- Email: [email protected]
SnO2 obtained by the sol–gel processing method: application as electrochemical sensor for ascorbic acid
Journal of Colloid and Interface Science 274 (2004) 579–586 www.elsevier.com/locate/jcis
Copper (II) adsorbed on SiO2 /SnO2 obtained by the sol–gel processing method: application as electrochemical sensor for ascorbic acid Lincoln A. Kurihara,a Sergio T. Fujiwara,a Rení V.S. Alfaya,b Yoshitaka Gushikem,a Antonio A.S. Alfaya,b,∗ and Sandra C. de Castro c a Instituto de Química, Unicamp, CP 6154, 13083-970 Campinas, SP, Brazil b Departamento de Química, Universidade Estadual de Londrina—UEL, CP 6001, 86051-990 Londrina, PR, Brazil c Instituto de Física Gleb Wataghin, CP 6165, 13083-970 Campinas, SP, Brazil
Received 4 September 2003; accepted 20 January 2004 Available online 24 March 2004
Abstract With the objective of producing a material showing better conductive properties to be used as a support for electroactive species, a SiO2 /SnO2 mixed oxide was prepared. The procedure for SiO2 /SnO2 mixed oxide preparation using the sol–gel processing method, starting from tetraethylorthosilicate and SnI4 as precursor reagents, is described. SiO2 /SnO2 with composition Sn = 15.6 wt% and SBET = 525 m2 g−1 , Vp = 0.28 ml g−1 , and Dp = 1.5 nm, where SBET , Vp , and Dp are the specific surface area, the average pore volume, and the average pore diameter, respectively, was obtained. The X-ray photoelectron spectroscopy showed that the mixed oxide was thermally very stable for samples heat-treated at up to 1073 K. The Brønsted acid sites, probed with pyridine molecules for samples heat-treated at various temperatures, were chemically stable up to 473 K. Segregation of SnO2 crystalline phase was observed at 1473 K but no crystalline phase was verified for SiO2 at this temperature. The porous SiO2 /SnO2 matrix was used as base for Cu(II) immobilization and an electrode was developed for application in electrochemical detection of vitamin C in tablets. 2004 Elsevier Inc. All rights reserved. Keywords: SiO2 /SnO2 mixed oxide; Sol–gel method; Brønsted and Lewis acid sites; SiO2 /SnO2 thermal stability; Ascorbic acid
1. Introduction SiO2 /Mx Oy mixed oxides have been successfully used as substrates for covalently bonded organic phase for use as alkaline-solution-resistant packing material in HPLC-RP columns [1–5] and as base for immobilization of chemical species for further use as adsorbents [6] and for immobilization of electroactive species for fabrication of electrochemical sensors and biosensors [7–13]. All these substrates have been prepared by coating a thin film of the metal oxide Mx Oy on the porous silica gel surface by a grafting technique [14]. More recently, interest in using SiO2 /Mx Oy prepared by sol–gel processing has gained considerable importance [15–18]. The main difference between materials synthesized by this process is that the sol–gel method provides a mixed oxide in which the metal oxide is dispersed * Corresponding author.
E-mail address: [email protected] (A.A.S. Alfaya). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.01.040
throughout the bulk while in materials obtained by the grafting technique the metal oxide covers only the matrix surface as a monolayer film. The sol–gel process is advantageous when a solid containing a high density of metal ion centers is required. The high specific surface area provided by the silica matrices, mainly those obtained by the sol–gel method, associated with high chemical stability and mechanical resistance, constitute an interesting substrate for preparing a new class of chemically modified electrodes. In order to test the potentiality of this kind of substrate the use of Cu(II), adsorbed by means of an ion exchange reaction on a SiO2 /SnO2 matrix obtained by the sol–gel processing method, in the preparation of an electrode for determining ascorbic acid is described. Methods of ascorbic acid analysis based on Cu(II)-mediated electrochemical detection in homogeneous solution has been described [19]. However, for the first time the study of its oxidation at a solid–solution interface mediated by Cu(II) strongly entrapped in a highly porous matrix is described.
580
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
2. Experimental 2.1. Preparation of the xerogel The precursor reagent SnI4 was prepared following a procedure described in the literature [20]. The solid SnI4 was purified by recrystallization from chloroform and the crystal thus obtained presented a melting point of 416 K, the same as described in the literature. To 100 ml of a tetraethylorthosilicate (TEOS) ethanol 1:1 (v/v) solution was added 5 ml of H2 O and 0.3 ml of concentrated HNO3 and the mixture was refluxed for 2.5 h. To the resulting solution, 200 ml of SnI4 in an ethanol 6.9% (m/v) solution was carefully added and stirred at 298 K for 2 h. Following this, 25 ml of water and 1.1 ml of concentrated HNO3 were added and the solution was stirred for another 2 h at the same temperature. The temperature of the solution was then raised to 333 K and the solvent was slowly evaporated until complete gelation of the mixture occurred. The xerogel obtained was exhaustively washed with pure ethanol until elimination of all iodide present in the solid took place and was heated to 353 K. The solid was ground and sieved into particles with sizes between 0.15 and 0.25 mm.
40 kV, with a current of 30 mA and a sweeping velocity of 2◦ min−1 . The infrared spectra of the sample previously heat-treated for 4 h at various temperatures from 353 up to 1473 K were obtained as KBr pellets (1 wt%). The Lewis and Brønsted acid sites, LAS and BAS, respectively, were studied using pyridine as the molecular probe. The measurements were carried out on a Bomen FTIR MB series spectrophotometer with 4 cm−1 resolution, using 200 cumulative scans. The thermogravimetric analyses were carried out on an thermogravimetric analyzer from TA instruments (Model ITA-5100). The samples were heated to between 300 and 1223 K with heating rate of 5 K min−1 under a flux of argon gas. The X-ray photoelectron spectra were obtained on a McPherson Esca-36 apparatus using an AlKα = 1486.6 eV anode as the excitation source under a pressure of 10−7 Torr. The measurements were calibrated against the C1s binding energy (BE) at 284.6 eV [23]. The atomic ratios were calculated by integrating the areas under the peaks and correcting for the analyzer transmission and the photoionization cross section; the mean free path was assumed to be a function of the kinetic energy [24,25].
2.2. Chemical analysis and physical measurements
2.3. Electrochemical experiments
The amount of Sn(IV) in the matrices was determined by treating the samples (0.5 g) with 40% HF solutions until complete dissolution occurred. About 150 ml of water was added, the metal was precipitated with concentrated NH4 OH solution, and the residue was fired to 1173 K and weighed as SnO2 . N2 adsorption isotherms were determined on an ASAP 2010 apparatus at liquid nitrogen temperature. Initially the sample was degassed at 373 K for 48 h. The specific surface area of the material was obtained using the MP method [21] and the average pore volume was obtained using the Horvath–Kavazoe approximation [22]. About 1 g of SiO2 /SnO2 was immersed in 0.5 mol L−1 CuCl2 ethanol solution and shaken for 24 h at room temperature. The suspension was filtered, washed with 1 × 10−3 mol L−1 HCl aqueous solution and pure water, and dried under vacuum (10−3 Torr) at room temperature. The chemical analysis of the adsorbed copper was made after its release from the solid surface by heating the solid with concentrated HNO3 solution and the metal ion in the solution phase determined using an atomic absorption spectrometer (Perkin–Elmer Model 5100). The SEM images of the samples were obtained by dispersing the powders on double-faced conducting tape fixed on a brass support. The microscope used was a JEOL JSM T-300 connected to a secondary electron detector and to an X-ray dispersive energy analyzer from Northern. X-ray diffraction patterns were obtained on a Shimadzu XRD-6000 using CuKα radiation (λ = 0.15406 nm) at
The electrochemical measurements were carried out on a potentiostat–galvanostat Autolab PGSTAT 20. The carbon paste electrode was prepared by mixing SiO2 /SnO2 /Cu(II) with high-purity-grade graphite (Fluka) in a 1:1 proportion and liquid hydrocarbon was used as the binder. This electrode was used as the working electrode. A platinum wire and saturated calomel were used as counter and reference electrodes, respectively. The electrode was tested in determining ascorbic acid in vitamin C tablets.
3. Results and discussion SiO2 /SnO2 with the following characteristics was obtained: Sn = 15.6 wt%, specific surface area SBET = 525 m2 g−1 , average pore volume of 0.28 ml g−1 , and average pore diameter of 1.5 nm. The amount of Cu2+ adsorbed in the matrix is 10 µmol g−1 . 3.1. Degree of SnO2 dispersion in the xerogel Fig. 1a shows the SEM image of a SiO2 /SnO2 particle, and Fig. 1b shows the corresponding EDS image. The white points in the Fig. 1b are emission lines of SnLα at 3.44 keV [26]. Within the magnification used (×5000), it is clear that the oxide particles are highly dispersed and there is no evidence of metal oxide island formation is detected.
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
581
(a)
(b) Fig. 1. Scanning electron microscopy image of SiO2 /SnO2 (a); the corresponding energy-dispersive scanning (EDS) image of tin (b).
3.2. X-ray photoelectron spectra Table 1 shows the XPS results obtained for the samples previously heat-treated at various temperatures. Under the O1s BE peak it is possible to distinguish between the oxygen atom bonded to the Si atom, designated as O(i) , whose BE is around 532.4 eV, and the oxygen atom bonded to Sn, designated as O(ii) , with a BE around 530.4 eV. The O1s BE for bulk SiO2 (O(i) ) is observed at 532.5 eV and for bulk SnO2 (O(ii) ) around 530.3 eV [27,28]. The Sn3d5/2 BE for SiO2 /SnO2 is observed around 487.1 eV and it is slightly shifted to a higher energy region with respect to the bulk SnO2 material, whose energy is observed at 486.2 eV [28]. This behavior has been observed for highly dispersed SnO2 films on SiO2 substrate [29] and presumably can be explained due to the stronger Sn–O bond polarization upon formation of Sn–O–Si bond. No peak at around 485.6 eV
Table 1 XPS data for SiO2 /SnO2 mixed oxide calcined at various temperatures Samples
Temp.
Binding energy (eV) O1s a
(K) SiO2 /SnO2
Sn3d5/2
Si2p
530.3 (7) 530.1 (6) 530.3 (7) 530.6 (9) 530.4 (6) 530.5 (7) 530.0 (4)
486.9 487.0 487.0 487.0 487.2 487.2 487.3
103.2 102.9 103.2 103.3 103.2 103.4 103.4
530.4 – –
486.2 485.6 –
O(i)
O(ii)
353 473 673 873 1073 1273 1473
532.3 (93) 532.3 (94) 532.3 (93) 532.4 (91) 532.4 (94) 532.4 (93) 532.6 (96)
– – –
– – 532.5
SnO2 b SnOb SiO2 b
a In parentheses: relative percentage of O and O . (i) (ii) b Ref. [28,29].
582
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
Fig. 2. Sn/Si atomic ratios for SiO2 /SnO2 previously heated at various temperatures. Fig. 4. Specific surface area (SBET ) changes of previously heat-treated SiO2 /SnO2 at various temperatures.
Fig. 3. Thermogravimetric analysis curve of SiO2 /SnO2 . Heating rate of 5 K min−1 under argon.
[28], due to the Sn(II) species, normally produced by reduction of Sn(IV) by organic residues entrapped in the matrix during the calcination procedure [30], was detected. The changes of the observed Sn/Si atomic ratios for the various temperatures at which the samples were heat-treated are presented in Fig. 2. It was observed that the Sn/Si atomic ratios remained nearly constant between 353 and 1100 K, indicating a small thermal mobility of the atoms in this range of the temperature for both samples. Above 1100 K, a decrease of the atomic ratio is observed, due to the diffusion of Sn(IV) into the matrix.
Fig. 5. X-ray diffraction patterns of SiO2 /SnO2 previously heat-treated at various temperatures (K): (a) 353, (b) 473, (c) 673, (d) 873, (e) 1073, (f) 1273, and (g) 1473.
Dehydration of the solids is followed by a decrease of the specific surface areas, as observed in Fig. 4. The change is more significant in the temperature region where hydration and structural water are lost. Between 650 and 1100 K, the areas remain nearly constant and, above this region, the solids become non porous. 3.4. XRD patterns
3.3. Thermal stability and specific surface area The TGA curve (Fig. 3), illustrated in the present case for SiO2 /SnO2 , shows a weight loss of about 23% for temperatures from 353 up to 800 K. In this range of the thermal treatment, the weight decrease is due to hydration water and structural water losses, the latter arising from SiOH and SnOH polycondensation [31]. Above 800 K, the weight loss remained constant up to 1200 K.
The X-ray diffraction patterns show that the samples as obtained (Fig. 5a) and heat-treated up to 1073 K (Figs. 5b– 5e) remain amorphous. At 1273 K (Fig. 5f), some diffraction peaks appear, and at 1473 K (Fig. 5g), they become more intense. These diffraction peaks match those of the separated SnO2 crystalline tetragonal phase (JCPDS (21-1250)) [32]. No peaks assignable to the SiO2 crystalline phase are observed.
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
Fig. 6. Infrared spectra of SiO2 /SnO2 previously heat-treated at various temperature (K): (a) 353, (b) 473, (c) 673, (d) 873, (e) 1073, (f) 1273, and (g) 1473.
3.5. IR spectra The spectra for the SiO2 /SnO2 sample, as presented in Fig. 6, shows absorption bands (in cm−1 ) assigned as follows: 1070 (νSi–O–Sisym), 793 (νSi–O–Siasym ), 962 (νSi–O of SiOH group), and 452 (δSi–O–Si) [33]. The weak band at 565 cm−1 (Figs. 6a), whose intensity decreases as the sample is treated at higher temperatures and practically disappears at 1073 K (Figs. 6e), was assigned to νSn–O of the Sn–OH bond [34,35]. This assignment is consistent with the SnOH polycondensation that occurs above this temperature. Similarly, the band at 960 cm−1 (Figs. 6a, 6b) practically disappears at 673 K (Fig. 6c) due to SiOH polycondensation. At temperatures of 1273 K (Fig. 6f), a band at 672 cm−1 is developed and it is intensified at 1473 K (Fig. 6g). This band is assigned to the νSn–O–Sn [34] of the SnO2 crystalline phase, as was shown by the XRD patterns (Fig. 5g). 3.6. Acid properties The Sn(IV) in the silica matrix can present both Brønsted and Lewis acid sites, the first being due mainly to the SnOH groups and the second one arising from the highly dispersed coordinatively unsaturated metals in the matrix. Pyridine was used as a molecular probe to map the Lewis and Brønsted acid sites, LAS and BAS, respectively. Fig. 7 shows the pyridine-adsorbed vibrational modes for samples heat-treated at various temperatures. The absorption band at 1547 cm−1 is assigned to the 19b mode of a pyridine mole-
583
Fig. 7. Infrared spectra of pyridine adsorbed on SiO2 /SnO2 heated at various temperatures (K), for 30 min, under pressure of 1.3 × 10−2 Pa: (a) 353, (b) 373, (c) 423, and (d) 473.
cule adsorbed on BAS. This mode is observed for samples heated up to 473 K, indicating that BAS are thermally very stable. The band at 1614 cm−1 , which is observed for samples heated up to 423 K (Fig. 7c), is assigned to the 8a mode of a pyridine adsorbed on BAS [36–38]. The band at 1450 cm−1 is assigned to the 19b mode of a pyridine bonded to Si–OH groups by hydrogen bonding. This band is observed for pyridine adsorbed on a pure silica surface and normally disappears when the sample is heat-treated at 373 K. In the present case, the 19b mode for samples heat-treated at temperatures below 473 K (Figs. 7a, 7b) is overlapped with that at 1460 cm−1 , assigned to the 19b mode of pyridine adsorbed on LAS, which is clearly observable in Fig. 7d upon heat-treatment at 473 K. The band observed at 1490 cm−1 is assigned to the 19b mode of pyridine and this band does not shift when this molecule is adsorbed on different kinds of acid sites. 3.7. Electrochemical studies The SiO2 /SnO2 is a stable solid state Brønsted acid, as demonstrated by pyridine adsorbed IR spectra, and thus, the adsorption of Cu2+ on the surface can be represented by the ion exchange reaction 2SnOH + Cu2+ (SnO)2 Cu + 2H+ , where SnOH stands for hydrated SnO2 bonded to the silica matrix. The ions exchange may involve only SnOH groups, pKa = 4.8, since SiOH is a very weak acid, pKa = 7.1 [39]. Fig. 8 show the cyclic voltammetry curves obtained by using the SiO2 /SnO2 /Cu(II) carbon paste electrode immersed in a solution containing variable concentration of
584
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
Fig. 8. Cyclic voltammetry curves for SiO2 /SnO2 /Cu(II) in KCl 0.5 mol L−1 solutions at pH 7 (buffer of TRIS) in different concentrations of ascorbic acid: (a) absence of ascorbic acid, (b) 9.0 × 10−4 mol L−1 , (c) 1.7 × 10−3 mol L−1 , and (d) 2.3 × 10−3 mol L−1 .
Fig. 9. Cyclic voltammetry curves for SiO2 /SnO2 /Cu(II) in KCl 0.5 mol L−1 solutions at different pH. Concentration of ascorbic acid is 1.7 × 10−3 mol L−1 . Solutions pH: (a) 8.6, (b) 7.6, (c) 6.7, (d) 5.2, (e) 4.3, (f) 3.4, and (g) 2.5.
ascorbic acid. The experiment was carried out at pH 7.0 [tris(hydroxymethyl)aminemethane buffer] in 0.5 mol L−1 KCl electrolyte solution. In the absence of the ascorbic acid, the anodic current intensity corresponding to the Cu(I) → Cu(II) oxidation at 0.11 V is very weak and it is enhanced as the ascorbic acid is introduced in the solution. The oxidation potential of the ascorbic acid on platinum electrode, in a similar condition, occurs at 0.4 V vs SCE [40] indicating that on the SiO2 /SnO2 /Cu(II) electrode, the oxidation is mediated by Cu(II). The anodic peak potential of the ascorbic acid is largely affected by the solution pH. Its pKa is 4.1 [41] and the oxidation to the dehydroascorbic acid on the electrode surface is represented by the equation
Fig. 10. Anodic peak current, ipa , against number of redox cycles for SiO2 /SnO2 /Cu(II) in solution at pH 7, ascorbic acid is 1.7×10−3 mol L−1 and electrolyte solution of KCl is 0.5 mol L−1 .
Between pH 8.6 and 5.2, the anodic peak potential practically is not affected, but at lower pH (Fig. 9), it shifts to higher values in accordance with the electrode surface reaction equation. An important characteristic of the electrode is that it is very stable under the various redox cycles. When the anodic peak currents, ipa , are plotted against number of redox cycles (Fig. 10), the intensities of the peak currents practically remain the same, indicating that Cu2+ ion is not released to the solution phase, at least up to 140 cycles. Considering that the experiments are carried out in KCl 0.5 mol L−1 solution, we conclude that the metal ion is strongly entrapped in the matrix pores. When anodic peak current, ipa , is plotted against the square root of the scan rate (Fig. 11) a linear relationship is obtained, a result that is similar to those observed for diffusion-controlled processes [42], indicating that the
process is controlled by the diffusion of ascorbic acid to the electrode–solution interface [43,44]. This mechanism is consistent since the electron transfer process must involve metal-to-ascorbic acid interaction. 3.8. Amount of ascorbic acid in real samples In order to test the analytical potentiality of the carbon paste electrode as an electrochemical sensor for ascorbic acid in real samples, vitamin C in tablets were determined by using the chronoamperometry technique. Fig. 12 shows the plot of current intensity change (i) against concentration of the ascorbic acid and the inserted figure shows the plot of the current against time for various additions of the ascorbic acid. A linear correlation i = 0.0182 + 0.268[Ascorbic acid] with r = 0.999 was obtained. For determination of ascorbic acid in the tablets, the samples were prepared as indicated by the standard procedure
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
585
Table 2 Determination of ascorbic acid tablets using the SiO2 /SnO2 /Cu(II) carbon paste electrode and compared with the standard method Samples
Declared amounta (g)
Standard methodb (g)
Using the electrodec (g)
A B C D E
1.0 1.0 1.0 1.0 0.5
1.041 ± 0.004 1.038 ± 0.004 0.990 ± 0.005 0.989 ± 0.005 0.476 ± 0.004
1.05 ± 0.04 1.04 ± 0.05 1.01 ± 0.04 0.95 ± 0.04 0.48 ± 0.02
a Declared amount (for five different samples on the market). b Standard method using dpip reagent. c Mean values of five determinations for each sample with confidence
interval of 95%.
Fig. 11. ipa against v 1/2 (v is the scan velocity) for SiO2 /SnO2 /Cu(II) immersed in KCl 0.5 mol L−1 and ascorbic acid is 1.7 × 10−3 mol L−1 .
Fig. 12. Variation of the anodic peak current, i, for SiO2 /SnO2 /Cu(II) immersed in KCl 0.5 mol L−1 buffered solution with TRIS at pH 7. The chronoamperograms were obtained for successive addition of 100 µl of ascorbic acid (1 × 10−2 mol L−1 ). In the inserted figure, i is plotted against time, t.
method and the values obtained compared with that obtained by using 2,6-dichlorophenilindofenol reagent [45] for validation. The results are summarized in Table 2. The results obtained agree well with those obtained using standard methods for validation.
4. Conclusions The SiO2 /SnO2 mixed oxide, homogeneously distributed, was obtained as a very stable solid Brønsted acid. The specific surface areas decreased upon heat-treatment but the
Sn/Si atomic ratios remained nearly constant up to a temperature of 1073 K. Decreases of the surface areas are observed with the temperature increase but diffusion of SnO2 to the interior of the SiO2 matrices was observed only at temperatures above 1273 K. The BAS, probed with the pyridine molecule, are thermally very stable up temperatures of 473 K. The pyridine mode adsorbed on a LAS, clearly observable at this temperature, is very stable because SnO2 particles are highly and homogeneously dispersed in the matrices and are not very thermally mobile as shown by XPS measurements. The SiO2 /SnO2 material presents high porosity and Brønsted acid sites with high thermal stability due to the uniform distribution of the SnO2 particles in the silica matrix, making possible Cu(II) ion entrapment by ion-exchange reaction, both in the surface and in the pores of the material. Electrochemical studies show that the Cu(II) species immobilized directly in the matrix is bound strongly by the Brønsted acid sites. The utilization of this material as a sensor for ascorbic acid is very interesting, as this is the first time that the study of ascorbic acid oxidation has been done with the Cu(II) species immobilized in this form. Considering the analysis made in real samples to determine of ascorbic acid with the carbon paste electrode and by the standard method, it is observed that the material SiO2 /SnO2 /Cu(II) is potentially useful as a new sensor.
Acknowledgments YG is indebted to FAPESP for financial support and Professor Carol H. Collins for English revision. AASA and LAK are indebted to CNPq, STF to FAPESP, and RVSA to Capes/PICD, for fellowships.
References [1] R.B. Silva, C.H. Collins, J. Chromatogr. A 845 (1999) 417. [2] L.F.C. Melo, I.C.S.F. Jardim, J. Chromatogr. A 845 (1999) 423. [3] R.B. Silva, K.E. Collins, C.H. Collins, J. Chromatogr. A 869 (2000) 137.
586
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
[4] L.F.C. Melo, C.H. Collins, K.E. Collins, I.C.S.F. Jardim, J. Chromatogr. A 869 (2000) 129. [5] R.B. Silva, Y. Gushikem, C.H. Collins, J. Sep. Sci. 24 (2001) 1. [6] S.T. Fujiwara, Y. Gushikem, R.V.S. Alfaya, Colloids Surf. A 178 (2001) 135. [7] E.F. Perez, G.D. Neto, L.T. Kubota, Sens. Actuators B 72 (2001) 80. [8] S.S. Rosatto, L.T. Kubota, G.D. Neto, Anal. Chim. Acta 390 (1999) 65. [9] L.T. Kubota, Y. Gushikem, J. Perez, A.A. Tanaka, Langmuir 11 (1995) 1009. [10] Y. Gushikem, C.R.M. Peixoto, U.P. Rodrigues Filho, L.T. Kubota, E. Stadler, J. Colloid Interface Sci. 184 (1996) 236. [11] G.A.P. Zaldivar, Y. Gushikem, L.T. Kubota, J. Electroanal. Chem. 318 (1991) 247. [12] G.A.P. Zaldivar, Y. Gushikem, J. Electroanal. Chem. 337 (1992) 165. [13] G.A.P. Zaldivar, Y. Gushikem, E.V. Benvenutti, S.C. de Castro, A. Vasquez, Electrochim. Acta 39 (1994) 33. [14] Y. Gushikem, S.S. Rosatto, J. Braz. Chem. Soc. 12 (2001) 695. [15] A.A.S. Alfaya, Y. Gushikem, S.C. de Castro, Chem. Mater. 10 (1998) 909. [16] A. Walcarius, Electroanalysis 10 (1998) 1217. [17] A. Walcarius, Chem. Mater. 13 (2001) 3351. [18] A. Walcarius, Electroanalysis 13 (2001) 701. [19] A. Sano, T. Kuwayama, M. Furukawa, S. Takitani, H. Nakamura, Anal. Sci. 11 (1995) 405. [20] T. Moeller, D.C. Edwards, in: J.C. Bailar Jr. (Ed.), Inorganic Syntheses, McGraw–Hill, London, 1953, p. 119. [21] P.A. Webb, C. Orr, Analytical Methods in Fine Particle Technology, Micromeritics Instrument Corp., Norcross, 1997, p. 71. [22] G. Horvath, K. Kawazoe, J. Chem. Eng. Jpn. 16 (1983) 470. [23] P. Swift, Surf. Interface Anal. 4 (1982) 47. [24] J.H. Scofied, J. Electron Spectrosc. Relat. Phenom. 8 (1976) 129. [25] C. Powell, J. Surf. Sci. 44 (1974) 29. [26] P.J. Goodhew, F.J. Humpheys, Electron Microscopy and Analysis, second ed., Taylor & Francis, London, 1988.
[27] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in: J. Chastain (Ed.), Handbook of X-ray Photoelectron Spectroscopy, Perkin–Elmer Corporation, Physical Electronics Division, Eden Prairie, 1992, p. 126. [28] J.M. Themlin, M. Chtaib, L. Henrard, P. Lambin, J. Darville, J.M. Gilles, Phys. Rev. B Condens. Matter Mater. Phys. 46 (1992) 2460. [29] V.M. Jimenez, J.A. Mejias, J.P. Espinos, A.R. Gonzalez-Elipe, Surf. Sci. 366 (1996) 545. [30] G. Carturan, R. Ceccato, G. Principi, U. Russo, J. Radioanal. Nucl. Chem. 190 (1995) 419. [31] C.J. Brinker, G.W. Schere, Sol–Gel Science: The Physics and Chemistry of Sol–Gel Processing, Academic Press, San Diego, 1990. [32] JCPDS—International Centre for Diffraction Data, PCPDFWIN, version 1.30, 1997. [33] Y. Kawano, S. Denofre, Y. Gushikem, Vib. Spectrosc. 7 (1994) 293. [34] B. Orel, U. Lavrencic-Stangar, Z. Crnjak-Orel, P. Bukovec, M. Kosec, J. Non-Cryst. Solids 167 (1994) 272. [35] B. Yu, L. Guo, Z. Yang, C. Zhu, F. Gan, G. Zhang, G. Tang, X. Wu, W. Chen, Phys. Lett. A 251 (1999) 67. [36] E.P. Parry, J. Catal. 2 (1963) 371. [37] E.V. Benvenutti, Y. Guhikem, C.U. Davanzo, S.C. de Castro, I.L. Torriani, J. Chem. Soc. Faraday Trans. 88 (1992) 3193. [38] C. Morterra, G. Cerrato, Langmuir 6 (1990) 1810. [39] K. Tanabe, M. Misono, Y. Ono, H. Hattori, in: B. Delmon, J.T. Yates (Eds.), New Solid Acids and Bases—Their Catalytic Properties, Kodansha–Elsevier, Tokyo, 1989. [40] A. Ivaska, Z. Gao, T. Zha, G. Wang, P. Li, Z. Zhao, Talanta 40 (1993) 399. [41] S.P. Mushran, M.C. Agrawal, J. Sci. Ind. Res. India 36 (1977) 274. [42] A.J. Bard, L.R. Faulkner, Electroanalytical Chemistry, Dekker, New York, 1982, p. 53. [43] C.A. Pessoa, Y. Gushikem, L.T. Kubota, L. Gorton, J. Electroanal. Chem. 431 (1997) 23. [44] C.A. Pessoa, Y. Gushikem, L.T. Kubota, Electroanalysis 9 (1997) 800. [45] J. Lindquist, Analyst 100 (1975) 339.
Copper (II) adsorbed on SiO2 /SnO2 obtained by the sol–gel processing method: application as electrochemical sensor for ascorbic acid Lincoln A. Kurihara,a Sergio T. Fujiwara,a Rení V.S. Alfaya,b Yoshitaka Gushikem,a Antonio A.S. Alfaya,b,∗ and Sandra C. de Castro c a Instituto de Química, Unicamp, CP 6154, 13083-970 Campinas, SP, Brazil b Departamento de Química, Universidade Estadual de Londrina—UEL, CP 6001, 86051-990 Londrina, PR, Brazil c Instituto de Física Gleb Wataghin, CP 6165, 13083-970 Campinas, SP, Brazil
Received 4 September 2003; accepted 20 January 2004 Available online 24 March 2004
Abstract With the objective of producing a material showing better conductive properties to be used as a support for electroactive species, a SiO2 /SnO2 mixed oxide was prepared. The procedure for SiO2 /SnO2 mixed oxide preparation using the sol–gel processing method, starting from tetraethylorthosilicate and SnI4 as precursor reagents, is described. SiO2 /SnO2 with composition Sn = 15.6 wt% and SBET = 525 m2 g−1 , Vp = 0.28 ml g−1 , and Dp = 1.5 nm, where SBET , Vp , and Dp are the specific surface area, the average pore volume, and the average pore diameter, respectively, was obtained. The X-ray photoelectron spectroscopy showed that the mixed oxide was thermally very stable for samples heat-treated at up to 1073 K. The Brønsted acid sites, probed with pyridine molecules for samples heat-treated at various temperatures, were chemically stable up to 473 K. Segregation of SnO2 crystalline phase was observed at 1473 K but no crystalline phase was verified for SiO2 at this temperature. The porous SiO2 /SnO2 matrix was used as base for Cu(II) immobilization and an electrode was developed for application in electrochemical detection of vitamin C in tablets. 2004 Elsevier Inc. All rights reserved. Keywords: SiO2 /SnO2 mixed oxide; Sol–gel method; Brønsted and Lewis acid sites; SiO2 /SnO2 thermal stability; Ascorbic acid
1. Introduction SiO2 /Mx Oy mixed oxides have been successfully used as substrates for covalently bonded organic phase for use as alkaline-solution-resistant packing material in HPLC-RP columns [1–5] and as base for immobilization of chemical species for further use as adsorbents [6] and for immobilization of electroactive species for fabrication of electrochemical sensors and biosensors [7–13]. All these substrates have been prepared by coating a thin film of the metal oxide Mx Oy on the porous silica gel surface by a grafting technique [14]. More recently, interest in using SiO2 /Mx Oy prepared by sol–gel processing has gained considerable importance [15–18]. The main difference between materials synthesized by this process is that the sol–gel method provides a mixed oxide in which the metal oxide is dispersed * Corresponding author.
E-mail address: [email protected] (A.A.S. Alfaya). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.01.040
throughout the bulk while in materials obtained by the grafting technique the metal oxide covers only the matrix surface as a monolayer film. The sol–gel process is advantageous when a solid containing a high density of metal ion centers is required. The high specific surface area provided by the silica matrices, mainly those obtained by the sol–gel method, associated with high chemical stability and mechanical resistance, constitute an interesting substrate for preparing a new class of chemically modified electrodes. In order to test the potentiality of this kind of substrate the use of Cu(II), adsorbed by means of an ion exchange reaction on a SiO2 /SnO2 matrix obtained by the sol–gel processing method, in the preparation of an electrode for determining ascorbic acid is described. Methods of ascorbic acid analysis based on Cu(II)-mediated electrochemical detection in homogeneous solution has been described [19]. However, for the first time the study of its oxidation at a solid–solution interface mediated by Cu(II) strongly entrapped in a highly porous matrix is described.
580
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
2. Experimental 2.1. Preparation of the xerogel The precursor reagent SnI4 was prepared following a procedure described in the literature [20]. The solid SnI4 was purified by recrystallization from chloroform and the crystal thus obtained presented a melting point of 416 K, the same as described in the literature. To 100 ml of a tetraethylorthosilicate (TEOS) ethanol 1:1 (v/v) solution was added 5 ml of H2 O and 0.3 ml of concentrated HNO3 and the mixture was refluxed for 2.5 h. To the resulting solution, 200 ml of SnI4 in an ethanol 6.9% (m/v) solution was carefully added and stirred at 298 K for 2 h. Following this, 25 ml of water and 1.1 ml of concentrated HNO3 were added and the solution was stirred for another 2 h at the same temperature. The temperature of the solution was then raised to 333 K and the solvent was slowly evaporated until complete gelation of the mixture occurred. The xerogel obtained was exhaustively washed with pure ethanol until elimination of all iodide present in the solid took place and was heated to 353 K. The solid was ground and sieved into particles with sizes between 0.15 and 0.25 mm.
40 kV, with a current of 30 mA and a sweeping velocity of 2◦ min−1 . The infrared spectra of the sample previously heat-treated for 4 h at various temperatures from 353 up to 1473 K were obtained as KBr pellets (1 wt%). The Lewis and Brønsted acid sites, LAS and BAS, respectively, were studied using pyridine as the molecular probe. The measurements were carried out on a Bomen FTIR MB series spectrophotometer with 4 cm−1 resolution, using 200 cumulative scans. The thermogravimetric analyses were carried out on an thermogravimetric analyzer from TA instruments (Model ITA-5100). The samples were heated to between 300 and 1223 K with heating rate of 5 K min−1 under a flux of argon gas. The X-ray photoelectron spectra were obtained on a McPherson Esca-36 apparatus using an AlKα = 1486.6 eV anode as the excitation source under a pressure of 10−7 Torr. The measurements were calibrated against the C1s binding energy (BE) at 284.6 eV [23]. The atomic ratios were calculated by integrating the areas under the peaks and correcting for the analyzer transmission and the photoionization cross section; the mean free path was assumed to be a function of the kinetic energy [24,25].
2.2. Chemical analysis and physical measurements
2.3. Electrochemical experiments
The amount of Sn(IV) in the matrices was determined by treating the samples (0.5 g) with 40% HF solutions until complete dissolution occurred. About 150 ml of water was added, the metal was precipitated with concentrated NH4 OH solution, and the residue was fired to 1173 K and weighed as SnO2 . N2 adsorption isotherms were determined on an ASAP 2010 apparatus at liquid nitrogen temperature. Initially the sample was degassed at 373 K for 48 h. The specific surface area of the material was obtained using the MP method [21] and the average pore volume was obtained using the Horvath–Kavazoe approximation [22]. About 1 g of SiO2 /SnO2 was immersed in 0.5 mol L−1 CuCl2 ethanol solution and shaken for 24 h at room temperature. The suspension was filtered, washed with 1 × 10−3 mol L−1 HCl aqueous solution and pure water, and dried under vacuum (10−3 Torr) at room temperature. The chemical analysis of the adsorbed copper was made after its release from the solid surface by heating the solid with concentrated HNO3 solution and the metal ion in the solution phase determined using an atomic absorption spectrometer (Perkin–Elmer Model 5100). The SEM images of the samples were obtained by dispersing the powders on double-faced conducting tape fixed on a brass support. The microscope used was a JEOL JSM T-300 connected to a secondary electron detector and to an X-ray dispersive energy analyzer from Northern. X-ray diffraction patterns were obtained on a Shimadzu XRD-6000 using CuKα radiation (λ = 0.15406 nm) at
The electrochemical measurements were carried out on a potentiostat–galvanostat Autolab PGSTAT 20. The carbon paste electrode was prepared by mixing SiO2 /SnO2 /Cu(II) with high-purity-grade graphite (Fluka) in a 1:1 proportion and liquid hydrocarbon was used as the binder. This electrode was used as the working electrode. A platinum wire and saturated calomel were used as counter and reference electrodes, respectively. The electrode was tested in determining ascorbic acid in vitamin C tablets.
3. Results and discussion SiO2 /SnO2 with the following characteristics was obtained: Sn = 15.6 wt%, specific surface area SBET = 525 m2 g−1 , average pore volume of 0.28 ml g−1 , and average pore diameter of 1.5 nm. The amount of Cu2+ adsorbed in the matrix is 10 µmol g−1 . 3.1. Degree of SnO2 dispersion in the xerogel Fig. 1a shows the SEM image of a SiO2 /SnO2 particle, and Fig. 1b shows the corresponding EDS image. The white points in the Fig. 1b are emission lines of SnLα at 3.44 keV [26]. Within the magnification used (×5000), it is clear that the oxide particles are highly dispersed and there is no evidence of metal oxide island formation is detected.
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
581
(a)
(b) Fig. 1. Scanning electron microscopy image of SiO2 /SnO2 (a); the corresponding energy-dispersive scanning (EDS) image of tin (b).
3.2. X-ray photoelectron spectra Table 1 shows the XPS results obtained for the samples previously heat-treated at various temperatures. Under the O1s BE peak it is possible to distinguish between the oxygen atom bonded to the Si atom, designated as O(i) , whose BE is around 532.4 eV, and the oxygen atom bonded to Sn, designated as O(ii) , with a BE around 530.4 eV. The O1s BE for bulk SiO2 (O(i) ) is observed at 532.5 eV and for bulk SnO2 (O(ii) ) around 530.3 eV [27,28]. The Sn3d5/2 BE for SiO2 /SnO2 is observed around 487.1 eV and it is slightly shifted to a higher energy region with respect to the bulk SnO2 material, whose energy is observed at 486.2 eV [28]. This behavior has been observed for highly dispersed SnO2 films on SiO2 substrate [29] and presumably can be explained due to the stronger Sn–O bond polarization upon formation of Sn–O–Si bond. No peak at around 485.6 eV
Table 1 XPS data for SiO2 /SnO2 mixed oxide calcined at various temperatures Samples
Temp.
Binding energy (eV) O1s a
(K) SiO2 /SnO2
Sn3d5/2
Si2p
530.3 (7) 530.1 (6) 530.3 (7) 530.6 (9) 530.4 (6) 530.5 (7) 530.0 (4)
486.9 487.0 487.0 487.0 487.2 487.2 487.3
103.2 102.9 103.2 103.3 103.2 103.4 103.4
530.4 – –
486.2 485.6 –
O(i)
O(ii)
353 473 673 873 1073 1273 1473
532.3 (93) 532.3 (94) 532.3 (93) 532.4 (91) 532.4 (94) 532.4 (93) 532.6 (96)
– – –
– – 532.5
SnO2 b SnOb SiO2 b
a In parentheses: relative percentage of O and O . (i) (ii) b Ref. [28,29].
582
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
Fig. 2. Sn/Si atomic ratios for SiO2 /SnO2 previously heated at various temperatures. Fig. 4. Specific surface area (SBET ) changes of previously heat-treated SiO2 /SnO2 at various temperatures.
Fig. 3. Thermogravimetric analysis curve of SiO2 /SnO2 . Heating rate of 5 K min−1 under argon.
[28], due to the Sn(II) species, normally produced by reduction of Sn(IV) by organic residues entrapped in the matrix during the calcination procedure [30], was detected. The changes of the observed Sn/Si atomic ratios for the various temperatures at which the samples were heat-treated are presented in Fig. 2. It was observed that the Sn/Si atomic ratios remained nearly constant between 353 and 1100 K, indicating a small thermal mobility of the atoms in this range of the temperature for both samples. Above 1100 K, a decrease of the atomic ratio is observed, due to the diffusion of Sn(IV) into the matrix.
Fig. 5. X-ray diffraction patterns of SiO2 /SnO2 previously heat-treated at various temperatures (K): (a) 353, (b) 473, (c) 673, (d) 873, (e) 1073, (f) 1273, and (g) 1473.
Dehydration of the solids is followed by a decrease of the specific surface areas, as observed in Fig. 4. The change is more significant in the temperature region where hydration and structural water are lost. Between 650 and 1100 K, the areas remain nearly constant and, above this region, the solids become non porous. 3.4. XRD patterns
3.3. Thermal stability and specific surface area The TGA curve (Fig. 3), illustrated in the present case for SiO2 /SnO2 , shows a weight loss of about 23% for temperatures from 353 up to 800 K. In this range of the thermal treatment, the weight decrease is due to hydration water and structural water losses, the latter arising from SiOH and SnOH polycondensation [31]. Above 800 K, the weight loss remained constant up to 1200 K.
The X-ray diffraction patterns show that the samples as obtained (Fig. 5a) and heat-treated up to 1073 K (Figs. 5b– 5e) remain amorphous. At 1273 K (Fig. 5f), some diffraction peaks appear, and at 1473 K (Fig. 5g), they become more intense. These diffraction peaks match those of the separated SnO2 crystalline tetragonal phase (JCPDS (21-1250)) [32]. No peaks assignable to the SiO2 crystalline phase are observed.
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
Fig. 6. Infrared spectra of SiO2 /SnO2 previously heat-treated at various temperature (K): (a) 353, (b) 473, (c) 673, (d) 873, (e) 1073, (f) 1273, and (g) 1473.
3.5. IR spectra The spectra for the SiO2 /SnO2 sample, as presented in Fig. 6, shows absorption bands (in cm−1 ) assigned as follows: 1070 (νSi–O–Sisym), 793 (νSi–O–Siasym ), 962 (νSi–O of SiOH group), and 452 (δSi–O–Si) [33]. The weak band at 565 cm−1 (Figs. 6a), whose intensity decreases as the sample is treated at higher temperatures and practically disappears at 1073 K (Figs. 6e), was assigned to νSn–O of the Sn–OH bond [34,35]. This assignment is consistent with the SnOH polycondensation that occurs above this temperature. Similarly, the band at 960 cm−1 (Figs. 6a, 6b) practically disappears at 673 K (Fig. 6c) due to SiOH polycondensation. At temperatures of 1273 K (Fig. 6f), a band at 672 cm−1 is developed and it is intensified at 1473 K (Fig. 6g). This band is assigned to the νSn–O–Sn [34] of the SnO2 crystalline phase, as was shown by the XRD patterns (Fig. 5g). 3.6. Acid properties The Sn(IV) in the silica matrix can present both Brønsted and Lewis acid sites, the first being due mainly to the SnOH groups and the second one arising from the highly dispersed coordinatively unsaturated metals in the matrix. Pyridine was used as a molecular probe to map the Lewis and Brønsted acid sites, LAS and BAS, respectively. Fig. 7 shows the pyridine-adsorbed vibrational modes for samples heat-treated at various temperatures. The absorption band at 1547 cm−1 is assigned to the 19b mode of a pyridine mole-
583
Fig. 7. Infrared spectra of pyridine adsorbed on SiO2 /SnO2 heated at various temperatures (K), for 30 min, under pressure of 1.3 × 10−2 Pa: (a) 353, (b) 373, (c) 423, and (d) 473.
cule adsorbed on BAS. This mode is observed for samples heated up to 473 K, indicating that BAS are thermally very stable. The band at 1614 cm−1 , which is observed for samples heated up to 423 K (Fig. 7c), is assigned to the 8a mode of a pyridine adsorbed on BAS [36–38]. The band at 1450 cm−1 is assigned to the 19b mode of a pyridine bonded to Si–OH groups by hydrogen bonding. This band is observed for pyridine adsorbed on a pure silica surface and normally disappears when the sample is heat-treated at 373 K. In the present case, the 19b mode for samples heat-treated at temperatures below 473 K (Figs. 7a, 7b) is overlapped with that at 1460 cm−1 , assigned to the 19b mode of pyridine adsorbed on LAS, which is clearly observable in Fig. 7d upon heat-treatment at 473 K. The band observed at 1490 cm−1 is assigned to the 19b mode of pyridine and this band does not shift when this molecule is adsorbed on different kinds of acid sites. 3.7. Electrochemical studies The SiO2 /SnO2 is a stable solid state Brønsted acid, as demonstrated by pyridine adsorbed IR spectra, and thus, the adsorption of Cu2+ on the surface can be represented by the ion exchange reaction 2SnOH + Cu2+ (SnO)2 Cu + 2H+ , where SnOH stands for hydrated SnO2 bonded to the silica matrix. The ions exchange may involve only SnOH groups, pKa = 4.8, since SiOH is a very weak acid, pKa = 7.1 [39]. Fig. 8 show the cyclic voltammetry curves obtained by using the SiO2 /SnO2 /Cu(II) carbon paste electrode immersed in a solution containing variable concentration of
584
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
Fig. 8. Cyclic voltammetry curves for SiO2 /SnO2 /Cu(II) in KCl 0.5 mol L−1 solutions at pH 7 (buffer of TRIS) in different concentrations of ascorbic acid: (a) absence of ascorbic acid, (b) 9.0 × 10−4 mol L−1 , (c) 1.7 × 10−3 mol L−1 , and (d) 2.3 × 10−3 mol L−1 .
Fig. 9. Cyclic voltammetry curves for SiO2 /SnO2 /Cu(II) in KCl 0.5 mol L−1 solutions at different pH. Concentration of ascorbic acid is 1.7 × 10−3 mol L−1 . Solutions pH: (a) 8.6, (b) 7.6, (c) 6.7, (d) 5.2, (e) 4.3, (f) 3.4, and (g) 2.5.
ascorbic acid. The experiment was carried out at pH 7.0 [tris(hydroxymethyl)aminemethane buffer] in 0.5 mol L−1 KCl electrolyte solution. In the absence of the ascorbic acid, the anodic current intensity corresponding to the Cu(I) → Cu(II) oxidation at 0.11 V is very weak and it is enhanced as the ascorbic acid is introduced in the solution. The oxidation potential of the ascorbic acid on platinum electrode, in a similar condition, occurs at 0.4 V vs SCE [40] indicating that on the SiO2 /SnO2 /Cu(II) electrode, the oxidation is mediated by Cu(II). The anodic peak potential of the ascorbic acid is largely affected by the solution pH. Its pKa is 4.1 [41] and the oxidation to the dehydroascorbic acid on the electrode surface is represented by the equation
Fig. 10. Anodic peak current, ipa , against number of redox cycles for SiO2 /SnO2 /Cu(II) in solution at pH 7, ascorbic acid is 1.7×10−3 mol L−1 and electrolyte solution of KCl is 0.5 mol L−1 .
Between pH 8.6 and 5.2, the anodic peak potential practically is not affected, but at lower pH (Fig. 9), it shifts to higher values in accordance with the electrode surface reaction equation. An important characteristic of the electrode is that it is very stable under the various redox cycles. When the anodic peak currents, ipa , are plotted against number of redox cycles (Fig. 10), the intensities of the peak currents practically remain the same, indicating that Cu2+ ion is not released to the solution phase, at least up to 140 cycles. Considering that the experiments are carried out in KCl 0.5 mol L−1 solution, we conclude that the metal ion is strongly entrapped in the matrix pores. When anodic peak current, ipa , is plotted against the square root of the scan rate (Fig. 11) a linear relationship is obtained, a result that is similar to those observed for diffusion-controlled processes [42], indicating that the
process is controlled by the diffusion of ascorbic acid to the electrode–solution interface [43,44]. This mechanism is consistent since the electron transfer process must involve metal-to-ascorbic acid interaction. 3.8. Amount of ascorbic acid in real samples In order to test the analytical potentiality of the carbon paste electrode as an electrochemical sensor for ascorbic acid in real samples, vitamin C in tablets were determined by using the chronoamperometry technique. Fig. 12 shows the plot of current intensity change (i) against concentration of the ascorbic acid and the inserted figure shows the plot of the current against time for various additions of the ascorbic acid. A linear correlation i = 0.0182 + 0.268[Ascorbic acid] with r = 0.999 was obtained. For determination of ascorbic acid in the tablets, the samples were prepared as indicated by the standard procedure
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
585
Table 2 Determination of ascorbic acid tablets using the SiO2 /SnO2 /Cu(II) carbon paste electrode and compared with the standard method Samples
Declared amounta (g)
Standard methodb (g)
Using the electrodec (g)
A B C D E
1.0 1.0 1.0 1.0 0.5
1.041 ± 0.004 1.038 ± 0.004 0.990 ± 0.005 0.989 ± 0.005 0.476 ± 0.004
1.05 ± 0.04 1.04 ± 0.05 1.01 ± 0.04 0.95 ± 0.04 0.48 ± 0.02
a Declared amount (for five different samples on the market). b Standard method using dpip reagent. c Mean values of five determinations for each sample with confidence
interval of 95%.
Fig. 11. ipa against v 1/2 (v is the scan velocity) for SiO2 /SnO2 /Cu(II) immersed in KCl 0.5 mol L−1 and ascorbic acid is 1.7 × 10−3 mol L−1 .
Fig. 12. Variation of the anodic peak current, i, for SiO2 /SnO2 /Cu(II) immersed in KCl 0.5 mol L−1 buffered solution with TRIS at pH 7. The chronoamperograms were obtained for successive addition of 100 µl of ascorbic acid (1 × 10−2 mol L−1 ). In the inserted figure, i is plotted against time, t.
method and the values obtained compared with that obtained by using 2,6-dichlorophenilindofenol reagent [45] for validation. The results are summarized in Table 2. The results obtained agree well with those obtained using standard methods for validation.
4. Conclusions The SiO2 /SnO2 mixed oxide, homogeneously distributed, was obtained as a very stable solid Brønsted acid. The specific surface areas decreased upon heat-treatment but the
Sn/Si atomic ratios remained nearly constant up to a temperature of 1073 K. Decreases of the surface areas are observed with the temperature increase but diffusion of SnO2 to the interior of the SiO2 matrices was observed only at temperatures above 1273 K. The BAS, probed with the pyridine molecule, are thermally very stable up temperatures of 473 K. The pyridine mode adsorbed on a LAS, clearly observable at this temperature, is very stable because SnO2 particles are highly and homogeneously dispersed in the matrices and are not very thermally mobile as shown by XPS measurements. The SiO2 /SnO2 material presents high porosity and Brønsted acid sites with high thermal stability due to the uniform distribution of the SnO2 particles in the silica matrix, making possible Cu(II) ion entrapment by ion-exchange reaction, both in the surface and in the pores of the material. Electrochemical studies show that the Cu(II) species immobilized directly in the matrix is bound strongly by the Brønsted acid sites. The utilization of this material as a sensor for ascorbic acid is very interesting, as this is the first time that the study of ascorbic acid oxidation has been done with the Cu(II) species immobilized in this form. Considering the analysis made in real samples to determine of ascorbic acid with the carbon paste electrode and by the standard method, it is observed that the material SiO2 /SnO2 /Cu(II) is potentially useful as a new sensor.
Acknowledgments YG is indebted to FAPESP for financial support and Professor Carol H. Collins for English revision. AASA and LAK are indebted to CNPq, STF to FAPESP, and RVSA to Capes/PICD, for fellowships.
References [1] R.B. Silva, C.H. Collins, J. Chromatogr. A 845 (1999) 417. [2] L.F.C. Melo, I.C.S.F. Jardim, J. Chromatogr. A 845 (1999) 423. [3] R.B. Silva, K.E. Collins, C.H. Collins, J. Chromatogr. A 869 (2000) 137.
586
L.A. Kurihara et al. / Journal of Colloid and Interface Science 274 (2004) 579–586
[4] L.F.C. Melo, C.H. Collins, K.E. Collins, I.C.S.F. Jardim, J. Chromatogr. A 869 (2000) 129. [5] R.B. Silva, Y. Gushikem, C.H. Collins, J. Sep. Sci. 24 (2001) 1. [6] S.T. Fujiwara, Y. Gushikem, R.V.S. Alfaya, Colloids Surf. A 178 (2001) 135. [7] E.F. Perez, G.D. Neto, L.T. Kubota, Sens. Actuators B 72 (2001) 80. [8] S.S. Rosatto, L.T. Kubota, G.D. Neto, Anal. Chim. Acta 390 (1999) 65. [9] L.T. Kubota, Y. Gushikem, J. Perez, A.A. Tanaka, Langmuir 11 (1995) 1009. [10] Y. Gushikem, C.R.M. Peixoto, U.P. Rodrigues Filho, L.T. Kubota, E. Stadler, J. Colloid Interface Sci. 184 (1996) 236. [11] G.A.P. Zaldivar, Y. Gushikem, L.T. Kubota, J. Electroanal. Chem. 318 (1991) 247. [12] G.A.P. Zaldivar, Y. Gushikem, J. Electroanal. Chem. 337 (1992) 165. [13] G.A.P. Zaldivar, Y. Gushikem, E.V. Benvenutti, S.C. de Castro, A. Vasquez, Electrochim. Acta 39 (1994) 33. [14] Y. Gushikem, S.S. Rosatto, J. Braz. Chem. Soc. 12 (2001) 695. [15] A.A.S. Alfaya, Y. Gushikem, S.C. de Castro, Chem. Mater. 10 (1998) 909. [16] A. Walcarius, Electroanalysis 10 (1998) 1217. [17] A. Walcarius, Chem. Mater. 13 (2001) 3351. [18] A. Walcarius, Electroanalysis 13 (2001) 701. [19] A. Sano, T. Kuwayama, M. Furukawa, S. Takitani, H. Nakamura, Anal. Sci. 11 (1995) 405. [20] T. Moeller, D.C. Edwards, in: J.C. Bailar Jr. (Ed.), Inorganic Syntheses, McGraw–Hill, London, 1953, p. 119. [21] P.A. Webb, C. Orr, Analytical Methods in Fine Particle Technology, Micromeritics Instrument Corp., Norcross, 1997, p. 71. [22] G. Horvath, K. Kawazoe, J. Chem. Eng. Jpn. 16 (1983) 470. [23] P. Swift, Surf. Interface Anal. 4 (1982) 47. [24] J.H. Scofied, J. Electron Spectrosc. Relat. Phenom. 8 (1976) 129. [25] C. Powell, J. Surf. Sci. 44 (1974) 29. [26] P.J. Goodhew, F.J. Humpheys, Electron Microscopy and Analysis, second ed., Taylor & Francis, London, 1988.
[27] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in: J. Chastain (Ed.), Handbook of X-ray Photoelectron Spectroscopy, Perkin–Elmer Corporation, Physical Electronics Division, Eden Prairie, 1992, p. 126. [28] J.M. Themlin, M. Chtaib, L. Henrard, P. Lambin, J. Darville, J.M. Gilles, Phys. Rev. B Condens. Matter Mater. Phys. 46 (1992) 2460. [29] V.M. Jimenez, J.A. Mejias, J.P. Espinos, A.R. Gonzalez-Elipe, Surf. Sci. 366 (1996) 545. [30] G. Carturan, R. Ceccato, G. Principi, U. Russo, J. Radioanal. Nucl. Chem. 190 (1995) 419. [31] C.J. Brinker, G.W. Schere, Sol–Gel Science: The Physics and Chemistry of Sol–Gel Processing, Academic Press, San Diego, 1990. [32] JCPDS—International Centre for Diffraction Data, PCPDFWIN, version 1.30, 1997. [33] Y. Kawano, S. Denofre, Y. Gushikem, Vib. Spectrosc. 7 (1994) 293. [34] B. Orel, U. Lavrencic-Stangar, Z. Crnjak-Orel, P. Bukovec, M. Kosec, J. Non-Cryst. Solids 167 (1994) 272. [35] B. Yu, L. Guo, Z. Yang, C. Zhu, F. Gan, G. Zhang, G. Tang, X. Wu, W. Chen, Phys. Lett. A 251 (1999) 67. [36] E.P. Parry, J. Catal. 2 (1963) 371. [37] E.V. Benvenutti, Y. Guhikem, C.U. Davanzo, S.C. de Castro, I.L. Torriani, J. Chem. Soc. Faraday Trans. 88 (1992) 3193. [38] C. Morterra, G. Cerrato, Langmuir 6 (1990) 1810. [39] K. Tanabe, M. Misono, Y. Ono, H. Hattori, in: B. Delmon, J.T. Yates (Eds.), New Solid Acids and Bases—Their Catalytic Properties, Kodansha–Elsevier, Tokyo, 1989. [40] A. Ivaska, Z. Gao, T. Zha, G. Wang, P. Li, Z. Zhao, Talanta 40 (1993) 399. [41] S.P. Mushran, M.C. Agrawal, J. Sci. Ind. Res. India 36 (1977) 274. [42] A.J. Bard, L.R. Faulkner, Electroanalytical Chemistry, Dekker, New York, 1982, p. 53. [43] C.A. Pessoa, Y. Gushikem, L.T. Kubota, L. Gorton, J. Electroanal. Chem. 431 (1997) 23. [44] C.A. Pessoa, Y. Gushikem, L.T. Kubota, Electroanalysis 9 (1997) 800. [45] J. Lindquist, Analyst 100 (1975) 339.