A Study of Electrochemical Sensor Based on BHb-imprinted Magnetic ...

ANALYTICAL SCIENCES OCTOBER 2017, VOL. 33

1105

2017 © The Japan Society for Analytical Chemistry

A Study of Electrochemical Sensor Based on BHb-imprinted Magnetic Nanoparticles Yanxia LI,† Lu HUANG, Xiuping WANG, and Yiting CHEN† Department of Chemistry and Chemical Engineering, Minjiang University, Fuzhou, Fujian 350108, P. R. China

A study of an electrochemical sensor based on bovine hemoglobin (BHb) imprinted magnetic nanoparticles (MIPs) was investigated. First, BHb MIPs were successfully synthesized with magnetic Fe3O4@Au nanoparticles as carrier by surface modification of mercapto propionic acid for introducing carboxyl groups, combined with dopamine as the functional monomer and BHb as the template protein. Then, the MIPs were modified to the surface of a glassy carbon electrode (GCE) for electrochemical analysis by using cyclic voltammetry in the potassium ferricyanide solution. Results show that there was a good dynamic response between electrochemical signals and the adsorption amount of protein in the range of 0.05 to 0.5 mg mL–1. Results show that the sensor exhibits a significant specific recognition toward the template protein via selective test and can be used for analysis of serum samples. The synthesized MIPs are suitable for the removal and enrichment of template protein in proteomics. At the same time, the proposed electrochemical sensor can be used for recognition of BHb. Keywords Electrochemical sensor, molecular imprinting, magnetic nanoparticles, bovine hemoglobin (Received April 9, 2017; Accepted June 5, 2017; Published October 10, 2017)

Introduction Molecular recognition elements play an important role in all life processes, such as the ligand-receptor interaction, the immuno response, and enzyme catalysis.1–3 However, applications are significantly limited due to low stability and the high cost of biological elements.4 Therefore, the development of artificial receptors with an affinity and specificity approaching biological receptors is a fundamental challenge in molecular recognition.5–7 New strategies tested in this challenge could have an important long-term significance. Molecular imprinting is a powerful technology with the property of special recognition of the target molecule.8–10 As a new generation of artificial receptors, molecularly imprinted polymers (MIPs) are a kind of predesigned functional material of a tailor-made fashion. MIPs are capable of molecular recognition and other bio relevant functions and can be prepared by copolymerization in the presence of a crosslinker and a template molecule conjugated with a functional monomer.11,12 Compared with natural recognition counterparts like enzymes and antibodies, MIPs offer the significant advantages of low cost, easy preparation, structure predictability, mechanical, thermal and chemical stability, and highly selective recognition capabilities.13 To date, MIPs have found a wide range of applications in various fields, such as in separations, biosensors, catalysis, medical diagnostics and drug delivery.14–16 As a low-cost apparatus and direct analytical technique for high sensitivity, good selectivity and rapid identification of different molecules ranging from small molecules to biomacromolecules,17,18 electrochemical sensors are widely used To whom correspondence should be addressed. E-mail: [email protected] (Y. L.); [email protected] (Y. C.)

†

Electrochemical in molecular imprinting technology.19,20 polymerization is often used due to its simple operation and high sensitivity.21,22 However, low adsorption capacity and fewer polymeric monomers may limit the application of the sensor. Electrochemical sensors based on molecularly imprinted materials have gradually found unique applications.23 The sensors exhibit effective separation and enrichment of template molecules, and at the same time, possess highly sensitive recognition and detection. Sun et al. developed a fast and selective electrochemical sensor for determination of hemoglobin based on magnetic molecularly imprinted nanoparticles modified on the magnetic GCE with satisfactory results.24 In recent decades, artificial nanostructured materials have attracted increasing scientific interest due to some remarkable properties such as low cost, high surface-to-volume ratio and straightforward preparation. Such materials show high potential in highly selective recognition on chemical sensor fields.25 Magnetic nanoparticles (NPs) with a small dimension, unique magnetic responsivity, low toxicity and good biocompatibility have been widely applied to various fields, such as bioseparation, biosensor, and enzyme immobilization.26 For many of these applications, surface modification of magnetic NPs is a key challenge. A magnetic core coated with organic or inorganic materials can be accomplished by physical/chemical functionalization to introduce the required groups for specific applications.27 Combinations of Fe3O4 with Au NPs offer possibilities for the development of interesting advanced composite materials as a result of bringing together magnetic response, plasmonic properties, and biocompatibility.28 Furthermore, Au modification can also bring some other benefits, such as protecting the stability of Fe3O4 under harsh conditions, which is convenient for surface chemical modification of composite materials.29,30 In our previous study,

1106


Scheme 1 Construction route of electrochemical sensor.

we had synthesized a variety of functionalized Fe3O4@Au nanoparticles, which were successfully applied in molecular imprinting technology as a carrier with good results.31–33 The conductivity of nanoparticles can effectively promote the electron transfer on the surface of the electrode, which provided feasibility for electrochemical sensing based on molecularly imprinted materials. In the present work, we used BHb as the template, dopamine as the monomer, and functional Fe3O4@Au NPs as the carrier to prepare BHb imprinted magnetic nanoparticles. The MIPs were assembled to the surface of the GCE for electrochemical analysis by using cyclic voltammetry in the potassium ferricyanide solution. Owing to BHb effectively blocking the electron transfer on the electrode surface, there was a good dynamic response between electrochemical signals and the adsorption amount of protein. A kind of electrochemical sensor was constructed making full use of the selectivity of MIPs and high sensitivity of electrochemistry. The proposed sensor is not only used for recognition of BHb but also applied in the removal and enrichment of the template protein.

Experimental Materials Bovine hemoglobin (BHb), lysozyme (Lys), bovine serum albumin (BSA), cytochrome c (Cyt c), ovalbumin (OVA), horseradish peroxidase (HRP), and fetal bovine serum were purchased from Beijing Dingguo changsheng Biotechnology. Phosphate-buffered saline (PBS, 10 mmol L–1, pH 7.4), FeCl3·6H2O, FeCl2·4H2O, NH3·H2O (25 – 28%), sodium citrate, HCl, glacial ethanoic acid, gold chloride tetrahydrate (HAuCl4·4H2O), KCl, potassium ferricyanide [K3Fe(CN)6)], mercaptopropionic acid (MPA), and dopamine hydrochloride (DA·HCl, 98%) of analytical grade were received from Xinyuhua (Fuzhou, China). Milli-Q purified water was used for all experiments described here. Synthesis of functional Fe3O4@Au NPs The sodium citrate dispersed Fe3O4@Au NPs were prepared according to the procedure described in our previous work.33 The Fe3O4@Au NPs dispersed in 40 mL water were added to 0.5 mL MPA under stirring for 30 min, and washed several times by water, then dried at 60° C overnigh. A kind of functional Fe3O4@Au NPs modified with carboxyl group was successfully synthesized.

Synthesis of BHb-imprinted MIPs and NIPs The above functional Fe3O4@Au NPs were dispersed in 5 mL PBS followed by dissolved 15 mg of BHb. After pre-assembly for 1 h under stirring, 20 mg of dopamine was added to the aboved suspension, and kept stirring for 24 h at room temperature. The final precipitate was obtained by centrifuging, and washed thoroughly with water to remove the excess reactants. The precipitate was then incubated by acetic acid (10%, v/v) for 1 h, and washed thoroughly with water to remove remaining acetic acid. The MIPs were obtained and then dried at 60° C for 24 h. In comparison, the non-imprinted polymer coated Fe3O4@Au NPs (NIPs) were also prepared by the same procedure, only without using the template protein in the polymerization process. The saturated adsorption of MIPs with BHb are denoted as BHb-MIPs. Electrochemical analysis All electrochemical measurements were performed using the CHI 660D electrochemical workstation (Shanghai, China) with three-electrode system, where a self assembled glassy carbon electrode (GCE) (d = 3 mm) was used as a working electrode, an Ag/AgCl electrode as a reference electrode, and a platinum wire as a counter electrode. The bare GCE was polished with 0.05 mm alumina slurry followed by rinsing with purified water. The dispersed materials (3 mg mL–1) and chitosan solution (1.0 mg mL–1) were mixed in equal volume. Next, 5 μL of the mixture was dropped onto the surface of the GCE, then dried at room temperature. Electrodes modified with different kinds of materials were constructed and measured via cyclic voltammetry (CV: –0.2 – 0.6 V, 100 mV s–1) in 5.0 m mol L–1 ferricyanide aqueous solutions containing 0.1 mol L–1 KCl. Protein adsorption experiments For protein adsorption experiments, 3.0 mg MIPs or NIPs was immersed into 1.0 mL solutions of different BHb concentrations in PBS (10 mmol L–1, pH 7.4). The samples were incubated under slight stirring for 1 h. The precipitates were obtained by centrifuging, and washed three times with water to remove the unbound protein, and then were made to a constant volume of 1.0 mL with water. The effect of protein adsorbed on the MIPs or NIPs was determined by electrochemical analysis. The adsorption effect of protein is directly related to the peak current.


1107

Fig. 1 TEM images of Fe3O4@Au NPs (a), MIPs (b).

Results and Discussion Construction of electrochemical sensor The procedure for the construction of the electrochemical sensor is illustrated in Scheme 1. First, sodium citrate dispersed magnetic Fe3O4@Au NPs were synthesized. Magnetic Fe3O4 NPs are easy to magnetically separate; the wrapped aurum on the surface of Fe3O4 NPs, on the one hand, is to facilitate further surface modification, and on the other hand, is to protect the stability of Fe3O4 NPs in a harsh environment. Secondly, MPA modified Fe3O4@Au NPs are to introduce carboxyl groups, which can produce electrostatic interactions with the template protein. Finally, PDA thin films were coated onto the functional Fe3O4@Au NPs by a self polymerization process.34,35 Therefore, the carboxyl groups, amino-group and benzene-diol, have the effect of orientation imprinting on the template protein. After the MIPs formation, the adsorption effect of protein is directly related to the peak current in electrochemical measurement, which is due to the fact that most of the proteins can prevent electron transfer on the electrode surface. So a kind of electrochemical sensor based on BHb-imprinted magnetic In the process of sensor nanoparticles was obtained. construction, the synthesis of BHb-imprinted materials is carried out independently, combined with electrochemical analysis, which is easy to operate with high sensitivity. Accordingly, the synthesized MIPs are suitable for the removal and enrichment of template protein in proteomics. At the same time, the proposed electrochemical sensor can be used for recognition of BHb. Characterization of the synthesized nanoparticles The morphologies and structures of the magnetic nano materials were examined by TEM (Tecnai F30 G2 S-TWIN 300 kV). As shown in Fig. 1a, Fe3O4@Au NPs with an average diameter of 3 – 6 nm are well dispersed on the copper grid. From the TEM of MIPs (Fig. 1b), it can be seen that the diameter of 25 nm is obtained without considering the agglomeration phenomenon. The agglomerated particles may be derived from irregular polymerization of dopamine resulting in a small amount of particles with size exceeding 100 nm. Most particles exhibit uniformity. Therefore, it can be considered that a polydopamine layer with protein imprinting. The polydopamine layer should be calculated by a radius of about 10 nm. The size of a protein is usually about several nanometers. So the BHb should belong to the surface molecular imprinting, but does not rule out the possibility of BHb embedded in the polydopamine layer. Furthermore, free small size nano particles of Fe3O4@Au NPs were not found, which indicates that Fe3O4@Au NPs were successfully embedded in

Fig. 2 FTIR spectra of functional Fe3O4@Au NPs (a), MIPs (b).

the polymerization process. As shown in Fig. 2, FT-IR spectra of functional Fe3O4@Au NPs (a) and MIPs (b) were characterized by Fourier transform infrared (FT-IR) spectroscopy. FT-IR spectra of magnetic nano materials were recorded using the Nicolet 360 FT-IR spectrophotometer (Nicolet, USA). The peak around 3400 cm–1 was assigned to the –OH vibrations on the surface of magnetic nano materials (curves a, b). The absorption band at 1696 cm–1 corresponds to the carbonyl group of functionalized carboxyl groups, which means carboxyl groups were successfully modified on the surface of Fe3O4@Au NPs (curve a). And the peaks at 1411 to 1238 cm–1 were attributed to C–O stretching vibration and O–H in-plane deformation, respectively. After being coated with PDA, the characteristic peaks in curve a are covered and replaced by many new continuous peaks in curve b. The peak at 1648 cm–1 in curve b is attributed to the aromatic ring stretching vibration of polydopamine. The peak at 1260 cm–1 is attributed to C–N stretching vibration. Results confirm that a layer of molecularly imprinted polymer was successfully prepared. Electrochemical characteristics In order to verify the successful synthesis of MIPs, electrochemical analysis can provide useful information on peak current of MIPs and NIPs. Therefore, cyclic voltammetry (CV) was carried out at an NIPs electrode (Fig. 3a), an MIPs electrode (Fig. 3b) and a BHb-MIPs electrode (Fig. 3c) in 5.0 mM ferricyanide aqueous solutions containing 0.1 mol L–1 KCl. As we can see from Fig. 3a, the NIPs electrode shows the fastest electron-transfer kinetics of [Fe(CN)6]3–/4– compared with the MIPs electrode and BHb-MIPs electrode. The cavities of MIPs after removal of template molecules should promote the redox reaction of [Fe(CN)6]3–/4– on the electrode surface. But due to a small amount of protein residues in MIPs, which often hinder the electron transfer of electrode surface, the peak current of MIPs electrode slightly decreased compared with the NIPs electrode. A significant decreased peak current of BHb-MIPs electrode shows BHb imprinted to the MIPs. This further proved that a large number of BHb hindered the electron transfer on the surface of the BHb-MIPs electrode. In order to investigate the electrochemical behavior of magnetic nanomaterials, the MIPs electrode’s electrochemical response was also studied as a comparison. Figure 3d showed CV curve of the MIPs electrode in 0.1 mol L–1 KCl. It can be seen that the MIPs modified electrode itself does not produce an electrochemical signal.

1108


Fig. 3 Cyclic voltammograms of different modified electrodes in 5.0 mM ferricyanide aqueous solutions containing 0.1 mol L–1 KCl, (a) NIPs, (b) MIPs, (c) BHb-MIPs, (d) MIPs modified electrode in 0.1 mol L–1 KCl. Scan rate: 100 mV s–1.

On the other hand, it was also proven that the signal change of the nanomaterial modified electrode in potassium ferricyanide solution only came from the amount of adsorbed protein, and was not related to the nanomaterial under the same experimental conditions. Rebinding capacity The BHb imprinting effectiveness of the MIPs modified electrode was evaluated by the measurement of BHb rebinding capacities. The rebinding tests were performed at different initial concentrations of BHb in the range of 0 – 1.0 mg mL–1. In order to reflect the direct relationship between the adsorbed amount of protein and electrochemical response, the amount of the MIPs were equivalent to 3.0 mg mL–1. Unadsorbed protein is not involved in the electrochemical analysis through the removal of supernatant and preparing constant volume to 1.0 mL with water. So the electrochemical response depends strongly on the amount of rebound BHb into MIPs due to the electrochemical inertness of proteins. With the increasing of BHb concentration, more imprinted cavities are filled with template proteins leading to the decrease of peak current because most of the proteins can prevent electron transfer on the electrode surface. Figure 4 demonstrates CVs of the MIPs after rebinding a series of concentrations of modified electrodes in 5.0 mM potassium ferricyanide solution containing 0.1 M KCl. As we can see from Fig. 4A, the changed current is related to the concentration of BHb. Figure 4B shows the adsorption isotherms of the MIPs modified electrode at different equilibrium concentrations. From the isotherms, the redox peak current reduction decreases gradually with increasing concentration of BHb. When the initial concentration of BHb reached 0.5 mg mL–1, a saturation adsorption was achieved for MIPs. The MIPs modified electrode shows good dynamic response in the range of 0.05 to 0.5 mg mL–1. The results showed high rebinding capacity and wide dynamic range compared with common molecularly imprinted polymer materials and BHb imprinted electrochemical sensors indicating the proposed sensor can be used for identification of target protein. Rebinding selectivity To evaluate the selectivity of MIPs toward the template proteins of BHb, another five proteins with different isoelectric points (PI) and molecular weights (Mw), specifically Lys

Fig. 4 CVs (A) of MIPs modified electrode in 5.0 mmol L–1 potassium ferricyanide solution containing 0.1 mol L–1 KCl after MIPs rebinding for 60 min in different initial concentrations of BHb solutions. From outside to inside: 0, 0.05, 0.1, 0.2, 0.4, 0.5 mg mL–1. The adsorption isotherms (B) of CV reduction current for MIPs modified electrode changing with different equilibrium concentrations.

(14.4 kDa, PI = 11), BSA (66 kDa, PI = 4.8), Cyt c (12.4 kDa, PI = 10.2), OVA (45 kDa, PI = 10.7) and HRP (40 kDa, PI = 7.2), were chosen as contrast substrates. Figure 5 shows the variations of peak current after MIPs and NIPs’ capture for different proteins in a final concentration of 0.5 mg mL–1, and the imprinting factor (α) is calculated according to the binding ratio of MIPs to NIPs. Obviously, the MIPs clearly exhibited high selectivity for template protein BHb in comparison to Lys, BSA, Cyt c, OVA and HRP. However, the adsorption capacity of NIPs for BHb was quite close to the five competitive proteins. The α value for BHb of 3.9 is higher than those for the other proteins employed, where the α values for binding Lys, BSA, Cyt c, OVA and HRP were 1.1, 1.5, 1.5, and 1.0, respectively. The high adsorption selectivity and specificity of the MIPs toward the template protein can be attributed to the specific binding sites in the imprinted PDA layer. The specific interaction involves two roles, including multiple weak interactions provided by the synergistic effects of the multiple specific binding sites and shape complementarity. The amino and hydroxyl groups as well as π–π bonds of PDA can provide specific binding sites to form hydrogen bonding and electrostatic interaction with the template protein. However, the noncovalent interaction of NIPs was not strong enough for the selected proteins. The difference suggested that the specific recognition cavities complementary in shape, size, and functional groups with the template protein were formed in the MIPs.


1109

Fig. 6 Variations of peak current of MIPs and NIPs modified electrodes after MIPs and NIPs’ capture BHb in the 20-fold diluted fetal bovine serum. Fig. 5 Variations of peak current of different proteins on the MIPs and NIPs. Adsorption conditions: V, 1.0 mL; m, 3.0 mg; Ci, 0.5 mg mL–1; time, 100 min; temperature, RT.

Real sample analysis The MIPs were applied in a real bovine serum sample to further evaluate their applicability and separation effectiveness. The samples were investigated using electrochemical analysis after MIPs’ selective capture of BHb in the 20-fold diluted fetal bovine serum. Serum albumin is the main component of fetal bovine serum. In addition, it is important that fetal bovine serum contains a small amount of BHb, a large number of growth factors and a number of small molecules, such as amino acids, carbohydrates, lipids and hormones. Figure 6 shows the variations of peak current after MIPs and NIPs’ capture for these samples in a final concentration of 0.5 mg mL–1 BHb without and with fetal bovine serum. As we can see from Fig. 6, the peak current shows no significant difference after NIPs’ capture in three different solutions. However, MIPs showed specific adsorption leading to changes of peak current. It has a weak variation of peak current after MIPs’ capture in fetal bovine serum which shows fetal bovine serum contains a small amount of BHb. At the same time, MIPs showed a significant variation in the solution after adding 0.5 mg mL–1 BHb. The recovery is 79.86% after MIPs’ capture of 0.5 mg mL–1 BHb in fetal bovine serum. Results suggested that the MIPs had specific recognition and adsorption for the template protein BHb in the fetal bovine serum.

Conclusions In summary, a kind of electrochemical sensor was successfully developed based on modified electrode of BHb-imprinted nanomaterials. functional Fe3O4@Au NPs were chosen as the support material and DA as the monomer to fabricate BHbimprinted nanomaterials. The synthesis of MIPs is convenient, high-yielding and low-cost, and at the same time, the modified process of the electrode is simple and quick. The present electrochemical sensor showed significant selectivity, high binding concentration and applicability for serum sample analysis. The method can be used for separation, enrichment and recognition of the target protein. These features are encouraging for applications of protein capturing MIPs in electrochemical sensors, biotechnology and the life sciences.

Acknowledgements This project was financially supported by NSFC (21405075), Industrial Technology Key Project of Fujian Province (2014H0040), Fujian Province Natural Science Foundation (2017J01418, 2016J05040), Fujian Provincial Youth Natural Fund Key Project (JZ160468), the Talent Introduction Research Start-up Funds of Minjiang University (MJY14001), the Plan of College Youth Outstanding Research Talents in Fujian Province (2015), Program for New Century Excellent Talents of University in Fujian Province (2017), and Science and Technology Project of Minjiang University (MYK15001).

References 1. W. Liu, C. L. Cassano, X. Xu, and Z. H. Fan, Anal. Chem., 2016, 85, 10270. 2. P. Li, L. Wang, J. Zhu, Y. Wu, and W. Jiang, Biosens. Bioelectron., 2015, 72, 107. 3. A. Warshel and R. P. Bora, J. Chem. Phys., 2016, 144, 180901. 4. H. Q. Zhang, Polymer, 2014, 55, 699. 5. F. Yang, S. Lin, and X. Dong, Chem. Commun., 2015, 51, 7673. 6. R. Dutta and P. Ghosh, Chem. Commun., 2015, 51, 9070. 7. S. Dasgupta and P. S. Mukherjee, Org. Biomol. Chem., 2017, 15, 762 . 8. L. Chen, X. Wang, W. Lu, X. Wu, and J. Li, Chem. Soc. Rev., 2016, 45, 2137. 9. W. S. Lee and T. Takeuchi, Anal. Sci., 2005, 21, 1125. 10. H. Shiigi, M. Kishimoto, H. Yakabe, B. Deore, and T. Nagaoka, Anal. Sci., 2002, 18, 41. 11. T. Takeuchi, T. Hayashi, S. Ichikawa, A. Kaji, M. Masui, H. Matsumoto, and R. Sasao, Chromatogr., 2016, 37, 43. 12. L. Zhang, H. Zhang, L. Li, P. Zuo, F. L. Zhao, M. H. Liu, B. C. Ye, and Y. C. Li, Anal. Sci., 2016, 32, 923. 13. V. Kamble, P. A. Mahanwar, A. V. Rane, and V. K. Abitha, Chem. Int., 2015, 1, 202. 14. T. Greibrokk and B. Sellergren, J. Sep. Sci., 2015, 32, 3263. 15. G. Ertürk, M. Hedström, and B. Mattiasson, Biosens. Bioelectron., 2016, 86, 557. 16. W. Li and S. Li, Adv. Polym. Sci., 2006, 206, 191. 17. M. M. Barsan, M. E. Ghica, and C. M. Brett, Anal. Chim. Acta, 2015, 881, 1.

1110 18. J. N. Tiwari, V. Vij, K. Christian Kemp, and K. S. Kim, ACS Nano, 2015, 10, 46. 19. M. Yola and N. Atar, Curr. Anal. Chem., 2016, 13, 13. 20. M. Cieplak, K. Szwabinska, M. Sosnowska, B. K. Chandra, P. Borowicz, K. Noworyta, F. D’Souza, and W. Kutner, Biosens Bioelectron., 2015, 74, 960. 21. Z. Y. Shang, Y. L. Liu,Y. Liu, and Q. J. Song, J. Instrumental Anal., 2013, 32, 1401. 22. H. Yang, L. Li, Y. P. Ding, D. X. Ye, Y. Z. Wang, S. Q. Cui, and L. F. Liao, Biosens. Bioelectron., 2017, 92, 748. 23. B. B. Prasad and K. Singh, Sens. Actuators, B, 2017, 244, 167. 24. B. H. Sun, X. J. Ni, Y. H. Cao, and G. Q. Cao, Biosens. Bioelectron., 2017, 91, 354. 25. J. Wackerlig and P. A. Lieberzeit, Sens. Actuators, B, 2015, 207, 144. 26. W. Wan, Q. Han, X. Q. Zhang, Y. M. Xie, J. P. Sun, and M. Y. Ding, Chem. Commun., 2015, 51, 3541. 27. M. Zhang, X. Y. He, L. X. Chen, and Y. K. Zhang, J. Mater. Chem., 2010, 20, 10696.

ANALYTICAL SCIENCES OCTOBER 2017, VOL. 33 28. J. F. Ren, S. Shen, Z. Q. Pang, X. H. Lu, C. H. Deng, and X. G. Jiang, Chem. Commun., 2011, 47, 11692. 29. Y. Hu, J. Yang, P. Wei, J. C. Li, L. Ding, G. X. Zhang, X. Y. Shi, and M. Y. Shen, J. Mater. Chem. B, 2015, 3, 9098. 30. S. V. Salihov, Y. A. Ivanenkov, S. P. Krechetov, M. S. Veselov, N. V. Sviridenkova, A. G. Savchenko, N. L. Klyachko, Y. I. Golovin, N. V. Chufarova, E. K. Beloglazkina, and A. G. Majouga, J. Magn. Magn. Mater., 2015, 394, 173. 31. Y. X. Li, M. Hong, M. Miao, Q. Bin, Z. Y. Lin, Z. W. Cai, and G. N. Chen, J. Mater. Chem. B, 2013, 1, 1044. 32. Y. X. Li, Q. Bin, Z. Y. Lin, Y. T. Chen, H. H. Yang, Z. W. Cai, and G. N. Chen, Chem. Commun., 2015, 51, 202. 33. Y. X. Li, Y. T. Chen, L. Huang, B.Y. Lou, and G. N. Chen, Analyst, 2017, 142, 302. 34. X. P. Jia, M. L. Xu, Y. Z. Wang, D. Ran, S. Yang, and M. Zhang, Analyst, 2013, 138, 651. 35. R. X. Gao, L. L. Zhang, Y. Hao, X. H. Cui, D. C. Liu, M. Zhang, and Y. H. Tang, Talanta, 2015, 144, 1125.