Microchimica Acta

, Volume 171, Issue 3, pp 423–429

Molecular imprinting polymer electrosensor based on gold nanoparticles for theophylline recognition and determination


    • College of Chemistry and Materials Science, Anhui Key Laboratory of Chemo-BiosensingAnhui Normal University
  • Tingting Liu
    • College of Chemistry and Materials Science, Anhui Key Laboratory of Chemo-BiosensingAnhui Normal University
  • Hong Zhou
    • College of Chemistry and Materials Science, Anhui Key Laboratory of Chemo-BiosensingAnhui Normal University
  • Chen Li
    • College of Chemistry and Materials Science, Anhui Key Laboratory of Chemo-BiosensingAnhui Normal University
  • Bin Fang
    • College of Chemistry and Materials Science, Anhui Key Laboratory of Chemo-BiosensingAnhui Normal University
Original Paper

DOI: 10.1007/s00604-010-0455-5

Cite this article as:
Kan, X., Liu, T., Zhou, H. et al. Microchim Acta (2010) 171: 423. doi:10.1007/s00604-010-0455-5


An electrochemical sensor for theophylline (ThPh) was prepared by electropolymerizing o-phenylenediamine on a glassy carbon electrode in the presence of ThPh via cyclic voltammetry, followed by deposition of gold nanoparticles using a potentiostatic method. The effects of pH, ratio between template molecule and monomer, number of cycles for electropolymerization, and of the solution for extraction were optimized. The current of the electro-active model system hexacyanoferrate(III) and hexacyanoferrate(IV) decreased linearly with successive addition of ThPh in the concentration range between 4.0 × 10−7 ~ 1.5 × 10−5 mol·L−1 and 2.4 × 10−4 ~ 3.4 × 10−3 mol·L−1, with a detection limit of 1.0 × 10−7 mol·L−1. The sensor has an excellent recognition capability for ThPh compared to structurally related molecules, can be regenerated and is stable.


In this paper, an electrochemical sensor for theophylline (ThPh) was prepared by electropolymerizing o-phenylenediamine (o-PD) on a glassy carbon electrode in the presence of ThPh via cyclic voltammetry, followed by deposition of gold nanoparticles to enhance the sensitivity of the sensor. Therefore, the sensor showed a high sensitivity for ThPh determining. Peak current of [Fe(CN)6]3−/[Fe(CN)6]4− varied linearly with the concentration of ThPh in the range of 4.0×10-7~1.5×10-5 mol·L-1 and 2.4×10-4~3.4×10-3 mol·L-1, and the detection limit reached 1.0×10-7 mol·L-1. Compared to structurally related molecules, the sensor also has a high recognition capability for ThPh. With excellent regeneration property and stability, the present sensor maybe provides a new class of polymer modified electrodes for sensor applications.


Molecular imprinted polymerElectrochemical sensorAu nanoparticlesTheophyllineRecognition


Theophylline (1, 3-dimethyl-1H-purine-2, 6-dione, ThPh), a xanthine derivative with diuretic, cardiac stimulant, and smooth muscle relaxant activities, has been widely used for the treatment of asthma and bronchospasm in adult [1, 2]. However, overdose of this kind of drugs would cause side effect, such as tremor tachycardia. Thus, the pharmaceutical preparations and their therapeutic drug monitoring are necessary for ThPh. The methods reported for the determination of ThPh include Surface plasmon resonance [3], gas chromatography-mass spectrometry [4], Liquid chromatographic [5], high-performance liquid chromatography [69], liquid chromatography-mass spectrometry [10], and solid phase extraction [11] and so on. The above mentioned motheds suffered disadvantages of long procedures and more other interferences of the similar structure molecules except as reported by Dove JW. etc. [3]. Different substrate materials such as ITO film [12, 13] and carboxylic multi-wall carbon nanotubes [14, 15] were also utilized to determination of ThPh. As an alternative, electrochemical sensors have been used to analyze inorganic ions, drugs, proteins, DNA, and other biomolecules due to their advantages of easy preparation, sensitive determination, and time saving [1619]. The electrochemical methods for determination of ThPh by square-wave voltammetry in tea and drug formulation at a Nafion®/lead-ruthenium oxide pyrochlore modified electrode had been reported [20]. As well known, ThPh is coexisted with caffeine and other purin alkaloids in real samples [21]. Therefore, many efforts have been employed to improve the selectivity of the modified electrodes.

One of the most promising materials in the field of artificial molecular recognition systems would be molecular imprinting polymer (MIP), which has increasingly attracted considerable attention in recent years. MIP is synthesized by copolymerizing functional and cross monomer in the presence of template molecule. Subsequent extraction of template molecules reveals binding sites in the polymer that are complementary in size and shape to the template molecule, resulting in the prepared MIP with the capabilities of specific rebinding and recognition to template molecule [2226]. Therefore, as recognition elements, MIP has been strongly developed in wide fields, such as liquid chromatography, capillary electrochromatography, solid phase extraction, drug controlled release, and electrochemical sensor [2733]. The recognition of ThPh also has been achieved by using molecular imprinted technique [14 15]. But these MIPs could not be used to determine the concentration of ThPh at the same time. The molecular imprinting technique and electrochemical sensor has been combined to develop the novel electrochemical sensor to achieve the determination and recognition simultaneously [3438].

Electrochemical method based on MIP for ThPh recognition and determination has been reported [3941]. In order to further improve the recognition property of sensor, herein, we reported a new MIP electrochemical sensor combined with Au-nanoparticles to improve the current response. In situ electropolymerization method was employed by using ThPh and o-phenylenediamine (o-PD) as template molecule and functional monomer, respectively. Au nanoparticles (AuNPs) were deposited on the above polymers film surface. The prepared electrochemical sensor was characterized by scanning electron microscopy (SEM), cyclic voltammetry (CV), electrochemical impedance spectrum (EIS), and differential pulse voltammetry (DPV). As expected, the electrochemical sensor exhibited high recognition capacity toward ThPh, as well as broader linear range and lower detection limit.



o-Phenylenediamine (o-PD) was purchased from Shanghai Chemical Reagent Company, China (http://www.reagent.com), and purified by sublimation before use. Theophylline (ThPh) and theophylline-7-acetic acid (ThPh-7) were purchased from Alfa Aesar (http://www.alfa.com). Guanine (GUAN) and adenine (ADEN) were obtained from Sanland-chem International Inc. HAuCl4·4H2O was obtained from Aaladdin Reagent Co., Ltd.(http://www.aladdin-reagent.com). All other reagents were of analytical grade and used without further purified. All solutions were prepared with triple distilled water.


All electrochemical measurements were performed on a CHI 660A electrochemical workstation (Chenhua Instruments Co., Shanghai, China) with a standard three-electrode configuration. A modified glassy carbon electrode (GCE) was used as the working electrode (3.0 mm diameter), a saturated calomel electrode (SCE) was used as the reference electrode and a platinum electrode was employed as the auxiliary electrode. The actual pH values were determined with a pH/Ion Analyser model pHS-3CT (Da Pu Instrument Co., Ltd. Shanghai, China). Scanning electron microscopy (SEM) was performed with Hitachi S-4800 SEM (operated at 10 kV).

Preparation of Au-nanoparticles and imprinted modified electrode

Au nanoparticles and molecular imprinted modified electrode (AuNPs/MIP/GCE) was constructed by using a three-step procedure, as shown in Scheme 1. The polymer modified electrode (ThPh-PoPD/GCE) was obtained by electropolymerizing of o-PD and ThPh on the surface of GCE using cyclic voltammetry in the potential range between 0 and +0.80 V during 20 cycles (scan rate: 50 mV·s−1) in 0.1 mol·L−1 HAc-NaAc buffer solution (pH 5.2) [42]. The concentrations of o-PD and ThPh were 5 × 10−3 mol·L−1 and 1.5 × 10−2 mol·L−1, respectively. Then potentiostatic deposition method was employed to prepare the Au nanoparticles modified ThPh-PoPD/GCE (AuNPs/ThPh-PoPD/GCE) by immersing of ThPh-PoPD/GCE into the 4 × 10−3 mol·L−1 HAuCl4 and 0.1 mol·L−1 KNO3 solution and applying a potential of −0.2 V during 180 s [43, 44]. After that the ThPh-PoPD/GCE and AuNPs/ThPh-PoPD/GCE were immersed in extraction solution under gentle stir to remove the template, obtaining MIP/GCE and AuNPs/MIP/GCE, respectively. Non molecular imprinting polymer modified electrode (NIP/GCE) and the Au nanoparticles modified NIP/GCE (AuNPs/NIP/GCE) were prepared under the same procedures except for the addition of ThPh.
Scheme 1

Scheme of fabrication of the AuNPs/MIP/GCE

Results and discussion

Characterization of the imprinted modified electrode

The SEM images of MIP/GCE and AuNPs/MIP/GCE surface were shown in Fig. 1. It was observed that the Au nanoparticles were distributed on the MIP film with the heterogenous diameter, which might be caused by the nonconducting polymer film.
Fig. 1

SEM imgaes of surface of ThPh-PoPD/GCE (a) and AuNPs/ThPh-PoPD/GCE (b)

Figure 2 showed the cyclic voltammograms of GCE (curve a), ThPh-PoPD/GCE (curve b), and AuNPs/ThPh-PoPD/GCE (curve c) recorded in [Fe(CN)6]3−/[Fe(CN)6]4− solution. No obvious oxide peak current could be found in curve b due to the nonconducting ThPh-PoPD film on the electrode surface. Compared to bare electrode, the peak current value of AuNPs/ThPh-PoPD/GCE increased remarkably, which would attributed to the enhancement of the conduction of electrode by deposited AuNPs.
Fig. 2

Cyclic voltammgrans for bare GCE (a), ThPh-PoPD/GCE (b), and AuNPs/ThPh-PoPD/GCE (c) in 0.01 mol·L−1 [Fe(CN)6]3−/[Fe(CN)6]4− solution

EIS is an effective method for probing the features of a surface-modified electrode. Figure 3 illustrated the Nyquist diagrams of GCE (curve a), MIP/GCE (curve b), and AuNPs/MIP/GCE (curve c) in the presence of 0.01 mol·L−1 [Fe(CN)6]3−/[Fe(CN)6]4−. As shown in curve a, there was a very low charge transfer resistance at bare GCE. After modifying with MIP, a semicircle of about 3300 Ω diameter was present, implying very high electron transfer resistance to the redox-probe and the block of charge transfer by MIP film. However, the diameter of the high frequency semicircle was obviously reduced to 120 Ω at AuNPs/MIP/GCE, suggesting that this modified electrode exhibited lower electron transfer resistance and greatly increases the electron transfer rate. The impedance change of the modified process indicated that MIP film and AuNPs had been modified to the GCE surface. After immersing the AuNPs/MIP/GCE in 2 × 10−4 mol·L−1 ThPh, the resistance substantially increased from 120 Ω to 370 Ω, which would be caused by rebinding of non-electroactive ThPh in imprinted cavities and blocking the arrival of probe [Fe(CN)6]3−/[Fe(CN)6]4− to electrode surface.
Fig. 3

Electrochemical impedance spectra of GCE (a), MIP/GCE(b), AuNPs/MIP/GCE (c) in 0.01 mol·L−1 [Fe(CN)6]3−/[Fe(CN)6]4−, and AuNPs/MIP/GCE in 0.01 mol·L−1 [Fe(CN)6]3−/[Fe(CN)6]4− containing 0.2 mmol·L−1 ThPh

Optimization of conditions for imprinted modified electrode preparation

In order to construct an efficient sensor, different influencing factors including pH value of electropolymerization solution, scan cycles of electropolymerization process, template molecule/monomer ratio, and extraction solution were investigated.

The effect of electropolymerization solution (0.1 mol·L−1 HAc-NaAc buffer) pH over the range of 4.0 ~ 6.7 on the current intensity of [Fe(CN)6]3−/[Fe(CN)6]4− on MIP/GCE was investigated. A maximum response was observed at about pH 5.2, indicating the formation of maximum imprinted cavities under this condition. Therefore, the electropolymerization was carried out at pH 5.2 HAc-NaAc solution.

The thickness of the polymer film would increase with the increase of scan cycles of electropolymerization, which would also affect the sensitivity of sensor. The number of scan cycles was varied from 10 to 40 in this research to determine the optimal film thickness. Polymer films that were formed less than 20 scan cycles were found to be unstable. Higher cycles lead to form the thicker sensing film with less accessible imprinted sites. The current response curves of [Fe(CN)6]3−/[Fe(CN)6]4− on MIP/GCE implied that the optimum polymerization cycles was to be 20.

To determine the effect of the ratio between ThPh and o-PD to the response of MIP/GCE, the MIP films were electropolymerized in solutions of constant o-PD (5 × 10−3 mol·L−1) concentration and varying ThPh concentrations in the range of 5–20 × 10−3 mol·L−1. After extracting the ThPh, the response of [Fe(CN)6]3−/[Fe(CN)6]4− to the modified electrode increased with the increase of the ratio of ThPh to o-PD from 1:1 to 3:1. However, when the ratio was increased to 4, the current response was almost negligible, which was possibly attributed to unstable MIP film on the electrode surface by using too little o-PD. Thus, the optimum ThPh/o-PD ratio was chosen as 3.

The removal of the template from the MIP is necessary to release the imprinted sites, which would selectively rebind the template molecule [45]. Citric-phosphate buffer (0.1 mol·L−1; pH 2.2, 3.0), HAc-NaAc buffer (0.2 mol·L−1; pH 3.0, 3.5, 4.0, 4.5), H2SO4 (0.5 mol·L−1), and ethanol were respectively applied to remove ThPh. The results indicated that ThPh could be removed almost completely by ethanol within 5 h, and the AuNPs/MIP/GCE could be regenerated for several times by this method. So soaking in ethanol for 5 h was chosen as the best condition for template removal.

Determination of theophylline

The dependence of the reduction peak current of [Fe(CN)6]3−/[Fe(CN)6]4− on ThPh concentration at AuNPs/MIP/GCE was investigated by using DPV methode. The results showed that the decrease of peak current of [Fe(CN)6]3−/[Fe(CN)6]4− was proportional to the concentration of ThPh in two ranges of 4.0 × 10−7 ~ 1.5 × 10−5 mol·L−1 and 2.4 × 10−4 ~ 3.4 × 10−3 mol·L−1 with correlation coefficient of 0.995, as shown in Fig. 4. The detection limit was determined to be 1.0 × 10−7 mol·L−1. On the contrary, the decrease of peak current of [Fe(CN)6]3−/[Fe(CN)6]4− was much lower and independent on ThPh concentration when AuNPs/NIP/GCE was used as working electrode, owing to no imprinting cavities in the polymer film.
Fig. 4

The plots of peak current versus ThPh concentration on AuNPs/MIP/GCE. Insets are the calibration curves of ThPh covering different concentration range: 4.0 × 10−7 ~ 1.5 × 10−5 mol·L−1 (a), and 2.4 × 10−4 ~ 3.4 × 10−3 mol·L−1 (b)

The results of different method and some other reported detection techniques for ThPh determination were summarized in Table 1. In comparison with chromatographic, spectrophotometry technique, and other electrochemical sensors based on MIP, our sensor had wider linear range or lower detection limit, indicating that the sensitivity of sensors could be improved by the deposited AuNPs.
Table 1

Comparison with other methods for the determination of ThPh


Calibration range

Detection limit

Selectivity factor


RNA biosensor

1.0 × 10−4 ~ 4.8 × 10−2 mol·L−1

2.0 × 10−4 mol·L−1



Surface plasmon resonance

0 ~ 3.3 × 10−2 mol·L−1

2.2 × 10−3 mol·L−1



Gas chromatography-mass spectrometry

1.1 × 10−6 ~ 5.6 × 10−5 mol·L−1



Reversed-phase high-performance liquid chromatographic

2.8 × 10−6 ~ 5.6 × 10−5 mol·L−1

1.4 × 10−6 mol·L−1



Ratio-spectra derivative spectrophotometry

1.1 × 10−4 ~ 1.0 × 10−3 mol·L−1

4.1 × 10−6 mol·L−1



Liquid chromatographic

2.7 × 10−6 mol·L−1



High-performance liquid chromatographic

1.1 × 10−5 ~ 2.8 × 10−5 mol·L−1

5.6 × 10−7 mol·L−1



5.0 × 10−5 ~ 1.7 × 10−4 mol·L−1



2.8 × 10−5 ~ 8.4 × 10−4 mol·L−1

3.3 × 10−6 mol·L−1



Liquid chromatography–mass spectrometry

4.25 × 10−5 mol·L−1



Solid phase extraction (MIP)

1.1 × 10−5 ~ 1.1 × 10−4 mol·L−1

5.6 × 10−6 mol·L−1



Electrochemial sensor

1 × 10−6 ~ 1 × 10−2 mol·L−1

5.5 × 10−7 mol·L−1



1.0 × 10−7 ~ 1.0 × 10−4 mol·L−1

1.0 × 10−7 mol·L−1



electrochemical sensor based on MIP

2.0 × 10−5 ~ 7.5 × 10−5 mol·L−1

3.0 × 10−9 mol·L−1





0 ~ 1.5 × 10−5 mol·L−1

1.0 × 10−6 mol·L−1








Our method

4.0 × 10−7 ~ 1.5 × 10−5 mol·L−1

1.0 × 10−7 mol·L−1



2.4 × 10−4 ~ 3.4 × 10−3 mol·L−1


Recognition of imprinted modified electrode

The special selectivity test of AuNPs/MIP/GCE was carried out by using Caffeine (CAFF), ThPh -7-acetic acid (ThPh -7), guanine (GUAN), and adenine (ADEN) as comparative compounds. It was also performed by DPV in [Fe(CN)6]3−/[Fe(CN)6]4−.

It was obvious that little difference of the current intensity subtraction on AuNPs/NIP/GCE could be observed before and after binding ThPh, CAFF, ThPh -7, GUAN, or ADEN into solution, as shown in Fig. 5. However, the obtained subtraction of the current intensity on AuNPs/MIP/GCE by adding ThPh was about 10 times as high as that on AuNPs/NIP/GCE, with 9, 13, 6, and 9 times as high as that of CAFF, GUAN, ADEN, and ThPh-7 on AuNPs/MIP/GCE, respectively. These differences indicated that binding capacity of AuNPs/MIP/GCE to ThPh was much higher than that to interferent. The results also further confirmed that AuNPs/MIP/GCE had an excellent selective recognition capacity toward template molecule due to the imprinting effect produced in the presence of ThPh.
Fig. 5

Selectivity of AuNPs/MIP/GCE

Regeneration and stability

After the first electrochemical determination of ThPh, AuNPs-MIP/GCE was immersed in ethanol to remove ThPh bounding in the polymer. AuNPs-MIP/GCE was then incubated in [Fe(CN)6]3−/[Fe(CN)6]4− solution containing the same concentration of ThPh for the next electrochemical measurement. The current of [Fe(CN)6]3−/[Fe(CN)6]4− decreased to about 95% of the original value after being used more than 10 binding/detection/extraction cycles. After a 15-day storage period at 4 °C in dry condition, the sensor retained 85% of its initial current response, which was similar to the result reported by Wang [41]. The results demonstrated that the prepared electrochemical sensor had excellent regeneration property and stability, which maybe provides a new class of polymer modified electrodes for sensor applications.

Serum analysis

In order to verify the reliability of the method, 3 volunteers’ blood was used for analysis. A 3 ml sample of human blood was centrifuged at 14000 rpm for 10 min, then 50 μL of the supernatant was injected into the test tube and diluted 1:9 (v/v) with phosphate buffer solution (0.1 mol·mL−1, pH 7.0). All the samples were determined by standard addition method. The recoveries were calculated from the corresponding calibration curve and the results are shown in Table 2. The results were in acceptable agreement and good reproducibility which suggested the sensor should be promising in clinic determination of ThPh.
Table 2

Analytical result and recovery


Theophylline (mol·L−1)

Recovery (%)


7.2 × 10−6



4.3 × 10−6



10.6 × 10−6



The present study describes the development of molecular imprinting polymer electrochemical sensor for special recognition and determination of ThPh. Sensor was fabricated by electropolymerization of o-PD in the presence of ThPh on GCE surface and electrodeposition Au NPs on MIP modified electrode surface successively. The prepared sensor not only exhibited a special recognition capacity to ThPh, but also had a high sensitivity for ThPh determining. Peak current of [Fe(CN)6]3−/[Fe(CN)6]4− varied linearly with the concentration of ThPh in the range of 4.0 × 10−7 ~ 1.5 × 10−5 mol·L−1 and 2.4 × 10−4 ~ 3.4 × 10−3 mol·L−1, and the detection limit reached 1.0 × 10−7 mol·L−1, which may be enhanced by the deposited AuNPs and could be used as a basis for ThPh determination.


We greatly appreciate the support of the National Natural Science Foundation of China for General program (20675001) and young program (21005002), Anhui University Provincial Natural Science Foundation Key program (KJ2010A138), Dr Start-up Fundation of Anhui Normal University (160-750834).

Copyright information

© Springer-Verlag 2010