Print version ISSN 0103-5053
J. Braz. Chem. Soc. vol.13 no.3 São Paulo June 2002
Cristiano Giacomellia, Karina Cklessb, Dayani Galatoa,c, Fabio S. Mirandaa and Almir Spinelli*a
b Departamento de Bioquímica - CCB, Universidade Federal de Santa Catarina, 88040-900, Florianópolis - SC, Brazil
c Curso de Farmácia, Universidade do Sul de Santa Catarina, 88704-900, Tubarão - SC, Brazil
O comportamento eletroquímico do ácido cafeico (H3CAF) em meio aquoso foi estudado na faixa de pH 2,0-8,5 aplicando-se as técnicas voltametria cíclica, eletrólise com potencial controlado e espectroscopia UV-vis. A eletro-oxidação envolve a transferência reversível de dois elétrons e de dois prótons em soluções de pH até 5,5, de acordo com o mecanismo uma etapa-dois elétrons. Em soluções de pH superiores a 5,5 a eletro-oxidação do H3CAF segue um mecanismo ECi. O principal produto desta oxidação é a o-quinona correspondente (o-HCAF), a qual decompõe-se rapidamente em soluções de pH superior a 7,4, obedecendo a uma cinética de primeira ordem. Na faixa de pH 2,0-8,5, o potencial formal (E0') varia linearmente com o pH, gerando uma reta com coeficiente angular de ¾60,83 mV/pH. Em paralelo, a corrente de pico anódica (ipa) diminui de modo não-linear. Os processos são controlados por difusão em toda a faixa de pH estudada.
The electrochemical behavior of caffeic acid (H3CAF) in aqueous solutions with pH 2.0 to 8.5 was studied by cyclic voltammetry, controlled potential electrolysis and UV-vis spectroscopy. The electro-oxidation of H3CAF involves a reversible transfer of two electrons and two protons in solutions of pH up to 5.5, in agreement with the one step-two electron mechanism. In solutions of pH higher than 5.5, the process of electro-oxidation of H3CAF follows an ECi mechanism. The main oxidation product is the corresponding o-quinone (o-HCAF), which is decomposed quickly at pH higher than 7.4 obeying a first order kinetics. In the pH range investigated, the formal potential (E0') varies linearly with pH, generating a straight line with an angular coefficient of -60.83 mV/pH. In parallel, the anodic peak current (ipa) decreases in a nonlinear mode. The processes are controlled by diffusion over the whole pH range studied.
Keywords: caffeic acid, oxidation, electrochemistry, UV-vis spectroscopy
Caffeic acid (H3CAF), 3-(3,4-dihydroxyphenyl)-2-propenoic acid (Figure 1), is the phenylpropenoid most encountered in nature and has proven medicinal properties, especially as an antioxidant agent.1,2 Despite this, few studies have been dedicated to the oxidation mechanism of this substance. The enzymatic oxidation is the most important reaction of H3CAF in the presence of polyphenoloxidase.3 However, non-enzymatic oxidation can take place in the presence of oxygen, particularly in alkaline medium.4 Previous studies3,5 showed that the chemical oxidation of H3CAF promoted by sodium periodate leads to the formation of the corresponding o-quinone (o-HCAF), 3-(cyclohex-1,5-dien-3,4-dione)-2-propenoic acid. In acidic conditions a disproportion of o-HCAF occurs, leading to the formation of two isomers of 2,5-(3',4'-dihydroxyphenyl) tetrahydrofuran-3,4-dicarboxylic acid. The electro-oxidation of H3CAF in non-aqueous solution5 - acetonitrile - occurs in two stages, leading to the formation of the semi-quinone and then to the corresponding o-quinone and involves the transfer of two electrons and two protons. In aqueous solution (pH 4.0) the electro-oxidation also involves the transfer of two electrons but there is little information about the products formed. As can be deduced from the structural formula (Figure 1), the electro-oxidation of H3CAF is pH-dependent, and this aspect has been poorly explored in the literature. Electro-oxidation of organic compounds is carried out almost always in non-aqueous solutions due to solubility characteristics. However, this is not the case of H3CAF. The objectives of the present work are, therefore, to study the electrochemical behavior of H3CAF in solutions with pH 2.0 to 8.5, to characterize the products and to propose a mechanism for its electro-oxidation in aqueous solution.
All the reagents used in this study are of analytical grade acquired from Aldrich (H3CAF) and Merck (H3PO4, K2HPO4, KH2PO4, KOH and citric acid). They were used without previous purification. Distilled and deionized water was used for all solution preparations.
Cyclic voltammograms were obtained in a 15-mL three-electrode cell, with a glassy carbon working electrode (A = 0.0314 cm2), a graphite rod counter electrode, and an aqueous saturated calomel reference electrode (SCE). All potentials in the text are quoted versus this reference electrode. The working electrode was carefully polished before each experiment with 0.05 mm alumina paste and ultrasonically rinsed in deionized water. IxE curves were recorded with an EG & G PAR model 263A potentiostat/galvanostat with M270 software coupled to a personal computer. Buffers formed by citric acid and K2HPO4 or KH2PO4 (both 0.05 mol L-1) with pH adjusted with KOH or H3PO4 for the studied values (2.0; 3.5; 5.5; 7.4; 8.5) were used as supporting electrolyte. H3CAF was added directly into the cell (final concentration 0.8 mmol L-1) after attainment of a cyclic voltammogram of the electrode immersed in the H3CAF free solution. The solutions were purged with nitrogen for 8 min and the experiments were done at room temperature.
Controlled Potential Electrolysis
Chronoamperometry was performed to determine the number of electrons transferred in the electro-oxidation process and to calculate, using the Cottrell equation,6,7 the diffusion coefficient of the species implicated. The experiments were carried out under the same conditions described for the cyclic voltammetry. The applied potential was 50 mV more positive than the anodic peak potential (Epa). For the determination of the number of electrons involved in the electro-oxidation process in solutions of different pH, the current decrease was observed until zero. The charge was obtained by integration of the resulting curve and the number of electrons was calculated. Controlled potential electrolysis was also carried out under the same conditions to identify the product of H3CAF oxidation, but graphite rods were used as working and auxiliary electrodes (the voltammetric response of H3CAF is equivalent at glassy carbon and graphite electrodes). Aliquots were extracted from the electrochemical cell and analyzed by UV-vis spectroscopy in a HITACHI U-3000 spectrophotometer before and after 10 min of electrolysis. The temperature of the solutions was controlled at 5 ºC (Microquímica MQBTC 99-20 thermostat).
Chemical oxidation of the H3CAF
The main objective of this experiment was to identify the product of H3CAF oxidation and to follow its disappearance kinetics, in view of its instability. The solutions used in this study were prepared in a way similar to the solutions used in the previous experiments, but with a constant ionic strength (m = 0.5 mol L-1) controlled with KCl. H3CAF was oxidized with KMnO4, which was added to oxidize 50% of the organic compound, admitting that the oxidation product was the corresponding o-quinone.3 An excess of KMnO4 was avoided to hinder the possible cleavage of the double bond of the propenoic group in acidic conditions. The identification of o-quinone was done by UV-vis spectroscopy and its disappearance was monitored at 400 nm (characteristic for o-HCAF absorption). The experiments were extended up to 1600 s after the addition of the oxidizing agent.
Results and Discussion
Figure 2 shows a typical cyclic voltammogram at a glassy carbon working electrode in a pH 2.0 solution containing 0.8 mmol L-1 H3CAF, for a scan rate (v) of 100 mV s-1. The cyclic voltammogram is characteristic of an electrochemically reversible reaction showing only one anodic peak (Epa = 507 mV) and one cathodic peak (Epc = 415 mV). The voltammogram profile does not change significantly even after 20 cycles. The ratio ipc ipa-1 = 0.954 confirms the reversibility of the system under these conditions. The formal potential (E0') of the electrochemically reversible couple that quickly changes electrons with the working electrode is centered between Epa and Epc.8 For the system in study (Epa + Epc)/2 = 461 mV. It is interesting to note that this value is very close to the standard potential (E0) and to the formal potential (E0') in different acidic conditions of the redox couple quinone-hydroquinone (E0 = 455 mV and E0' = 452 mV in 1.0 mol L-1 H2SO4, HClO4 and HCl).9 This result suggests that the couple implicated in the electro-oxidation of caffeic acid in pH 2.0 solutions is H3CAF-o-HCAF involving the transfer of two electrons and two protons. However the value of Epa ¾ Epa/2 = 51 mV is not expected for this type of electron transfer. This value should be close to 30 mV for a reversible two-electron transfer. The difference between the obtained and expected value indicates, however, that the charge transfer is not sufficiently fast under the studied conditions and that reaction is not completely reversible, as demonstrated below.
Influence of pH
The influence of pH on the cyclic voltammograms at a glassy carbon electrode in 0.8 mmol L-1 H3CAF solutions, v = 100 mV s-1, is shown in Figure 3. It is observed that the profile of the IxE curves is the same in the pH range studied, indicating that the electro-oxidation of H3CAF follows the same reaction mechanism. However, with the increase of pH, the anodic and cathodic peak potentials are shifted toward less positive values. The shift of Epa to less positive values indicates an increase in the nucleophilicity of the organic compound,10 and that its antioxidant activity is thermodynamically favored with the pH increase. Some authors have described the same dependence for oxidation of H3CAF, gallic acid and derivatives.11,12 On the other hand, in 50/50 (v/v) water-methanol solvent, the anodic peak potential shifts to more positive values with an increase in pH.5 In addition we observed that the separation between the peak potentials is also pH-dependent. The large separation between the anodic and cathodic peak potentials in the pH 8.5 cyclic voltammogram is a manifestation of sluggish heterogeneous electron transfer kinetics.8 Figure 4 shows the linear decrease of E0' as a function of pH with -60.83 mV pH-1 slope (correlation coefficient r = -0.9907). This value is close to that expected for Nernstian systems with electron transfer followed by deprotonation. Figures 3 and 4 show that ipa decreases in a nonlinear way with increasing pH. The decrease of ipa as a function of pH can be associated to the concentration and to the diffusion coefficient of the electro-active species that effectively are being oxidized, as demonstrated below. Figure 5 shows the composition of the H3CAF solutions (alpha values) as a function of pH. Under the conditions of air experiments we have: pH 2.0: 99% of H3CAF and 1% of H2CAF-; pH 3.5: 88% of H3CAF and 12% of H2CAF-; pH 5.5: 8% of H3CAF and 92% of H2CAF-; pH 7.4: 94% of H2CAF- and 6% of HCAF2-; pH 8.5: 58% of H2CAF- and 42% of HCAF2-. Thus, at this time, if one accepts that the oxidation processes of these species are represented by the reactions shown in Figure 6, the dependence of ipa with pH can be explained as follows: at pH 2.0 and 3.5, ipa practically does not change, therefore the species that is being oxidized is H3CAF, its concentration at the two pH being practically the same. The current for the electrochemical reaction (at a mass transport rate) of an electro-active material with a diffusion coefficient, D, is described by the Cottrel equation:6,7
where D is the diffusion coefficient (cm2 s-1) and Cb is the bulk concentration of substrate. n, F, A and t have their usual significance. Under diffusion (mass transport) control, a plot of i versus t-1/2 will be linear and, from the slope, the value of D can be obtained. Chronoamperometry was performed and a diffusion coefficient D = 2.3 x 10-5 cm2 s-1 (pH 2.0) was obtained. When the pH progressively changes from 3.5 to 8.5, it is observed that ipa becomes more pH dependent. As the number of electrons involved in the process is always the same, the decrease of ipa is certainly associated with a reduction of the concentration and/or of the diffusion coefficient of the involved species. As at pH 3.5 there are already more than one species in solution, it is impossible to associate the diffusion coefficient calculated from the Cottrell equation with the diffusion coefficient of one particular species. Under these conditions, the obtained value would be a compilation of the diffusion coefficients of all species present, as considered by Schifino et al.14 We also observed that the diffusion coefficient gradually diminishes from 2.3 x 10-5 cm2 s-1 to 0.49 x 10-5 cm2 s-1 when the pH is increased from 2.0 to 8.5. This indicates, in a first approach, that the species would have gradually smaller diffusion coefficients as they were being deprotonated.
It is important to note that the dependence of Epa with pH is attenuated when the pH changes from 7.4 to 8.5 (Figure 3). This behavior can be explained by the increase of the concentration of HCAF2- in solution. In the oxidation of HCAF2-, only one proton is involved (Figure 6, reaction 3), while in the oxidation of the other species two protons are involved.
In solutions of pH higher than 8.5 the reduction peak was not observed (data not shown), which indicates that the product of oxidation of H3CAF became unstable and an irreversible homogeneous chemical reaction occurs. For this reason we limited our studies to the pH range 2.0-8.5.
Influence of the scan rate
The influence of the scan rate on the anodic peak potential (Epa) is shown in Figure 7 for the pH range studied. It is verified that the system under study presents a reversible behavior for scan rates up to 100 mV s-1 and an increasing irreversibility for faster scan rates. In parallel, the anodic peak current (ipa) increases linearly with v1/2 (Figure 8), indicating that the process is diffusion limited over the whole pH range studied.
Influence of slow scan rates in alkaline solutions
Figure 9 shows cyclic voltammograms at a glassy carbon working electrode in pH 2.0 and 8.5 solutions containing 0.8 mmol L-1 H3CAF, v = 5 mV s-1. It is observed that at pH 2.0 the electro-oxidation of H3CAF is reversible (ipc.ipa-1 ~ 1). This ratio decreased as the pH increased and, at pH 8.5, the reduction peak was not observed, denoting that the product of H3CAF oxidation (that is the corresponding o-quinone (o-HCAF)) became unstable, resulting, therefore, in only partial reduction. We observed that o-HCAF is more stable in solutions of pH < 5.5. For solutions of pH > 7.4 the stability of o-HCAF diminished significantly, implying an expressive increase of its disappearance rate. The instability of o-HCAF at pH > 7.4 suggests the involvement of HCAF2- ions, which have a high nucleophilicity. The influence of the nucleophiles on the stability of o-quinones was also suggested by Nematollahi et al.15 At pH £ 5 (in the presence of H3CAF and H2CAF- ions) o-HCAF is stable, while at higher pH (in the presence of only HCAF2- ), o-HCAF is less stable.
Controlled potential electrolysis
Electrolyses with controlled potentials were carried out to confirm the number of electrons involved and to identify o-HCAF as the product of oxidation of H3CAF. The number of electrons involved in the oxidation of H3CAF in the pH range studied was found to be two (Table 1). The identification of o-HCAF was carried out by UV-vis spectroscopy. Electrolysis was performed with the cell temperature kept at 5 0C, in view of the instability of o-HCAF at higher temperatures.
Figure 10 shows the UV-vis spectra for 0.8 mmol L-1 H3CAF in pH 2.0 solution for t = 0 and after 10 min of electrolysis. No absorption band was observed before electrolysis (t = 0). A large band around 400 nm was observed after electrolysis which was attributed to o-HCAF, since this substance presents a characteristic absorption band at this wavelength.3 This absorption band was also observed in aliquots of solutions with pH higher than 2.0, presenting a small wavelength shift to lower energy (406 nm at pH 7.4).
These studies therefore confirmed that, in the process of H3CAF electro-oxidation, two electrons are involved and that the main product formed is the corresponding o-quinone.
Kinetic study of the disappearance of o-HCAF
During these studies we observed that the o-quinone resulting from the oxidation of H3CAF was fairly unstable at room temperature, which is a characteristic behavior of many quinones, mainly in the presence of nucleophiles. Therefore, this behavior was studied as a function of pH. Chemical oxidation of H3CAF with KMnO4 was promoted and an absorption band around 400 nm, attributed to o-HCAF, repeats.3 The decrease of the absorption band at 400 nm as a function of time allows one to study the kinetics of o-HCAF disappearance,16 as shown in Figure 11. The kinetics of disappearance of o-HCAF involves a first-order reaction, as recognized by other authors.3 In pH £ 5.5 solutions o-HCAF is sufficiently stable. When the pH is higher than 5.5 the o-HCAF instability increases, as it can be verified observing an enhancement of kinetics of its disappearance. These results are in accordance with those observed in the electrochemical experiments, where an increasing irreversibility at pH higher than 5.5 was shown. As mentioned previously, this behavior suggests that HCAF2- ions are related to the instability of o-HCAF. However this approach is beyond the objective of this work.
The results obtained in these studies show that H3CAF is reversibly oxidized in solutions of pH up to 5.5. The electro-oxidation follows a mechanism involving only one step with the transfer of two electrons and two protons. The H3CAF oxidation product was identified as being the corresponding o-quinone. For pH values higher than 5.5 o-HCAF becomes unstable, with a chemically homogeneous irreversible reaction with a first order kinetics occurring. Under these conditions the mechanism is of the ECi type. The proposed electro-oxidation mechanisms shown in Figure 12 take these comments into consideration. Reaction 1 can be accepted for H3CAF and H2CAF- in solutions of pH up to 5.5. A fast reaction follows a one step-two electron type mechanism. The intermediates cannot be identified by electrochemical methods under our experimental conditions, but the results obtained indicate that the reaction occurs with charge transfer, followed by deprotonation, supporting the proposed mechanisms. Although the H2CAF- species are present even at pH 8.5, from pH 7.4 (and even at slightly acid pH) there are also HCAF2- groups present. In the oxidation of HCAF2- two electrons are also involved but the reaction is less dependent on pH (Figure 12, reaction 2). We believe that HCAF2- is also responsible for the instability of the o-quinone formed and is the basis of a slow irreversible chemical reaction with first order kinetics.
The authors are grateful to CNPq for financial support.
1. Nardini, M.; Pisu, P.; Gentili, V.; Natella, F.; Di Felice, M.; Piccolella, E.; Scaccini, C.; Free Radical Biol. Med. 1998, 25, 1098. [ Links ]
2. Galato, D.; Ckless, K.; Susin M. F.; Giacomelli, C.; Ribeiro-do-Valle, R. M.; Spinelli, A.; Redox Report 2001, 6, 243. [ Links ]
3. Fulcrand, H.; Cheminat, A.; Brouillard, R.; Cheynier V.; Phytochemistry 1994, 35, 499. [ Links ]
4. Cilliers, J. J. L.; Singleton, V. L.; J. Agric. Food Chem. 1989, 37, 890. [ Links ]
5. Hapiot, P.; Neudeck, A.; Pinson, J.; Fulcrand, H.; Neta, P.; Rolando, C.; J. Electroanal. Chem. 1996, 405, 169. [ Links ]
6. Bard, A. J.; Faulkner, L. R.; Electrochemical Methods, Fundamentals and Applications, John Wiley & Sons Inc: New York, 1980, p 142. [ Links ]
7. Brett, C. M. A.; Brett, A. M. O.; Electrochemistry: Principles, Methods, and Applications, Oxford University Press: New York, 1993, p 87. [ Links ]
8. Sawyer, D. T.; Heineman, W. R.; Beebe, J. M.; Chemistry Experiments for Instrumental Methods, John Wiley & Sons: New York, 1984, p 79. [ Links ]
9. Skoog, D.A.; West, D. M.; Holler, F. J.; Fundamentals of Analytical Chemistry, 6th ed., Saunders College Publishing: New York, 1992, appendix 6. [ Links ]
10. Sawyer, D. T.; Sobkowiak, A.; Roberts Jr., J. L.; Electrochemistry for Chemists, 2nd ed., John Wiley & Sons Inc: New York, 1995, p 442. [ Links ]
11. Zare, H.R.; Golabi, S.M.; J. Solid State Electrochem. 2000, 4, 87. [ Links ]
12. Gunckel, S.; Santander, P.; Cordano, G.; Ferreira, J.; Munoz, S.; Nunez-Vergara, L. J.; Squella, J. A.; Chemico-Biological Interactions 1998, 114, 45. [ Links ]
13. Martell, A. E.; Smith, R. M.; Critical Stability Constants, Plenum Press: New York, 1976, vol. 4. [ Links ]
14. Schifino, J.; Basso, N. R. de S.; Olegário, R. M.; Quím. Nova 1991, 14, 254. [ Links ]
15. Golabi, S. M.; Nematollahi, D.; J. Electroanal. Chem. 1997, 420, 127. [ Links ]
16. Atkins, P. W.; Physical Chemistry, 5th ed., Oxford University Press: New York, 1994, p. 864. [ Links ]
Received: April 6, 2001
Published on the web: April 18, 2002