Boletín de la Sociedad Chilena de Química
versão impressa ISSN 0366-1644
Bol. Soc. Chil. Quím. v.45 n.2 Concepción jun. 2000
POTENTIOMETRIC RESPONSE OF A GRAPHITE ELECTRODE
MODIFIED WITH COBALT PHTHALOCYANINE FOR THIOLS AND
JOSÉ H. ZAGAL.* AND JAIME J.H.HENRIQUEZ
Departamento de Química de los Materiales, Facultad de Química y Biología,
Universidad de Santiago de Chile, Casilla 40, Correo 33, Santiago, Chile.
(Received: February 23, 2000 - Accepted: March 30,2000)
In memoriam of Doctor Guido S. Canessa C.
Se ha investigado la respuesta potenciométrica the electrodos de grafito pirolítico ordinario (OPG) modificados con ftalocianina de cobalto (Co-Pc) frente a los tioles (R-SH) 2-mecaptoetanol y L-cisteína y los correspondientes disulfuros (R-SS-R). Al agregar pequeñas cantidades de estos tioles a soluciones acuosas con pH en el rango 11-4, se obtienen respuestas en potencial estables dentro de unos pocos segundos despues de las adiciones. Gráficas de potencial vs. log [R-SH] dan rectas para todos los casos, con pendientes cercanas a -0,060 V para concentraciones del tiol en el rango de 10-5 a 10-2 M. Estas medidas se efectuaron en presencia de aire u oxígeno. En atmósfera de nitrógeno, las pendientes de las gráficas suben a -0,14 V. El potencial de circuito abierto de los electrodos OPG/Co-Pc resultó ser independiente de la concentración de los disulfuros lo que indica que el potencial del electrodo modificado es sólo sensible al tiol. Los electrodos de grafito sin modificar frente a los tioles dan respuestas en potencial irreproducibles, lo que muestra la importancia de la presencia Co-Pc para obtener respuestas potenciométricas estables.
PALABRAS CLAVES: Ftalocianina de cobalto, 2-mercaptoetanol, L-cisteína,disulfuros, respuesta potenciométrica.
We have investigated the potentiometric response of ordinary pyrolytic graphite electrodes (OPG) modified with cobalt phthalocyanine (Co-Pc) for thiols (R-SH) 2-mercaptoethanol, L-cysteine and their corresponding disulfides (R-SS-R). Stable potentials are achieved after a few seconds of additions of different amounts of thiols to aqueous solutions of pH values between 11 and 4. Plots of potential vs. log [R-SH] give straight lines for all cases with slopes ca. -0.060 V with concentrations of the thiol varying from 10-5 up to 10-2 M. These measurements were conducted in the presence of air or oxygen. Under nitrogen, the slopes increase to -0.14 V. The potential response of OPG/Co-Pc is independent of disulfide concentration, which shows that these modified electrodes are only sensitive to the thiol. Graphite electrodes without modification are not sensitive to the concentration of thiols in the solution and the potential response is erratic.
KEY WORDS: Cobalt phthalocyanine, 2-mercaptoethanol, L-cysteine, disulfides, potentiometric response.
Controlling the structure and reactivity of the electrode/solution interface by manipulation or "chemical modification" of the electrode surface is a subject of continuous interest in the literature since the 80's (1-5). These chemically modified electrodes (CMEs) have potential applications in electrocatalysis, electroanalysis, molecular recognition, photoelectrochemistry, electrochemical synthesis and many others.
One special class of CMEs has involved electrodes where molecules, which act as redox mediators, are confined or anchored on their surface. These molecules are able then to accelerate electrochemical reactions that on the unmodified surface do not occur at all or take place at very low rates, requiring then very large overpotentials. Among the redox mediators, metallo-phthalocyanines are now well known as catalyts of many electrochemical reactions (6). Of special interest in the present work is the electrochemical oxidation of thiols to disulfides. The electrocatalytic activity of metallophthalocyanines for the anodic oxidation of 2-mercaptoethanol, L-cysteine and glutathione has been demonstrated in a series of papers published by various groups (7-16). These properties have found applications in electroanalysis. Cobalt phthalocyanines, when confined to electrode surfaces or incorporated into carbon paste, produce modified electrodes that serve as amperometric sensors of sulfhydryl compounds like mercaptans, cysteine, glutathione and similar molecules (16-22). However, there is only one report on the potentiometric detection of thiols using metallophthalocyanines. There is also a report on the potentiometric detection of sulfide ion on graphite electrodes modified with tetramethyltetra-3,4-pyridinoporphyrazino cobalt (I) (23). In this paper we examine the potentiometric response of ordinary pyrolytic graphite electrodes (OPG) modified with cobalt phthalocyanine (Co-Pc) for the thiols 2-mercaptoethanol (ME), L-cysteine and the corresponding disulfides 2-hydroxyethyldisulfide and L-cystine.
All chemicals were reagent grade and used as provided. Water was deionized and then distilled. Buffer solutions were prepared from 0.2 M solutions of analytical grade NaOH, NaHCO3, Na2CO3 , NaH2PO3, NaHC8O4H4 and 0.1 M H2SO4 for pH 1. All electrolytes were obtained from Merck. Co-Pc was obtained from Aldrich and used as received. 2-mercaptoethanol and 2-hydroxyethyldisulfide were Riedel de Häen chemicals. L-cysteine and L-cystine were obtained from Merck. Oxygen and nitrogen were ultra pure from AGA. Buffer solutions were saturated with air, oxygen or nitrogen and maintained under 1 atm of the appropriate gas during measurements. Ordinary pyrolytic graphite (OPG, Union Carbide) was mounted in polyethylene to expose a rectangular area of 0.7 cm2. Before modification the OPG surface was polished with 1000 grit emery paper. Placing a drop of 10-4 M solution for 30 min. of Co-Pc in DMSO modified the electrode and the excess of the complex was removed with DMSO. The electrode was then washed thoroughly with ethanol and finally with bidistilled water.
Cyclic voltammetry was performed on a BAS 100W potentiostat. A three-compartment electrochemical glass cell with a SCE reference and Pt counter electrode was used for the electrochemical measurements. Open-circuit potential (OPC) measurements were conducted with two electrodes, the OPG working electrode and the SCE reference. Polarization of the electrodes was avoided by measuring the OPC on an X-t recorder connected two a high impedance potentiometer. A stock solution of 2-mercaptoethanol or L-cysteine was used in the electrolyte of the appropriate pH and different amounts of the solutions were injected to the cell containing the electrolyte of the same pH. The same procedure was used for measurements with 2-hydroxyethyldisulfide and L-cystine. The time base recorder followed the time for the steady response and stability of the signal. A steady potential of the OPG/Co-Pc electrode was attained within ca. 30 s after successive additions of the thiol.
RESULTS AND DISCUSSION
Fig.1-A illustrates the potentiodynamic response of a clean OPG electrode with (solid line) and without (dashed line) ME in the electrolyte. When no ME is present, no visible signals are observed and the background corresponds to capacitive currents. The irreversible wave at ca. 0.5 V is attributed to the oxidation of ME. Fig.1-B shows the potentiodynamic response of an OPG electrode in alkaline media, which has been previously modified with an adsorbed layer of Co-Pc. The reversible peak located at ca. -0.6 V corresponds to the Co(II)/Co(I) reversible couple in the adsorbed Co-Pc. This peak is not present when Co-Pc is absent from the OPG surface as seen in Fig.1-A. . When ME is added to the electrolyte a pair of current waves appear due to the electrochemical oxidation of ME to the disulfide, which is reduced back to ME during the negative potential scan. The catalytic effect of the adsorbed layer of Co-Pc is demonstrated by the fact that the oxidation wave for ME oxidation is shifted ca. 0.5 V to less positive potentials compared to the wave measured on clean OPG (see Fig. 1-A). This is well established from previous papers (9,12). Similar results to those illustrated in Fig.1 are obtained with L-cysteine (7,8,13,19).
|Fig.1 (A): Cyclic voltammogram of a clean OPG electrode in 0.5 M NaOH with (solid line) and without (dashed line) 0.0037 M. 2-mercaptoethanol. (B): Conditions similar to (A) but after modifying the OPG electrode with Co-Pc. Scan rate: 10 V min-1. N2 saturated solutions.|
Fig.2 shows the potentiometric response of the OPG electrode modified with Co-Pc after additions of different ME concentrations to the electrolyte. The experiments were conducted for different pH and under air-saturated conditions. In all cases straight lines are obtained. The slopes are -0.060 V for pH 10.0, 9.00, 8.00, 7.0 and 5.0 and -0.072 V for pH 0 7.0. For pH 6.0 the slope is -0.045 V and for pH 1.0 is slightly positive (0.022 V). This positive slope cannot be explained in simple terms using the Nernst equation. The results in Fig.2 essentially show that the response of the OPG/Co-Pc electrode is essentially Nernstian for most pH studied but the slope deviates from the theoretical value of 0.059 V for pH 6.0, 7.0 and 1.0. The small slope (slightly positive) for pH 1 indicates that OPG/Co-Pc electrode is rather insensitive to ME concentration in the electrolyte which eventually would prevent possible applications of this type of electrodes as sensors for very acidic pH values. This also illustrates that the presence of ME in its dissociated form (RS-) is required for potentiometric sensing purposes. pKa of ME is 9.45 (9) so the results in Fig.2 show that in spite of the rather large value of pKa, potentiometric detection of ME is possible in a wide range of pH values. Similar results to those obtained with ME are illustrated in Fig.3 for L-cysteine. The main difference is that a greater dispersion of slopes is observed. The slopes of the lines are: -0.060 V for pH 10, -0.058 V for pH 9.0; -0.063 V for pH 7.0; -0.031V for pH 6.0; -0.058 V for pH 5.0 and -0.031 for pH =1. Again as observed for ME, deviations from the theoretical value of -0.059 V are observed for pH 6.0 and 0.0. As discussed above for the ME, deviations at very low pH are expected as the concentration of RS- species is very low but at pH = 6.0 this explanation is not so valid since for pH 5.0 the slopes are very close to the theoretical value.
[2-mercaptoethanol] for different pH values. Air-saturated solutions.
Fig.4 shows the potential response of the OPG/Co-Pc electrode versus log [ME] under different conditions: electrolyte saturated with oxygen, with air and with nitrogen. Under oxygen or under air, the straight lines have slopes close to -0.060 V whereas under anoxic conditions, they change to -0.140 V. Fig.5 shows similar results obtained with L-cysteine. Under oxygen or air, the straight line have slopes close to -0.055 V whereas under anoxic conditions, they change to -0.073 V.
The potential of the OPG/Co-Pc modified electrode is given by the Nernst equation:
E = Eo - (RT/nF)ln Co(I)/Co(II)
Where reaction Co(I)/Co(II) depends on the concentration of RS- and oxygen according to the following equilibrium:
|0000000000000Co(II) + RS- D Co(I) + RS.||(2)|
|2Co(I) + O2 + H2O D 2Co(II) + HO2- + OH-||(3)|
The Nernstian behavior of the open circuit potential of the OPG/Co-Pc electrodes in the presence of oxygen can be explained on the basis of the equilibrium described above. The open-circuit potential of the same electrode in the absence of oxygen is more complicated. Indeed, a slope of -0.140 V for ME cannot be explained easily. With L-cysteine under nitrogen the slope is also larger (-0.073 V) than the Nernstian slope. According to recent studies of the open circuit potential of gold electrodes in the presence of alkane thiols under anoxic conditions, the open circuit potential of gold after additions of an alkane thiol shifts negatively due to the accumulation of negative charge on the electrode through the donation of electron-density from sulfur to gold with simultaneous reduction of the sulfhydryl proton to hydrogen (24). In the present case, the shift in potential to negative values by the presence of ME is attributed to the reduction of Co(II) to Co(I) and the shift, in the absence of oxygen is more pronounced than that predicted by the Nernst equation. Even though the reduction of the cobalt center from Co(II) to Co(I) by reaction of cobalt tetrasulfophthalocyanine (Co-TSPc) with a thiol is known from studies in homogeneous phase, UV-vis in situ electroreflectance spectroscopic studies conducted with Co-TSPc adsorbed on the basal plane of graphite have demonstrated that the adsorbed Co(II)TSPc is reduced to Co(I)TSPc by the addition of L-cysteine to the electrolyte under anoxic conditions (13). This was detected both by changes in the spectra and by changes in the open circuit potential. . So measurements under anoxic conditions are more difficult to interpret but for practical applications, measurements under oxygen or air are more reliable.
We tested the potentiometric response of the OPG/Co-Pc electrode for the presence of the corresponding disulfides, in the absence of the thiol, in order to determine if these product could affect the measurements of the thiols. This is important since thiols are oxidized by air to the corresponding disulfides, so samples of thiols are bound to contain small amounts of disulfides. Fig.6-A illustrates the potentiometric response of the OPG/Co-Pc electrode to different concentrations of 2-hydroxyethyldisulfide in the presence of a constant amount ME. As seen in the Figure, the response in potential is practically constant, which shows that the open circuit potential does not depend on the concentration of 2-hydroxyethyldisulfide in the bulk. Similar results are obtained with L-cystine (Cys-SS-Cys) in the presence of a constant amount of L-cysteine (Fig.6-B). The results in Fig.6 essentially show that the presence of the corresponding disulfides does not interfere with the open circuit potential measurements of the thiols. For practical applications, the OPG/Co-Pc is a good potentiometric sensor for the thiols. The lack of response of the OPG/Co-Pc electrode for the disulfides can be explained by equation 2. Even though the thiol is able to reduce the Co(II) centers on the adsorbed Co-Pc to Co(I) the disulfide does not oxidize the Co(I) centers back to Co(II). The open circuit potential of the modified electrode depends on the Co(I)/Co(II) ratio on the adsorbed Co-Pc and this ratio is not altered by the presence of the disulfide.
| ||Fig.6 A) Potential of the OPG/Co-Pc modified electrode plotted against log |
[2-hydroxyethyldisulfide] in the presence of 10-3 M 2-mercaptoethanol.
B) Potential of the OPG/Co-Pc modified electrode plotted against log [L-cystine] in the presence of 10-3 M L-cysteine. Air-saturated solutions.
This work has been supported by the Presidential Chair in Science 1996, by Fondecyt Projects 1970653 and by Dicyt-Usach.