Introduction

Metallic nanoparticles (MNPs) have attracted considerable attention recently because of several studies dealing with their synthesis, modification, properties, and applications. This interest is mainly due to the remarkable and unique properties presented by these particles that are not available in their bulk equivalent, e.g., high surface/volume ratios and quantum confinement effects. The use of these properties in technological applications has enabled exponential advancements in materials science and has led to the discovery of new ways of handling and manufacturing materials at the nano level [1–4].

The importance of gold nanoparticles (nAu) is highlighted by a wide range of applications in areas such as controlled drug delivery systems [5], electronic and optical devices [6], medical diagnosis [7], catalysis [1], sensors [8], and biosensors [9]. In particular, the physicochemical properties of nAu make them very attractive and suitable for the modification of conventional electrodes and the development of new electrode materials. The introduction of MNPs in electrochemical sensors and biosensors is a result of their large specific surface area, good biocompatibility, and excellent catalytic properties, which significantly decrease the overpotential of many electrochemical reactions [9–11].

Although NPs provide a significantly higher number of electroactive centers in electroanalysis than the bulk metals, the NP free surface energy often leads to particle growth or aggregation and the loss of important properties achieved at the nano level. To avoid the inherent instability associated with particles of this size, the combination of synthesis procedures and effective MNP immobilization methods is extremely important. Porous solids with rigid frameworks, such as carbonaceous materials and inorganic oxides like silica, alumina, and titania, have been considered as alternatives to support and stabilize the MNPs [12, 13]. Most studies featuring supported metal nanoparticles have been related to heterogeneous catalysis because the rigid framework of porous solids usually offers high thermal and chemical stability and a large NP surface area, which are important properties for catalytic applications.

In the field of electroanalysis, sol–gel methodology has proven to be a useful tool in the preparation of silica-based functional materials [14–16], ranging from pure inorganic and mixed oxides [17, 18] to the more chemically complex hybrid organic–inorganic solids [19], which have been successfully applied in electroanalysis as working electrode materials.

Ascorbic acid (H2AA), dopamine (DA), and uric acid (UA) are electroactive compounds of great biological and chemical interest that play important roles in human health. Sensing of these analytes is of great interest since H2AA, DA, and UA coexist in the extracellular fluid of the central nervous system and serum. As they have similar oxidation potentials at most conventional electrodes, separate determination of these species is difficult due to their overlapped signals [20, 21]. Thus, the development of efficient electrodes with enhanced characteristics to distinguish these species in mixtures is a very important subject.

Striving to combine the remarkable properties of MNPs and the benefits offered by porous solids, the present work describes the synthesis, characterization, and application of a system consisting of nAu supported on the thiol groups (−SH) of an organofunctionalized silica network synthesized by sol–gel process.

Experimental

Reagents and solutions

All reagents used were of analytical grade and were used without any further purification. Phosphate buffer solutions (0.10 mol L−1) were prepared using phosphoric acid and potassium hydroxide. The stock solutions of dopamine, ascorbic acid, and uric acid were prepared in the phosphate buffer solutions described above.

Preparation of gold nanoparticles

An aqueous chloroauric acid solution (95.0 mL) containing 5.0 mg of Au was brought to a boil and vigorously stirred in a round-bottom flask fitted with a reflux condenser. After it, 5.0 mL of aqueous 1 wt.% sodium citrate solution was added to the flask. This mixture was refluxed for 30 min. The solution changed color from pale yellow to deep red. The system was allowed to cool to room temperature with continuous stirring and was stored in a dark bottle until further use. This method yields quasi-spherical particles with an average diameter of about 10 nm [22, 23].

Synthesis of SiO2/MPTS/Au

The thiol-functionalized silica network was prepared using the sol–gel process according to the following procedure. First, 10.0 mL of tetraethylorthosilicate was initially pre-hydrolyzed in ethanol solution (1:1 v/v) with 3.2 mL of distilled water; this reaction was catalyzed by HCl. The solution was stirred and heated at 353 K for 5 h. Next, 4.8 mL of (3-mercaptopropyl)-trimethoxysilane was added to this mixture, and then the mixture was stirred for 3 h at 353 K. At room temperature, 2.0 mL of the sol containing nAu was added to this solution. The temperature of the system was raised to 333 K until gel formation and then heated to 353 K to complete the solvent evaporation. The xerogel was washed with ethanol in a Soxhlet extractor and dried under vacuum. The obtained material was designated as SiO2/MPTS/Au. Thiol-functionalized silica (SiO2/MPTS) was also prepared by the same procedure in the absence of the AuNP sol.

Characterization

The FT IR spectrum of the material was obtained using a pressed KBr disk (1.0 wt.%) on a Bomem MB series FT IR with a 4 cm−1 resolution and 50 cumulative scans. Solid-state nuclear magnetic resonance spectroscopy for 13 C (CP/MAS NMR) was performed on a Bruker AC300/P spectrometer using pulse sequences with 4 ms contact time, an interval of 1 s between pulses, and an acquisition time of 41 ms. UV–vis absorption spectra were collected on a UV–vis NIR Cary-5 2300 spectrometer equipped with a diffuse reflectance accessory, and barium sulfate was used as a reference. Transmission electron microscopy (TEM) images were obtained on JEOL JEM-1220 microscope operating at 120 kV.

The electrochemical measurements were carried out using a PGSTAT 128 N Autolab® potentiostat–galvanostat. All experiments were carried out in a conventional three-electrode system with a platinum wire and a saturated calomel electrode as the counter and reference electrodes, respectively. The working electrode was prepared by mixing SiO2/MPTS/Au and graphite powder in a 3:2 (wt) proportion to obtain an adequate consistency and electrochemical behavior. The mixture was deposited into a cavity (1.0 mm in depth) in contact with a platinum disk (0.50 cm in diameter) fused at the end of a glass tube. The gold, platinum, and vitreous carbon electrodes have a geometric area of 0.15 cm2. The following instrument parameters were used to record the square wave voltammograms: 10 mV square wave amplitude, 10 Hz frequency, and 19.5 mV step potential.

Results and discussion

Characterization of the SiO2/MPTS/Au

The main characteristic absorption bands of SiO2/MPTS/Au (Fig. 1a) are in the 3,020 − 2,800 cm−1 range and can be attributed to the C − H asymmetric and symmetric stretching vibrations of the − CH2 groups. The spectrum also shows C − H deformation of − CH2 groups at 1,440 cm−1 and a weak band at 2,560 cm−1 assigned to the S − H stretching vibration. The other absorption bands were the same for the absorption spectrum of silica gel (Fig. 1b) and can be assigned as follows: 1,200–1,100 cm−1 (ν AS Si − O − Si), 920 cm−1 (ν S Si − O of the silanol group), 795 cm−1 (ν S Si − O of the siloxane group), 690 cm−1 (ν S Si − O − Si), and 456 cm−1 (δ Si − O − Si). A broad absorption band between 3,750 and 3,000 cm−1 is assigned to both the O − H stretching mode of silanol groups and the remaining adsorbed water [24, 25].

Fig. 1
figure 1

Infrared spectra of (a) SiO2/MPTS/Au and (b) silica gel

Full size image

The solid state 13 C CP-MAS NMR spectra for SiO2/MPTS/Au show three signals with specific peak positions at 13.2, 19.2, and 29.3 ppm, which can be assigned according to the carbon atoms numbered in the inset [26, 27]. The weak signals at 49.2 and 61.5 ppm are assigned to unhydrolyzed ethoxy and methoxy groups, respectively [28] (Fig. 2).

Fig. 2
figure 2

13C CP-MAS NMR of SiO2/MPTS/Au

Full size image

The optical properties of metal nanoparticles can be correlated with nanoparticle size, shape, and dielectric environment [3]. The UV–vis spectrum of nAu sol was obtained before incorporation of nAu in the SiO2/MTPS material (Fig. 3a). This result displays a single surface plasmon resonance (SPR) band of Au nanoparticles, centered at 525 nm, which corresponds to a particle diameter smaller than 30 nm [29]. The inset in Fig. 3a presents the transmission electron microscopy of gold nanoparticles in colloidal solution. The image shows that nAu are quasi-spherical with a narrow size distribution. Diffuse reflectance spectroscopy measurements were carried out to detect the presence of nAu at prepared material. The UV–vis spectrum of SiO2/MPTS (Fig. 3b) shows no absorption band. However, the SiO2/MPTS/Au spectrum (Fig. 3b) displays a single SPR band centered at 540 nm. This result indicates that nAu were immobilized on the functionalized silica network. The verified shift, as compared to sol spectrum, can be attributed to a different medium permittivity at the solid state and the interaction between the nAu and sulfur in the functionalized silica network [29–31].

Fig. 3
figure 3

a UV–vis spectra for gold nanoparticles sol. Inset, TEM of nAu sol. b Diffuse reflectance absorption spectra of SiO2/MPTS (dashed line) and SiO2/MPTS/Au (solid line)

Full size image

Electrochemical study of the SiO2/MPTS/Au

The redox behavior of an electroactive species such as a [Fe(CN)6]3−/4− couple is a valuable tool for evaluating the kinetic barrier of the electrode/solution interface [32].

The cyclic voltammetry response shows that no faradaic current was observed at curves a and b in Fig. 4 because there is no species in the solution that can be oxidized or reduced within this potential range. The electrochemical responses obtained for the SiO2/MPTS and SiO2/MPTS/Au electrodes were similar. In contrast, curves c and d in Fig. 4 show the oxidation peak of [Fe(CN)6]4− ions at 0.29 V and the corresponding reduction peak at 0.06 V. The SiO2/MPTS/Au electrode showed a 0.23 V peak potential separation (ΔE p = E ANODIC PEAK − E CATHODIC PEAK) and a 0.978 current ratio (I ANODIC PEAK/I CATHODIC PEAK). A significant current increase related to the [Fe(CN)6]3−/4− redox processes was also observed when the nAu were bonded onto the functionalized silica network.

Fig. 4
figure 4

Cyclic voltammograms (CVs) for SiO2/MPTS (a, c) and SiO2/MPTS/Au (b, d) electrodes in the absence (a, b) and presence (c, d) of 5.0 mmol L−1 K3Fe(CN)6 obtained with a 20 mV s−1 scan rate in 0.10 mol L−1 KCl solution

Full size image

The peak current is directly proportional to the square root of the scan rate as shown in Fig. 5. This behavior indicates that the process is controlled by [Fe(CN)6]3−/4− diffusion to the electrode/solution interface. Therefore, the slope of the two curves obtained from the SiO2/MPTS/Au electrode (Fig. 5b) is larger than the slope of the curves from the SiO2/MPTS electrode (Fig. 5a). This can be related to the presence of nAu immobilized on a functionalized silica network that favors electron transfer at the electrode/solution interface. Also, one can observe that as the scan rate increased, the peak current related to the redox processes of [Fe(CN)6]3−/4− also increased. Moreover, the peaks at high scan rates are not well defined for the SiO2/MPTS electrode (Fig. 5a), unlike the peaks for the SiO2/MPTS/Au electrode (Fig. 5b).

Fig. 5
figure 5

Cyclic voltammograms for a SiO2/MPTS and b SiO2/MPTS/Au electrodes in 0.10 mol L−1 KCl containing 5.0 mmol L−1 K3Fe(CN)6 obtained with scan rates between 10 and 500 mV s−1. The insets show the dependence of the peak current on the square root of the scan rate

Full size image

Electrocatalytic oxidation of ascorbic acid

Electrocatalytic oxidation of H2AA was studied to evaluate the potential use of SiO2/MPTS/Au material for electrochemical sensor.

Significant diminution of the oxidation potential occurs at the SiO2/MPTS as compared to the values for conventional electrodes (Fig. 6a, b). The peak of current density in the anodic scan at approximately 0.190 V (Fig. 6b) can be assigned to the electro-oxidation of H2AA. [33–35]. In addition, the current density at oxidation potential was approximately ten times higher for the SiO2/MPTS electrode than for the bare platinum electrode, probably due to porosity of the material. Higher oxidation current density was observed for the electrode containing nAu, which can be attributed to an increase of the surface area due to the presence of the nanoparticles.

Fig. 6
figure 6

a Cyclic voltammograms for platinum (dashed line), glassy carbon (solid line), gold electrodes (dash-dotted line), and b SiO2/MPTS (dashed line) and SiO2/MPTS/Au (solid line) electrodes in 0.10 mol L−1 phosphate buffer solution (pH 2.0) containing 2.0 mmol L−1 H2AA. SiO2/MPTS/Au (dash-dotted line) without H2AA. All CV are obtained with a 10 mV s−1 scan rate

Full size image

In most cases, the solution pH is an important factor in the electrochemical reaction. Cyclic voltammetry was performed to study the solution pH effects on the electro-oxidation behavior of H2AA, DA, and UA at the SiO2/MPTS/Au electrode. The decrease of oxidation peak potential and current density with pH was observed (Fig.  7b–e). For the three analytes present simultaneously in solution, the highest current density for the oxidation process and the larger peak separation was observed at pH 2 (Fig.  7f). Considering these results, a 0.10 mol L−1 phosphate buffer solution with pH 2.0 was chosen as a supporting electrolyte for the experiments on electrochemical detection of H2AA. The redox process for the species studied involves proton and electron transfer, which can explain the changes of the peak potential and current density with solution pH [36–38].

Fig. 7
figure 7

a Cyclic voltammograms for SiO2/MPTS/Au electrode from pH 2 to 10, b containing 2.0 mmol L−1 of H2AA, c containing 1.0 mmol L−1 of DA, d containing 1.0 mmol L−1 of UA. e Oxidation peak current density for the three analytes in all studied pH. f Containing the three analytes simultaneously. All CVs are obtained with a 10 mV s−1 scan rate

Full size image

The oxidation current (Fig. 8) presented a linear behavior within the two H2AA concentration ranges: 0.050 to 4.0 and 4.0 to 10.0 mmol L−1. The inset in the figure depicts the analytical curve obtained by the SiO2/MPTS/Au electrode in the two concentration ranges, with correlation coefficients of 0.9959 for n = 13 and 0.9989 for n = 7. The decrease in sensitivity in the second linear range is attributed to a kinetic limitation at the electrode surface [39, 40].

Fig. 8
figure 8

Cyclic voltammograms for the SiO2/MPTS/Au electrode in 0.10 mol L−1 phosphate buffer solution (pH 2.0) with different H2AA concentrations: (a) 0, (b) 0.0060, (c) 0.020, (d) 0.050, (e) 0.10, (f) 0.20, (g) 0.40, (h) 0.60, (i) 0.80, (j) 1.0, (l) 2.0, (k) 3.0, (m) 4.0, (n) 5.0, (o) 6.0, (p) 7.0, (q) 8.0, (r) 9.0, and (s) 10.0 mmol L−1. A scan rate of 10 mV s−1 was used. Inset, plot of the anodic peak current density (j) versus ascorbic acid concentration [H2AA]

Full size image

The reference voltammogram measured in the absence of H2AA was subtracted from the voltammograms in the presence of the analyte to obtain a better resolved ascorbic acid oxidation peak. A detection limit of 0.0160 mmol L−1 was determined according to IUPAC recommendations [41]. The SiO2/MPTS/Au electrode showed an anodic potential and a detection limit for H2AA determination that was similar to other researchers’ work [40, 42–44].

Chronoamperometric study of ascorbic acid

The response time of the SiO2/MPTS/Au electrode observed in Fig. 9 was about 1.0 s. This prompt response is followed by decay in the current density as observed after each ascorbic acid addition. This behavior is associated with the slow diffusion of ascorbic acid to the electrode surface before the electron transfer process [45]. In Fig. 9b inset, a linear response of the SiO2/MPTS/Au electrode can be observed with a correlation coefficient of 0.9982 for n = 45 in the concentration range between 0.0050 and 0.423 mmol L−1. The detection limit determined was 2.842 μmol L−1 [41]. These values were comparable to results from other chronoamperometric sensors for H2AA [46–48].

Fig. 9
figure 9

a Chronoamperometric curves obtained for the SiO2/MPTS/Au electrode after successive additions of 20 μL of 0.025 mol L−1 H2AA in 50 mL of 0.10 mol L−1 phosphate buffer solution (pH 2.0). Applied potential = 0.180 V. Inset, chronoamperometric curve without H2AA. b Plot of anodic current density (j) versus ascorbic acid concentration [H2AA]

Full size image

Determination of ascorbic acid in the presence of DA and UA

For the conventional electrodes (Fig. 10a), the oxidation responses for the three analytes cannot be defined clearly; however, they are observed at higher potential than for SiO2/MPTS and SiO2/MPTS/Au electrodes. The three anodic peaks at 0.184, 0.494, and 0.626 V are well defined and symmetrical, which can be attributed to the oxidation processes of H2AA, DA, and UA, respectively (Fig. 10b). A peak around 0.380 V in the cathodic scan was attributed to the reduction of the byproduct resulting from the DA oxidation. The peaks of interest are well separated using the SiO2/MPTS and SiO2/MPTS/Au electrodes. The separation of the anodic peaks obtained for H2AA/DA and H2AA/UA are presented in Table 1 and compared with the data for other modified electrodes. The separation of peak potentials for H2AA, DA, and UA could be attributed to their different adsorption affinities at the electrode surface; the SiO2/MPTS/Au electrode yielded higher redox currents for the analytes probably due to the high surface area of the nAu.

Fig. 10
figure 10

a Cyclic voltammograms for platinum (dashed line), glass carbon (solid line), gold electrodes (dash-dotted line) and b for SiO2/MPTS (solid line) and SiO2/MPTS/Au (dashed line) electrodes in 0.10 mol L−1 phosphate buffer solution (pH 2.0) containing 2.0 mmol L−1 H2AA, 1.0 mmol L−1 DA, and 1.0 mmol L−1 UA. A scan rate of 10 mV s−1 was used

Full size image
Table 1 Comparison of ΔE pa(DA-H2AA) and ΔE pa(UA-H2AA) obtained by cyclic voltammetry for different modified electrodes
Full size table

Comparing Figs. 6b and 10b, one can notice a significant decrease in the oxidation current of H2AA in the presence of DA and UA. A possible explanation can be related to a stronger interaction between nAu and dopamine/uric acid due to the NH2 and NH groups in these compounds, respectively, leading to a higher affinity of these species at the electrode surface than for the ascorbic acid.

The determination of H2AA, DA, and UA in solution was performed by changing the concentration of one compound while the other two concentrations were kept constant. This study was performed using the square wave voltammetry technique to obtain higher current sensitivity and better resolution. The results are shown in Fig. 11.

Fig. 11
figure 11

Square wave voltammograms of a H2AA concentrations: 0.050 − 1.0 mmol L−1, with SiO2/MPTS/Au electrodes in the presence of 20.0 μmol L−1 DA and 20.0 μmol L−1 UA. b DA concentrations: 0.20 − 20.0 μmol L−1, with SiO2/MPTS/Au electrodes in the presence of 2.0 mmol L−1 H2AA and 20.0 μmol L−1 UA. c UA concentrations: 0.50 − 40.0 μmol L−1, with SiO2/MPTS/Au electrodes in the presence of 2.0 mmol L−1 H2AA and 20.0 μmol L−1 DA. Supporting electrolyte: 0.10 mol L−1 phosphate buffer solution (pH 2.0). Square wave voltammogram in the absence of the analytes (dash line)

Full size image

As can be seen in Fig. 11a, the H2AA current peak increases proportionally with the H2AA concentration when the concentrations of DA and UA were kept constant. No clear change in the DA and UA oxidation currents was observed while varying the H2AA concentration. Similarly, the oxidation peak current of DA and UA increased proportionally with the increase in the DA or UA concentration, keeping the concentration of the other two compounds constant (as shown in Fig. 11b, c). Interestingly, the oxidation processes of H2AA, DA, and UA at the SiO2/MPTS/Au electrode could be observed simultaneously, and no peak overlapping was observed. The analytical parameters for the simultaneous determination of H2AA, DA, and UA for the electrode prepared are listed and compared with the values for different modified electrodes in Table 2.

Table 2 Comparison between determination range and detection limit of different modified electrodes for the simultaneous determination of H2AA, DA, and UA
Full size table

The detection limits were compared with results obtained by other researchers, and this comparison indicates that the SiO2/MPTS/Au electrode can be used for simultaneous determination of H2AA, DA, and UA [42–44].

Conclusions

SiO2/MPTS with immobilized nAu was obtained by a simple preparation procedure using sol–gel chemistry. The gold nanoparticles in a functionalized silica network change the electrochemical behavior of the SiO2/MPTS/Au electrode, favoring the electron transfer process of the studied species. These characteristics are desirable when this type of electrode is prepared for use in electrocatalytic and electroanalytical studies.

The SiO2/MPTS/Au electrode was efficient in H2AA determination as verified by cyclic voltammetry, chronoamperometry, and square wave voltammetry techniques. The voltammetric response to the oxidation process of H2AA in the presence of DA and UA showed an adequate anodic peak separation for simultaneous detection of the analytes. The electrode prepared has shown promising characteristics to be used as an electrochemical sensor.