Pore size effect in the amount of immobilized enzyme for manufacturing carbon ceramic biosensor

Pore size effect in the amount of immobilized enzyme for manufacturing carbon ceramic biosensor

Microporous and Mesoporous Materials 247 (2017) 95e102 Contents lists available at ScienceDirect Microporous and Mesoporous Materials journal homepa...

2MB Sizes 0 Downloads 19 Views

Microporous and Mesoporous Materials 247 (2017) 95e102

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

Pore size effect in the amount of immobilized enzyme for manufacturing carbon ceramic biosensor Elisangela Muncinelli Caldas a, b, Dhjulia Novatzky a, Monique Deon a, Eliana Weber de Menezes a, Plinho Francisco Hertz c, Tania Maria Haas Costa a, Leliz Ticona Arenas a, Edilson Valmir Benvenutti a, * a b c

Instituto de Química, UFRGS, CP 15003, CEP 91501-970, Porto Alegre, RS, Brazil ~o, Ci^ Instituto Federal de Educaça encia e Tecnologia, CEP 95180-000, Farroupilha, RS, Brazil Instituto de Ci^ encia e Tecnologia de Alimentos, UFRGS, CEP 91501-970, Porto Alegre, RS, Brazil

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 February 2017 Received in revised form 16 March 2017 Accepted 27 March 2017 Available online 30 March 2017

Understanding the mechanism of enzyme immobilization in porous designed matrices is important issue to develop biosensors with high performance. Mesoporous carbon ceramic materials with conductivity and appropriated textural characteristics are promising candidates in this area. In this work, carbon ceramic materials were synthesized using the sol-gel method by planning the experimental conditions to obtain materials with different pore size, from 7 to 21 nm of diameter. The study of the influence of pore size in the biomacromolecules immobilization capacity was performed using glucose oxidase enzyme as probe. The influence of textural characteristics of material in the amount of enzyme immobilized, as well as, its performance as biosensor, was studied. On the surface of highest pore size matrix, it was possible to immobilize the highest amount of enzyme, resulting in better electrochemical response. With this simple material, composed only by silica, graphite and enzyme, which was improved by the amount of immobilized enzyme through the enlargement of matrix pore size, it was possible to prepare an electrode to be applied as biosensor for glucose determination. This electrode presents good reproducibility, sensitivities of 0.33 and 4.44 mA mM1 cm2 and detection limits of 0.93 and 0.26 mmol L1, in argon and oxygen atmosphere, respectively. Additionally, it can be easily reused by simple polishing its surface. © 2017 Published by Elsevier Inc.

Keywords: Textural properties Interconnected pores Silica xerogel Cyclic voltammetry Chronoamperometry GOD enzyme

1. Introduction The interest in developing materials containing enzymes has been rising in the last decades because of the numerous applications in enzymatic catalysis and in the preparation of biosensors [1e5]. These materials present high selectivity and specificity, minimized impurities, easier product separation and environmental acceptance, when compared with non-enzymatic systems [6]. An important aspect when enzymatic materials are being developed is the immobilization and stabilization of the biomolecules on adequate substrates. Although the adsorption of enzymes and proteins in solid matrices has been widely studied, the ability to control the amount adsorbed, the interaction between enzyme and matrix surface, and the location of the enzyme in the

* Corresponding author. E-mail address: [email protected] (E.V. Benvenutti). http://dx.doi.org/10.1016/j.micromeso.2017.03.051 1387-1811/© 2017 Published by Elsevier Inc.

pore structure are still an important field to be investigated [7]. There are several reports dealing with the enzyme immobilization on porous materials, however the immobilization is commonly accompanied by drastic reduction in the textural characteristics such as surface area and pore volume. This behavior can be consequence of fully entered enzyme inside of pores or also due to immobilization on external surface leading to pore blocking [7]. For immobilization inside of pores, it should be taking into account also the remained free space into the pore to provide sufficient enzyme mobility and retain its catalytic activity. Additionally, the free space eases the substrate diffusion to the catalytic sites [8]. Therefore, to synthesize porous materials with adequate pore structure allowing the immobilization of enzyme, without the loss of textural properties, is still a challenge [7,9]. Concerning electrochemical sensors, there is a recent interest in the preparation of devices based on carbon ceramic materials [10e12]. These materials are obtained by sol-gel synthesis method, based on hydrolysis and polycondensation of silicon precursor

96

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

reactants, in the presence of carbon source. The main feature of this approach is that the sol-gel synthesis method allows designing characteristics such as texture, morphology, composition and chemical reactivity of the synthesized materials [13,14]. In this way, it is possible to obtain carbon ceramic materials with essential characteristics to be used as matrices for enzyme immobilization, as mechanical rigidity and the possibility of chemical functionalization provided by silica. On the other hand, the conductivity provided by the carbon moiety allows their use as electrochemical devices. The surface of electrodes can be renewed by polishing, which affords the creation of a reproducible and reusable system with easy handling, improving its technological importance. Even presenting such advantages, the carbon ceramic electrodes have been not often applied as enzymatic biosensors [15e20]. Biosensors based on glucose oxidase enzyme are a significant share of publications in the field of biosensors [21e23] and even so, there are only a few reports of glucose oxidase immobilized on carbon ceramic materials [18e20]. Glucose plays essential role in diabetes disease and in food industry, justifying the need for quantification devices [24]. The papers that deal with carbon ceramic electrodes and glucose oxidase enzyme present studies of the electrochemical behavior and the biosensor performance. Nevertheless, a physicochemical characterization of the matrix is also necessary aiming to elucidate the influence of the textural characteristics of the matrix in the amount of enzyme immobilized and how these aspects are related to the electrochemical response. In this work, carbon ceramic materials were synthesized with planned textural characteristics by changing the experimental conditions in order to obtain materials with different pore size and surface area. These materials were used as matrices for glucose oxidase immobilization aiming to prepare carbon ceramic electrodes. The influence of textural characteristics in the amount of enzyme immobilized, as well as, in the performance as biosensor, was studied. 2. Materials and methods 2.1. Chemicals Tetraethylorthosilicate (TEOS) (Aldrich), ethanol (Merck), graphite powder (Aldrich), hydrofluoric acid (HF) (Merck, 40%), hydrochloric acid (HCl) (Merck, 37%), 3-aminopropyltrimethoxysilane (APTMS) (Aldrich), glutaraldehyde (GA) (SigmaAldrich, 25%), glucose oxidase enzyme (GOD) (Aldrich, E.C. 1.1.3.4 50 KU), glucose (Vetec), L-ascorbic acid (Synth), dopamine (Sigma), uric acid (Aldrich), all analytical grade, were used without previous purification. Phosphate buffer solution (PBS) (0.1 mol L1, pH 7.0) was prepared from NaH2PO4 (FMaia) and Na2HPO4 (FMaia). Mineral oil was used for the electrode manufacture. All the solutions were prepared in distilled water. 2.2. Synthesis of silica graphite matrices by sol-gel method (SG matrices) Three carbon ceramic matrices were synthesized in two steps, using TEOS as silica precursor, graphite powder and different quantities of a solution HF/HCl (6.0 mol L1) as catalyst. Initially, 5.0 mL of TEOS were pre-hydrolyzed in 5.0 mL of ethanol, in the presence of 0.22 mL of distilled water and 0.03 mL of catalyst. The solution was maintained under magnetic stirring, in reflux, at 80  C by 1 h and then, it was cooled to room temperature. Graphite powder (50 wt%, calculated from the expected SiO2 weight) was added to this solution and the system was sonicated for 2.5 h at 40  C. In the second step, 0.73 mL of distilled water were homogenized with 0.68 mL of catalyst and added to the mixture, under

constant stirring. The resultant matrix was named SG1. Two more matrices were synthesized changing the amount of catalyst and the water added in the second step. For SG2 matrix, 0.41 mL of distilled water and 1.04 mL of catalyst were added, while for SG3, just 1.59 mL of catalyst. The materials were covered without sealing and stored for solvent evaporation at room temperature. After 15 days, the monoliths were powdered, washed with distilled water and ethanol, and then they were vacuum-dried for 2 h at 80  C. 2.3. Modifications of SG matrices with APTMS by grafting The matrices were previously activated by heat treatment at 120  C, under vacuum for 2 h before modification with APTMS. Then, for each matrix, 1 g was added in a three-neck round bottom flask containing 1 mL of APTMS solubilized in 20 mL of ethanol, the reaction was performed under argon atmosphere, mechanical stirring, at 65  C. After 24 h, the supernatant was removed and the matrices were vacuum-dried for 2 h at 80  C. Finally, the modified matrices were washed thoroughly with water and ethanol, vacuum-dried at 80  C and denominated SG1-AP, SG2-AP and SG3AP, where AP specifies the aminopropyl group. 2.4. Glutaraldehyde activation Activation with glutaraldehyde was performed using 1 g of each SG-AP immersed in 10 mL of 5% glutaraldehyde solution (PBS) and kept in shaking for 3 h. After this, the materials were washed with PBS, vacuum-dried at 60  C for 2 h, and hereafter called SG1-AP-GA, SG2-AP-GA, SG3-AP-GA, where GA specifies glutaraldehyde. 2.5. GOD enzyme immobilization GOD was immobilized in the modified matrices using a PBS GOD solution containing 20.3 mg mL1 of protein. For this, 1 g of the materials was kept immersed in 8 mL of GOD solution, at 4  C, under shaking and rest overnight. Protein content in solution was determined by the Bradford assay [25]. The immobilized protein was estimated as the difference between the amount of protein offered to the material and the amount recovered in the supernatant and washing fractions. For SG3 matrix, there was no residual protein in the supernatant. Because of this, a more concentrated solution of glucose oxidase (33.8 mg mL1) was offered. All materials were washed with PBS, vacuum-dried in ice bath for 2 h and are hereafter called SG1-AP-GA-GOD, SG2-AP-GA-GOD, SG3-APGA-GOD. 2.6. Characterization The N2 adsorption-desorption isotherms were determined at liquid nitrogen boiling point, using a Tristar 3020 Kr Micromeritics equipment. The samples were previously degassed at 120  C (except for samples with enzymes, where degassed temperature was 60  C), under vacuum, for 12 h. The specific surface areas were determined by the BET (Brunauer, Emmett and Teller) multipoint technique and the pore size distribution was obtained by using the BJH (Barret, Joyner and Halenda) method [26]. The thermogravimetric analysis of materials were performed under argon flow on a Shimadzu Instrument model TGA-50 2, with a heating rate of 10  C min1, from room temperature up to 600  C. The temperature range used to estimate the organic contente was 150 to 600  C. 2.7. Electrochemical measurements Carbon ceramic electrodes were prepared by pressing (3 tons) 20 mg of the materials with 3 mg of mineral oil. The obtained disks

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

had 6 mm of diameter and c.a. 1 mm of thickness, and then were glued to a glass tube with gel glue. The electric contact was made with a copper wire. Pure graphite powder was added into the glass tube to improve the connection between the cooper wire and the disk. Electrochemical measurements were carried out on Autolab PGSTAT302N apparatus using an electrochemical cell with three electrodes. The electrodes were a saturated calomel electrode (SCE) and a platinum wire as reference and auxiliary electrodes, respectively, and as working electrode the previously prepared carbon ceramic electrode was used. All measures were made at room temperature. 3. Results and discussion 3.1. Textural analysis In this work, three materials were obtained by increasing the amount of catalyst used in the sol-gel synthesis (SG materials). Fig. 1a presents the N2 adsorption-desorption isotherms and in Fig. 1b it is shown the BJH pore size distribution of SG1, SG2 and SG3 materials. As it can be seen, all curves are type IV isotherms, typical of mesoporous materials, obtained by using hydrofluoric acid as solgel catalyst [27e29]. The use of fluoride catalyst leads to the formation of xerogels with close-packed spherical particles [14,26,30]. The mesoporosity is consequence of the interstitial space between this primary structure, forming a series of interconnected pores [26]. The increase in the catalyst amount produces an enlargement in pore size, which can be observed by a shift in the isotherm inflection positions to higher relative pressures as well as, in the BJH pore size distribution [30e32]. This behavior was clearly observed for SG1, SG2 and SG3 materials, where the maxima of BJH pore diameter distribution curves are located at 7, 14 and 21 nm, respectively. The SG materials were subsequently modified with APTMS, glutaraldehyde and GOD, and their textural analyses are shown in Figs. 2e4, for SG1, SG2 and SG3 materials, respectively. It is possible to observe for all materials, the subsequent modifications produce a decreasing in the adsorbed volume, since the saturation point of isotherms is attained with lower nitrogen amounts. Moreover, the pore size distribution curves show the same tendency. It is observed a pore volume decrease, as well as, a

97

pore diameter shift to lower values, along with the modifications. The pore volume and diameter are presented on Table 1, together with the surface area values. As a matter of fact, it is clearly observed a decreasing in pore volume, in the pore diameter and also in the specific surface area along with the modifications. 3.2. Protein content analysis The amount of immobilized enzyme was firstly estimated by using Bradford method [25] based on difference between the amount of protein offered to the material and the amount recovered in the supernatant and washing fractions. The results are presented on Table 2. It can be seen that for the same amount of offered protein (20.3 mg mL1), the amount of immobilized enzyme increases from SG1 to SG3 material. For SG3, the loading saturation was only attained using higher enzyme concentration. 3.3. Thermogravimetric analysis All SG materials were submitted to thermogravimetric analysis and the obtained curves are presented as supplementary material. It was observed, for all curves, in the range from 150 to 600  C, a weight loss due to the decomposition and desorption of available organics on accessible pores. The weight loss values undergo an increasing with the successive modifications. Based on these values, it was estimated the organic content incorporated, as well as the group densities after each modification step, which are shown on Table 1. It can be seen the aminopropyl (AP) incorporated amount decreases from 41.4 for SG1 to 11.6 mg g1 for SG3 material. The same trend can be observed for glutaraldehyde (GA), which incorporated amount reduces from 43.9 to 18.1 mg g1. However, when these values are converted in organic group per square nanometer, the aminopropyl and glutaraldehyde densities on surface were very similar, indicating that the incorporated organics are related to the available surface area during the grafting reaction, which was higher for SG1 and decreases toward SG3 matrix (see Table 1). It means that from SG1 to SG3 there is a reduction in the available bonding sites on matrix surface for covalent immobilization of enzyme, because there is a reduction in surface area. Even so, the amount of incorporated enzyme did not follow the same trend, i.e. the enzyme content did not decrease. In contrast, it was observed that the enzyme quantity increased in more than three times for

Fig. 1. Textural analysis of SG1, SG2 and SG3 materials. a) N2 adsorption-desorption isotherms; b) BJH pore size distributions.

98

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

Fig. 2. Textural analysis for SG1 and its modifications. a) N2 adsorption-desorption isotherms; b) BJH pore size distributions. Inset Fig. 2b shows the pore size distribution curves in detail.

Fig. 3. Textural analysis for SG2 and its modifications. a) N2 adsorption-desorption isotherms; b) BJH pore size distributions. Inset Fig. 3b shows the pore size distribution curves in detail.

Fig. 4. Textural analysis for SG3 and its modifications. a) N2 adsorption-desorption isotherms; b) BJH pore size distributions. Inset Figure 4a shows the hysteresis region in detail.

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

99

Table 1 Textural and thermogravimetric data of SG material series. Sample

SBET (±5 m2 g1)

BJH Pore Volume (±0.01 cm3 g1)

Pore sizea (nm)

SG1 SG1-AP SG1-AP-GA SG1-AP-GA-GOD SG2 SG2-AP SG2-AP-GA SG2-AP-GA-GOD SG3 SG3-AP SG3-AP-GA SG3-AP-GA-GOD

273 240 184 172 158 142 147 127 73 72 78 65

0.66 0.48 0.35 0.32 0.72 0.63 0.53 0.45 0.57 0.54 0.47 0.41

7.0 5.7 5.6 5.6 14.0 12.5 11.3 11.2 21.0 20.7 18.1 17.6

a b

Organic density (group per nm2)

41.4 43.9 11.7

1.57 1.10

19.0 40.6 20.5

1.26 1.69

11.6 18.1 39.3

1.65 1.50

maximum of BJH pore diameter distribution curve. Incorporated organic content estimated from thermogravimetric analysis in the range of 150e600  C.

Table 2 Bradford analysis results. Material

Offered protein (mg mL1)

Residual protein (mg mL1)

SG1-AP-GA SG2-AP-GA SG3-AP-GA SG3-AP-GA

20.3 20.3 20.3 33.8

19.0 2.6 N.D.a 4.6

a

Organic contentb (mg g1)

Non detected.

SG3 when compared to SG1 material. Therefore, this behavior can be interpreted taking into account the variation in the pore size of the matrices, which increases from 7 nm for SG1 to 21 nm for SG3 matrix, as can be seen in Fig. 1 and Table 1. It had been reported that for ordered cylindrical mesoporous materials, biomacromolecules, including GOD enzyme, diffuse successfully inside the pores [7,33,34]. For silica material with interconnected mesopores, there are few reports, however the diffusion and immobilization of GOD enzyme was also attained [35,36]. Thus, the pore size becomes more important since the narrow regions could difficult the biomacromolecule access to the pores. In the present work, whereas GOD has an approximate dimension of 7.0  5.5  8.0 nm [7], its diffusion was better attained in the largest pore size support, i.e. SG3 modified matrix. It is important to highlight that even presenting the highest enzyme content, SG3-AP-GA-GOD material had its porosity preserved. The appreciable surface area and pore volume of SG3-AP-GA-GOD material allow enzyme mobility and a fast mass transport of substrates, which is an important issue for biosensors development [7,8]. 3.4. Electrochemical characterization The materials modified with GOD, which are SG1-AP-GA-GOD, SG2-AP-GA-GOD and SG3-AP-GA-GOD, were used in the preparation of pressed disks for carbon ceramic electrodes, and they were assigned as EGOD-1, EGOD-2 and EGOD-3, respectively. Cyclic voltammetry was used to evaluate their electrochemical behavior and the results are shown in Fig. 5. Each electrode presented a pair of redox peaks that indicates glucose oxidase presence. The peaks were attributed to the redox pair FAD/FADH2 [37e39] according to the following reaction:

GODðFADÞ þ 2e þ 2Hþ #GODðFADH2 Þ By comparing the anodic and cathodic peak current of electrodes, it can be observed an increase from EGOD-1 to EGOD-3 and well defined peaks for EGOD-3. This result indicates that EGOD-3 has higher enzyme content when compared with the other two

electrodes, in agreement to the thermogravimetric results, and therefore, it was chosen to be applied as biosensor for glucose. In order to confirm that the redox peaks are due to the presence of GOD enzyme, a blank electrode was prepared following the same procedure, except the GOD immobilization step, i.e. using the SG3AP-GA material. The cyclic voltammograms of EGOD-3 and blank electrode are presented as Supplementary Material. As it was not observed peaks for blank electrode, it can be inferred that the peaks seen for EGOD-3 are due to the FAD/FADH2 redox pair of GOD enzyme. The effect of pH in the enzymatic response was evaluated for EGOD-3, by cyclic voltammetry, and the correlation between pH and peak current is presented in supplementary material. Considering anodic and cathodic peak currents are higher in pH 7.0 and also this pH is close to biological environment (pH ¼ 7.4), this condition was chosen for future measures. 3.5. Electrochemical behavior of GOD on EGOD-3 electrode Cyclic voltammograms of EGOD-3 electrode at different scan rates, from 0.020 to 1.9 V s1, are shown in Fig. 5b. The redox peak currents increase linearly with scan rate from 0.020 to 0.350 V s1 (inset Fig. 5b), which is typical of redox process controlled by electroactive species adsorbed on the electrode surface [40e43]. For values of scan rate above 0.350 V s1 the linearity is maintained but there is an increasing in the contribution of capacitive current, as clearly observed on cyclic voltammograms of Fig. 5b. The linear regression equations in the range from 0.020 to 0.350 V s1 were ipa (A) ¼ 3.599  105 þ 3.385  104 v (V s1) with R ¼ 0.999 and ipc (A) ¼  8.969  105  5.299  104 v (V s1) with R ¼ 0.998. The linear dependence observed between ip and scan rate allows estimating the amount of adsorbed GOD on electrode surface, as follow in Equation (1) [44]:

ip ¼

n2 F 2 v A г 4RT

(1)

Where n is the number of transferred electrons, F is the Faraday's constant (96.4885 103 C mol1), v is the scan rate, A is the electrode area (cm2), г is the amount of adsorbed GOD on electrode surface (mol cm2), R is the ideal-gas constant (8.314 J K1 mol1), T is the temperature (298 K). Therefore, the average value of GOD adsorbed on the electrode surface was 4.09  1010 mol cm2. The found value was higher than others already reported [38,44e47]. According to Laviron's theory [48], when nDEp > 200 mV, where n is the number of electrons involved (n ¼ 2 for GOD process), the charge transfer coefficient (a) and electron transfer rate constant (ks) can be calculated from the relationship between Ep and log v,

100

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

Fig. 5. Cyclic voltammograms obtained under argon atmosphere and PBS (pH ¼ 7.0): a) Carbon ceramic electrodes at 50 mV s1; b) EGOD-3 at different scan rate, from 0.020 to 1.900 V s1. Inset Figure 5b shows correlation between scan rate and current peaks (ipa and ipc) for v ¼ 0.020e0.350 V s1.

which in high scan rates, presents a linear behavior. The equations that describe this behavior are presented below as Equations (2)e(4):

Epa ¼ Ec. þ

Epc ¼ Ec. 

 2:3RT log v ð1  aÞnF

(2)

 2:3RT log v anF

(3)





  RT log ks ¼ a logð1  aÞ þ ð1  aÞ log a  log nFv 

að1  aÞnF DEp 2:3RT

(4)

Where E. c is the formal potential estimated from [(Epa þ Epc)/2], R is the ideal-gas constant, T is the temperature (298 K) and F is the Faraday constant. In supplementary material it is shown the relationship between Ep and log v for EGOD-3 electrode, when v > 0.60 V s1 and the Laviron theory requirements are obeyed. The equations describing the linear behavior are Epa (V) ¼  0.35263 þ 0.10044 log v (V s1) with R ¼ 0.980 and Epc (V) ¼  0.60745e0.16896 log v (V s1) with R ¼ 0.996. The a value was estimated as 0.44 and from it, the electron transfer rate constant ks was calculated as 0.28 s1, using Equation (4). This value is in the same range of several electrodes already presented [23], and also it is similar to that found for single-walled carbon nanotube/gold electrode (0.3 s1) [49e51], which contains nanostructures that facilitate the electron transfer. In the present system, where this kind of nanostructures is not present, the ks found value can be interpreted taking into account the efficiency of the system enzyme immobilized in the carbon ceramic electrode.

3.6. Effect of atmosphere Fig. 6 presents cyclic voltammograms of EGOD-3 electrode obtained in different atmospheres: a) argon, b) air and c) oxygen. It was observed, from a to c curve, an increase in cathodic peak current and a slight decrease in anodic peak current. In EGOD-3 electrode, a direct electrochemistry transfer occurs between GOD

Fig. 6. Cyclic voltammograms of EGOD-3 under different atmosphere: a) argon, b) air and c) oxygen, in PBS (pH ¼ 7.0), v ¼ 50 mV s1. Inset Figure shows the representation of GOD redox reactions in the presence of oxygen.

and electrode, leading to the conversion from GOD(FAD) to GOD(FADH2) [37,40,51]. In the presence of oxygen, GOD(FADH2) is oxidized to GOD(FAD) and oxygen is reduced to H2O2, indicating that EGOD-3 electrode exhibit electrocatalytic activity to oxygen [37] as represented in inset Fig. 6. 3.7. Amperometric response to glucose The response of EGOD-3 electrode for glucose was evaluated using chronoamperometry with successive additions of glucose. Fig. 7 shows the measure results under argon and oxygen atmosphere. Inset Figures show the linear range obtained between current and glucose concentration, and also the influence of interferents. The amperometric currents increased linearly with increasing glucose concentration in the range of 0.39e5.36 mmol L1 for argon atmosphere and 0.2e2.47 mmol L1 for oxygen atmosphere. Linear regression were expressed by i (mA) ¼  1.249 þ 0.094 [glucose]

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

101

Fig. 7. Amperometric response of EGOD-3, with successive additions glucose (PBS pH ¼ 7.0, applied potential ¼  0.366 V). Inset Figure a) calibration curve. Inset Figure b) Response in the presence of interferents (0.5 mmol L1): ascorbic acid (AA); dopamine (DOP); and uric acid (UA).

(mmol L1) with R ¼ 0.994 and i (mA) ¼  6.788 þ 1.257 [glucose] (mmol L1) R ¼ 0.997, respectively. Sensitivity of EGOD-3 were determined as the slope of curves, being 0.33 mA mM1 cm2 and 4.44 mA mM1 cm2, for argon and oxygen, respectively. Detection limit were calculated by the ratio (3  SDb/Slope), where SDb was the standard deviation of blank measurements (n ¼ 10), and the obtained value were 0.93 mmol L1 and 0.26 mmol L1. Actually, EGOD-3 exhibits better performance in oxygen atmosphere. However, in argon atmosphere, the electrode becomes more selective, as can be seen in inset Fig. 7. In oxygen atmosphere, the electrode presented response also for the tested interferences (ascorbic acid, dopamine and uric acid). Five individual glucose measures were made with the same EGOD-3, after subsequent surface renewing by polishing with sandpaper grade 1200 and the results are presented in Fig. 8. It is possible to observe in Fig. 8 that there is a minimal loss of enzyme activity even after five cycles. The relative standard deviation of these measurements was 6%. These results indicate that the carbon ceramic electrode with GOD immobilized is an interesting

alternative in the area of biosensors development, because it is an easy prepared system, dispenses the addition of nanostructures and is able to be reused by simple polishing its surface. 4. Conclusion Using the sol-gel synthesis method, it was possible to obtain mesoporous carbon ceramic materials composed by silica and graphite, with variable textural characteristics, such as, surface area, pore volume and pore diameter, aiming to the immobilization of biomolecules for biosensing applications. GOD enzyme was used as probe, and it was observed that the amount of immobilized GOD was related to the pore size of the matrix. On the surface of highest pore size matrix (SG3), it was possible to immobilize the highest amount of enzyme, even presenting the lowest surface area. The textural properties were preserved even after the enzyme immobilization, allowing enzyme mobility and diffusion of analytes, making the material promising to be applied as biosensor. Electrode made with highest enzyme content material presented better defined peaks of FAD/FADH2 pair redox of enzyme, by cyclic voltammetry measurements, and because of this, it was applied as biosensor for electrochemical glucose determination in argon and oxygen atmosphere. Although, the electrode was more selective in argon atmosphere, it was more sensible in oxygen. The found sensitivities were 0.33 and 4.44 mA mM1 cm2, for argon and oxygen, respectively. The detection limits were 0.93 and 0.26 mmol L1, respectively. In this way, our large pores carbon ceramic material allows the immobilization of a greater amount of enzyme, when compared with other enzymatic electrodes, resulting in easy prepared and simple biosensor, with good reproducibility. Acknowledgements The authors are grateful to CNPq, FAPERGS and CAPES for financial support and grants. Appendix A. Supplementary data

Fig. 8. Relative peak current intensity obtained from a same EGOD-3 electrode, obtained for subsequent measurements (PBS pH ¼ 7.0, applied potential ¼  0.366 V).

Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.micromeso.2017.03.051.

102

E.M. Caldas et al. / Microporous and Mesoporous Materials 247 (2017) 95e102

References [1] M. Etienne, L. Zhang, N. Vil, A. Walcarius, Electroanalysis 27 (2015) 2028e2054. € Sag lam, B.K. ızılkaya, H. Uysal, Y. Dilgin, Talanta 147 (2016) 315e321. [2] O. [3] J.M. Palomo, M. Filice, Biotechnol. Adv. 33 (2015) 605e613. [4] Y. Zhang, J. Ge, Z. Liu, ACS Catal. 5 (2015) 4503e4513. [5] C.M. Maroneze, G.P. dos Santos, V.B. de Moraes, L.P. da Costa, L.T. Kubota, Biosens. Bioelectron. 77 (2016) 746e751. [6] L.P. Christopher, H. Kumar, V.P. Zambare, Appl. Energy 119 (2014) 497e520. [7] S. Hudson, J. Cooney, E. Magner, Angew. Chem. Int. Ed. 47 (2008) 8582e8594. [8] A. Walcarius, M.M. Collinson, Annu. Rev. Anal. Chem. 2 (2009) 121e143. [9] P.R.A.F. Garcia, R.N. Bicev, C.L.P. Oliveira, O.A. Sant’Anna, M.C.A. Fantini, Microporous. Mesoporous. Mater 235 (2016) 59e68. [10] C.M. Maroneze, L.T. Arenas, R.C.S. Luza, E.V. Benvenutti, R. Landers, Y. Gushikem, Electrochim. Acta 53 (2008) 4167e4175. [11] T. Zhan, X. Sun, X. Wang, W. Sun, W. Hou, Talanta 82 (2010) 1853e1857. [12] L.T. Arenas, P.C.M. Villis, J. Arguello, R. Landers, E.V. Benvenutti, Y. Gushikem, Talanta 83 (2010) 241e248. land, Laura M. Ilharco, [13] R. Ciriminna, A. Fidalgo, V. Pandarus, F. Be Mario Pagliaro, Chem. Rev. 113 (2013) 6592e6620. [14] E.V. Benvenutti, C.C. Moro, T.M.H. Costa, M.R. Gallas, Quim. Nova 32 (2009) 1926e1933. [15] B. Habibi, M. Jahanbakhshi, Microchim. Acta 182 (2015) 957e963. [16] R. Mohammad-Rezaei, H. Razmi, Electroanalysis 24 (2012) 2094e2101. [17] B. Haghighi, A. Rahmati-Panah, S. Shleev, L. Gorton, Electroanalysis 19 (2007) 907e917. [18] H. Razmi, R. Mohammad-Rezaei, Biosens. Bioelectron. 41 (2013) 498e504. [19] F. Tian, G. Zhu, Anal. Chim. Acta 451 (2002) 251e258. [20] X. Yang, L. Hua, H. Gong, S.N. Tan, Anal. Chim. Acta 478 (2003) 67e75. [21] H. Ciftci, Y. Oztekin, U. Tamere, A. Ramanaviciene, A. Ramanavicius, Colloids Surf. B Biointerfaces 123 (2014) 685e691. ^a, Sens. Actuators [22] C.G. de Jesus, D. Lima, V. dos Santos, K. Wohnrath, C.A. Pesso B 186 (2013) 44e51. [23] A. Rabti, W. Argoubi, N. Raouafi, Microchim. Acta 183 (2016) 1227e1233. [24] S. Ding, M. Schumacher, Sensors 16 (2016) 589. [25] M.M. Bradford, Anal. Biochem. 254 (1976) 248e254. [26] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, second ed., Academic Press, London, 1982. [27] J.V.H. Ramos, F.M. Morawski, T.M.H. Costa, S.L.P. Dias, E.V. Benvenutti, E.W. de Menezes, L.T. Arenas, Microporous. Mesoporous. Mater 217 (2015) 109e118.

[28] E.W. de Menezes, M.R. Nunes, L.T. Arenas, S.L.P. Dias, I.T.S. Garcia, Y. Gushikem, T.M.H. Costa, E.V. Benvenutti, J. Solid State Electrochem 16 (2012) 3703e3713. [29] E.W. de Menezes, E.C. Lima, B. Royer, F.E. de Souza, B.D. dos Santos, rio, T.M.H. Costa, Y. Gushikem, E.V. Benvenutti, J. Colloid Interface J.R. Grego Sci. 378 (2012) 10e20. [30] F.A. Pavan, S.A. Gobbi, C.C. Moro, T.M.H. Costa, E.V. Benvenutti, J. Porous Mater 9 (2002) 307e311. [31] L. He, K. Power, R.H. Baney, L. Gower, R.S. Duran, P. Sheth, S.R. Carino, J. Non Cryst. Solids 289 (2001) 97e105. [32] R.F. Silva, W.L. Vasconcelos, Mater Res. 2 (1999) 197e200. [33] P.R.A.F. Garcia, R.N. Bicev, C.L.P. Oliveira, O.A. Sant’Anna, M.C.A. Fantini, Microporous. Mesoporous. Mater 235 (2016) 59e68. [34] P. Majumdar, A.Y. Khan, R. Bandyopadhyaya, Biochem. Eng. J. 105 (2016) 489e496. [35] K. Yao, Y. Zhu, P. Wang, X. Yang, P. Cheng, H. Lu, Mater. Sci. Eng. C 27 (2007) 736e740. [36] S. Wu, H. Ju, Y. Liu, Adv. Funct. Mater 17 (2007) 585e592. [37] M. Cui, B. Xu, C. Hu, H.B. Shao, L. Qu, Electrochim. Acta 98 (2013) 48e53. [38] H. Zhang, Z. Meng, Q. Wang, J. Zheng, Sens. Actuators B 158 (2011) 23e27. [39] B. Unnikrishnan, S. Palanisamy, S.-M. Chen, Biosens. Bioelectron. 39 (2013) 70e75. [40] Q. Liu, X. Lu, J. Li, X. Yao, J. Li, Biosens. Bioelectron. 22 (2007) 3203. [41] M. Deon, E.M. Caldas, D.S. da Rosa, E.W. de Menezes, S.L.P. Dias, M.B. Pereira, T.M.H. Costa, L.T. Arenas, E.V. Benvenutti, J. Solid State Electrochem 19 (2015) 2095e2105. [42] F. Hu, S. Chen, C. Wang, R. Yuan, Y. Chai, Y. Xiang, C. Wang, J. Mol. Catal. B 72 (2011) 298e304. [43] E.M. Caldas, E.W. de Menezes, T.M. Pizzolato, S.L.P. Dias, T.M.H. Costa, L.T. Arenas, E.V. Benvenutti, J. Sol-Gel Sci. Technol. 72 (2014) 282e289. [44] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, John Wiley & Sons, New York, 2001. [45] Y. Song, H. Liu, Y. Wang, L. Wang, Electrochim. Acta 93 (2013) 17e24. [46] C. Qiu, X. Wang, X. Liu, S. Hou, H. Ma, Electrochim. Acta 67 (2012) 140e146. [47] Y. Li, F. Wang, F. Huang, Y. Li, S.J. Feng, J. Electroanal. Chem. 685 (2012) 86e90. [48] E. Laviron, J. Electroanal. Chem. 101 (1979) 19e28. [49] K. Hyun, S.W. Han, W.-G. Koh, Y. Kwon, Int. J. Hydrogen Energ 40 (2015) 2199e2206. [50] J. Liu, A. Chou, W. Rahmat, M.N. Paddon-Row, J.J. Gooding, Electroanalysis 17 (2005) 38e46. [51] W. Wu, J. Huang, Y. Tsai, Mater. Sci. Eng. C 32 (2012) 983e987.