Highly stable magnetite modified with chitosan, ferrocene and enzyme for application in magneto-switchable bioelectrocatalysis

ARTICLE

Highly stable magnetite modified with chitosan, ferrocene and enzyme for application in magneto-switchable bioelectrocatalysis

Antonio F. A. A. Melo^I; Roberto A. S. Luz^II; Rodrigo M. Iost^I; Iseli L. Nantes^II; Frank N. Crespilho^I,*

^IInstituto de Química de São Carlos, Universidade de São Paulo, 13560-970 São Carlos-SP, Brazil

^IICentro de Ciências Naturais e Humanas, Universidade Federal do ABC, Rua Santa Adélia, 166, Bairro Bangu, 09210-170 Santo André-SP, Brazil

ABSTRACT

Magnetic fields have been used in Bioelectrochemistry to carry enzymes or redox mediators immobilized on magnetite (Fe₃O₄) to the electrodes surface, providing a switchable control of faradaic current from biocatalysis. In this work, it is reported an advance in the magnetic control of bioelectrochemical reactions, by construction of a system containing simultaneously a magnetic particle (for controlled driving), an enzyme (for biocatalysis) and a redox mediator (for mediation of electron transfer). The advance was attained by synthesis of a new material (Fe₃O₄-Chi-Fc/GOx) that consists of Fe₃O₄ particles modified with insoluble ferrocene (Fc) and chitosan (Chi) cross-linked with glucose oxidase (GOx). When this material was used in electrochemical studies, an increase of 70% was observed in the catalytic current of glucose oxidation when 0.24 T was applied perpendicularly to electrode plane. This is the first time that a control of the bioelectrocatalytic process was achieved using enzyme, mediator and magnetite in a unique system switched by a magnetic field.

Keywords: magneto-switchable bioelectrocatalysis, magnetite, chitosan, ferrocene, glucose oxidase

RESUMO

O direcionamento por campo magnético tem sido aplicado em Bioeletroquímica, principalmente com intuito de transportar enzimas ou mediadores redox imobilizados em partículas de magnetita (Fe₃O₄) para a superfície de eletrodos, proporcionando o controle comutável das correntes faradaicas provenientes da biocatálise. Neste trabalho, relata-se um avanço no controle magnético de reações bioeletroquímicas por meio da construção de um sistema que contém simultaneamente partículas magnéticas (para transporte controlado), enzima redox (para biocatálise) e um mediador redox (para mediação da transferência eletrônica). Para isso, sintetizou-se um novo material que consiste em partículas de Fe₃O₄ modificadas com ferroceno insolúvel (Fc) e quitosana (Chi) entrecruzada com glicose oxidase (GOx). Quando esse material foi utilizado em suspensão em estudos eletroquímicos, observou-se um aumento de 70% da corrente catalítica de oxidação de glicose quando um campo de 0,24 T foi aplicado perpendicularmente ao plano do eletrodo. Esta é a primeira vez que um controle de processos bioeletrocatalíticos foi relatado utilizando enzima, mediador e magnetita em um sistema único controlado por campo magnético.

Introduction

The control of bioelectrochemical reactions using a magnetic field was first explored by Willner et al.¹ In that study, magnetite particles (Fe₃O₄) modified with N-(ferrocenylmethyl) aminohexanoic acid were employed to mediate biocatalysis of enzyme glucose oxidase (GOx). In a simple manner, two switchable modes, "switch on" and "switch off", were used in order to induce the electrochemical current from enzymatic reaction. Katz et al.² have reported the simultaneous control of biocatalytic reactions using two enzymes, GOx and lactate dehydrogenase (LDH). For the latter, PQQ-functionalized magnetic particles were attracted to electrode surface, leading to oxidation of lactate and the blocking the reaction catalyzed by GOx. In another study, Willner et al.³ reported magneto-switchable charge transport and bioelectrocatalysis using maghemite-Au core-shell nanoparticle/polyaniline composites. Dong et al.⁴ reported the capacity to modulate the electrochemical signal without the removal of the Fe₃O₄-enzyme biocomposite from the working electrode surface. In this condition, the biocomposite was oriented in alignment with the magnetic field lines resulting in a wire that could be oriented by the angle of magnetic field application. More recently, Liang et al.⁵ reported a magneto-controlled bioelectrocatalytic system for glucose oxidation, in which ferrocene was grafted to the thiol-terminated Fe₃O₄ nanoparticles via UV-induced thiol-ene click reaction.

In general, for magnetic control of bioelectrocatalytic process, some important points should be addressed: (i) ferromagnetic particles should exhibit colloidal stability in aqueous suspension, at least, during the time spent for the electrochemical experiments; (ii) it is not necessary to obtain magnetite in the form of nanoparticles, because enzyme diameter (e.g., dimeric-GOx) is around of nanometers, and the enzymes should be strongly immobilized on solid surface (magnetic microparticles); (iii) a mild microenvironment must be found at the surface of the solid particles on which the enzyme will be immobilized to avoid denaturation and loss of activity. This requirement can be satisfied by the use of appropriate methods for modification of the surface of the particles; (iv) the method of enzyme immobilization should guarantee no significant detachment of the protein during the bioelectrochemical experiments; (v) the magnetic particle should contain a redox mediator with high electrochemistry stability in order to provide conditions for a fast electron transfer.

As observed in the above mentioned points, several experimental requisites should be addressed to obtain an efficient system for magnetic control of bioelectrocatalytic processes. In the present work, it is reported a simple and effective method to obtain a new material (Fe₃O₄-Chi-Fc) from the physical adsorption between the redox mediator ferrocene (Fc) and the surface of Fe₃O₄ particles modified natural polymer chitosan (Chi). In order to demonstrate a possible application of Fe₃O₄-Chi-Fc, the oxidation of H₂O₂ from the glucose oxidation by using Fe₃O₄-Chi-Fc/GOx in the presence and absence of a magnetic field was evaluated. The best of our knowledge, this is the first time that an enzymatic reaction is achieved through magnetic particle, redox mediator and enzyme in a single system, and providing a fine control of the bioelectrocatalytic processes.

Experimental

Reagents and samples

All solutions were prepared from chemical of high purity and ultra-pure water (Milli-Q, 18.2 MΩ cm^-1). Ferric chloride 98% (anhydrous FeCl₃), ferrous chloride tetrahydrate 98% (FeCl₂.4H₂O) and ammonium hydroxide 98% (NH₄OH) were purchased from Vetec. Chitosan (Chi) with deacetylation degree of 85% was acquired from the company Polymar Nutrition Science S/A. The redox mediator Fc, acetic acid (glacial) 100%, glutaraldehyde 25% (GA), GOx (from Aspergillus niger), bovine serum albumin (BSA) and β-D-glucose, hexane 95% were purchased from Sigma-Aldrich. Sodium phosphate dihydrate (NaH₂PO₄) and sodium phosphate monohydrate (Na₂HPO₄) used in the preparation of the 0.1 mol L^-1 phosphate buffer solution (PBS, pH 7.0) were also purchased from Vetec.

Preparation of Fe₃O₄ particles modified with chitosan and ferrocene

Fe₃O₄ particles were prepared by co-precipitation method⁶ using a mixed aqueous solution of ferric chloride (0.1 mol L^-1) and tetrahydrous ferrous chloride (0.05 mol L^-1) in a 2:1 ratio, respectively. An excess of NH₄OH was added until pH 9.0 and a black precipitate was formed. During the addition of reagents, the system was kept in nitrogen atmosphere under magnetic stirring. Fe₃O₄ particles in suspension were decanted by magnetic separation and then washed with distilled water until the supernatant presented a pH 7.0. After washing, Fe₃O₄ particles were dried at a temperature of 60 ºC and macerated to produce a fine powder of homogeneous size. Respecting a stoichiometric ratio of 1:10 (Fe₃O₄:Chi), a powder mass of Fe₃O₄ particles was weighed and re-dispersed in an acid solution of chitosan (2% acetic acid) 1.0 g L^-1,⁷ producing Fe₃O₄-Chi. After 24 h, all Fe₃O₄-Chi in suspension was decanted by magnetic separation and then vacuum-dried. Once dry, Fe₃O₄-Chi was modified with Fc molecules by physical adsorption, in which a mass of Fe₃O₄-Chi (4.0 mg) was added to an excess of 0.1 mol L^-1 ferrocene dissolved in hexane. The solvent was fully removed by evaporation, producing Fe₃O₄-Chi-Fc. The new material obtained was washed with water distilled in order to remove non-adsorbed crystals of Fc and then vacuum-dried at room temperature (25 ºC).

Preparation of biocomposite Fe₃O₄-Chi-Fc/GOx

After the preparation of the Fe₃O₄-Chi-Fc, GOx was immobilized by crosslinking with GA (2.5%) in the presence of BSA.⁸ For this purpose, 10.0 mg of Fe₃O₄-Chi-Fc were dispersed in 2.0 mL of PBS 0.1 mol L^-1 (pH 7.0) containing BSA (15.0 mg mL^-1) and GOx (5.0 mg mL^-1). Then, it was added the solution of 110 µL of GA. This method provides a strong immobilization of GOx on the Fe₃O₄-Chi-Fc surface, since the GA acts as a bifunctional cross-linking agent in which bonds can be established between the amino groups present in the structure of the Chi and the mixture BSA/GOx. Before magneto-switchable studies, Fe₃O₄-Chi-Fc/GOx was stored for a period of 24 h at 4 ºC in order to ensure that all enzymes were bound to the surface of Fe₃O₄-Chi-Fc.

Characterization of Fe₃O₄ particles and modified-Fe₃O₄

The formation of Fe₃O₄ particles in suspension was monitored by UV-Vis spectroscopy using a Varian Cary 50 spectrophotometer. The measurements were performed in a quartz cuvette (1 cm of optical path). The crystal structure of Fe₃O₄ particles and modification of Fe₃O₄-Chi with Fc were evaluated by X-ray diffractometry (XRD) in a Bruker D8 Focus diffractometer with copper radiation (Cu K_α, λ = 0.15406 nm). The scan angle was understood in a range of 10º < 2θ < 70º with steps of 0.05º, measured at each 5 s. Fourier transform infrared (FTIR) spectra for powder of Fe₃O₄ particles, chitosan and Fe₃O₄-Chi were obtained in attenuated total reflectance mode (ATR) on ZnSe crystal using a Varian-600 FTIR spectrometer.

Magneto-switchable electrochemistry

Chronoamperometric and cyclic voltammetric experiments were carried out in a potentiostat coupled to a system of switchable magnetic field generation, with "switch on" and "switch off" modes, denominated MagnetoElectron (homemade equipment, Figure 1).

 

The MagnetoElectron equipment was fabricated with magnets of neodymium-iron-boron and all magnets were grouped and positioned perpendicular to the working electrode providing a maximum field of 0.24 T (0.5 cm from the center of the electrochemical cell). The measurement of magnetic field was carried out using a gaussimeter and then used in the electrochemical experiments. For all the electrochemical experiments, a conventional electrochemical cell configuration formed by three electrodes was used: ITO (glass coated by tin oxide doped with indium) electrode was utilized as working electrode (1.0 cm²), platinum as counter electrode (1.5 cm³) and Ag/AgCl_sat as reference electrode. For this purpose, a potentiostat/galvanostat µAutolab III was used.

It is noteworthy that during the electrochemical experiments, the two commutation states of the magnetic field modulator ("switch on" and "switch off" modes) were used in two configurations. In the configuration I, the "switch off" mode corresponds only to the suspension electrochemistry (absence of external magnetic), while in the "switch on" mode, all magnets are positioned behind work electrode. Different than what was proposed in the configuration I, the "switch off" mode used in the configuration II was employed when all magnets are positioned in front of the working electrode (inverse position). Thus, the two switched states ("switch on" and "switch off" modes) corresponds to a magnetic field of the same intensity (0.24 T).

Results and Discussion

Characterization of Fe₃O₄ particles and the Fe₃O₄-Chi-Fc

As described in the experimental section, the formation of Fe₃O₄ particles occurs according to the following equation:⁹

Electronic spectroscopy (UV-Vis) was used to evaluate the formation of Fe₃O₄ particles in suspension. Figure 2a shows the electronic spectra for 3 mmol L^-1 FeCl₃ solution (gray line) and for the suspension of Fe₃O₄ particles (black line). FeCl₃ solution is characterized by the presence of a band with maximum absorption at 300 nm (overlapped to the absorption band of FeCl₂, data not shown) attributed to metal to ligand charge transfer (MLCT) transition in aqua-iron complexes in solution. The disappearance of MLCT after the formation of Fe₃O₄ particles results from the removal of the aqua-iron complexes by hydrolysis of ferric ions promoted by NH₄OH. The spectrum of Fe₃O₄ suspension presents a broad featureless absorption tail typical of transition in the band gap of semiconductor materials.¹⁰ In this condition, the increase of the spectrum baseline is evident due to the light scattering effect¹¹ generated by colloids present in suspension. Figure 2b shows the diffractograms obtained for Fe₃O₄ particles (black line) and ICSD database (card No. 26410) attributed to the magnetite pattern (gray line). Diffraction peaks at 2θ (30.1, 35.5, 43.1, 53.75, 57.2 and 62.85) were consistent with the magnetite pattern, corresponding respectively to the crystal planes (2 2 0), (3 1 1), (4 0 0), (422), (5 1 1) and (4 4 0) of inverse spinel structure.¹²

For magneto-switchable studies, particles with micrometer size are attractive because of the large surface area offered by each particle for the immobilization of large biomolecules. According to Willner and co-workers,¹ Fc-functionalized magnetic microparticles offer a surface coverage of ca. 1000-3000 Fc units per particle, available to be used in magneto-switchable bioelectrocatalysis, as it will be described. To obtain the composite for GOx immobilization, the Fe₃O₄ particles were modified with Chi and Fc. The former modification consisted in the adsorption of Chi by 24 h on surface of suspended Fe₃O₄ particles in 2% acetic acid. The adsorption of Chi was characterized by FTIR in ATR mode. Figure 3a shows the FTIR spectra for Fe₃O₄ particles (black line), Chi (light gray line) and the Fe₃O₄-Chi (gray line). For the Fe₃O₄ particles, the presence of two peaks at 598 and 659 cm^-1 corresponding to the vibrations of the Fe-O and Fe-O-Fe bonds, respectively, was observed.

For Chi, the presence of peaks at 1650 cm^-1 (C=O stretch of amide I), 1558 cm^-1 (C-N stretching of amide II coupled the N-H deformation of primary amine) and 1317 cm^-1 (amide III) was observed and related to the acetamide groups of chitin deacetylation.^13,14 The Fe₃O₄-Chi spectrum shows the characteristic peaks of chitosan and a peak at 660 cm^-1 typical of magnetite. The interaction Chi with magnetite was evidenced by the shift to a lower wavenumber (1545 cm^-1) of the peak related to the N-H deformation of primary amine concomitant with the appearance of a peak at 1403 cm^-1 that is characteristic of this composite.^15,16 These results are consistent with the amino groups present in chitosan as the functional groups responsible for interacting with the surface of Fe₃O₄ particles.

Fe₃O₄-Chi was then modified with Fc using also physical adsorption process promoted by the solvent evaporation of Fc hexane solution. The obtained material was washed with distilled water several times in order to remove non-adsorbed Fc molecules. This method ensures that Fe₃O₄-Chi-Fc contains exclusively molecules of Fc strongly adsorbed on the surface of Fe₃O₄-Chi. Figure 3b shows the diffractograms obtained for Fe₃O₄ particles (black line), Fe₃O₄-Chi (light gray line) and Fe₃O₄-Chi-Fc (gray line). The same crystallographic profile was observed for Fe₃O₄ particles and Fe₃O₄-Chi indicating no change in the crystalline structure after the modification with Chi.^17,18 Furthermore, Fe₃O₄-Chi-Fc diffractogram shows peaks related to Fc (2θ from 10 to 20º ), indicating that the redox mediator was successfully adsorbed on the surface of Fe₃O₄-Chi. To verify whether Fc was effectively adsorbed on the Fe₃O₄-Chi, cyclic voltammetric experiments were performed using two suspensions: one containing 10 mg mL^-1 of Fe₃O₄-Chi (without Fc) and the other with 10 mg mL^-1 of Fe₃O₄-Chi-Fc. The suspension containing Fe₃O₄-Chi did not show faradaic process. Otherwise, the Fe₃O₄-Chi-Fc suspension exhibited two well-defined redox processes.

Magneto-switchable electrochemistry of Fe₃O₄-Chi-Fc

Magneto-switchable voltammetric experiments determined the electrochemical behavior of Fe₃O₄-Chi-Fc (10 mg mL^-1) in the presence and absence of an external magnetic field. For this set of experiments, configuration I was used, in which "switch off" mode corresponds to the electrochemistry of Fe₃O₄-Chi-Fc suspension at zero magnetic field. Figures 4a e 4b show the cyclic voltammograms obtained at different scan rates (10-100 mV s^-1) using "switch off" and "switch on" modes, respectively. At 100 mV s^-1 scan rate, the "switch on" mode increased 86.2 and 42.3% the current density values (3.64 and 1.63 µA cm^-2) for the anodic and cathodic currents, respectively. The external magnetic field attracts Fe₃O₄-Chi-Fc to the electrode surface in a condition of forced convection that results in a larger amount of Fe₃O₄-Chi-Fc deposited onto ITO surface. This leads to higher values of current densities. The optimization of Fe₃O₄-Chi-Fc deposition onto ITO surface was achieved by starting the cyclic voltammetry measurement in "switch on" mode one minute after the beginning of magnetic field application. This strategy ensures that all Fe₃O₄-Chi-Fc present in the suspension is conducted to be adsorbed on the electrode surface. The application of magnetic field for longer than one minute did not improve the yielding of current density, indicating that this time is enough for saturation of the electrode surface with Fe₃O₄-Chi-Fc. Thus, only a short time interval is required for electrochemical measurements in a set of successive magneto-switchable field of Fe₃O₄-Chi-Fc.

 

The quasi-reversible redox processes that are present in the voltammograms can be attributed to Fc, because Fe₃O₄ particles and Chi did not show electroactivity at the same potential range. Interestingly, the potential shifts obtained for a "switch on" and "switch off" set of commutative experiments were different. In the "switch on" mode, they were observed with E_pc = -0.16 V and E_pa = 0.32 V that results in ΔE_{switch on} = 0.47 V at 100 mV s^-1. In the "switch off" mode, they were observed with E_pc = -0.18 V and E_pa = 0.39 V, that results in ΔE_{switch off} = 0.57 V. Considering ΔE_{switch on} lower than ΔE_{switch off}, one can assume that the proximity of the Fc/Fc⁺ redox pair at the electrode surface affects the electrochemical reversibility and, consequently, the charge-transfer energy. This assumption could be corroborated by comparing the ΔE_{switch on} and ΔE_{switch off} values obtained at all scan rates, as shown in Figure 4c. In the whole scan rate range used here (from 10 to 100 mV s^-1), the ΔE_{switch on} values were always lower than ΔE_{switch off} values. While the result of reversibility affects the kinetic of the redox couple, no catalytic effect was observed for "switch on" case, as shown in Figure 4d.

The application of magnetic field ("switch on" mode) affects not only the thermodynamic properties of Fe₃O₄-Chi-Fc but also the kinetics of charge transport of Fc. The insets of Figures 4a and 4b show a linear increase of the anodic and cathodic peak currents as function of the square root of the scan rate consistent with electrochemical reactions limited by diffusion. In "switch off" mode, some factors influence the diffusion of the Fe₃O₄-Chi-Fc. One effect to be considered is the gravity force acting perpendicularly to the diffusion plane towards the electrode surface (Fick's law). In this regard, precipitation of microparticles was visible only after 15-20 min of suspension at rest. This time is significantly higher than that necessary for a voltammogram measurement and enabled us to disregard the precipitation process for electrochemical results. In the "switch on" mode, with the deposition of Fe₃O₄-Chi-Fc on the electrode surface, a linear increase of the faradaic current is expected with increasing of scan rate, typical of a process limited by charge transfer. However, a hyperbolic profile was also observed in the "switch on" mode.

It is possible that the limited diffusional process of redox-modified magnetite adsorbed on the electrode surface may related to the interface structure or/and migration of ions. For instance, considering that after Fe₃O₄-Chi-Fc deposition on the electrode surface, there is the formation of an interface resembling a thin film-electrode system, the migration of ions in the internal plane of electrical double layer (inner Helmholtz plane) should occur efficiently in order to maintain the electro-neutrality on the electrode surface after oxidation of Fc with formation of species Fe₃O₄-Chi-Fc⁺. The inner Helmholtz plane is the area where charges not solvated and specifically adsorbed are located and it is known that the adsorption of non-charged species can block the electrode surface and prevent the efficient electron transfer. Thus, it is possible that the non electroactive components of Fe₃O₄-Chi-Fc (Chi and Fe₃O₄ particles) may be contributing for the impairment of the electron transfer process. This phenomenon can also be discussed in terms of possible deviations of the Cottrell equation for "switch on" mode that describes the change in current in chronoamperometry, as shown in Figure 5.

In this condition, the current measured for Fe₃O₄-Chi-Fc depends on the rate at which the electron diffuses to electrode surface. The theoretical electron diffusion constant (D_ct), was determined for the "film" of Fe₃O₄-Chi-Fc deposited on electrode; here, it was considered film-like the Fe₃O₄-Chi-Fc deposited in "switch on" mode. The Cottrell equation for an ITO electrode that has planar geometry and boundary conditions in conjunction with Fick's second law of diffusion was considered. The smoothly decaying current-time curve was obtained when plotted according to the Cottrell equation for linear diffusion (i vs. t^-1/2) for 600 s experiment time (equation 2).

In equation 2, D_ct is the electron diffusion constant, c_i is the concentration of electroactive ferrocene sites in the "film" (Fe₃O₄-Chi-Fc adsorbed), F is the Faraday constant, A is the electrode surface area and t is the time, respectively. Figure 5a shows the simulated (black line) and experimental (gray line) chronoamperometry for "switch on" mode during 600 s. In order to determine the amount of Fc physically adsorbed onto Fe₃O₄ particles, the charge (Q = ne) was determined from the integration of experimental chronoamperogram, with lim(t).¹⁹ First, the determination of the adjusted curve leads to the determination of Fc charge and the proportion related to the Fe₃O₄ mass used for the electrochemical experiments. For Fc species was attributed a charge value of 53.4 µC cm^-2 and an amount of 5.5 × 10^-7 mol was determined, which corresponds to 0.103 mg of Fc in 10.0 mg of Fe₃O₄-Chi-Fc (1.0% of Fc in mass adsorbed on Fe₃O₄-Chi). The modeling of the obtained chronoamperometry curves was also used to determine the theoretical value of D_ct = 4.7 × 10^-6 cm s^-1; by using equation 2. However, as observed in the chronoamperometric experiments shown in Figure 5a, there is a deviation from theoretical chronoamperogram. The deviation is more clearly seen in the plot i vs. t^-1/2 (Figure 5b) with a rapid growth curve that reaches a finite diffusion up to t^-1/2 = 0.2 s (black curve).

This result can be associated with the concentration-distance profiles of Fc⁺ sites within Fe₃O₄-Chi. The incorporation of Fe₃O₄ particles and Chi organic polymer can be the cause of the electron transfer resistance, as observed in previously reported studies using copolymer film at different times during the potential step electrolysis.²⁰ For the latter, Nakahama and Murray²⁰ showed the effect of composition of a ferrocene-containing redox polymer on the electrochemistry of its thin film coatings on electrodes. In principle, for Fe₃O₄-Chi-Fc/solution interface, the diffusion of electrons should follow infinite diffusion laws, e.g., the Cottrell. At longer times, the "supply" of ferricenium sites becomes limited by the finite thickness of the non-redox species, and consequently the current falls more rapidly than anticipated from equation 2. This is called finite diffusion, and the current-time relation in this regime is ideally represented by an equation identical to that employed for thin-layer electrochemistry.²⁰ Thus, this result can be consistent with the case of "switch on" mode, with the deposition of Fe₃O₄-Chi-Fc on the surface, in which a linear increase of the faradaic current with increasing of scan rate was not observed (Figure 4), indicating a system limited by diffusion process.

High electrochemical stability of Fe₃O₄-Chi-Fc

After evaluating the electrochemical behavior of Fe₃O₄-Chi-Fc in the presence and absence of an external magnetic field, by chronoamperometry (configuration II), a more refined control of the redox processes was observed (Figure 6). As illustrated in Figure 1b, "switch off" mode consists in the total removal of redox species from the electrode surface by the application of the magnetic field contrary to the plane of the electrode. For chronoamperometric experiment, a potential of 0.34 V was applied because Fc is in reduced form in Fe₃O₄-Chi-Fc. Similarly to the voltammetric experiments, in the "switch on" mode there was an increase in oxidation currents due to Fe₃O₄-Chi-Fc deposited onto ITO surface. During 50 s, when the quasi-stationary current was reached, an increase of current density of 1.0 µA cm^-2 was observed (Figure 6a).

In order to check the electrochemical stability of Fe₃O₄-Chi-Fc, the "switch on" and "switch off" modes were applied successively, and no significant loss of current was observed. The same experiments were reproduced more than 10 times with different electrodes and the chronoamperometric currents remained unchanged, a result consistent with high stability and reproducibility of the Fe₃O₄-Chi-Fc. In addition to the several cycles of "switch on" and "switch off" modes, the long term stability of this system was examined.

Figure 6b shows chronoamperograms obtained at different intervals, such as 1 h (black line), 5 h (gray line), 12 h (light gray line) and 24 h (dashed line). A stable system was clearly shown with more than 50% of current density values maintained even after several switching cycles collected after 24 h of operation. The decrease in current density was observed only for the chronoamperometric experiments performed in the absence of the enzyme glucose oxidase (GOx). This decrease is related to the consumption of Fc due the application of an oxidation potential on the ITO electrode (E = 0.34 V). Thus, at each switching cycle performed, more species are reacting. Also, it is important to mention that the same suspension of Fe₃O₄-Chi-Fc (0.103 mg Fc) was used in all electrochemical experiments described in this work, supporting the rationale described above.

Magneto-switchable bioelectrocatalysis

In order to show an application for Fe₃O₄-Chi-Fc, the catalytic activity of GOx when immobilized on Fe₃O₄-Chi-Fc was evaluated. For this purpose, Fe₃O₄-Chi-Fc/GOx was subjected to magneto-switchable bioelectrocatalysis. GOx is an enzyme-model commonly used in the fabrication of amperometric biosensors and also in bioelectrocatalysis studies.^21-24 Among all studies reported in the literature^1-5 about magneto-switchable bioelectrocatalysis, there is no one describing the immobilization of redox enzymes attached directly to magnetic particles modified with redox mediator. For instance, Willner and co-workers^1-3 reported the use of GOx dispersed in aqueous solution and exclusively the redox mediator was immobilized on surface of Fe₃O₄ particles. Dong and co-workers⁴ used the GOx immobilized on the surface of Fe₃O₄ particles while the mediator molecules are dissolved in the electrolyte.

From the best of our knowledge and differently of all other previous studies, for the first time it was used both GOx and Fc immobilized on the surface of Fe₃O₄-Chi particles. The objective here is the construction of a system containing the three components together: magnetic particle, mediator and enzyme. For the latter component, the amino groups of Chi provide a mild environment for enzyme immobilization. Also, as described in experimental section, cross-linking procedure utilizing GA molecules in the presence of BSA provides an effective immobilization of GOx on the Fe₃O₄-Chi-Fc surface. The latter can be observed in Figure 7a, in which the system Fe₃O₄-Chi-Fc/GOx is easily attracted by the magnetic field. Figure 7b shows the response of Fe₃O₄-Chi-Fc/GOx in the absence of glucose (black line) and in the presence of glucose (20 mmol L^-1) using a magnetic field modulator in "switch off" (light gray line) and "switch on" (gray line) modes (configuration I).

In the absence of glucose, a redox process related to oxi-reduction of Fc mediator was observed. In the presence of glucose, the oxidation current of H₂O₂ was observed in the two switchable modes. The oxidation of glucose in the presence of Fe₃O₄-Chi-Fc/GOx can be represented by equations 3, 4 and 5.

The oxidation currents attain the maximum around 10.7 µA cm^-2 at 0.8 V for "switch on" mode, leading to an increase of 3.2 µA cm^-2 when compared with "switch off" mode. In fact, the current improvement caused by the increased of amount of enzyme on the electrode is very attractive for further studies in biocatalysis.

Conclusions

A novel material composed by glucose oxidase enzyme, chitosan, ferrocene and magnetite (Fe₃O₄-Chi-Fc/GOx) was obtained. Fe₃O₄-Chi-Fc showed high stability, retaining 50% of the values of current density after successive cycles of switching run during 24 h. By cyclic voltammetry, it was observed that in the presence of the external magnetic field, the mass transport of Fe₃O₄-Chi-Fc is intensified, resulting in the increase of 86.3 and 42.3% for the anodic and cathodic currents, respectively, attributed for the redox couple Fc/Fc⁺. Also, the redox potential shift is different for the "switch on" and "switch off" commutative experiments. In addition, the kinetics of charge transfer of Fc/Fc⁺ is also influenced, because the electrochemical reactions are limited by diffusion for both "switch on" and "switch off" modes. This unexpected behavior was attributed to the non-electroactive components (chitosan and iron oxide) of Fe₃O₄-Chi-Fc that may contribute to hinder the electron transfer process. For Fe₃O₄-Chi-Fc/GOx suspended in 20 mmol L^-1 of glucose solution, that catalytic current of glucose oxidation was 70% higher for "switch on" than "switch off" mode. The oxidation currents attained the maximum about 10.7 µA cm^-2 at 0.8 V for "switch on" mode, leading to an increase of 3.2 µA cm^-2 when compared with "switch off" mode. Based on these results, it is possible to achieve a fine control of the bioelectrocatalysis, because redox mediator and enzyme components are both assembled in a magnetic particle.

Acknowledgments

The financial support from FAPESP (F. N. Crespilho, Project No. 2009/15558-1 and 2011/01541-0); CAPES, CNPq (Project No. 2011/01541-0), INEO; and Rede NanoBioMed-Brasil (CAPES) is acknowledged.

Submitted: October 10, 2012

Published online: February 18, 2013

FAPESP has sponsored the publication of this article.

• 1. Hirsch, R.; Katz, E.; Willner, I.; J. Am. Chem. Soc. 2000,122,12053.
• 2. Katz, E.; Ichia, L. S. H.; Bückmann, A. F.; Willner, I.; Angew. Chem., Int. Ed. 2002,41,1343.
• 3. Riskin, M.; B. Basnar; Huang, Y.; Willner, I.; Adv. Mater. 2007,19,2691.
• 4. Zhai, Y.; Zhai, J.; Wen, D.; Zhou, M.; Zhang, L.; Dong, S.; Electrochem. Commun 2008,10,1172.
• 5. Liang, C.; Jing, L.; Shi, X.; Zhang, Y.; Xian, Y.; Electrochim. Acta 2012,69,167.
• 6. Elmore, W.C.; Phys. Rev. 1938,54,309.
• 7. Kaushik, A.; Khan, R.; Solanki, P. R.; Pandey, P.; Alam, J.; Ahmad, S.; Malhotra, B. D., Biosens. Bioelectron. 2008,24,676.
• 8. Ghica, M.E.; Brett, C.; Anal. Chim. Acta 2005,532,145.
• 9. Laurent, S.; Forge, D.; Port, M.; Roch, A.; Robic, C.; Vander Elst, L.; Muller, R. N.; Chem. Rev. 2008,108,2064.
• 10. Correa-Duarte, M. A.; Giersig, M.; Kotov, N. A.; Liz-Marzán, L. M.; Langmuir 1998,14,6430.
• 11. van de Hulst, H. C.; Light Scattering by Small Particle; Dover Publications: New York, USA, 1981.
• 12. Li, G.; Jiang, Y.; Huang, K.; Ding, P.; Chen, J.; J. Alloys Compd. 2008,466,451.
• 13. Pearson, F. G.; Marchessault, R. H.; Liang, C. Y.; J. Polym. Sci 1960,43,101.
• 14. Roeges, N. P. G.; A Guide to the Complete Interpretation of Infrared Spectra of Organic Structures; John Wiley: Chichester, UK, 1994.
• 15. Wu, Y.; Guo, J.; Yang, W.; Wang, C.; Fu, S.; Polymer 2006,47,5287.
• 16. An, X.; Su, Z.; J. Appl. Polym. Sci. 2001,81,1175.
• 17. Tran, H.V.; Tran, L. D.; Nguyen, T.N.; Mater. Sci. Eng. C 2010,30,304.
• 18. Yuwei, C.; Jianlong, W.; Chem. Eng. J. 2011,168,286.
• 19. Holze R.; Experimental Electrochemistry: A Laboratory Textbook; Wiley-VCH: Weinheim, Germany, 2009.
• 20. Nakahama, S.; Murray, R.W.; J. Electroanal. 1983,158,303.
• 21. Pereira, A. R.; Iost, R. M.; Martins, M. V. A.; Yokomizo, C. H.; Silva, W. C.; Nantes, I. L.; Crespilho, F. N.; Phys. Chem. Chem. Phys. 2011,13,12155.
• 22. Iost, R. M.; Crespilho, F. N.; Biosens. Bioelectron. 2012,31,1.
• 23. Goncalves, W. D.; Lanfredi, A. J. C.; Crespilho, F. N.; J. Phys. Chem. C 2011,115,16172.
• 24. Crespilho, F. N.; Ghica, M.; Gouveia, C. C.; Oliveira jr, O.; Brett, C.; Talanta 2008,76,922.

Publication Dates

• Publication in this collection
  20 May 2013
• Date of issue
  Feb 2013

History

• Received
  10 Oct 2012
• Accepted
  18 Feb 2013