Reduction of diprotonated form of aryl hydrazones

Journal of

Electroanalytical Chemistry Journal of Electroanalytical Chemistry 589 (2006) 7–14 www.elsevier.com/locate/jelechem

Reduction of diprotonated form of aryl hydrazones M.S. Baymak a, H. Celik a, H. Lund b, P. Zuman a

a,*

Department of Chemistry, Clarkson University, Box 5810, 8 Clarkson Avenue, Potsdam, NY 13699-5810, USA b Department of Organic Chemistry, Aarhus University, Aarhus, DK, Denmark Received 5 September 2005; received in revised form 17 November 2005; accepted 24 November 2005 Available online 20 March 2006

Abstract Hydrazones derived from aromatic aldehydes and ketones are reduced at pH 2 to about 8 in a four-electron step. The species reduced in this step bears two positive charges on adjacent nitrogen atoms. This has been proved by the pH-dependence of half-wave potentials of N,N,N-trialkylhydrazonium ions, which indicates a protonation of the azomethine nitrogen prior to the first electron uptake. Similar species with two adjacent positive charges is generated by diprotonation of hydrazones, adsorbed at the electrode surface. The existence of such species as reactive intermediates at electrode surface has been in some instances confirmed, based on steep plots of imax/ id = f(pH). The steep shape of these plots has been confirmed for acetophenone (II) and fluorenone (III) hydrazones using conventional buffers and for benzophenone hydrazone (IV) after extrapolation of buffer concentration to zero. The shape of the imax/id = f(pH) can be namely achieved by protonation not only by H+ ions, but also by acid buffer components, as in general acid catalysis.  2005 Elsevier B.V. All rights reserved. Keywords: Hydrazones; N,N,N-trialkylhydrazonium ions; Electroreduction; Diprotonation; Adsorption; Polarography

1. Introduction Most hydrazones derived from aromatic aldehydes and ketones are reduced at pH lower than about 7 or 8 in a single four-electron step [1,2]. It has been proposed and recently confirmed [3] that the reduction involves an initial cleavage of the N–N bond followed by a reduction of the imine. Limiting currents of arylhydrazones, which are up to about pH 7 pH-independent, decrease gradually at higher pH-values. This behavior resembles that of the more extensively studied oximes [3–6] and was attributed [1–6] to a proton transfer preceding the uptake of the first electron. Nevertheless, no evidence was presented for the number of protons transferred. The role of an antecedent protonation of the azomethine bond was also indicated by the observation of the pH-dependence of reduction potentials of quarternized hydrazones [7]. *

Corresponding author. Tel.: +1 315 268 2340; fax: +1 315 268 6610. E-mail address: [email protected] (P. Zuman).

0022-0728/$ - see front matter  2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jelechem.2005.11.038

It is the aim of this contribution to demonstrate that the reduction of arylhydrazones can occur in the dicationic form, with two positive charges on two adjacent nitrogen atoms. Such species are generated at pH lower than about 8 in a heterogeneous process at or close to the surface of the electrode – either by the diprotonation of hydrazones unsubstituted on the amino nitrogen, or by the protonation of the azomethine nitrogen in N,N,N–trialkylhydrazonium ions. 2. Experimental 2.1. Instrumentation Current–voltage curves were recorded by using Sargent Model 4001 Polarograph and IBM EC/225 Voltammetric Analyzer combined with IBM 7424 MT X-Y-T Recorder as well as capillary electrodes with characteristics of m = 2.5 mg s1, t1 = 3.0 s at h = 64 cm. A two-electrode electrolytic cell is used with a S.C.E. separated by a liquid junction (Kalousek cell).

8

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

2.2. Chemicals

Table 1 Composition of common buffers used

Benzaldehyde hydrazone (I) was prepared in solution in situ by reacting 0.2 mM solution of benzaldehyde with 2 mM solution of hydrazine in an acetate buffer, pH 5.7, for 60 min. Under these conditions the formation of azine was negligible. Prepared reaction mixture was used as a stock solution in electroanalytical experiments. Acetophenone hydrazone (II), fluorenone hydrazone (III) and benzophenone hydrazone (IV) were supplied by Aldrich. Benzaldehyde N,N,N-trimethyl hydrazonium iodide (V), acetophenone N,N,N-trimethyl hydrazonium iodide (VI), benzophenone N,N,N-trimethyl hydrazonium iodide (VII) and fluorenone N,N,N-trimethyl hydrazonium perchlorate (VIII) were prepared at the Department of Organic Chemistry at the University of Aarhus (Denmark) by reaction of the dimethylhydrazones with methyl iodide in acetonitrile.

pH

Acid

[Acid] (M)

Base

[Base] (M)

3.10 3.60 4.20 4.70 5.20 5.70 6.10 6.30 6.50 6.70 7.10 7.25 8.00 8.35 8.50 9.10 9.30 9.75 8.00 8.90 9.25 10.4 10.8 11.2

H3PO4 CH3COOH

0.010 0.250 0.200 0.100 0.050 0.300 0.150 0.150 0.100 0.060 0.030 0.020 0.033 0.020 0.095 0.100 0.075 0.036 0.100 0.100 0.100 0.012 0.100 0.100

NaH2PO4 CH3COONa

0.160 0.025 0.063 0.100 0.160 0.050 0.050 0.075 0.100 0.100 0.100 0.100 0.033 0.080 0.015 0.050 0.075 0.140 0.100 0.032 0.010 0.137 0.020 0.063

N N H2

R

C I II

N N H2

N N H2 R=H R=CH3

III + N N R3

R

IV + N N R3

+ N N (C H 3 ) 3

C

a

NaH2PO4

BarbHa H3BO3

NH4Cl

H3BO3 Na2HPO4

Na2HPO4

Barb H2 BO 3

NH3 H2 BO 3 Na3PO4

Barbital.

For the initial information about the dependence of current–voltage curves of studied hydrazones on pH, a set of buffers was used denoted in text below as ‘‘common buffers’’. Instead of universal buffers sometimes used, which present some disadvantages for mechanistic studies, a series of simple buffers were used (Table 1). Each of these buffers contained only a single weak acid component. The chemicals used for preparation of buffers were of analytical quality. For quantitative evaluation of the role of concentration of buffer components, that act as proton donors in the electroreduction of hydrazones, several series of buffers were prepared (Table 2). In each series the analytical concentration of the buffer was varied, but the ratios of concentrations of the acid and base component of the buffer was kept constant. Ionic strength was kept constant by addition of a solution of sodium chloride. Thus the pH in each series remained constant.

ried out in buffered solutions containing concentrations of acetonitrile lower than 30%. Presence of acetonitrile up to 30% v/v as a co-solvent was needed to secure the solubility of 0.02–0.2 mM solutions of some of the investigated hydrazones. The pH-values were measured using a glass electrode standardized with using buffers containing 15% and 30% acetonitrile [8]. A stock solution of the investigated hydrazone derivative was added to the buffered solution after deaeration, with gelatin added in some cases to prevent streaming maxima. The following final concentrations were used: For the investigation of compound I the reaction mixture prepared as described above was diluted 10-fold by the individual buffer solution to have a hydrazone concentration of 0.02 mM. For compound II 0.02 mM in 30% acetonitrile–water mixture, for III 0.05 mM in 15% acetonitrile–water mixture, for IV 0.2 mM in water, for V 0.1 mM in water with 0.004% gelatin and 1% acetonitrile, for VI 0.2 mM in water with 0.005% gelatin, for VII 0.1 mM in 15% acetonitrile–water mixture with 0.005% gelatin and for compound VIII 0.05 mM in the presence of 15% acetonitrile. After brief final purging by nitrogen the current–voltage curves were recorded. Unless otherwise stated, highest mean current (imax) was measured.

2.4. Procedure

3. Results and discussion

The stock solutions (0.01 M) were prepared for compounds IV, VI and VII in water, for compounds II, III, V, and VIII in acetonitrile. Simple buffer solutions (phosphate, acetate, borate, veronal, ammonia–ammonium ions and glycine) were used. Recording of i–E curves was car-

The first evidence of a hydrazone derivative bearing two positive charges on adjacent nitrogen atoms is based on pH-dependence of half-wave potentials of the reduction of N,N,N-trialkylhydrazonium ions. The shift of half-wave potentials (E1/2) of such compounds has been mentioned

V VI

R=H R=CH3

V II

V III

2.3. Buffers used

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

9

Table 2 Buffers used for verification of number of protons in the reaction preceding the first electron uptake ;ionic strength, l, kept constant by addition of a solution of NaCl Acid

Concentration varied From

Base

To

Concentration varied

pH varied

From

To

From

l To

Acetophenone hydrazone NaH2PO4 0.005 H3BO3 0.050 BarbHa 0.0016 0.068 GlyH+b NH4+ 0.010

0.200 0.200 0.036 0.270 0.400

Na2HPO4 H2 BO 3 Barb Gly NH3

0.020 0.005 0.004 0.014 0.100

0.250 0.100 0.051 0.170 0.400

6.70 7.55 7.55 8.20 8.60

7.60 8.60 8.45 9.00 9.55

0.5 0.5 0.5 0.5 0.5

Benzophenone hydrazone 0.006 NaH2PO4 NH4+ 0.003 H3BO3 0.030 0.022 GlyH+b Na2HPO4 0.020

0.300 0.300 0.300 0.220 0.350

Na2HPO4 NH3 H2 BO 3 Gly Na3PO4

0.010 0.003 0.003 0.004 0.007

0.300 0.300 0.030 0.240 0.127

6.05 8.10 8.35 8.60 10.8

7.25 10.1 – 10.45 11.3

0.5 0.5 0.5 0.5 1.0

a b

Barbital. Glycine.

early [7] and was observed and confirmed for several compounds of this type in this research (Table 3). The observed shifts indicate that compounds with grouping ˜C@NHþ NRþ 3 are the electroactive species. For an irreversible reduction, such as the electroreduction of hydrazonium ions, the shifts of E1/2 to more negative values with increasing pH can only be interpreted as due to a rapid þ þ conversion of ˜C@N–NRþ 3 into ˜C@NH NR3 , which – as all conjugate acids – is more readily reduced than the base form. As long as the rate of protonation of the conjugate base predominating in the bulk of the solution is sufficiently rapid to convert all of the grouping ˜C@N–NRþ 3 into ˜C@NHþ –NRþ 3 , the E1/2 will be shifted to more negative potentials with increasing pH at a pH considerably higher than the pKa-value. Such shifts are observed in the medium pH-range. The equilibrium concentration of the protonated form bearing grouping ˜C@NHþ NRþ 3 remains very low, as the pKa of the conjugate acid is <2. But electrolysis is a dynamic method. When the concentration of the electroactive species is depleted by its reduction in the vicinity of the electrode, it is replenished in the vicinity of the electrode surface both by the diffusion from the bulk of the solution and by chemical reactions. It has been recently confirmed [3] that the first step in the four-electron reductions of protonated hydrazones is a cleavage of the N–N bond. Hence the protonation on the a-nitrogen facilTable 3 Dependence of half-wave potentials of some hydrazones and hydrazonium ions on pH Compound

pH-range

dE1/2/dpH (V/pH)

I II III IV V VI VII VIII

5.0–7.0 6.5–8.0 3.5–10 4.5–6.5 2.0–7.0 2.0–5.5 1.0–3.0 0.0–3.0

0.088 0.112 0.075 0.127 0.092 0.120 0.077 0.090

itates reductive cleavage of the N–N bond. Furthermore, the positive charge on the b-nitrogen assures that the NR3 will be a good leaving group. In analogy with the reducibility of the grouping ˜C@NHþ NRþ 3 it can be assumed that similar factors, namely facilitation of the cleavage of the N–N by a positive charge on a-nitrogen and conversion of the b-amino group into a good leaving group by protonation, will facilitate reductive cleavage of the N–N bond in the diprotonated form of hydrazones with a grouping ˜C@NHþ NHþ 3. The second approach to confirmation of the diprotonated form as the electroactive form in the reduction of hydrazones is based on the shape of i = f(pH) plots of reduction waves of hydrazones. The situation here is complicated by the fact that the reduction of hydrazones is a heterogeneous process. This is indicated by the shape of i–E waves: For currents governed by diffusion, once the limiting current is reached, it remains independent of potential until the next reduction or oxidation occurs. Electrochemical processes involving species adsorbed within potential range in which the limiting current is observed are often manifested by a decrease, a dip, of the limiting current [9]. The potential range in which this dip occurs depends on the potential region in which the compound is adsorbed. As this potential range increases with increasing concentration of the studied compound, the width of the dip is also a function of this concentration. Such dips are observed on current–voltage curves of hydrazones (Fig. 1). The current at a given potential depends on pH and the i = f(pH) plots have a shape of a dissociation curve (Fig. 2). The shape of this curve remains similar for currents measured at different potentials, only the curves are shifted along the pH axis (Fig. 2). As the decrease in current corresponds to a decrease in the rate of protonation, it can be concluded that the rate of protonation, which is lower in the presence of an adsorbate, decreases with increasingly negative potentials. Thus the antecedent chemical reaction gradually varies with increasingly more

10

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

Fig. 1. Dependence of polarographic current–voltage curves of benzophenone hydrazone (IV) on pH. Curves recorded in buffers containing 2% v/v acetonitrile. (1) Acetate buffer, pH 5.1; (2) and (3) phosphate buffers, pH 6.1 and 7.3; (4) and (5) borate buffers, pH 8 and 8.5. Curves starting at 0.4 V (SCE).

1.0

i/id

4 3 2 1

protonated by one or two protons (see below) and as long as the rate of this protonation is sufficiently rapid, all of the base is converted into the acid form. The acid form is – as in all known other cases – reduced more easily, at more positive potentials than its conjugate base. The rate of protonation depends on concentration of the base at the electrode surface, which in turn depends on adsorption. At sufficiently negative potentials the rate of formation of the protonated form becomes in some cases so slow, that a reduction of the conjugate base may take place. As stated above the plots of the current measured at different potentials, plotted as a function of pH have a shape of a dissociation curve, which are shifted along the pH axis, depending on the potential, at which the current was measured. The inflection point of such dissociation curve, denoted pK 0 , is observed at lowest pH for the current measured at a potential corresponding to the lowest point of the dip on the i–E curve, as at that potential the rate of protonation is slowest. The value of the rate constant of protonation (kr) decreases namely with decreasing difference between pK 0 and thermodynamic pKa [10,11], as follows from the equation 2pK 0  pKa = const. + log kr. Lower pK 0 at a given pKa indicates a lower rate of protonation in the potential range, where the dip is observed. For homogeneous reactions, where a single acid is in equilibrium with a single base as in reactions: HA ¢ H+ + Aor BH+ ¢ B + H+ and the acid form HA or BH+ is reduced in current i, the decrease in i with pH follows [10,11]: i=id ¼

0.5

0:886ðt1 k 1 =K 1 Þ1=2 ½Hþ  1=2

1 þ 0:886ðt1 k 1 =K 1 Þ

½Hþ 

ð1Þ

;

where id is the diffusion controlled current, K1 the acid dissociation constant of the reduced conjugate acid, k1 is the rate constant of protonation of A or B and t1 is the drop time. For dibasic acids, where the form H2A or BH2þ is 2 reduced, the reduction current i of H2A or BH2þ 2 depends on acidity as follows from [12,13]: 1=2

i=id ¼

0.0 6

7

8 pH

9

10

Fig. 2. Theoretical plots of the dependence of i/id on pH for currents measured at various potentials for current–voltage curves with a dip. Currents i measured at the following potentials: (1) at maximum current (pK 0 8.3); (2) at potential by 0.1 V more negative (pK 0 8.1); (3) at potential by 0.2 V more negative (pK 0 7.9); (4) at potential by 0.3 V more negative (pK 0 7.7).

negative potentials from a predominantly homogeneous to a predominantly heterogeneous process. The pH-dependence of i–E curves of hydrazones can be thus interpreted as follows: The reduction at pH > 7 takes place in a solution where in the bulk of the solution the concentration of the conjugate base of the hydrazone predominates. Base form adsorbed at the electrode surface is

0:886ðt1 k 1 =K 1 Þ

1 þ ð0:886ðt1 k 1 =K 1 Þ

2

f½Hþ  =ð½Hþ  þ K 2 Þg

1=2

2

f½Hþ  =ð½Hþ  þ K 2 ÞgÞ

.

ð2Þ

In this equation K1 is the first and K2 the second dissociation constant of H2A or BH2þ 2 and k1 the rate of protonation of HA or BH+ and t1 the drop-time. The plots of i/id = f(pH) for both Eqs. (1) and (2) have a shape of a dissociation curve, but with different slopes. For the reduction of a monobasic acid following Eq. (1), the dissociation curve is less steep. In this case the decrease of the current from i/id = 0.9 to i/id = 0.1 takes place within 2 pH-units. For the reduction of the acid form H2A or BH2þ 2 of the dibasic acid, when Eq. (2) is applicable and pK1  pK2 and pK 0 > pK2, the decrease of the current as a function of pH from i/id = 0.9 to i/id = 0.1 takes place within a single pH-unit (Fig. 3). To verify, if the variations of reduction currents of hydrazones with pH follows Eqs. (1) or (2), the highest

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

reducible acid form and the limiting current is controlled by the rate of diffusion, imax = id. At pH > (pK 0  2) where the decreased rate of protonation converts only part of the base form into the more easily reducible acid, the current imax decreases with increasing pH (Fig. 1). The plots of imax/id = f(pH) for currents measured in commonly used buffers (cf. Section 2) (Figs. 4 and 5) are compared first. For acetophenone (Fig. 4a) and fluorenone (Fig. 4b) hydrazones the measured values of imax/id = f(pH) fitted well plots corresponding to Eq. (2), proving that a diprotonated form is reduced. The situation was complicated for fluorenone hydrazone (Fig. 4b) by the fact that for this compound the decrease of the four-electron reduction of the diprotonated form is accompanied by an increase of a two-electron wave attributed to the reduction of the monoprotonated form. Hence the current imax decreases from a four-electron to a two-electron wave and verification of Eq. (2) was restricted to pH < pK 0 (Fig. 4b). When imax values were obtained in buffers of commonly used compositions (cf. Section 2) the plots of imax/id = f(pH) for benzaldehyde and benzophenone hydrazones, on the other hand, have shown a better fit to Eq. (1) than to Eq. (2) (Fig. 5). This is, at least partly, attributed to the fact that values of imax depend not only on pH, but also on the kind and concentration of the buffer used. Such dependence indicates that the diprotonated form can be generated in the vicinity of the electrode not only by addition of H+ ions, but also by a reaction with the acid component of the buffer (HA). The role of the kind and concentration of the acid buffer component HA on the dependence of imax = f(pH) has been tested by following the dependence of imax on concentration

1.0 2 1

i/id

0.5

0.0 pH Fig. 3. Plots of theoretical dependences of i/id on pH. (1) Dependence for the reduction of the acid form of a monobasic acid, following Eq. (1); (2) dependence of the fully protonated form of a dibasic acid, if pK1  pK2 and pK 0 > pK2, following Eq. (2).

current on the i–E curve was measured and denoted imax. At pH < (pK 0  2), where a sufficiently high rate of protonation converts all the base form present into the more easily

1.0

1.0

i/id

i/id

11

0.5

0.0

0.5

0.0

6

7

8

9 pH

a 4

10

11

5 b

7

9

11

13

15

pH

Fig. 4. Dependence of currents of (a) 2 · 10 M acetophenone hydrazone in 30% CH3CN and (b) 5 · 105M fluorenone hydrazone in the presence of 15% CH3CN on pH. Theoretical curves corresponding to Eq. (2), experimental points: currents measured in the following buffers: r, phosphate; m, borate; h, barbital, and j, NH3–NH4Cl.

12

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

1.0

i/id

i/id

1.0

0.5

0.5

0.0

0.0 5

7

9

11

pH

a

6

8

10 pH

b

12

Fig. 5. Dependence of currents of 2 · 104M solution of (a) benzaldehyde hydrazone (I) in the presence of 30% CH3CN; (b) benzophenone hydrazone (IV) in the presence of 2% CH3CN. Theoretical curves corresponding to Eq. (1), experimental points: currents measured in the following buffers: r, phosphate; d, borate; j, barbital, and h, NH3–NH4Cl.

1.75

i (μA)

i (μA)

1.0

0.5

0.0

0.75 0.0

a

1.25

0.1

0.2 [Buffer]

0.0

0.3 M b

0.2

0.4 M [Buffer]

Fig. 6. Dependence of maximum currents on buffer concentration (a) for 2 · 104M acetophenone hydrazone (II) in borate buffers pH: r, 7.55; j, 7.85; m, 8.1; d, 8.6; (b) for 2 · 104M benzophenone hydrazone (IV) in glycine buffers pH: r, 8.6; j, 8.95; m, 9.1; d, 9.3.

of the buffer at a constant pH (Fig. 6). The linearity of the imax = f([HA]) plots indicates that only one acid component of the buffer is involved in the protonation of the investigated hydrazones. Such effect of acid buffer components as proton donors have been reported for reduction of some aldehydes [14], and deoxybenzoins [15], but not for hydraz-

ones. The value of pK 0 , and hence the position of the plot of imax/id = f(pH) on pH axis, thus depends also on the nature of the acid buffer component. To express the dependence of imax on pH for the protonated form of hydrazone generated in the vicinity of the electrode solely by a reaction with H+ ions, the value of imax was

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

1.0

i/id

i/id

1.0

0.5

0.5

0.0

0.0 6 a

13

8 pH

5

10 b

7

9

11

pH

Fig. 7. Dependence of i/id for the reduction of (a) 2 · 104M acetophenone hydrazone (II) and of (b) 2 · 104M benzophenone hydrazone (IV) for currents extrapolated to zero buffer concentration. Buffers used: (a) m, glycine; d, phosphate; n, barbital; (b) h, NH3–NH4Cl; j, borate buffers; r, glycine; d, phosphate.

determined at several buffer concentrations. These values were extrapolated to zero buffer concentration ði0max Þ. For acetophenone hydrazone (II) the values of i0max , obtained for phosphate, barbital and glycine buffers fitted a single curve, corresponding to Eq. (2) (Fig. 7a). Taken into account the limited accuracy of finding the value of pK 0 the pK 0 value obtained from this curve (7.85) is in acceptable agreement with the pK 0 = 7.80 obtained from the imax/id = pH plot, when imax values were measured in conventional buffers (Fig. 4a). A different pK 0 value was observed for imax measured in ammonia–ammonium ion buffers (Fig. 4a). A linear dependence on concentration of ammonia–ammonium ion buffers, as well as the unexceptional behavior of glycine buffers, where the proton donor is a ANHþ 3 group, rule out an exceptionally strong effect of NHþ ions as proton donors, not following the Brønsted 4 relationship. The possibility of addition of NH3 to the azomethine bond was excluded by determination of imax in buffers, containing a constant concentration of NH3 and pH varied by changing the concentration of NHþ 4. Thus, a strong effect of double-layer composition on the rate of protonation in the vicinity of the electrode seems to be the most probable interpretation of the role of ammonia–ammonium ion buffers. The values of imax obtained in borate buffers differed from those obtained in all other buffers. This indicates a specific interaction of borate ion with the hydrazone that was not further investigated. For benzophenone hydrazone (IV) values of imax obtained in conventional buffers were better fitting Eq. (1), but when values i0max , obtained by extrapolation to zero buffer concentration were used, they were best fitted by Eq. (2) (Fig. 7b). The pK 0 value obtained for imax in conven-

tional buffers (8.25) also differed from the value pK 0 = 9.0 obtained for i0max at zero buffer concentration. Both for acetophenone hydrazone (II) (Fig. 7a) and benzophenone hydrazone (IV) (Fig. 7b) deviations from theoretical plot of i0max =id ¼ f (pH) were observed at pH higher than about (pK 0 + 0.5). These are ascribed to a decreased accuracy in measurement of small waves and role of adsorption, which plays a larger role at smaller currents i0max . The reduction of the diprotonated species of II, III and IV is thus supported also by the imax/id = f(pH), respectively i0max =id ¼ f (pH) plots. Similar treatment of the i0max =id ¼ f (pH) plots for the benzaldehyde hydrazone (I) (Fig. 5b) was prevented by the preparation of this hydrazone in situ. Presence of a needed excess of hydrazine limited investigation of the role of acid buffer component, as NH2 NHþ 3 was also present as proton donor. The dilution of the reaction mixture varied both pH and equilibrium concentration of hydrazine. The apparent fit of imax/id = f(pH) Eq. (1), when common buffers were used, is attributed to the role of acid buffer components as additional proton donors. Transfer of two protons taking place before the uptake of the first electron is further supported by the size of the slope of the plot of E1/2 = f(pH). Values of dE1/2/ dpH > 0.060 (Table 3) strongly indicate presence of more than one antecedent proton transfer. Quantitative treatment of the shifts of half-wave potentials is prevented by the heterogeneous nature of the electrode process. Presence of two adjacent positively charged nitrogen atoms in hydrazine derivatives has been assumed to be present in the reactive species in benzidine rearrangement

14

M.S. Baymak et al. / Journal of Electroanalytical Chemistry 589 (2006) 7–14

[16] and in some reactions in the gas phase [17], but to our best knowledge our experimental evidence offers a first proof of existence of diprotonated hydrazonium ions or protonated N-trialkylhydrazonium ions, briefly mentioned in our preliminary report [18]. 4. Conclusions Electroreductions of hydrazones and N,N,N-trialkylhydrazonium ions takes place in species bearing two positive charges an adjacent nitrogen atoms. Such species are formed at the electrode surface either by protonation of the N,N,N-trialkylhydrazonium ions or by diprotonation of hydrazones. Whereas species with –N+H–N+H– grouping have been reported for hydrazinium ions, our observation seems to be the first reported of diprotonated hydrazonium ions. Protonation of the conjugate base can be achieved not only by H+ ions, but also by acid components of used buffers. Acknowledgments M.S. Baymak acknowledges generous financial support by the Marmara University, Istanbul, Turkey. Helpful comments by Dr. J. Ludvik are also acknowledged.

References [1] H. Lund, Acta Chem. Scand. 13 (1959) 249. [2] T.V. Troepolskaya, G.K. Budnikov, The Electrochemistry of Azomethines, Nauka, Moscow, 1989, p. 224 (in Russian). [3] M.S. Baymak, H. Celik, H. Lund, P. Zuman, J. Electroanal. Chem. 581 (2005) 284. [4] P. Souchay, S. Ser, J. Chim. Phys. 49 (1952) C172. [5] H.J. Gardner, W.P. Georgans, J. Chem. Soc. (1956) 4180. [6] H. Lund, O. Hammerich (Eds.), Organic Electro chemistry, fourth ed., Marcel Dekker, New York, 2001, p. 435. [7] H. Lund, Discussion Faraday Soc. 45 (1968) 193. [8] J. Barbosa, V. Sanz-Nebot, J. Chem. Soc., Faraday Trans. 90 (21) (1994) 3287. [9] S.G. Mairanovskii, Catalytic and Kinetic Waves in Polarography, Plenum Press, New York, 1968. [10] R. Brdicˇka, K. Wiesner, Collect. Czech. Chem. Commun. 12 (1947) 139. [11] J. Koutecky´, Collect. Czech. Chem. Commun. 18 (1953) 597. [12] V. Hanusˇ, R. Brdicˇka, Khimija 2 (1951) 28. [13] J. Koutecky´, Collect. Czech. Chem. Commun. 19 (1954) 1093. [14] R. Brdicˇka, Collect. Czech. Chem. Commun. 20 (1955) 387. [15] P. Zuman, B. Turcsanyi, Collect. Czech. Chem. Commun. 33 (1968) 3090. [16] H.J. Shine, J. Phys. Org. Chem. 2 (1989) 491. [17] V.G. Nenajdenko, N.E. Shevchenko, E.S. Balenkova, I.V. Alabugin, Chem. Rev. 103 (2003) 229–282. [18] M.S. Baymak, H. Celik, J. Ludvik, H. Lund, P. Zuman, Tetrahedron Lett. 45 (2004) 5113.