Potentiometric determination of aqueous dissociation constants of flavonols sparingly soluble in water

Available online at www.sciencedirect.com

Talanta 74 (2008) 1008–1013

Potentiometric determination of aqueous dissociation constants of flavonols sparingly soluble in water Jos´e M. Herrero-Mart´ınez, Carme Repoll´es, Elisabeth Bosch, Mart´ı Ros´es, Clara R`afols ∗ Departament de Qu´ımica Anal´ıtica, Universitat de Barcelona, Mart´ı i Franqu´es, 08028 Barcelona, Spain Received 20 March 2007; received in revised form 30 July 2007; accepted 9 August 2007 Available online 15 August 2007

Abstract Two different approaches were evaluated and used to estimate the aqueous pKa values of some flavonols sparingly soluble in water (morin, fisetin and quercetin) from their pKa values in methanol/water mixtures obtained by potentiometry. Both approaches lead to similar results, although one of them requires only one pKa value at one unique methanol/water mixture, whereas the other (the classical Yasuda–Shedlovsky plot) requires several pKa data at different methanol/water mixtures. Thus, the first approach is recommended because it is faster and simpler. The automated potentiometric method is strongly recommended for pKa determination of these types of compounds because of its simplicity and speed of operation. © 2007 Elsevier B.V. All rights reserved. Keywords: Flavonols; Dissociation constant; Potentiometry; Methanol/water mixtures

1. Introduction The acidic dissociation constant, pKa , is an important parameter in ADMET (absorption, distribution, metabolism, excretion, toxicity) research because it helps to explain chemical phenomena such as absorption, distribution and elimination of substances. Potentiometry [1], spectrophotometry [2] and capillary electrophoresis (CE) [3,4] are well-established current methods for pKa determination of compounds of pharmaceutical or biological interest. Potentiometry and spectrophotometry require purity and stability of the sample. Potentiometry is a more general method because it does not require the presence of a chromophore, although the compound must be soluble enough in water. When the compounds are sparingly soluble in water, the pKa determination is commonly done in organic/water mixtures and the obtained pKa value extrapolated to the aqueous one. Methanol/water mixtures [5–8] are often used because they have a lower polarity than pure water, but keeping a similar environment. CE methods do not require high purity of solute since CE is a separation technique. The pKa of compounds insoluble in water can be determined by dissolving the sample in an organic

∗

Corresponding author. Tel.: +34 934034874; fax: +34 934021233. E-mail address: [email protected] (C. R`afols).

0039-9140/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2007.08.007

solvent (e.g. methanol), which is directly injected into the aqueous buffer inside the capillary. However, it is an open question whether the pKa value obtained is in pure water or in an aqueous solvent that may contain a small but unknown amount of organic solvent from the sampling solvent. The disadvantage of CE methods as compared with automated potentiometric titration methods is that they are not yet fully automated and since they require the preparation of several buffers the total time spent in the determination of several consecutive pKa values can be quite high. The aqueous pKa determination of most compounds with a pKa value between 4 and 10 is usually straightforward, but when the compound has very low or very high pKa or it shows several and close pKa values the determination of pKa may be much more complex. This is the case of flavonoids. These compounds are polyphenols and because of their strong antioxidant activity they play an important role in human diet [9,10]. They show a variety of pharmacological activities, including anticancer, antiinflamatory, antibacterial, anti-allergic, and antiviral activity and reduce the risk of coronary heart disease [9]. There are also reports of flavonoids inhibiting the activities of an array enzymes, including lipoxygenase, cyclooxygenase and DADHoxidase. The inhibition of the enzymes by flavonoids could be due to a reaction of the flavonoid with free radical generated at the active site of the enzymes [11]. The role of flavonoids is

J.M. Herrero-Mart´ınez et al. / Talanta 74 (2008) 1008–1013

1009

ionisable hydroxyl groups with pKa values higher than 7 and relatively close one to each other. Moreover, most of them are sparingly soluble in water and, consequently, accurate aqueous pKa determination is not an easy task. There are a few pKa values of these compounds reported in the literature, but they show a great disparity. In fact, some of them are determined in mixtures of water with an organic solvent [4,15–24]. The aim of this work is the pKa determination of flavonols by potentiometry in order to establish and test procedures for the accurate determination of dissociation constants of compounds with several high pKa values and sparingly soluble in water. Two procedures are used to estimate the aqueous pKa values from the pKa values determined in methanol/water mixtures. Yasuda–Shedlovsky equation relates the pKa values determined in methanol/water systems, ss pKa , with the reverse of the dielectric permittivity of the binary solvent. This extrapolation method involves the determination of ss pKa in at least three or four methanol/water mixtures with a percentage of methanol lower than 60% in weight [5,6,25]. The other estimation equation used, is the linear correlation we proposed [8,26]. This equation allows the estimation of aqueous pKa values, w w pKa , from the ss pKa values in only one methanol/water mixture and works for any percentage of methanol. The extrapolated values are compared with those obtained by CE in a previous work [4], as well as with a few ones that can be directly obtained in water. 2. Experimental 2.1. Reagents and chemicals Methanol (Merck for chromatography) and water purified by Milli-Q plus system from Millipore were used to prepare the solvent mixture. Hydrochloric acid (p.a. 25%, w/w) and potassium chloride (AR grade) were purchased from Merck and potassium hydroxide (1 M titrisol) from Carlo Erba. Quercetin dihydrate were purchased from Fluka (Buchs, Switzerland), fisetin from Sigma (St Louis, MO, USA) and morin from Riedel de Ha¨en (Seelze, Germany). Fig. 1. Molecular structures of the studied flavonols.

related to their chemical structure, for instance their antioxidant efficiency is related to their hydrogen radical donating abilities and to the number of hydroxyl groups in the molecule [11–14]. The hydroxyl moiety deprotonation has influence in the intrinsic antioxidant potential of the flavonoid, deprotonation generally enhances the antioxidant action of the flavonoid [14], thus the knowledge of their physicochemical parameters, such as pKa , is important to predict their antioxidant capacity. In addition, the pKa values determine the precise species present in a biological medium and their knowledge is fundamental in bioclinical and pharmacological research studies. Flavonols are one class of flavonoids and the basic structure of these compounds are given in Fig. 1. They present several

2.2. Apparatus and procedure All titrations were carried out by means of an automatic titrator PCA 101 instrument (Sirius Analytical Instruments, East Sussex, England) equipped with a Sirius 010604 combined electrode. The electrode system was standardized according to the specifications of literature [5]. About 0.01 or 0.005 g of flavonol dissolved in 15 ml of water or methanol/water mixture (methanol content between 5 and 60% in weight) were preacidified to pH 1.8 with 0.5 M standardized aqueous HCl and titrated with 0.5 M standardized aqueous KOH until pH 12.2. In the case of morin and for aqueous determination, the solution was first basified (pH ∼ = 12.2) and titrated with HCl. All titrations were performed in solutions of constant ionic strength (0.15 M KCl) under nitrogen atmosphere at 25 ± 0.1 ◦ C.

1010

J.M. Herrero-Mart´ınez et al. / Talanta 74 (2008) 1008–1013

2.3. Computation

3. Results and discussion

The ss pKa values were obtained using the pKa LOGPTM software (v5.2, Sirius), where ss pKa refers to the pKa value in the particular pH scale of the working methanol/water mixture. PCA 101 instrument works at a constant ionic strength, and then the pKa values obtained were in the molar concentration scale. To obtain the thermodynamic values the activity coefficients of each species must be considered. Debye–H¨uckel equation was used to compute the activity coefficients: √ Az2 I √ log γ = − (1) 1 + a0 B I

The flavonol compounds shown in Fig. 1 differ in the number and position of the hydroxyl groups attached to the rings, but all of them have more than three ionizable hydroxyl groups. The first pKa value is higher than 7 except for morin, and the others are relatively close together [4]. Moreover, they are sparingly soluble in water and their pKa values cannot be directly determined in aqueous solution by potentiometry. Then, several titrations have been carried out at different methanol/water mixtures for the studied flavonols using the automatic PCA101 titrator. The Sirius methodology measures at the physiological ionic strength (I ∼ = 0.15 M) and the pKa values obtained are referred to this working ionic strength (pKa (I = 0.15 M)). The thermodynamic ones (pKa (I = 0 M)) are obtained from these values and the values of the activity coefficients. Table 1 shows the s pK values obtained experimentally at different methanol/water a s mixtures ss pKa (I = 0.15 M) and their corresponding thermodynamic ones (ss pKa (I = 0 M)). The solvent mixtures with lowest methanol content were 10%, 25% or 40% (in weight) depending on the solubility of the studied flavonol. Morin is the most soluble one and quercetin the most insoluble. We have been able to determine three pKa values for morin and quercetin and two for fisetin. The third pKa of fisetin is higher than 12 [15,31] and thus it could not be determined by potentiometry. As expected, the s pK (I = 0.15 M) of flavonols, like all polyphenols, increases a s when the percentage of methanol increases [26], except for s pK (I = 0.15 M) which remains practically constant in all a1 s the range of methanol percentages studied. Table 1 also reports the w w pKa (I = 0 M) estimated from each methanol/water mixture using Eq. (5). The mean w w pKa (I = 0 M) values and its standard deviations are also given. In all cases the standard deviations are lower than 0.11. This fact shows the good agreement s in the w w pKa (I = 0 M) values estimated from single s pKa (I = 0.15 M) determination at any methanol/water composition. w pK (I = 0.15 M) values have also been estimated by a w means of Eq. (4). Fig. 2, shows a straight line in all cases according to the Yasuda–Shedlovsky plots. Table 2 summarizes the values extrapolated to aqueous solution and corrected to zero ionic strength using Eqs. (4) and (1). Table 2 also shows the w w pKa (I = 0 M) estimated by Eq. (5) and those obtained by capillary electrophoresis (CE) [4]. Morin is slightly soluble in neutral and basic aqueous solutions (anionic morin forms), but it is sparingly soluble in acidic aqueous solution (neutral form). Then, potentiometric titrations could be carried out directly in water adding KOH in excess and using HCl as a titrant. The titration curves showed a jump close pH 5 attributed to the poor solubility of the neutral morin and the aqueous pKa1 could not be directly determined in w water. The w w pKa2 (I = 0.15 M) and w pKa3 (I = 0.15 M) values obtained were 7.73 ± 0.02 and 9.53 ± 0.04, respectively. The thermodynamic ones were obtained applying activity coefficient correction and they are also presented in Table 2. For morin, the extrapolations by means of Eq. (4) have been carried out from only 10% in weight of methanol and the w pK (I = 0 M) values estimated by this equation are in very a w good agreement with those obtained by Eq. (5). Moreover, the

where γ is the activity coefficient of the involved species, z the ion charge of the ion and I is the ionic strength of the solution. A and B parameters are functions of temperature and dielectric constant of the medium. For aqueous solution at 25 ◦ C, A = 0.51 and B = 0.33. The a0 value depends strictly on the solvated radius of the ions but, since this is unknown, the Bates–Guggenheim convention is usually applied [27]. This convention assigns ˚ for aqueous solutions, and thus a0 B = 1.5. A and a0 B a0 = 4.56 A parameters for each methanol/water mixture can be calculated from the empirical equations previously reported: A = 0.51 + 0.14w + 2.01w2 − 3.18w3 + 2.39w4

(2)

a0 B = 1.50 + 0.19w + 0.74w2 − 1.04w3 + 0.67w4

(3)

where w is the weight fraction in methanol [8]. The software of PCA 101 includes the Yasuda–Shedlovsky extrapolation equation to estimate aqueous pKa values (w w pKa ) from those in methanol/water. It relates ss pKa to the reverse of the dielectric permittivity of the binary solvent by means of the equation: s s pKa

+ log[H2 O] =

aε + bε ε

(4)

where ε is the dielectric permittivity, and aε and bε are the empirical fitting constants [5,6,7,25,28]. In earlier papers [8,26,29,30] we proposed linear equations such as Eq. (5) for neutral and cationic acids to correlate their pKa values in methanol/water mixtures (ss pKa ) to the aqueous pKa (w w pKa ): s s pKa

= aw w w pKa + bw

(5)

where aw and bw are related with the solvent composition. Twenty-eight phenols with literature pKa data in 6–53 different methanol/water mixtures were analyzed in order to relate aw and bw to solvent composition. The following equation were obtained: aw =

1 − 0.305w − 0.195w2 1 − 0.542w + 0.002w2

(6)

bw =

−0.573w + 1.227w2 1 + 0.498w − 1.311w2

(7)

where w is the weight fraction of methanol in the binary solvent.

J.M. Herrero-Mart´ınez et al. / Talanta 74 (2008) 1008–1013

1011

Table 1 (I = 0.15 M) and ss pKa (I = 0 M) of flavonoids at several methanol/water mixtures and pKa in water (w w pKa (I = 0 M))

s pK a s

MeOH (wt.%) Morin 10.7 16.1 21.1 21.4 31.1 31.9 40.5 41.5 44.9 49.8 52.9 61.0

(I = 0.15 M)

s pK a1 s

4.843 4.709 4.845 4.787 5.024 4.877 5.163 4.895 5.155 5.181 5.100 5.095

± ± ± ± ± ± ± ± ± ± ± ±

0.014 0.023 0.016 0.021 0.011 0.018 0.021 0.020 0.009 0.010 0.039 0.032

(I = 0.15 M)

s pK a2 s

7.818 7.876 7.919 7.970 8.136 8.055 8.252 8.224 8.322 8.388 8.423 8.484

± ± ± ± ± ± ± ± ± ± ± ±

0.010 0.016 0.011 0.016 0.009 0.011 0.019 0.015 0.007 0.008 0.027 0.025

(I = 0.15 M)

s pK a3 s

9.556 9.786 9.752 9.825 9.923 9.938 9.905 10.158 10.161 10.241 10.280 10.366

± ± ± ± ± ± ± ± ± ± ± ±

0.010 0.016 0.012 0.016 0.010 0.012 0.019 0.016 0.008 0.010 0.029 0.032

s pK a1 s

0 M) 5.11 4.99 5.15 5.08 5.36 5.21 5.53 5.27 5.54 5.59 5.51 5.55

(I =

s pK a2 s

0 M) 8.35 8.44 8.52 8.56 8.80 8.71 8.99 8.96 9.10 9.20 9.24 9.40

(I =

w pK a1 (I = w 0 M)

w pK a2 w

0 M)

0 M)

w pK a3 (I = w 0 M)

10.36 10.63 10.65 10.71 10.93 10.93 11.01 11.27 11.32 11.46 11.50 11.73

5.03 4.87 4.98 4.91 5.08 4.93 5.14 4.89 5.10 5.09 4.97 4.90

8.20 8.20 8.20 8.24 8.32 8.22 8.34 8.30 8.37 8.39 8.37 8.37

10.16 10.31 10.24 10.29 10.31 10.29 10.21 10.43 10.41 10.45 10.43 10.49

4.99 ± 0.09

8.29 ± 0.08

10.33 ± 0.11

7.60 7.56 7.55 7.58 7.58 7.51 7.63 7.58 7.55 7.56 7.54

9.32 9.37 9.37 9.37 9.33 9.28 9.40 9.34 9.31 9.31 9.38

7.57 ± 0.03

9.34 ± 0.04

7.61 7.54 7.66 7.65 7.59 7.67 7.60 7.65 7.60 7.54 7.65 7.66 7.57 7.63 7.60 7.55

9.33 9.26 9.37 9.32 9.33 9.36 9.30 9.39 9.32 9.28 9.36 9.39 9.39 9.38 9.39 9.39

– – 11.42 11.46 11.47 11.53 11.42 11.66 11.54 11.51 11.62 11.57 11.70 11.72 11.66 11.63

7.59 ± 0.06

9.33 ± 0.04

11.56 ± 0.10

s pK a3 s

Mean Fisetin 26.6 31.9 37.2 42.5 51.0 52.9 56.4 58.1 58.2 58.2 63.6

7.664 7.683 7.734 7.836 7.929 7.891 8.049 8.030 8.005 8.015 8.067

± ± ± ± ± ± ± ± ± ± ±

0.022 0.019 0.012 0.018 0.007 0.009 0.013 0.021 0.013 0.010 0.031

9.169 9.290 9.344 9.402 9.437 9.423 9.577 9.544 9.514 9.509 9.643

± ± ± ± ± ± ± ± ± ± ±

0.023 0.017 0.012 0.018 0.007 0.010 0.018 0.021 0.015 0.011 0.035

7.98 8.01 8.08 8.20 8.34 8.30 8.48 8.46 8.44 8.45 8.53

9.74 9.95 10.04 10.13 10.25 10.24 10.44 10.41 10.38 10.37 10.56

Mean Quercetin 31.9 37.1 39.1 39.4 40.6 43.6 44.0 47.9 49.0 49.9 52.2 52.5 54.5 59.1 59.7 60.6

7.734 7.728 7.870 7.857 7.813 7.929 7.861 7.968 7.925 7.873 8.022 8.030 7.964 8.091 8.069 8.020

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.013 0.013 0.004 0.005 0.006 0.003 0.006 0.015 0.003 0.013 0.005 0.004 0.013 0.005 0.010 0.007

9.242 9.231 9.346 9.299 9.320 9.382 9.329 9.464 9.406 9.369 9.479 9.514 9.542 9.575 9.588 9.519

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.013 0.013 0.004 0.005 0.007 0.003 0.006 0.015 0.003 0.013 0.005 0.004 0.013 0.005 0.024 0.008

– – 11.197 ± 11.246 ± 11.262 ± 11.361 ± 11.242 ± 11.548 ± 11.429 ± 11.399 ± 11.533 ± 11.489 ± 11.649 ± 11.708 ± 11.643 ± 11.626 ±

0.013 0.011 0.019 0.014 0.015 0.098 0.010 0.013 0.041 0.026 0.013 0.042 0.072 0.032

8.06 8.08 8.24 8.22 8.18 8.31 8.25 8.37 8.33 8.28 8.44 8.45 8.39 8.54 8.52 8.47

Mean

w estimated w w pKa2 (I = 0 M) and w pKa3 (I = 0 M) values agree with those directly obtained in water, but they are slightly lower than the ones obtained by CE. The differences found with the two techniques could be attributed to the difficulties in determining pKa values very close together or to the presence of some residual methanol from the sampling solvent in the CE determination. For w w pKa1 (I = 0 M), which differs by more than 3 units from the w w pKa2 (I = 0 M), the values estimated by potentiometry are consistent with the ones obtained by CE. This value is lower than the ones obtained for fisetin and quercetin. In gen-

9.90 9.93 10.08 10.03 10.06 10.15 10.10 10.26 10.21 10.18 10.32 10.35 10.39 10.47 10.49 10.43

– – 12.29 12.34 12.37 12.50 12.39 12.74 12.64 12.62 12.79 12.74 12.93 13.05 13.00 12.99

(I =

(I =

eral, the most acidic group of flavonols corresponds to the OH group in position A [4], but morin is an exception, the first ionization corresponds to the OH group in position F [4,32]. This fact can be attributed to the anion stabilization by formation of an intramolecular hydrogen bond [24]. For fisetin and quercetin there is a small difference between the w w pKa (I = 0 M) values estimated by Eqs. (4) and (5). The w pK (I = 0 M) values estimated by Eq. (4) are slightly higher a w than the ones estimated by Eq. (5) (except for w w pKa3 (I = 0 M) of quercetin). This fact was already observed by Avdeef et al. [5]

1012

J.M. Herrero-Mart´ınez et al. / Talanta 74 (2008) 1008–1013

for neutral acids. When Yasuda–Shedlovsky extrapolation was carried out from the methanol rich region (>30% in weight), which is the case of fisetin and quercetin, the extrapolated pKa values are slightly higher that the ones directly obtained in pure water. Anyway, the values obtained from potentiometric data are consistent and lead to reliable w w pKa (I = 0 M) values. For fisetin the w w pKa1 (I = 0 M) estimated by potentiometry is slightly higher and the w w pKa2 (I = 0 M) slightly lower than the ones obtained by CE. For quercetin the w w pKa1 (I = 0 M) estimated by potentiometry is also slightly higher than the one w obtained by CE whereas the w w pKa2 (I = 0 M) and w pKa3 (I = 0 M) are similar with the CE ones. In general the pKa values estimated by Eq. (5) are closer to the pKa values obtained by CE than those estimated by Eq. (4). 4. Conclusions

Fig. 2. The Yasuda–Shedlovsky plots of the studied flavonols. () Plot for pKa1 ; () plot for pKa2 ; () plot for pKa3 .

Table 2 (I = 0 M) values of studied flavonols

w pK a w

w pK determined a w in water

Morin pKa1 pKa2 pKa3

8.25 10.30

Fisetin pKa1 pKa2 Quercetin pKa1 pKa2 pKa3 a

From Ref. [4].

Estimated w w pKa From Eq. (4).

From Eq. (5).

4.99 ± 0.00 8.23 ± 0.03 10.34 ± 0.04

4.97 ± 0.09 8.29 ± 0.08 10.33 ± 0.11

CEa w w pKa values

5.06 8.64 10.62

7.70 ± 0.03 9.54 ± 0.04

7.57 ± 0.03 9.34 ± 0.04

7.36 9.71

7.71 ± 0.06 9.44 ± 0.04 11.46 ± 0.07

7.59 ± 0.06 9.33 ± 0.04 11.56 ± 0.10

7.19 9.36 11.56

The aqueous pKa values (w w pKa (I = 0 M)) of three flavonols sparingly soluble in water and with several ionizables hydroxyl groups have been estimated by potentiometry. Because of their poor solubility in water, they have been titrated in methanol/water mixtures and two equations (Eqs. (4) and (5)) have been used to estimate the aqueous values. For morin, which is the most soluble studied flavonol, the aqueous pKa values obtained by the two equations are practically the same and equal to the ones obtained by titration in pure water. For fisetin and quercetin, the estimations were done from a high percentage of methanol, and the aqueous pKa values obtained by the two equations are slightly different, but not by more than in 0.15 units. Eq. (5) allows a very good estimation of w w pKa (I = 0 M) from only one experimental ss pKa (I = 0.15 M) value, determined in any methanol/water mixture, whereas Eq. (4) requires at least three (in fact five or six) independent titrations in different methanol/water mixtures. Thus, we recommend the use of Eq. (5). These compounds present two or three pKa values higher than 7 and very close one to each other, then the determination of these pKa values is a difficult task by any technique (i.e. potentiometry or CE). The differences in w w pKa (I = 0 M) values obtained by potentiometric method with respect to the ones obtained previously by CE [4] (not more than 0.4 units) are in an acceptable agreement, given the difficulty of obtaining these pKa values. Since the pKa values are very close, about 15 buffers with close pH were necessary to obtain accurate determinations by CE method. pKa determination by potentiometry can be done by a direct and automated acid–base titration, previous appropriate calibration, and thus is much faster than the CE determination. Acknowledgements We thank financial support from the Spanish Government (Project CTQ2004-00965/BQU). References [1] A. Albert, E.P. Serjeant, Potentiometry and Potentiometric Titrations, John Wiley and Sons, New York, 1984.

J.M. Herrero-Mart´ınez et al. / Talanta 74 (2008) 1008–1013 [2] A. Albert, E.P. Serjeant, The determination of Ionization Constant, Chapman and Hall, London, 1971. [3] Y. Ishihama, Y. Oda, N. Asakawa, J. Pharm. Sci. 83 (1994) 1500. [4] J.M. Herrero-Mart´ınez, M. Sanmartin, M. Ro´ses, E. Bosch, C. R`afols, Electrophoresis 26 (2005) 1886. [5] A. Avdeef, J.E.A. Comer, S.J. Thomson, Anal. Chem. 65 (1993) 42. [6] A. Avdeef, K.J. Box, J.E. Comer, M. Gilges, M. Hadley, C. Hibbert, W. Patterson, K.Y. Tam, J. Pharm. Biomed. Anal. 20 (1999) 631. [7] K. Takacs-Novak, K.J. Box, A. Avdeef, Int. J. Pharm. 151 (1997) 235. [8] R. Ruiz, M. Ro´ses, C. R`afols, E. Bosch, Anal. Chim. Acta 550 (2005) 210. [9] C.A. Rice-Evans, L. Parker, Flavonoids in Health and Diseases, Marcel Dekker, New York, 1998. [10] L. Magnani, E.M. Gaydou, J.C. Hubaud, Anal. Chim. Acta 411 (2000) 209. [11] G. Cao, E. Sofic, R.L. Prior, Free Radic. Biol. Med. 22 (1997) 749. [12] W. Bors, W. Heller, C. Michel, M. Saran, in: Packer Lester, Alexander N. Glazer (Eds.), Flavonoids as antioxidants: Determination of Radicalscavenging Efficiencies. Methods in Enzymology, Academic Press, 1990, pp. 343–355. [13] C.A. Rice-Evans, N.J. Miller, G. Paganga, Free Radic. Biol. Med. 20 (1996) 933. [14] K. Lemanska, H. Szymusiak, B. Tyrakowska, R. Zielinski, A.E.M.F. Soffers, I.M.C.M. Rietjens, Free Radic. Biol. Med. 31 (2001) 869. [15] S.V. Jovanovic, S. Steenken, M. Tosic, B. Marjanovic, M.G. Simic, J. Am. Chem. Soc. 116 (1994) 4846.

1013

[16] Y. Fan, Y. Zheng, Anal. Chim. Acta 281 (1993) 353. [17] A. Niazi, A. Yazdanipour, J. Ghasemi, M. Kubista, Collect. Czech. Chem. Commun. 71 (2006) 1. [18] B. Bujonek, Pol. J. Chem. 67 (1993) 1339. [19] S.R. Jagasis, B.C. Haldar, J. Indian Chem. Soc. 40 (1963) 287. [20] V. Kuntic, N. Pejic, S. Micic, D. Malesev, Z. Vujic, Pharmazie 58 (2003) 439. [21] N. Sauerwald, M. Schwenk, J. Polster, E. Bengsch, Z. Naturforsch. B: Chem. Sci. 53 (1998) 315. [22] H. Liu, G. Yang, D. Wang, S. Sun, J. Ma, Chin. J. Chem. 19 (2001) 675. [23] M. Thompson, C.R. Williams, G.E.P. Elliot, Anal. Chim. Acta 85 (1976) 375. [24] E.N. Muzafarov, B.N. Ivanov, A.N. Mal’yan, E.K. Zolotareva, Biochem. Physiol. Pflanz. 181 (1986) 381. [25] M. Yasuda, Bull. Chem. Soc. Jpn. 32 (1959) 429. [26] F. Rived, I. Canals, E. Bosch, M. R´oses, Anal. Chim. Acta 439 (2001) 315. [27] R.G. Bates, Determination of pH: Theory and Practice, 2nd ed., Wiley, New York, 1973. [28] T. Shedlovsky, Electrolytes, Pergamon Press, New York, 1962. [29] R. Ruiz, C. R`afols, M. Ro´ses, E. Bosch, J. Pharm. Sci. 92 (2003) 1473. [30] M. Ros´es, F. Rived, E. Bosch, J. Chromatogr. A 867 (2000) 45–56. [31] Sparc, http://ibmlc2.chem.uga.edu/sparc/. [32] P.K. Agrawal, H. Schneider, Tetrahedron Lett. 24 (1983) 177.