was 0.8 mL

min−1. The detector wavelength was operated at 245 and 272 nm for paracetamol and caffeine,
 ACCEPTED MANUSCRIPT

respectively. and the column temperature was maintained at 40 ◦C. The volume of injection in

the auto sampler was 10 µL for all determinations.

The column was flushed with 50 % methanol to completely elute the remaining surfactants and

the other organic compounds for 60 min after each day’s work. Peaks in the chromatograms

were identified by comparison with retention times and UV spectra of standards. Peak area was

considered for quantification. All data were recorded using the above described

PT
chromatographic conditions.

RI
2.5. The proposed method

SC
10 mL of aliquot sample containing paracetamol and caffeine in the range of 5-800 and 10-

1000 ng mL-1 was placed in a screw-cap glass centrifuge tube including 70 mg of hybrid
NU
material. Then 2.0 mL of pH 8.0 buffer was added into tube and completed to 30 mL with water.

The extraction mixture was kept on an orbital shaker at 60 rpm for 30 sec. After analyte
MA

molecules were retained on magnetic nanocomposite, and the material was easily separated by

a niobium magnet and collected on side of tubes. The aqua phase was removed by a simple
D

pipette. Then, 1 mL of methanol was added and kept on vortex for 60 s. The samples were
E
PT

submitted to HPLC vials after they filtrated by a 0.45 µm filter. A calibration curve was created

by using a series standard solutions after the proposed method was applied at optimal
CE

conditions. The obtained chromatogram was shown in the Fig. 1.

AC

2.6. Sample Collection and Application of the proposed method

The application of proposed method was carried out by synthetically urine and wastewater

samples. Content of urine samples was prepared as mentioned in literature [32, 33]. 25.00 g urea,

1.08 g CaCl2·2H2O, 1.00 g NH4Cl, 1.60 g KCl, 1.40 g Na2SO4, 1.40 g KH2PO4, and 2.92 g NaCl were

dissolved in 1 L of ultra-pure water. The of synthetic urine solution was adjusted to 6.0 using NaOH

(0.1 M) or HCl (0.1 M). The standard synthetic urine was chosen to avoid contamination of

male and or female hormones. The mixture was stirred on a magnetic stirrer for 15 min and
 ACCEPTED MANUSCRIPT

kept in an ultrasonic water-bath. Then, the solution was diluted at 1:2 and 1:4 ratios. The

obtained solutions were stored in amber glass bottles until analysis.

Wastewater sample was obtained from main wastewater discharge line of university in Sivas,

Turkey. Wastewater samples were collected in amber glass bottles and immediately filtered

through 0.45 µm cellulose nitrate membrane. Subsequently, pH of samples were adjusted to 3

to reduce biological activity [34] and were stored in the dark at +4 °C until analysis.

PT
3. Results and Discussion

RI
3.1.Structure characterization

SC
The few layer planar sheets cause to GO easier to aggregate in water due to π–π stacking
NU
interactions between GO nanosheets, significantly decrease space between GO sheets and

further affect their adsorption efficiency. CNTs can act as between the layers of GO sheets due
MA

to their tubular structure. One-dimensional CNTs were introduced into GO to obtain a three-

dimensional hierarchical structure which could obstruct the stacking of GO sheets. CNTs attach
D

to GO sheets as spacer that effectively enlarged layer spacing and Fe3O4 particles are placed
E

these spaces.
PT

In this instance, the extraction efficiency of GO based materials increases. As a result, three
CE

dimensional new hybrid material consist of one dimensional MWCNT and two dimensional

GO has been obtained. In this way, a new network structure cause higher adsorption space and
AC

with higher adsorption capacity has been thus obtained. Both graphene oxide and multi walled

carbon nanotube are effective in adsorption. Surface of the GO/MWCNT/Fe3O4 material was

covered with SiO2 core shell structure. Addition of adsorption features of GO and MWCNTs,

the SiO2 coating shell has an abundance of surface hydroxyl groups which offers hydrogen

bonding between adsorbent and analytes such as paracetamol and caffeine.

The FT-IR spectrum of the GO and GO/MWCNT/Fe3O4/SiO2 materials were illustrated in the

(Fig. 2). It was shown that from FT-IR spectrum of GO, -OH (~ 3112 cm−1), C=O (~ 1713 cm−1)
 ACCEPTED MANUSCRIPT

and –C–O–C (1219 cm−1) functional groups in GO can be observed (Fig. 2A). For GO/

MWCNT/Fe3O4/SiO2 material, different shifts were observed in peaks and some differences at

peak shapes and intensities. The new peaks were obtained around 1036 cm−1 and 796 cm−1.

These bands are characteristic for Si-O and also a peak around 558 cm−1 is characteristic for

Fe-O. These new peaks indicate that SiO2 and Fe3O4 molecules are successfully formed on the

GO-MWCNT (Fig. 2B).

PT
Raman spectroscopy is a widely applied method for the characterization of carbon products,

RI
which has conjugated and double carbon-carbon bonds cause to high Raman intensities. The

SC
Raman spectrums of the MWCNTs, GO and GO/MWCNT/Fe3O4/SiO2 was shown in (Fig. 3).

A prominent G band peak at around 1570 cm−1 and a weak D band peak at around 1355 cm-1
NU
are specific peaks for the Raman spectrum of the pristine graphite, In the Raman spectrum of

GO, the D band grows in intensity, which is associated with the formation of GO and the G
MA

band broadens significantly and shows a shift to higher frequencies, (Fig. 3B). The Raman

spectrum of the pristine MWCNTs was fitted two specific peaks at 1330 cm-1 for D band and
D

1581 cm-1 for G band. The Raman spectrum of GO/MWCNT/Fe3O4/SiO2 confirmed that when
E

compared with Raman spectrum of MWCNTs and GO, significant differences were shown in
PT

Raman spectrum of the GO/MWCNT/Fe3O4/SiO2, Integrated area under the D band, with the
CE

ID/IG ratio going from 1.00 to 1.20, and the G band shifts from 1567 to 1572 cm-1.

Fig. 4 shows the XRD patterns of the GO and GO/MWCNT/Fe3O4/SiO2 materials. The XRD
AC

pattern of GO gives a reflection peak at 2θ=11.2°, corresponding to a d-space of 0.79 nm (Fig.

4A). There are four types of iron oxides commonly formed including magnetite (Fe3O4),
maghemite (γ-Fe2O3), hematite (α-Fe2O3) and goethite (FeO(OH)). XRD spectrums of GO and
GO/MWCNT/Fe3O4/SiO2 measured again. The XRD pattern of the GO/ MWCNT /Fe3O4/SiO2
material shows five main diffraction peaks at 2θ= 30 0, 360 and 580 that can be assigned to
Fe3O4, 2θ = 54.00 and 630 that can be assigned to α-Fe2O3 (Fig. 4B). It was seen from Fig. 4B
that the main iron oxide form is Fe3O4 due to intensities of peaks at (2θ = 30 0, 360 and 580).
The other iron oxide species are at impurity levels. The peak at 2θ =19° for GO/ND/Fe3O4/SiO2
proved the SiO2 formation.
 ACCEPTED MANUSCRIPT

The morphology of pristine MWCNT, as-obtained GO and GO/MWCNT/Fe3O4/SiO2 materials

were characterized by SEM, mapping photographs and was shown in Fig. 5A-5F. A large

amount of GO nano sheets were obtained (Fig. 5A, B), indicating the successful preparation of

GO by modified Hummers procedure. The creased and transparent GO sheets have exhibited

mono-or few layer planar sheet (Fig. 5A). The few layer planar sheets include many vacancies

and defects which allow the dispersion of MWCNTs and Fe3O4 particles (Fig 5A, B).

PT
In mapping photographs of GO/MWCNT/Fe3O4/SiO2, the iron (Fig. 5E) and silica particles

RI
(Fig. 5F) are uniformly dispersed in the GO/MWCNT matrix. SEM image of

SC
GO/ND/Fe3O4/SiO2 (Fig. 3D) and mapping photographs of GO/MWCNT/Fe3O4/SiO2 (Fig. 5E,

F), shows that GO/CNT surface was completely covered by Fe3O4 and SiO2 particles. Hence,
NU
the surface of GO/MWCNT/Fe3O4/SiO2 is seen different than that of GO and MWCNTs.

The BET surface area, pore size and pore volume for the GO/MWCNT/Fe3O4/SiO2 material
MA

was found to be 79.7 m2 g−1, 139.5 Å and 0.278 cm3 g-1, respectively. The BET isotherm result

of GO/MWCNT/Fe3O4/SiO2 material shows that the contribution of mesopores to the total

D

surface area and pore volume is significantly higher than that of macropores.
E
PT

3.2.Optimization of the MSPE Conditions

To achieve the optimal extraction efficiency of paracetamol and caffeine, the important
CE

parameters such as pH of the sample solution, amount of the sorbent, adsorption time, eluent
AC

type and volume, sample volume, salt concentration and buffer solution volume were

optimized.

3.3.Effect of pH and ionic strength

It is necessary to choose the optimal pH to achieve higher analyte adsorption. Hence, the

optimization of optimum pH should be carefully considered. The pH optimization was

performed in 10 mM phosphate matrix solution over the pH range of 2-10. As shown in Fig. 6,

the peak area reached a maximum at pH 8 for paracetamol and caffeine. At lower pH values,
 ACCEPTED MANUSCRIPT

the protonation of the functional groups of paracetamol and caffeine which may have cause

breakdown of the hydrogen bonds between analytes and adsorbent, and cause the lower

adsorption at the pH <8.0. But, at higher pH values than 8.0, downward trend for adsorption of

paracetamol and caffeine due to decrease of the hydrated surface of sorbent and formation of

colloidal particles [35, 36]. Hence, pH of 8.0 was selected for further studies.

To check the effect of ionic strength on the MSPE efficiency of analytes, various concentrations

PT
of NaCl in the range of 0-7 % (w/v) was studied in the presence of pH 8 buffer. It was found

RI
that the extraction efficiencies for paracetamol and caffeine were almost independent of the

SC
ionic strength.

3.4.Effect of the amount of the adsorbent

NU
For SPE methods, the amount of adsorbent is an important parameter, which plays effective

role in adsorption and desorption process [21, 22]. So, it must be optimized before real sample
MA

application. To monitor the effect of the amount of adsorbent on the adsorption of paracetamol

and caffeine, different amounts of GO/MWCNT/Fe3O4/SiO2 ranging from 10 to 150 mg were

D

added into the sample solutions and the proposed MSPE method was applied these mixtures.
E

The peak areas increased with increasing amount of the adsorbent from 10 to 70 mg and then
PT

stabilized for paracetamol with further increases, while a little decrease was observed in the
CE

peak area values for caffeine (Fig. 7). It can be explained as follows; In solid phase extraction

methods, as the amount of adsorbent used increases, desorption process of the analytes which
AC

are attached on the adsorbent becomes difficult and therefore it is necessary to use a higher

volume eluent. This results in a decrease in extraction efficiency and sensitivity. The obtained

results showed that 70 mg of GO/MWCNT/Fe3O4/SiO2 adsorbent was enough for extraction of

analytes from 50 mL of a sample solution. For further studies, 70 mg of the

GO/MWCNT/Fe3O4/SiO2 material was used for the complete extraction of the paracetamol and

caffeine.
 ACCEPTED MANUSCRIPT

3.5.Effect of time on adsorption of target molecules

Extraction of analytes from a aqueous phase to the solid phase is a equilibrium process. When

this equilibrium is reached between the two phases, the adsorption process is completed. For

this process, the aqueous phase including analytes should be subjected to interaction with the

solid phase (adsorbent) for a certain period of time in order to reach this equilibrium moment

of the extraction process. Shaker, vortex mixing and ultrasonic irridation are used to increase

PT
the interactions between aqueous and adsorbent, which lead to more simple extraction of

RI
analytes [27-30]. In our study, the effect of extraction time was checked by varying the

SC
absorption time between 5 and 90 min. by orbital shaker. The results were evaluated and the

maximum adsorption efficiency was obtained at 30 min. for paracetamol and caffeine.
NU
Therefore, an adsorption time of 30 min. was selected for the subsequent studies.

3.6.Effect of the desorption conditions

MA

In order to obtain the highest recoveries of paracetamol and caffeine, a series of eluents such as

water, methanol, ethanol, acetone, acetonitrile and isopropanol were used to optimize the
D

elution condition. The results were depicted in Fig. 8. As can be seen in the Fig. 8, the highest
E

peak areas were obtained by methanol as eluent. Hence, methanol was used as the optimal
PT

desorption solvent. The eluent volume is an important parameter to obtain higher pre-
CE

concentration factor and reduce organic solvent consumption. The effect of methanol volume

was also checked and results were shown in Fig. 9. The results showed that 1 mL of methanol
AC

can completely elute the adsorbed paracetamol and caffeine adsorbed from

GO/MWCNT/Fe3O4/SiO2. Hence, 1.0 mL of methanol was used for desorption of analytes from

adsorbent for next experiments.

3.7.Reusability and stability of adsorbent

Magnetic based sorbents have low chemical stability since Fe3O4 magnetic particles can be

damaged due to in harsh aqueus sample matrix and solvent used for elution. In order to
 ACCEPTED MANUSCRIPT

minimize this problem, modification of sorbents with a SiO2 core shell is a good way but not

the exact solution [31]. Hence, the reusability and stability test of magnetic sorbents is important

parameter. Finally, the reusability and stability of the GO/MWCNT/Fe3O4/SiO2 material was

checked by using model solutions including paracetamol and caffeine. The adsorption feature

of the GO/MWCNT/Fe3O4/SiO2 material was found to be apparently stable (less than 5 %) after

the repeated apply of more than 10 cycles of sorption and desorption of the caffeine and

PT
paracetamol.

RI
3.8.Analytical performance

SC
Under the optimum conditions, a series of analytical performance parameters with regard to the

linear range, correlation coefficient, limit of detection (LOD), limit of quantification (LOQ)
NU
and reproducibility were studied for the developed MSPE–HPLC procedure and the obtained

values were given in the Table 1. Linear regression analysis was conducted by using the peak
MA

areas against the concentrations of the caffeine and paracetamol, respectively. The limit of

detection (LOD) and limit of quantification (LOQ) were defined as 3 times and 10 times the
D

signal/slope of calibration curve, respectively. Pre-concentration factor (PF) is defined as the

E

ratio of the initial solution volume (30 mL) to the volume of eluent phase. The enhancement
PT

factor (EF) was calculated by using the ratio of the slope of calibration curve of the analytes
CE

after and before pre-concentration. The relative standard deviation (RSD) was calculated by

applying the developed method for seven repetition analysis, which include 250 µg L-1 of
AC

caffeine and paracetamol.

3.9.Application of the proposed method to synthetically urine and wastewater samples

Under optimized conditions, the developed MSPE-HPLC method was applied to the analysis

of caffeine and paracetamol by using model samples. For this purpose, the samples were

prepared as explained in sample preparation section and measurements were carried out as five

reparative analysis. Results were presented in Table 2. The recoveries were calculated by
 ACCEPTED MANUSCRIPT

comparing the pre-concentrated concentrations of analytes from those of the samples with the

corresponding added concentrations on calibration curves. Quantitative recoveries (between

92.6 and 108.6 %) with low relative standard deviation between 4.7 and 8.4 % were obtained.

The obtained results proved that the developed MSPE-HPLC method is the sensitive, precise

and accurate for the extraction and determination of caffeine and paracetamol in samples.

4. Conclusion

PT
To the best of our knowledge, there is no any report for the simultaneous separation and pre-

RI
concentration of paracetamol and caffeine. Hence, in this study, magnetic graphene

SC
oxide/multiwalled carbon nanotube core shell (GO/MWCNT/Fe3O4/SiO2) hybrid material was

successfully synthesized, characterized and subsequently used as a promising magnetic sorbent

NU
for MSPE of caffeine and paracetamol for the first time.
MA

Most of the published methods in literature describes [4,37–42] a direct determination method

of paracetamol and caffeine at higher concentrations (µg mL-1 level). More sensitive analysis
D

can be carried out only more expensive and complex tools including MS systems [43] or using
E

a pre-concentration technical [44]. As can be seen in table 3, the proposed method shows
PT

important analytical advantages. For instance, simultaneous analysis of paracetamol and

caffeine is possible as sensitive with ideal linear range for real samples. A conventional HPLC
CE

system and easily applicable pre-concentration method are enough for the proposed method.
AC

The methodology described is faster than classical pre-concentration methods such as SPE with

a few adsorbent requirements, minimum sample manipulation, lower chemical consumption,

and consequently minimum cost. Combined with HPLC, a rapid, efficient and simple method

for the determination of caffeine and paracetamol in model solutions was established

successfully. After GO/MWCNT/Fe3O4/SiO2 were reused and regenerated eight times, the

ninth recoveries of caffeine and paracetamol were still quantitative. Additionally, we believed
 ACCEPTED MANUSCRIPT

that the MSPE method is also a good alternative for combined with other spectroscopic and

chromatographic methods.

Acknowledge

This study has been supported by Cumhuriyet University Scientific Research Projects

Commission as the research project with the ECZ-021 code.

References

PT
[1] E. Dinç, A comparative study of the ratio spectra derivative spectrophotometry,

RI
Vierordt’s method and high-performance liquid chromatography applied to the

SC
simultaneous analysis of caffeine and paracetamol in tablets, J. Pharm. Biomed. Anal.

21(4) (1999) 723-730.

NU
[2] B.C. Lourenção, R.A. Medeiros, R.C. Rocha-Filho, L.H. Mazo, O. Fatibello-Filho,

Simultaneous voltammetric determination of paracetamol and caffeine in

MA

pharmaceutical formulations using a boron-doped diamond electrode, Talanta 78(3)

(2009) 748-752.
D

[3] C. Wang, C. Li, F. Wang, C. Wang, Covalent modification of glassy carbon electrode
E
PT

with L-cysteine for the determination of acetaminophen, Microchim. Acta 155(3-4)

(2006) 365-371.
CE

[4] J.T. Franeta, D. Agbaba, S. Eric, S. Pavkov, M. Aleksic, S. Vladimirov, HPLC assay of

acetylsalicylic acid, paracetamol, caffeine and phenobarbital in tablets, Farmaco 57(9)

AC

(2002) 709-713.

[5] E. Yilmaz, H.İ. Ulusoy, Ö. Demir, M. Soylak, A new magnetic nanodiamond/graphene

oxide hybrid (Fe<inf>3</inf>O<inf>4</inf>@ND@GO) material for pre-

concentration and sensitive determination of sildenafil in alleged herbal aphrodisiacs

by HPLC-DAD system, J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 1084

(2018) 113-121.
 ACCEPTED MANUSCRIPT

[6] M. Locatelli, N. Tinari, A. Grassadonia, A. Tartagli, D. Macerola, S. Piccolantonio, E.

Sperandio, C. D’Ovidio, S. Carradori, H.I. Ulusoy, K.G. Furton, A. Kabir, FPSE-

HPLC-DAD method for the quantification of anticancer drugs in human whole blood,

plasma, and urine, J. Chromatogr. B 1095 (2018) 204–213.

[7] M. Kartal, LC method for the analysis of paracetamol, caffeine and codeine phosphate

in pharmaceutical preparations, J. Pharm. Biomed. Anal. 26(5-6) (2001) 857-864.

PT
[8] A. Chafer, M.C. Pascual, A. Salvador, A. Berna, Supercritical fluid extraction and

RI
HPLC determination of relevant polyphenolic compounds in grape skin, J. Sep. Sci. 28

SC
(2005) 2050–2056.

[9] D. Martínez, M.J. Cugat, F. Borrull, M. Calull, Solid-phase extraction coupling to

NU
capillary electrophoresis with emphasis on environmental analysis, J. Chromatogr. A.

902 (2000) 65–89.

MA

[10] R.N. Goyal, S.P. Singh, Voltammetric determination of paracetamol at C60-modified

glassy carbon electrode, Electrochim. Acta 51(15) (2006) 3008-3012.

D

[11] E. Dinç, G. Kökdil, F. Onur, Derivative ratio spectra-zero crossing spectrophotometry

E

and LC method applied to the quantitative determination of paracetamol,

PT

propyphenazone and caffeine in ternary mixtures, J. Pharm. Biomed. Anal. 26(5-6)

CE

(2001) 769-778.

[12] V. Vichare, P. Mujgond, V. Tambe, S.N. Dhole, Simultaneous spectrophotometric

AC

determination of paracetamol and caffeine in tablet formulation, Int. J. PharmTech Res.

2(4) (2010) 2512-2516.

[13] N. Erk, Y. Özkan, E. Banolu, S.A. Özkan, Z. Zentürk, Simultaneous determination of

paracetamol and methocarbamol in tablets by ratio spectra derivative

spectrophotometry and LC, J. Pharm. Biomed. Anal. 24(3) (2001) 469-475.

[14] V. Pucci, R. Mandrioli, M.A. Raggi, S. Fanali, Reversed-phase capillary

 ACCEPTED MANUSCRIPT

electrochromatography for the simultaneous determination of acetylsalicylic acid,

paracetamol, and caffeine in analgesic tablets., Electrophoresis 4,5 (2004) 615-621.

[15] V. Lobo, A. Patil, A. Phatak, N. Chandra, Free radicals, antioxidants and functional

foods: Impact on human health, Pharmacogn. Rev. 4 (2010) 118-126.

[16] P. Mahesh, K. Swapnalee, M. Aruna, B. Anilchandra, S. Prashanti, Analytical method

development and validation of acetaminophen, caffeine, phenylephrine hydrochloride,

PT
dextromethorphan hydrobromide and chlorpheniramine maleate in tablet dosage form

RI
by RP-HPLC, International Journal of Pharmaceutical Science Invention 2(2) (2013) 9-

SC
15.

[17] M. Topkafa, H.F. Ayyildiz, F.N. Memon, H. Kara, New potential humic acid stationary
NU
phase toward drug components: Development of a chemometric-assisted RP-HPLC

method for the determination of paracetamol and caffeine in tablet formulations, J. Sep.
MA

Sci. 39(13) (2016) 2451-2458.

[18] V.D. Singh, S.J. Daharwal, Development and validation of multivariate calibration
D

methods for simultaneous estimation of Paracetamol, Enalapril maleate and

E

hydrochlorothiazide in pharmaceutical dosage form, Spectrochim. Acta - Part A Mol.

PT

Biomol. Spectrosc. 171 (2017) 369-375.

CE

[19] Q. Han, Z. Wang, J. Xia, S. Chen, X. Zhang, M. Ding, Facile and tunable fabrication of

Fe3O4/graphene oxide nanocomposites and their application in the magnetic solid-

AC

phase extraction of polycyclic aromatic hydrocarbons from environmental water

samples, Talanta 101 (2012) 388-395.

[20] J. Ding, Q. Gao, D. Luo, Z.G. Shi, Y.Q. Feng, n-Octadecylphosphonic acid grafted

mesoporous magnetic nanoparticle: Preparation, characterization, and application in

magnetic solid-phase extraction, J. Chromatogr. A. 1217(47) (2010) 7351-7358.

[21] Q. Gao, D. Luo, M. Bai, Z.-W. Chen, Y.-Q. Feng, Rapid Determination of Estrogens in
 ACCEPTED MANUSCRIPT

Milk Samples Based on Magnetite Nanoparticles/Polypyrrole Magnetic Solid-Phase

Extraction Coupled with Liquid Chromatography–Tandem Mass Spectrometry, J.

Agric. Food Chem. 59 (16) (2011) 8543–8549.

[22] M. Wierucka, M. Biziuk, Application of magnetic nanoparticles for magnetic solid-

phase extraction in preparing biological, environmental and food samples, TrAC -

Trends Anal. Chem. 59 (2014) 50-58.

PT
[23] Y. Moliner-Martinez, A. Ribera, E. Coronado, P. Campins-Falco, Preconcentration of

RI
emerging contaminants in environmental water samples by using silica supported

SC
Fe3O4 magnetic nanoparticles for improving mass detection in capillary liquid

chromatography, J. Chromatogr. A 1218 (16) (2011) 2276-2283.

NU
[24] J.H. Jang, H.B. Lim, Characterization and analytical application of surface modified

magnetic nanoparticles, Microchem. J. 94(2) (2010) 148-158.

MA

[25] L. Zhu, D. Pan, L. Ding, F. Tang, Q. Zhang, Q. Liu, S. Yao, Mixed hemimicelles SPE

based on CTAB-coated Fe3O4/SiO2 NPs for the determination of herbal bioactive

D

constituents from biological samples, Talanta 80(5) (2010) 1873-1880.

E

[26] I.S. Ibarra, J.A. Rodriguez, J.M. Miranda, M. Vega, E. Barrado, Magnetic solid phase
PT

extraction based on phenyl silica adsorbent for the determination of tetracyclines in

CE

milk samples by capillary electrophoresis, J. Chromatogr. A 1218 (16) (2011) 2196-

2202.
AC

[27] E. Yılmaz, M. Soylak, Preparation and characterization of magnetic carboxylated

nanodiamonds for vortex-assisted magnetic solid-phase extraction of ziram in food and

water samples, Talanta. 158 (2016) 152–158.

[28] M. Khan, E. Yilmaz, B. Sevinc, E. Sahmetlioglu, J. Shah, M.R. Jan, M. Soylak,

Preparation and characterization of magnetic allylamine modified graphene oxide-

poly(vinyl acetate-co-divinylbenzene) nanocomposite for vortex assisted magnetic

 ACCEPTED MANUSCRIPT

solid phase extraction of some metal ions, Talanta. 146 (2016) 130–137.

[29] E. Yilmaz, R.M. Alosmanov, M. Soylak, Magnetic solid phase extraction of lead(II)

and cadmium(II) on a magnetic phosphorus-containing polymer (M-PhCP) for their

microsampling flame atomic absorption spectrometric determinations, RSC Adv. 5

(2015) 33801–33808.

[30] L. Bai, B. Mei, Q.Z. Guo, Z.G. Shi, Y.Q. Feng, Magnetic solid-phase extraction of

PT
hydrophobic analytes in environmental samples by a surface hydrophilic carbon-

RI
ferromagnetic nanocomposite, J. Chromatogr. A 1217(47) (2010) 7331-7336.

SC
[31] W. Li, G. Li, Q. Lv, W. Xu, L. Liu, Chromate adsorption on amine-functionalized

core/shell magnetic mesoporous material, J. Porous Mater. 22(4) (2015) 965–972.

NU
[32] S. Cotillas, E. Lacasa, C. Sáez, P. Cañizares, M.A. Rodrigo, Electrolytic and electro-

irradiated technologies for the removal of chloramphenicol in synthetic urine with

MA

diamond anodes, Water Res. 128 (2018) 383-392.

[33] L.O. Paula, A.C. Sene, L.A. Manfroi, A.A. Vieira, M.A.R. Ramos, N.K. Fukumasu,
D

P.A. Radi, L. Vieira, Tribo-Corrosion and Corrosion Behaviour of Titanium Alloys

E

with and Without DLC Films Immersed in Synthetic Urine, J. Bio- Tribo-Corrosion.
PT

4(51) (2018) 50-55. doi:10.1007/s40735-018-0166-8.

CE

[34] E.M. Golet, C. Alder, Hartmann, T. Ternes, W. Giger, Trace determination of

fluoroquinolone antibacterial agents in urban wastewater by solid-phase extraction and

AC

liquid chromatography with fluorescence detection. Anal. Chem. 73(15) (2001) 3632-

3638. doi:10.1021/ac0015265.

[35] S. Azodi-Deilami, A.H. Najafabadi, E. Asadi, M. Abdouss, D. Kordestani, Magnetic

molecularly imprinted polymer nanoparticles for the solid-phase extraction of

paracetamol from plasma samples, followed its determination by HPLC. Microchim.

Acta 181 (15-16) (2014) 1823-1832.

 ACCEPTED MANUSCRIPT

[36] H. Parham, N. Rahbar, Solid phase extraction–spectrophotometric determination of

salicylic acid using magnetic iron oxide nanoparticles as extractor. J. Pharmaceut.

Biomed. 50(1) (2009) 58-63.

[37] A.P. Dewani, B.B. Barik, V.D. Chipade, R.L. Bakal, A. V. Chandewar, S.K. Kanungo,

RP-HPLC-DAD method for the determination of phenylepherine, paracetamol,

caffeine and chlorpheniramine in bulk and marketed formulation, Arab. J. Chem. 7(5)

PT
(2014) 811-816. doi:10.1016/j.arabjc.2012.07.010.

RI
[38] M. Sultan, Simultaneous hplc determination and validation of amphetamine,

SC
methamphetamine, caffeine, paracetamol and theophylline in illicit seized tablets, Int.

J. Pharm. Pharm. Sci. 6(4) (2014) 294-298.

NU
[39] R. Panchumarthy, R.A.O. Devala Garikapati, A. Akaula, S.B. Puttagunta, A validated

RP HPLC method for simultaneous estimation of paracetamol, diclofenac sodium and

MA

chlorzoxazone in combination tablet dosage form, J. Chem. Pharm. Sci. 6(3) (2013)

181-188.
D

[40] M.L. Altun, T. Ceyhan, M. Kartal, T. Atay, N. Özdemir, Ş. Cevheroǧlu, LC method for
E

the analysis of acetylsalicylic acid, caffeine and codeine phosphate in pharmaceutical

PT

preparations, J. Pharm. Biomed. Anal. 25(1) (2001) 93-101.

CE

[41] B.C. Lourenção, R.A. Medeiros, R.C. Rocha-Filho, L.H. Mazo, O. Fatibello-Filho,

Simultaneous voltammetric determination of paracetamol and caffeine in

AC

pharmaceutical formulations using a boron-doped diamond electrode, Talanta 78(3)

(2009) 748-752.

[42] D. Šatínský, I. Neto, P. Solich, H. Sklenářová, M. Conceição, B.S.M. Montenegro,

A.N. Araújo, Sequential injection chromatographic determination of paracetamol,

caffeine, and acetylsalicylic acid in pharmaceutical tablets, J. Sep. Sci. 7-8 (2004) 529-

536.
 ACCEPTED MANUSCRIPT

[43] H. Li, J. Wang, Y. Jiang, J.P. Fawcett, J. Gu, Simultaneous quantitation of paracetamol,

caffeine, pseudoephedrine, chlorpheniramine and cloperastine in human plasma by

liquid chromatography–tandem mass spectrometry, J. Pharm. Biomed. Anal. 51(3)

(2010) 716-722.

[44] S. Azodi-Deilami, A.H. Najafabadi, E. Asadi, M. Abdouss, D. Kordestani, Magnetic

molecularly imprinted polymer nanoparticles for the solid-phase extraction of

PT
paracetamol from plasma samples, followed its determination by HPLC, Microchim.

RI
Acta 15-16 (2014) 1823–1832.

SC
NU
MA
E D
PT
CE
AC
 ACCEPTED MANUSCRIPT

Figure 1. The obtained chromatogram after pre-concentration of trace analyte species

Figure 2. (A) FT-IR spectra of graphene oxide (GO), (B) FT-IR spectra of
GO/MWCNT/Fe3O4/SiO2 composite.

Figure 3. (A) Raman spectra of MWCNTs, (B) Raman spectra of GO, (C) Raman spectra of
GO/MWCNT/Fe3O4/SiO2 composite.

PT
Figure 4. (A) XRD spectrum of GO, (C) XRD spectrum of GO/MWCNT/Fe3O4/SiO2

RI
composite.

SC
Figure 5. (A, B) SEM image of GO, (C) SEM image of MWCNTs, (D) SEM image of
GO/MWCNT/Fe3O4/SiO2 composite, (E, F) mapping photographs of
NU
GO/MWCNT/Fe3O4/SiO2 composite.
MA

Figure 6. The effect of pH on the extraction efficiency of caffeine and paracetamol (N: 3).

Figure 7. The effect of the amount of the adsorbent on the extraction efficiency of caffeine and
D

paracetamol (N: 3).

E
PT

Figure 8. The effect of eluent type on the extraction efficiency of caffeine and paracetamol (N:
3).
CE

Figure 9. The effect of eluent volume on the extraction efficiency of caffeine and paracetamol
AC

(N: 3).
 ACCEPTED MANUSCRIPT

Table 1. Analytical merits of the proposed method.

Parameter Caffeine Paracetamol

Before SPE After SPE Before SPE After SPE

PT
Linear Range, ng mL-1 500-20000 5-800 1000-30000 10-1000

LODa, ng mL-1 149.3 1.5 340.0 3.3

RI
LOQb ng mL-1 492.5 4.9 1.0 0.9

SC
RSD, % 4.4 NU2.2 5.5 2.5

Calibration Sensitivity 78.9 2052.3 70.3 1124.9

R2 0.9924 0.9991 0.9972 0.9949

MA

PFc - 30.0 - 30.0

EFd - 26.1 - 16.0

D

a
Based on statistical 3Sblank/m-criterion for ten replicate blank absorbance measurements
E

b
Based on statistical 10 Sblank/m-criterion for ten replicate blank absorbance measurements
PT

c
Preconcentration factor is defined as the ratio of the initial solution volume (30 mL) to the volume of surfactant
rich phase 1.0 mL)
d
Enhancement Factor is defined as ratio of slope of calibration before and after CPE
CE
AC
 ACCEPTED MANUSCRIPT

Table 2. The amounts of caffeine and paracetamol in various samples after the proposed MSPE-HPLC method (N: 5)

Found b
Added RSD % Recovery %
ng mL-1
Samples a ng mL-1
P T
Paracetamol Caffein Paracetamol
I
Caffein

R
Paracetamol Caffein

Synthetic Urine Solution

(1:1 Urine solution)
50.0
100.0
46.3±3.4
93.5±6.5
48.6±2.9
95.6±3.3
S C
7.3
6.9
5.9
3.4
92.6
93.5
97.2
95.6

U
300.0 288.1±17.7 292.3±10.4 6.1 3.6 96.0 97.4

Synthetic Urine Solution

50.0
100.0
53.2±2.5
95.4±5.5
51.2±3.1
98.1±4.4
A N 4.7
5.8
6.0
4.4
106.4
95.4
102.4
98.4

M
(1:2 diluted urine solution)
300.0 310.5±18.1 306.8±16.2 5.9 5.2 103.5 102.3

Synthetic Urine Solution 50.0

E D
54.3±2.5 53.2±3.5 4.6 6.5 108.6 106.4
(1:4 diluted urine solution) 100.0

P
300.0 T 105.0±5.6
315.9±17.8
104.3±2.8
306.4±11.9
5.3
5.6
2.7
3.9
105.0
105.3
104.3
102.1

C E
50.0 46.9±3.7 46.8±2.4 8.4
6.7
5.1 93.8
94.9
43.6

a
Waste water Sample

A C 100.0
300.0
94.9 ±6.4
290.3±14.2
95.4±3.5
287.3±19.3 4.9
3.6
6.7 96.7
95.4
95.7
The preparation of samples was explained in the related section.
b
The average value of five replicates ± standard deviation

23
 ACCEPTED MANUSCRIPT

Table 3. Comparison of the presented procedure with literature

Target Samples Determination RSD% LOD Linear Range Applications Reference
Method
250-750 µg mL-1
T
Paracetamol, RP-HPLC-DAD
< 5.0 No data Bulk and marketed formulations [37]
Caffeine method 7.5-22.5 µg mL-1
Paracetamol
-1
0.39-100 µg mL-1
0.39-100 µg mL-1 I P
Illicit seized

R
Caffeine HPLC-UV < 3.0 0.18 µg mL [38]
tablets

SC
Paracetamol Pharmaceutical
HPLC-UV <1.7 10 ng/mL 100-600 μg mL-1 [39]
formulation
Paracetamol
Caffeine
Paracetamol
HPLC-UV
0.58-
2.18
9×10-5 -
1.7×10-4 M
N U
0.45-57.6 µg mL-1 Pharmaceutical tablets [4]

Caffeine
Paracetamol
Caffeine
HPLC-UV

Voltammetric
0.35

<3.0
0.075 μg mL-1

3.5×10-8 M
M A 0.4-1500 μg mL 1

5.0×10-7 -8.3×10-5 M
Pharmaceutical preparations

Pharmaceutical formulations
[40]

[41]

Paracetamol,
Caffeine
Sequential injection
chromatographic <1.82
E D
0.5 µg mL-1
0.3 µg mL-1
0.5–75 μg mL 1
0.5–50 μg mL 1
Pharmaceutical tablets [42]

PT
(SIC)
5.0–2000 ng mL-1
Human plasma LC–MS/MS <11.0 No data Human plasma [43]

Magnetic
C E 10–4000 ng mL-1

Paracetamol

A C
molecularly
imprinted polymer
nanoparticles
0.17 ng mL- 1-300 ng mL-1 Plasma samples [44]

(m-MIPs)/HPLC
3.30 ng mL-1 10-1000 ng mL-1
Paracetamol, MSPE Synthetic urine and waste
<2.50 1.50 ng mL-1 5-800 ng mL-1 This study
Caffeine HPLC-DAD water samples

24
 ACCEPTED MANUSCRIPT

Highlights
 A sensitive and simple method was developed for simultaneously determination of

caffeine and paracetamol.

 A new solid phase material (GO/MWCNT/Fe3O4/SiO2) was synthetized and

PT
characterized in detail.

 All experimental merits of new method were optimized and analytical merits of method

RI
was determined.

SC
 The proposed method was applied successfully to synthetically urine and waste water

samples.
NU
MA
E D
PT
CE
AC

25
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9