1 Introduction

Mesoporous silica materials (MSMs) have widely been used as adsorbents, thin films, low-k materials, nanowires, and catalysts due to their large specific surface areas and pore sizes ranging from 2 to 50 nm [1]. Some of the most common MSMs known in the market are SBA-15 and M41S-family due to their high surface area, large pore volume, regular pore distribution, and flexible synthesis conditions [2,3,4]. There are different variations of MSMs available with various pore shapes and sizes of both families providing opportunities in catalysis, drug delivery and imaging applications.

MSMs with diverse pore configurations include MCM-41 (hexagonal pores), MCM-48 (cubic pores), and MCM-50 (unstable lamellar pores) [5]. Furthermore, well-known MSMs have space groups of p6mm (MCM-41), Ia3d (MCM-48), p6mm (SBA-15), and Im3m (SBA-16) [6]. Silica materials have a high number of silanol groups on their surface, we can readily adjust the physicochemical nature of the surface through the functionalization process. On the other hand, MSMs have a large surface area and a high pore volume, which means they have a high loading capacity and are unique prospects for various applications [7] such as biomaterials [8], catalyst [9], adsorbent [10], sensor [11], and drug delivery [12].

In the M41S family, the pores of MCM-41 type catalysts are in a two-dimensional hexagonal structure, while the pores of MCM-48 are in a three-dimensional cubic structure [13, 14]. SBA-15, on the other hand, has high thermal stability and a wheat chain structure and SBA-16 is considered the most interesting mesothelium among SBA-type silica materials and is widely used in the field of biomaterials due to its spherical morphology [8, 15]. Different methods are used in the synthesis of MSMs such as impregnation, precipitation/co-precipitation, sol–gel, ion exchange, thermal fusion, solid–liquid leaching, and wet impregnation [16]. However, depending on the thermal strength and the amount of material obtained after synthesis, the hydrothermal synthesis method is widely used in the synthesis of silica-derived mesoporous materials since it can eliminate the use of high-grade materials.

Silica-based materials are known as poor electrical conductors, but they have been utilized in electrochemical systems. There are several examples of the way silica-based materials were employed in these systems such as deposited on conductive electrode surfaces or as thin films, manufactured as metal or carbon composites, and dispersed into conductive composites [17]. Therefore, they can be utilized as an immobilization matrix for selective ligands, electrocatalysts, metal particles, and enzymes [18]. There are also examples of MSMs reinforced with carbon-based materials such as carbon nanomembranes [19] and mesoporous carbon [20] mainly for capacitor applications. This potential of MSMs in electrochemical systems opened up possibilities for their use in immunological sensors, aptasensors, and enzymatic biosensors [21,22,23]. Such systems can offer a variety of opportunities for different applications such as diagnosis of bacterial infections, cancer therapy, cancer detection, detection of heavy metals, contaminated food detection, the detection of viruses, and enzymatic biosensors [22].

The use of MSMs in enzymatic biosensing is a promising application since the critical aspects of enzymatic systems can be supported by MSMs due to their critical properties. MSMs are shown to be effective in preserving the activity of enzymes due to their porous structure providing a protective environment for the enzymes [24, 25]. Furthermore, the high specific pore volume of silica particles can accommodate high enzyme loadings, thus resulting in high current density from enzymatic electrochemical reactions. There are several studies demonstrating the use of MSMs in enzymatic electrochemical systems for the detection of cholesterol [26], catechol [27], ethanol [28], hydrogen peroxide [29], lactic acid [30], uric acid [31], and glucose [28]. Different composites were prepared to increase the conductivity of the MSM-modified electrodes and to retain the enzyme activity for glucose detection such as gold nanoparticles [32, 33], Prussian-blue [34], Nafion [35, 36], and single-walled carbon nanotubes [37].

There are a few studies in the literature demonstrating the use of enzyme-incorporated MSMs for biosensing applications using glucose oxidase (GOx) and laccase. In these studies, Yusan et. al. demonstrated that nanoparticle selenium incorporated MCM-41 could be very effective in retaining enzyme activity [38]. On the other hand, another study by Tvorynska et al. also showed that the MCM-41 incorporated sensor was found to be the most stable sensor configuration for laccase-based biosensors [39]. Most of the studies utilizing MSMs for enzymatic biosensors showed promising performance and good stability, yet there is still a need for comprehensive and systematic studies to demonstrate the effect of the conducting composite materials and the performance of the MSMs on retaining enzyme activity fundamentally.

Herein, four different MSMs from two different types (M41S and SBA) were first synthesized using a hydrothermal method. Then, graphene oxide (GO)-MSM composites were prepared as dispersions of as-synthesized material of optimized amounts into a conductive aqueous GO matrix. Screen-printed electrodes (SPEs) were modified with different material loadings of the prepared composite dispersions and electrochemically characterized using voltammetry and amperometry. The electroanalytical performance of the GOx immobilized electrodes was investigated in terms of the sensitivity and stability of the enzyme. As a result, this study aims to present a comprehensive investigation of the performance of MSMs reinforced with GO in enzymatic electrochemical applications for the first time in the literature. Therefore, it can provide insight into the use of MSMS as additives to carbon-modified electrochemical electrodes (such as graphene, nanotubes, graphite, etc.) to retain enzyme activity.

2 Experimental

2.1 Materials

Pluronic p 123, Pluronic p 127, and the chemicals used in electrochemical studies were obtained at analytical grade from Sigma-Aldrich. Cetyltrimethylammonium bromide (CTMAbr), tetraethyl orthosilicate, (TEOS), and sodium silicate were obtained from Merck.

2.2 Synthesis and Characterization of Mesoporous Silica

The synthesis of MCM-41, SBA-15, MCM-48, and SBA-16 was carried out using the hydrothermal method according to the literature [40,41,42,43]. Briefly, the surfactant was dissolved at different temperatures and durations (MCM-41 and MCM-48; 30 °C, 2–6 h, SBA-15; 40 °C, 2 h, SBA-16; 38 °C, 2 h) using a magnetic stirrer (Elektro-mag, Turkey) following by filtration and drying. The synthesis is then completed with the calcination process at different temperatures to remove impurities in the structure of materials (MCM-41 and MCM-48: 550 °C and 6 h, SBA-15 and SBA-16: 540 °C and 5 h). Although the same hydrothermal synthesis method is used in the synthesis of these mesoporous materials, different chemicals are used as surfactants (CTMAbr, pluronic p 123, and pluronic p 127) and silica sources (TEOS and sodium silicate).

Fourier-transform infrared spectroscopy (FT-IR) analyses of mesoporous materials were performed using the Perkin Elmer IR USA (Attenuated Total Reflectance (ATR) technique) device between 380 and 4000 cm−1. The Panalytical Empryan HT (Netherlands) instrument was used to perform X-ray crystallography (XRD) analyses to determine the structural phases of mesoporous materials using CuK (= 1.540) radiation, 0.066 step size (sensitivity), 30 V (tension), 40 kV (current), and 0° < 2θ < 70° range. The surface morphologies of the samples were analyzed using scanning electron microscopy/energy dispersive X-ray (SEM/EDX, Zeiss SUPRA V40, Germany) analysis. In addition, the MAPPING analysis method was used to determine the distributions of C and Si elements in the structure of the catalyst sample.

2.3 Electrode Preparation and Characterization

All electrochemical experiments were carried out at 23 ± 1 °C using Ivium Potentiostat (Ivium Technologies, Netherlands) and carbon SPE (Model: Dropsens DRP-X1110 with a carbon working electrode surface area of 0.059 cm2, obtained from Metrohm AG, Switzerland). Carbon and silver paste electrodes were used as the counter, and reference electrodes, respectively. SPEs were pre-treated using linear sweep voltammetry (LSV) in a solution containing 0.1 M KCl to remove impurities on the working electrode surface and obtain reproducible results before any experiments [44]. Aqueous dispersions of GO (1 mg/mL, Ultra-pure water, 18.2 MΩ-cm, 4–10% edge oxidized exfoliated graphene nanoplatelets, Sigma-Aldrich) and GO-mesoporous silica mixtures with different mass ratios (1:1, 1:1.5, and 1:2) were prepared and sonicated (Bandelin RK 100 H, Germany) until homogenous mixtures were obtained. Then, the working electrode of the SPEs was drop-coated until a material loading of 0.15 mg/cm2 was achieved [45]. This value was chosen due to physical constraints of the working electrode surface area and above this level, the coating failed. The coated electrodes with different material ratios were then electrochemically characterized using cyclic voltammetry (CV, 50 mV/s) in a solution containing 2 mM K3Fe(CN)6/K4Fe(CN)6 redox couple in 0.1 M KCl. Bare electrode (BE) and GO-only coated SPE were also tested as control experiments. The optimal GO to mesoporous silica amount ratio was chosen based on the anodic and cathodic current changes in CV experiments.

2.4 Enzyme Immobilization and Electrochemical Glucose Oxidation

First, drop-coating 1 µL of ethanoic Nafion solution (0.05% w/w in ethanol) as a supporting layer and dried at room temperature for 15 min. Subsequently, drop-coating 1 µL of GOx [1, 5, and 10 mg/mL in 0.1 M phosphate buffer (PBS), pH 7.4] on Nafion-modified SPEs created an active enzymatic layer for electrochemical glucose oxidation. All prepared electrodes were kept at 4 °C for 24 h and immersed in PBS for 15 min following consecutive washing steps at least 3 times to remove weakly adsorbed species before use. The prepared enzymatic electrodes were denoted as SPE/GO/GOx, SPE/GO:MCM-41/GOx, SPE/GO:MCM-48/GOx, SPE/GO:SBA-15/GOx, and SPE/GO:SBA-16/GOx. The electrochemical glucose oxidation was tested using chronoamperometry (CA) with an applied voltage of 0.14 V for 120 s in 1 mM ferrocene carboxylic acid (FcCOOH, in 0.1 M PBS, pH 7.4) containing 0 and 5 mM glucose. The enzyme concentration for the immobilization was also optimized using LSV (5 mV/s) and used for the electrochemical performance tests. All electrochemical tests were conducted with three independently prepared electrodes unless otherwise stated (N = 3 samples). A schematic representation of electrode preparation steps is shown in Fig. 1.

Fig. 1
figure 1

Schematic representation of electrode preparation steps using MSMs

Full size image

3 Results and Discussion

3.1 Characterization Studies of the Mesoporous Silica Materials

Characterizations of the synthesized MCM-48, MCM-41, SBA-16, and SBA-15 were first performed using FT-IR, XRD, and SEM. FT-IR analysis results of silica-derived mesoporous materials showed that Si–O–Si bands of the silica structure were seen at wavelengths of 1059 cm−1 [46], 1053 cm−1 [47], 1065 cm−1 [15], and 1064 cm−1 [48], respectively (Fig. 2a). The bands of MCM-41 at 965 and 789 cm−1 correspond to the Si–OH and Si–O structures, respectively (Fig. 2a) [49]. In addition, the peaks at 2927 cm−1 and 2857 cm−1 belong to the expansion of the CH(CH2CH2CH2NH2) structure of MCM-48 (Fig. 2a) [46]. –OH stretch can be seen at the 1635 cm−1 band and the water in the structure shows a wide bandgap of around 3368 cm−1 (especially for MCM-41 and MCM-48; Fig. 2a) [50]. The symmetrical stretching vibration mode of O–H of isolated central silanol (Si–OH) groups are represented by the peaks at 3725 cm−1 (SBA-15; Fig. 2a) and 3749 cm−1 (MCM-48; Fig. 2a) [50, 51]. Although shifts were observed in the Bragg fundamental peaks from the XRD analysis result of MCM-41, the basic Bragg peaks of d(100), d(110), and d(200) reflections were obtained. These baseline Bragg peak values obtained showed that MCM-41 had a regular hexagonal structure (Fig. 2b) [50].

Fig. 2
figure 2

a FT-IR analysis b low-angle XRD patterns and c SEM images of MCM-41, MCM-48, SBA-15, SBA-16, GO and GO:MCM-41

Full size image

Due to the SiO2 groups in the structures of silica-based materials (MCM-41, MCM-48, SBA-15, and SBA-16), no clear differences may be observed in the FTIR results. In addition, precise determinations may not be possible due to the wide peaks of OH and Si–O–Si structures. After the modification processes, clear differences can be observed in the FT-IR analysis results depending on the functional groups. A more narrow wavelength result for the FT-IR analyses are also given in Figs. S7 and S8 to emphasize some of the peaks in Fig. 2a.

MCM-48’s main Bragg peaks (d(211) and d(220)) were measured at 2Θ:1.33 and 2.4, respectively (Fig. 2b) [42, 46]. Low-angle XRD analysis showed that the main Bragg peak (d110) of the SBA-16 support material was obtained at 2θ:0.84 (Fig. 2b) [15, 41]. XRD analysis for SBA-15 showed that d(100) and d(110) reflections were observed showing the main Bragg peaks of the mesoporous structure (Fig. 2b) [52].

According to the SEM analysis results, the cubic structure of MCM-48 [53], the hexagonal structure of MCM-41 [53], the spherical structure of SBA-16 [15], and the wheat chain structures of SBA-15 [54] were confirmed (Fig. 2c). Moreover, the EDX and MAPPING analysis were used to determine the distributions of C and Si elements in the structure (Figures S1–4). The SEM image of GO shows the randomly aggregated and crumpled sheets align with the previously reported characteristics [55]. Finally, the surface characteristics of the GO:MCM-41 composite were confirmed with SEM images demonstrating the crumbled GO sheets were wrapped around the hexagonal structure of MCM-41. Thereby, the MCM-41 particles are shown to be covered with GO flakes that would provide electrical conductivity for the prepared films (further confirmed by electrochemical measurements).

3.2 Electrochemical Characterization and Optimization of GO-Mesoporous Silica Ratio

To investigate the behaviour of the GO-mesoporous silica-coated SPEs at different material ratios, a series of CV experiments have been conducted in 0.1 M KCl containing 2 mM K3Fe(CN)6/K4Fe(CN)6. Figure 3 shows the voltammograms and respective anodic and cathodic peak current values with changing GO-mesoporous silica ratio.

Fig. 3
figure 3

CVs (50 mV/s) and anodic and cathodic peak current values of a SPE/GO:MCM-41, b SPE/GO:MCM-48, c SPE/GO:SBA-15, d SPE/GO:SBA-16 in 2 mM K3Fe(CN)6/K4Fe(CN)6 redox couple in 0.1 M KCl, (N = 3 samples)

Full size image

A reversible redox response was observed for BE at ca. 0.18 V and ca. 0.075 V (vs Ag/Ag+) for oxidation and reduction processes, respectively. The voltage separation between anodic and cathodic peaks was ca. 0.1 V (vs Ag/Ag+) and the anodic to cathodic peak current ratio (ipa/ipc) was ca. 1.02 suggesting chemical reversibility and quasi-reversible electron transfer [56]. It can also be seen that the modification of SPEs with GO and GO-MSM at different ratios didn’t cause a significant difference in the electrochemical parameters.. However, different modifications on SPEs made a difference in the catalytic response between electrodes. GO-modified electrode (SPE/GO) showed a significant increase in anodic and cathodic peak current values showing higher peak current values than BE. On the other hand, electrodes modified with the silica materials for all mixing ratios showed lower peak current values than SPE/GO as the integration of electrically insulating silica with GO would be expected the lower the overall conductivity of the composite [57]. However, GO:MCM-41 modified electrode with a 1:1 ratio was the only configuration that resulted in the highest current values. As the pore volume and active surface area of the mesoporous silica increase, the interactions between GO and the silica material would change as well as the electrochemical response due to an increase in total electroactive area. MCM-41 was reported to have a relatively larger surface area and pore-loading capacity than other silica materials used in this study which could be the reason for the better response [40].

3.3 Electrochemical Glucose Oxidation Studies

After the electrochemical investigation of the GO-mesoporous silica materials, enzyme adsorption was performed to evaluate the performance of these compositions in enzymatic glucose oxidation. In this work, the molecular aspects of mesoporous silica materials were not taken into account in the performance of the enzymatic performance, rather it was aimed to compare the performance of the different types of silica materials in the performance of enzymatic glucose oxidation. Therefore a series of CA experiments were performed in 0.1 PBS solution containing 1 mM FcCOOH as an electron transfer mediator. FcCOOH was chosen as a reliable electron transfer mediator in aqueous electrochemistry that was widely used in the literature [58, 59]. The optimization studies for enzyme concentration using LSV revealed that increasing the enzyme concentration for immobilization didn’t cause a significant change in the performance of electrochemical glucose oxidation (Fig. S5). This could be due to the saturation that might be reached in the mesopores of the silica materials, hence further increasing the enzyme concentration wouldn’t make a significant contribution to the catalytic response due to mass transfer limitations. As a result, 1 mg/mL was chosen for the electrochemical experiments. After enzyme loading optimization, CA experiments were performed using modified SPEs with GO, GO:MCM-41, GO:MCM-48, GO:SBA-15, and GO:SBA-16 with a material loading of 0.15 mg/cm2 at a 1:1 ratio to investigate the analytical performance of the prepared electrodes. Figure 4 shows the CA response of SPE/GO/GOx, SPE/GO:MCM-41/GOx, and SPE/GO:MCM-48/GOx with increasing glucose concentration. SPE/GO:SBA-15/GOx and SPE/GO:SBA-16/GOx electrodes were also tested but didn’t show a linear response (Fig. S6). However, MCM-modified electrodes showed a relatively good response to different glucose concentrations up to 7 mM. SPE/GO:MCM-41/GOx showed the best performance among all silica-modified electrodes showing similar performance to GO-only electrodes. In terms of analytical performance the sensitivity of GO, GO:MCM-41, and GO:MCM-48-modified electrodes were calculated as 0.51 µA/mM, 0.39 µA/mM, and 0.17 µA/mM, respectively. These suggest that MCM-modified electrodes showed promising performance compared to other GO-modified electrodes without giving up a significant analytical performance similar to previous investigations without the incorporation of the enzyme. On the other hand, although this study didn’t aim to develop a functional enzymatic biosensor, the following limit of detection (LOD) values were calculated for GO, GO:MCM-41, and GO:MCM-48-modified electrodes as 0.43, 2.62, and 0.80 mM, respectively (LOD: 3.3xSD/Slope) [60].

Fig. 4
figure 4

CA curves (applied voltage: 0.14 V for 120 s) and calibration graphs of (a) and d SPE/GO, b and e SPE/GO:MCM-41, and c and d SPE/GO:MCM-48 electrodes tested in 0.1 M PBS (pH 7.4) containing 1 mM FcCOOH for glucose concentrations between 0 and 7 mM, (N = 3 samples)

Full size image

Enzyme-modified electrodes usually suffer from low stability in vitro and significant effort has been spent to improve the stability of enzymes immobilized on electrodes [61, 62]. In this study, silica was used to help enhance enzyme stability, therefore, a series of inter-day and intra-day experiments have been performed. The anodic peak current values of the LSV experiments were used to investigate the stability of the enzyme immobilized on different modified electrodes. The tests were conducted with SPE/GO/GOx and SPE/GO:MCM-41/GOx electrodes in 0.1 M PBS containing 1 mM FcCOOH and glucose concentrations of 0 and 5 mM.

Figure 5a shows the intraday experiment results consisting of 6 measurements with 1 h intervals. The electrodes were kept in 0.1 M PBS between experiments at room temperature for intraday and + 4 °C for interday experiments and washed with 0.1 PBS before tests. The intraday experimental results show that both electrodes show similar performance, and the results didn’t show a significant change in the current response regardless of glucose concentration. On the other hand, interday experiments for 14 days (tested on days 0, 7, and 14) revealed that the performance of the SPE/GO:MCM41/GOx electrode showed superior performance over the SPE/GO/GOx electrode (Fig. 5b). Table 1 also summarizes the percentage change for intraday and interday stability measurements. The SPE/GO/GOx electrode showed a 40.57% decrease in current whereas the current change for SPE/GO:MCM-41/GOx was ca. 13.75%. All intraday and interday experiments showed a good degree of repeatability with relative standard deviation (RSD %) values less than 10% except for SPE/GO/GOx is 23.65%. This could be due to the unstable behaviour of the SPE/GO/GOx electrode compared to SPE/GO:MCM-41/GOx supporting that silica-based GO composite might provide a more suitable environment for coating and enzyme stability. The RSD values of the intraday and interday stability experiments are presented in Table S1.

Fig. 5
figure 5

Intraday (a) and interday (b) stability experiments of SPE/GO and SPE/GO:MCM-41 electrodes for 0 and 5 mM glucose concentrations. Electrodes were tested in 0.1 M PBS (pH 7.4) containing 1 mM FcCOOH, (N = 3 samples)

Full size image
Table 1 Percentage change for inter-day and intra-day stability experiments
Full size table

4 Conclusion

In this study, an electrochemical investigation of the performance of different GO composite materials prepared using mesoporous silica materials (MCM-41, MCM-48, SBA-15, and SBA-16) on enzymatic glucose oxidation. Optimization studies were conducted to find the optimal GO-to-mesoporous silica ratio and the effect of enzyme loading on the performance of the prepared electrodes. The results showed that the GO to mesoporous silica ratio was found to be 1:1 and the optimum enzyme working concentration was 1 mg/mL. Among all mesoporous silica materials, MCM:41 showed the most promising performance, therefore it was used for the enzymatic investigations. Enzymatic glucose oxidation experiments showed that GO:MCM-41-modified electrodes showed very promising results in detecting different glucose levels at similar sensitivity values. The sensitivity of enzyme-immobilized GO, GO:MCM-41, and GO:MCM-48-modified electrodes were calculated as 0.51 μA/mM, 0.39 μA/mM, and 0.17 μA/mM, respectively. Furthermore, it showed superior stability compared to GO-only modified enzymatic electrodes. Interday experiments revealed that SPE/GO/GOx electrode showed a 40% decrease in current whereas the current change for SPE/GO:MCM-41/GOx was about 14%. This study shows that an optimized amount of composite materials (1:1 for this study) consisting of GO and mesoporous silica, especially MCM-41 due to its relatively larger surface area and pore-loading capacity, can provide a suitable environment for enzyme retention on the surface. These findings could be important for long-term applications of enzymes such as enzymatic biofuel cells and continuous monitoring of glucose where the stability of the enzymes on silica supports such as MCM-family is one of the key factors for the desired performance.