A

bwopoTeoa©

K

1

p[ecpwplatkt

t

0d

Electrochimica Acta 51 (2006) 5845–5852

Two reduction waves of oximes and imine formation in acidic media

H. Celik a,1, J. Ludvik b,1, P. Zuman a,∗,1

a Department of Chemistry, Clarkson University, Potsdam, NY, 13699-5810, USAb J. Heyrovsky Institute of Physical Chemistry, Academy of Sciences of the Czech Republic,

Dolejskova 3, 182 23 Prague 8, Czech Republic

Received 17 January 2006; received in revised form 9 March 2006; accepted 12 March 2006Available online 18 April 2006

bstract

The separation of the two two-electron waves in acidic media in solutions of some aromatic aldoximes and ketoximes is attributed to differencesetween the pKa-values of oximes (which are smaller than about 1.0) and those of protonated forms of corresponding imines. Dependences of half-ave potentials on pH show that the lower limit pKa of imine varies from −0.25 for p-formylbenzaldehyde oxime to 2.2 for p-fluorobenzaldehydexime. For the benzaldehyde and acetophenone oximes bearing substituents, which have pKa higher than −1.50, a good agreement was observed forKa-values obtained spectrophotometrically and pKox values obtained from the change in the slope of E1

1/2 = f (H0) plots. Past misrepresentationsf the electroreduction mechanism of protonated forms of imines is attributed to insufficiently controlled compositions of supporting electrolytes.

he reduction potential of protonated forms of oximes in the more positive wave i1 is affected by the concentration of halides in the supportinglectrolyte. This is attributed to an ion-pair formation, between these anions and protonated forms of oximes. The reduction of the protonated formf imines at acidity, where in the bulk predominates the same protonated form that is reduced, is independent of the nature and concentration ofnions of the supporting electrolyte.

stren

a>bbda

irbbiCi

2006 Elsevier Ltd. All rights reserved.

eywords: Oximes; Electroreduction; Polarography; Imine intermediate; Ionic

. Introduction

The four-electron reduction of oximes and its dependence onH has been recognized early [1,2] and reviewed more recently3–5]. The four-electron process may be in principle initiatedither by the hydrogenation of the azomethine bond or by theleavage of the N–O bond. In his seminal paper Lund [6] pro-osed that the second alternative is the first step. This propositionas based on the known electrochemical behavior of the twoossible reduction intermediates—the imine and a hydroxy-amine derivative. The only experimental evidence was indirectnd based on identification of testosterone propionate as an elec-rolysis product of testosterone propionate oxime. The isolatedetone was assumed to be formed by hydrolysis of the testos-

erone propionate imine [6].

It has been demonstrated recently [7] that, at pH higherhan 2.0 and lower than about 7.0 or 8.0, aromatic oximes

∗ Corresponding author. Tel.: +1 315 268 2340; fax: +1 315 268 6610.E-mail address: zumanp@clarkson.edu (P. Zuman).

1 Active member of ISE.

odtuy

t

013-4686/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.oi:10.1016/j.electacta.2006.03.020

gth; Halide ions

re reduced in diprotonated form. In the resulting groupingC NH+OH2

+ the cleavage of the N–O bond is facilitatedy the presence of OH2 as a good leaving group as well asy the positive charge on the azomethine nitrogen. Similariprotonated form has been identified as the reducible specieslso in the reduction of hydrazones [8].

For the reduction of related hydrazones the formation ofmines as intermediates in their four-electron reduction hasecently been proved [9]. The proof has been made possibley comparison of current–voltage curves of hydrazones ofenzophenone and fluorenone with those of correspondingmines, that are sufficiently stable between pH 6 and 10.omparison of reduction potentials confirmed formation of

mine intermediates.For oximes Lund [10] reported a similar approach, based

n comparison of the oximes with the imine of 2,4-ihydroxybenzophenone at pH < 3.0. The reported stability ofhe imine at these acidities is remarkable, since the imine of

nsubstituted benzophenone underwent acid catalyzed hydrol-sis [9] already at pH < 4.6.

Hence for the proof of formation of an imine intermediate inhe reduction of oximes, another kind of experimental evidence

5 ica A

wbsSbciosrbhiamti

ctwrmt>

tmtHact

2

2

oIwKjma

wfve

2

hA

nrt(

N

IIIIVVVVIX

2

a(ksaamcsttsRioostb

3

3

itd

846 H. Celik et al. / Electrochim

as needed. Such evidence is confirmed in this contributionased on behavior of oximes at pH < 3.0. In these acidicolutions some oximes are reduced in two two-electron waves.uch separation of the four-electron process was first reportedy Lund [10] for 2,4-dihydroxybenzophenone oxime, whoorrectly attributed the second step to the reduction of themine, and later also reported for 2,4-dihydroxyacetophenonexime [11] and for oximes of 2-carboxybenzophenones [12]ubstituted in position 4. In this later contribution, the sepa-ation of two waves was attributed to formation of hydrogenond between the azomethine nitrogen and the oxygen of theydroxy or carboxy group. Nevertheless, the observation thatn 0.1 M HClO4 such separations were observed for 4-fluoro-nd 4,4′-difluorobenzophenone [13] and were reported also forethyl p-tolylketoxime in 10–50% H2SO4 and for unsubsti-

uted benzophenone oxime in 50% H2SO4 [2], ruled out thisnterpretation.

It was indicated [13] that the second reduction wave mayorrespond to the reduction of the imine and was observed thathe dE1/2/dpH slopes of the first and second wave differ. Thatas erroneously attributed to changes in the reversibility of the

eduction of the imine with acidity. Also the statement that theerging of the two waves with increasing pH “is determined by

he difference in the degree of electroreduction reversibility ofC NH+ and [>C NOH] H+ groups” [5] is incorrect.

That none of the above offered satisfactory explanation forhe separation of the two reduction waves of oximes in acidic

edia is due to insufficiently wide acidity range, particularly inhe region where pH-scale is replaced by the acidity function

0. The ionic strength in these studies was not kept constantnd no attention has been paid to the role of the nature of andoncentration of the anion of the strong acid. These factors areaken into consideration in the present contribution.

. Experimental

.1. Instrumentation

The polarographic current–voltage curves were recordedn an IBM EC/225 Voltammetric Analyzer combined withBM 7424 MT X-Y-T Recorder as well as capillary electrodesith characteristics of m = 2.5 mg s−1, t1 = 3.0 s at h = 70 cm. Aalousek cell [14] with a reference electrode separated by liquid

unction was used for DC polarography. The reference electrodeost cases was a saturated calomel electrode (SCE), in strongly

cidic solutions were used Hg2SO4/H2SO4.A Hewlett-Packard Agilent 8453 UV–vis spectrophotometer

as used in recording the UV–vis absorption spectra. A 10 mmused quartz cell was used during the measurements. A Den-er Instrument Model UB-10 pH-meter was used with a glasslectrode to determine the pH.

.2. Chemicals

The compounds studied, benzaldehyde oxime (I) and 2-ydroxyimino-2-phenylacetonitrile (XIX) were supplied byldrich, acetophenone oxime (XI) by Alfa Aesar, and benzophe-

Fnct

cta 51 (2006) 5845–5852

one oxime (XVIII) by Avocado Research Chemical Ltd. Theest of the oximes were prepared at the Department of Elec-rochemistry at J. Heyrovsky Institute of Physical ChemistryCzech Republic).

o. X No. Y No. R

H XI H XVIII PhI CH3 XII CH3 XIX CNII OCH3 XIII Cl XX CF3

V COOCH3 XIV OHBr XV COCH3

I F XVI CF3

II CN XVII NO2

III CF3

X CHONO2

.3. Procedures

The stock solutions (0.01 M) were prepared freshly forll studied compounds in acetonitrile. Simple buffer solutionsphosphate and acetate) were used, and ionic strength wasept constant by addition of a solution of sodium salts. Introngly acidic media solutions of sulfuric, perchloric, nitricnd hydrochloric acids were used either without and or with anddition of a salt. Potentials measured in such solution against aercurous sulfate electrode in concentrated sulfuric acid were

orrected to SCE scale using the potential of Tl+ (−0.450 V ver-us SCE). Acidity of these solutions was characterized usingabulated H0 functions. The above reference electrode was usedo minimize the liquid junction potential and to prevent diffu-ion of sulfuric acid into the investigated electrode compartment.ecording of i–E curves was carried out in the solutions contain-

ng concentrations of acetonitrile of about 1%. A stock solutionf the investigated oxime derivative was added to the solutionf supporting electrolyte after deaeration, with gelatin added inome cases to prevent streaming maxima. The final concentra-ions of studied oximes were 0.1 mM. After brief final purgingy nitrogen the current–voltage curves were recorded.

. Results and discussion

.1. Reason for separation of two two-electron processes

The basic information about the nature of processes, result-ng in separation of two two-electron waves in acidic media inhe electroreduction of oximes (Fig. 1), is obtained from theependence of half-wave potentials of individual waves on pH.

or irreversible reduction processes a shift of potentials to moreegative values indicate that the transfer of the electron is pre-eded by a rapidly established acid–base equilibrium, in whichhe conjugate acid is reduced. As long as the rate of protonation

H. Celik et al. / Electrochimica Acta 51 (2006) 5845–5852 5847

Fig. 1. Separation of two two-electron polarographic waves of 0.1 mM p-methoxybenzaldehyde oxime on acidity, expressed as H0 function or pH. Curve1ei

ieaepctt

wpaodrt

otta

pFaspFdppo

dopii

Fig. 2. Theoretical plot of the dependence of half-wave potentials of a reducibleacid on pH in the pH-range where the rate of protonation is fast enough to convertall the conjugate base into the reducible acid. The pH-value at the intersectionoms

Aiottpenpserbutcoctof

, −0.25 H0; 2, pH 0.5; 3, pH 1.0; 4, pH 2.0. Composition of the supportinglectrolyte: 1, 1 M H2SO4; 2, 0.3 M H2SO4; 3, 0.1 M H2SO4; 4, 0.01 M H2SO4

n 0.1 M NaCl. Polarographic curves starting at −0.2 V, and t1 = 3 s.

s faster than the rate of transport of the reduced species to thelectrode, the plot of the E1/2 = f(pH) plot remains linear, withslope of (RT)/(αnaF). With increasing pH increasingly more

nergy is needed to convert the basic form of the reduced speciesredominating in the bulk of the solution into the electroactiveonjugate acid. The increase of energy involved is reflected byhe increasingly more negative potentials needed for the reduc-ion.

Sometimes it is possible to reach a sufficiently low pH-range,here in absence of an acid–base bulk reaction—the half-waveotential becomes pH-independent. This occurs in a range ofcidity, where the reducible acid form predominates in the bulkf the solution. As the same species is reduced as the one pre-ominating in the solution, no additional energy is needed in theeduction process and hence the potential is governed only byhe change in the free activation energy of the electrode process.

The plots of E1/2 = f(pH) thus manifest two linear segments,ne pH-dependent and one pH-independent (Fig. 2). The pH athe intersection of these two linear segments is practically equalo the pKa value of the dissociation constant of the conjugatecid of the reducible species.

For reductions of oximes the observed shapes of E1/2–pHlots (Figs. 3–5) are in some instances different from those inig. 2. Here in some cases two two-electron waves i1 and i2re observed in more acidic media (Figs. 3 and 4), whereas aingle four-electron wave i1+2 at higher pH-values. The inter-retation of the plot of i1 and i1+2 in Fig. 3 is similar to that inig. 2. The pH at the intersection of the two linear segments,enoted pKox, corresponds to the pKa of the dissociation of therotonated oxime. The values of pKox of oximes obtained fromolarographic data are in an acceptable agreement with valuesf pKa obtained spectrophotometrically (Table 1).

The interpretation of the linear segment of the E–pH plotenoted i2, corresponding to the reduction of the protonated form

f the imine, is more complex. The pH at the intersection of thelots of pH-dependences of potentials of wave i2 and that of wave

1 is denoted pHx (Fig. 3a). An example of such E1/2 = f(pH) plotss the plot obtained for p-hydroxyacetophenone oxime (Fig. 3b).

Hf

H

f the two linear segments corresponds to the pKa-value of the acid. Such deter-ination of pKa is possible provided that the slope dE1

1/2/dpH at pH < pKa ismaller than the slope at pH > pKa.

t pH < pHx the half-wave potential of wave i2 remains pH-ndependent, as in this pH range is reduced the protonated formf the imine, predominating in the bulk of the solution. No addi-ional energy is hence needed to convert the predominating intohe electroactive form. But the value of pHx obtained from theH at the intersection of the linear segments for i1 and i2 is notqual to pKa of the imine. This is due to the fact that the proto-ated form of the imine is reduced at pH > pHx at more positiveotentials than those of the protonated form of the oxime ashown in Fig. 6. Thus any imine formed is immediately at thelectrode further reduced to amine. A single four-electron waveesults. From this follows that pKa of the imine may be equal,ut is usually higher than pHx (Fig. 6). To demonstrate the sit-ation, when the value of pKa of the imine is higher than pHx,wo hypothetical pKa-values of the imine, pKim

1 or pKim2 , were

hosen. The theoretical plots of E1/2–pH for imines with pKim1

r pKim2 are indicated by dotted lines in Fig. 6. For hypothetical

ompounds with pKim1 or pKim

2 the potential of the reduction ofhe imine at pH > pHx remains more positive than the potentialf the first two-electron steps of the reduction of the protonatedorm of the oxime and a single four-electron wave is observed.

ence only the lower limit of pKa of the imine can be obtained

rom plots of E1/2 = f(pH) and pKa (imine) ≥ pHx.In all observed cases the value of pHx is higher than pKox.

ence pKa (imine) is higher than pKox and it can be concluded

5848 H. Celik et al. / Electrochimica Acta 51 (2006) 5845–5852

Fig. 3. Dependence of half-wave potentials of oximes belonging to type A(Table 1) on pH, when both pKox and pHx are within the examined acidity range,when the lowest H0 or pH investigated is lower than both pKox and pHx. (a) The-oretical plot, where intersection of the two segments of wave i1 corresponds topKox and that of waves i2 and i1+2 corresponds to pHx. (b) Experimental plotfor half-wave potentials of i1 (♦), i2 (�) and i1 + i2 (�) for 0.1 mM solution ofp–hydroxyacetophenone oxime in buffers and in H2SO4 solutions.

Fig. 4. Dependence of half-wave potentials of oximes belonging to type B(Table 1) on pH, when pKox is more negative than the most acidic solutionstudied, but pHx is within the studied acidity range. (a) Theoretical plot wherepH at the intersection of the linear segment i2 and that of i1+2 is denoted pHx.(b) Experimental plot for half-wave potentials of i1 (♦), i2 (�) and i1 + i2 (�)for 0.1 mM solution of benzaldehyde oxime in buffers and in H2SO4 solutions.

H. Celik et al. / Electrochimica Acta 51 (2006) 5845–5852 5849

Fig. 5. Dependence of half-wave potentials of oximes belonging to type C(Table 1) on pH, when the values of both pKox and pHx are smaller thanthe most acidic solution studied. Plot of the best fitting line and experi-mental values of half-wave potentials obtained for 0.1 mM solution of p-c

wninlo

tlodihpo(

owtsTltpc

Fig. 6. Demonstration that pKim ≥ pHx. The reduction of imine occurs at poten-tial more positive than E1 . Theoretical E2 = f (pH) plots are indicated forpea

otnfilsimocoMasa

ihfCeasrq

3s

o

iing the slope of dE1/2/dpH differ, particularly about the slope

2

arbmethoxybenzaldehyde oxime in buffers and in H2SO4 solutions.

ith certainty from electrochemical data that pKa of the proto-ated oxime is lower than pKa of the corresponding protonatedmine. This conclusion agrees with data obtained by other tech-iques, indicating that pKa values of imines can be estimated toay between 4.0 and 6.0, whereas those of protonated forms ofximes are lower than 2.0.

This interpretation of the pH-dependence of half-wave poten-ials of the two two-electron waves of oximes in acidic mediaeads to an important conclusion: The reason for the separationf the two reduction waves of oximes in acidic media is theifference in pKa-values of protonated oximes and protonatedmines. In this context it may be pointed out that protonatedydroxylamine derivatives are stronger acids than correspondingrotonated amines, for example, the acid dissociation constantf phenylhydroxylamine (pKa = 3.2) is lower than that of anilinepKa = 4.6) [16].

The same interpretation applies to the most commonlybserved type of E1/2–pH plots in Fig. 4a, which are observed,hen the pKa value of the protonated form of the oxime is lower

han the lowest pH or H0 function investigated. An example ofuch plot is given in Fig. 4b, obtained for benzaldehyde oxime.he accessible value of pHx in these cases is also the lower

imit of pKa of the imine. Consequently in these cases pKa of

he protonated form of the oxime is lower than the pKa of therotonated imine. The observed separation of i1 and i2 in theseases is again caused by the difference in pKa-values.

dsc

1/2 1/2Kim = pHx, pKim = pK1, pKim = pK2. Experimentally accessible pHx can bequal or be lower than the thermodynamic value of pKim, corresponding todissociation of the protonated form of the imine.

The E1/2–pH plots in Fig. 5 are observed, when pKa valuesf protonated forms of both the oximes and the imines are lowerhan the lowest pH or H0 function investigated. Consequently,o separation of two waves is observed. This has been observed,or example, for p-carbmethoxybenzaldehyde oxime, as shownn Fig. 5. For the investigated oximes the E1/2–pH plots fol-ow the type A–C (Table 1). From this table it follows that theeparation of two waves in acidic media occurs predominatelyn acetophenone oximes. If the �-methyl group is replaced byore polar hydrogen, separation of two waves in benzaldehyde

ximes is less favored, as corresponds to a variation in Taft σ*

onstants. For the benzophenone oxime, with even larger valuef σ*, no separation of two waves is observed up to H0 − 1.0.ore electron withdrawing substituents in the side chain, such

s CN or CF3, have also a pronounced substituent effect and theeparation of i1 and i2 is not observed even in more stronglycidic media (Type C, Table 1).

For para-substituted benzaldehyde oximes the separations observed for less electron withdrawing substituents thanalides (σp < 0.26). No separation of two waves was observedor strongly electron withdrawing substituents, like p-CN, p-F3, or p-COOCH3 (σp > 0.45). pKa-values of oximes, obtainedither spectrophotometrically or for polarographic data, shown increasing trend with electron-donating character of the sub-tituent, as manifested by a more negative value of σp−X. Accu-acy and the number of available data are insufficient for auantitative correlation.

.2. Role of composition of the supporting electrolyte oneparation of waves

At pH or H0 lower than pHx the slopes of the dependencef half-wave potentials of the reduction of oxime E1

1/2 and

mine E21/2 differ. The reports in literature [2,10–13] concern-

E1/2/dpH. This led to erroneous interpretation [5,12] that theeparation is due to changes in mechanism of the electrode pro-ess. This conclusion was based on values of potentials obtained

5850 H. Celik et al. / Electrochimica Acta 51 (2006) 5845–5852

Table 1Determination of dissociation constant of monoprotonated form of oximes: Both current–voltage were recorded in a solution containing 1% acetonitrile and 0.1 mMoxime in solution of sulfuric acid

Oxime Typea Substituent pKoxb pKa

c Lowest H0d pK1

e

p-X ArCH NOH A H −0.92 −1.05 −1.80 0.25A CH3 −0.80 −0.85 −1.80 1.00A OCH3 −1.02 −0.93 −1.80 1.50C COOCH3 <−1.80 <−2.00 −1.80B Br <−1.50 −1.79 −1.50 1.00B F <−0.85 −1.23 −0.85 2.20C CN <−1.80 <−1.80 −1.80 −C CF3 <−1.80 <−1.80 −1.80 −B CHO <−1.80 <−1.80 −1.80 −0.25– NO2

e −3.10 – –

p-X ArC(CH3) NOH A H −0.30 −0.10 −0.85 1.00A CH3 0.55 0.49 −0.85 0.50A Cl −0.50 −0.50 −0.85 0.50A OH 0.60 0.56 −0.85 1.00A COCH3 −0.75 −0.74 −0.85 0.50C CF3 <−0.85 – −0.85 –– NO2

f −1.49 −1.80 –

PhC(X) NOH C Ph <−0.85 −0.90 −0.85 –C CN <−1.80 <−1.80 −1.80 –C CF3 <−1.80 <−1.80 −1.80 –

a Type A E1/2 = f(pH, H0) plots as in Fig. 2a; Type B E1/2 = f(pH, H0) plots as in Fig. 2b; Type C E1/2 = f(pH, H0) plots as in Fig. 2c.b Data obtained shifts of polarographic E1/2. Accuracy of determination of pKox from the intersection of two linear segments is at best ±0.1 pK-unit.c From spectrophotometric measurements.d

e.

issrhatw

ttpbtHb

dsfpd−1fwto

taaup2slfhTEpt

wwvifi(wrs

Lowest acidity (H0) at which i–E curves were recorded.e The pH-value at which wave i1 and i2 merge into a single four-electron wavf Reduction of the nitro group precedes the reduction of the oximino group.

n supporting electrolytes, the composition of which was notufficiently controlled. Therefore, the role of the nature of thetrong acid, of nature and concentration of anions, the possibleole of cations and of the ionic strength of investigated solutionsas been investigated in some detail. Using the same strong acidnd its salt at a constant ionic strength of the supporting elec-rolyte in all instances, the values of dE2

1/2/dpH or dE21/2/d(Ho)

ere equal to zero.For compounds, where the E1/2–pH plots follows the pat-

ern presented in Fig. 3a (Type A in Table 1), it was possibleo compare the values of pKox obtained from such plots withKa determined spectrophotometry (Table 1). The agreementetween the two sets of values can be considered as satisfactory,aking into account that the accuracy of the finding of the pH or

0 value of the intersection of the two linear segments is hardlyetter than ±0.1 pH-units.

This observation has an important consequence. Shapes of theependences of limiting current of oximes on pH and compari-on with reduction of nitrones [15] indicate that a diprotonatedorm of oxime is reduced at pH higher than about 4. As theKa of the diprotonated form can be estimated from the depen-ence of the acid catalyzed hydrolysis [15] to be lower than2.0, the rate constant of protonation would be of the order of

016 L mol−1 s−1. This value is much higher than that obtained

or fastest homogeneous reactions (3 × 1011 L mol−1 s−1) inhich each collision is effective. This indicates that the pro-

onation takes place as a heterogeneous reaction, involving anxime adsorbed at the electrode surface. This is supported by

cvrm

he observed decrease of limiting current with increasingly neg-tive potential [7]. The agreement of pKa of oximes obtainedt pH < 2.5 by comparing UV spectra at varying pH values andsing E1/2 = f(pH) plots indicates that at pH < 3 it is the mono-rotonated form which is reduced. Thus between pH about 4 anda change in reduction mechanism takes place. In this pH-range

ome of the oxime is present in monoprotonated form, which isess likely to be adsorbed at a positively charged electrode sur-ace. Hence at pH < 3 the protonation occurs predominantly as aomogeneous process and the monoprotonated form is reduced.his interpretation is supported by some changes of the slope of1/2 = f(pH) plots pH between 5 and 3. As in this pH-range theotentials depend somewhat on buffer kind and concentration,his aspect was not further followed.

Most of the measurements of potentials discussed belowere carried out at pH (or H0) values between pHx and pKox,here E1

1/2 is shifted with increasing acidity to more positivealues and where E2

1/2 remains independent of changes in acid-ty. To investigate the role of the supporting electrolyte used,rst the half-wave potentials of p-methoxybenzaldehyde oximeIII) were compared in 0.1 M solutions of various strong acids,here the acidity is comparable. The half-wave potentials of the

eduction of the protonated oxime, E11/2, were found to be the

ame (−0.560 V versus SCE) in 0.1 M solutions of sulfuric, per-

hloric, and nitric acid, but somewhat more negative (−0.570 Versus SCE) in 0.1 M hydrochloric acid. As the values of theeduction potentials of the protonated form of the imine inter-ediate were constant (−0.700 V versus SCE) for all four acids,

H. Celik et al. / Electrochimica Acta 51 (2006) 5845–5852 5851

Table 2Dependence of half-wave potentials of 0.1 mM p-tolualdehyde oxime and 1% acetonitrile on concentration of sulfuric and hydrochloric acid at varying ionic strength

[H2SO4] H0 −E11/2 (V) −E2

1/2 (V) H0 [HCl] −E11/2 (V) −E2

1/2 (V)

0.1 – 0.56 0.66 – 0.1 0.57 0.6550.3 – 0.525 0.66 – 0.3 0.525 0.6551.0 −0.25 0.48 0.66 −0.21 1.0 0.48 0.661.5 −0.56 0.465 0.66 −0.45 1.5 0.465 0.672

P d H2Sp

iolo

caaa

osAsE

iwaE

fWirstmntTtc

TDa

[

0011223

Ivc

nocAat

H

p

i

twsopositive values seem to be predominately caused by the effectof ionic strength on the activity of hydrogen ions.

In mixtures of 0.1 M H2SO4 and 0.1 M HCl that shouldhave comparable acidity, the values of E2

1/2 remain unchanged

Table 4Effect of addition of sodium salts on the half-wave potential solution of 0.1 mMof p-tolualdehyde oxime containing 1% acetonitrile

−E11/2 (V) −E2

1/2 (V)

[NaCl] in 0.1 M HClO4

0.00 0.55 0.640.10 0.55 0.64

.0 −0.85 0.46 0.66

otential vs. SCE measured against mercurous sulfate electrode in concentrateK1.

ncluding hydrochloric, it can be ruled out that the differencebserved in 0.1 M HCl is due to a difference in acidity or theiquid junction potential. The role of halide ions on the reductionf cations of protonated oximes is discussed below.

In 3.0 M solution of sulfuric and perchloric acid, which haveomparable acidity, where E1

1/2 is not any more dependent oncidity, its value (−0.460 V versus SCE) was the same in bothcids. To achieve the same acidity in hydrochloric acid solution3.8 M concentration would have been needed.

The role of concentration of sulfuric and hydrochloric acidsn the reduction of p-tolualdehyde oxime (II) was followed inolutions where the ionic strength was not controlled (Table 2).ll reported data were obtained under conditions where E1

1/2 ishifted with increasing acidity to more positive potentials and21/2 remains independent of acidity. The slope of dE1

1/2/dpHn both acids remains the same and equal to −0.062 V/pH, buthereas E2

1/2 in solutions of sulfuric acid remains unchanged,small shift towards more positive potentials is observed for21/2 (Table 2), corresponding to the reduction of the protonated

orm of the imine intermediate in hydrochloric acid solution.hen the concentration of perchloric acid was varied, but the

onic strength was kept constant by addition of sodium perchlo-ate (Table 3), E1

1/2 of p-methoxybenzaldehyde oxime (III) ishifted to more positive values up to H0 = −1.02 (as correspondso pKox given in Table 1) and becomes independent of acidity inore acidic solutions. On the other hand, the value of E2

1/2 doesot show any trend with increasing acidity and remains prac-

2

ically independent of acidity. Comparison of values for E1/2ables 2 and 3 allows conclusion that the potential of the reduc-

ion of the imine (E21/2) is independent of acidity (Table 3) but

an be affected by changes in the ionic strength (Table 2).

able 3ependence of half-wave potential of 0.1 mM p-methoxybenzaldehyde oxime

nd 1% acetonitrile on concentration of perchloric acid

HClO4] [NaClO4] −E11/2 (V) −E2

1/2 (V) µ

.1 2.4 0.57 0.755 2.5

.3 2.2 0.53 0.75

.0 1.5 0.485 0.745

.5 1.0 0.47 0.745

.0 0.5 0.465 0.75

.5 – 0.46 0.75

.0 – 0.46 0.755 3.0

onic strength (µ) was kept constant by addition of sodium perchlorate. Potentials. SCE measured against mercurous sulfate electrode in concentrated H2SO4,orrected against Tl+.

[

[

PH

−0.67 2.0 0.46 0.67

O4, corrected against Tl+ and reported vs. SCE. All H0 values are lower than

To demonstrate that cations of the supporting electrolyte playo significant role on the values of either E1

1/2 or E21/2 solution

f 1.0 M (final concentration) lithium, sodium, and potassiumhlorides were added to a 0.1 mM solution of oximes II (Type, Table 1), VIII and XVI (Type C, Table 1) in 1.0 M sulfuric

cid. Values of both E11/2 and E2

1/2 remained in the presence ofhese salts unchanged and varied by less than 0.005 V.

The addition of sodium chloride to 0.1 M HClO4 or 0.1 MCl has no effect on E2

1/2, similarly as the addition of sodium

erchlorate to 0.1 M HClO4 (Table 4). The shift of E11/2 has sim-

lar slopes dE11/2/d[anion] for addition of sodium chloride both

o 0.1 M HClO4 (0.023 with correlation coefficient r2 = 0.99) asell as to 0.1 M HCl (0.023 with r2 = 0.99) and for addition of

odium perchlorate to 0.1 M HClO4 (0.034 with r2 = 0.99). Thebserved shifts of half-wave potential in such solutions to more

0.30 0.545 0.640.50 0.54 0.640.70 0.535 0.641.00 0.53 0.64

NaClO4] in 0.1 M HClO4

0.00 0.545 0.640.10 0.54 0.640.30 0.535 0.640.50 0.53 0.640.70 0.52 0.641.00 0.51 0.635

NaCl] in 0.1 M HCl0.00 0.555 0.640.10 0.55 0.640.30 0.545 0.640.50 0.54 0.640.70 0.535 0.641.00 0.53 0.64

otential vs. SCE measured against mercurous sulfate electrode in concentrated

2SO4, corrected against Tl+.

5852 H. Celik et al. / Electrochimica A

Table 5Dependence of half-wave potential of 0.1 mM p-tolualdehyde oxime on concen-tration of chloride ions in mixtures of 0.1 M H2SO4 and 0.1 M HCl containing1% acetonitrile

[H2SO4] [HCl] −E11/2 (V) −E2

1/2 (V)

0.100 0 0.57 0.670.084 0.016 0.57 0.670.050 0.050 0.575 0.670.020 0.080 0.59 0.670

P

(pip

i

cuccbsbso

TEh

[

1000000

[

1000000

[

1110000

P

s

a

4

[tdoesistf(ooptset

0.100 0.61 0.67

otentials are reported vs. the SCE.

Table 5), but the potential E11/2 is shifted to more negative

otential with increasing concentration of chloride ions. Thisndicates an interaction, such as ion-pair formation, between therotonated form of the oxime and chloride ions.

At constant ionic strength, an addition of halide ion resultedn a shift of E1

1/2 to more positive potentials with increasing

oncentration of the anion, whereas E21/2 remains practically

nchanged (Table 6). This indicates that the observed shifts areaused by an ion-pair effect rather than due to a variation inomposition of the double layer, which would result in shifts ofoth E1

1/2 and E21/2. The shifts reach in all cases practically the

ame potential, namely −0.380 ± 0.007 V. For the more reactive

romide ions this limiting value is reached already in a 0.3 Molution of sodium bromide. The plot of the linear dependencef the E1

1/2 = f ([Cl−]) has a similar slope and intercept in both

able 6ffect of anions on half-wave potential of 0.1 mM p-trifluoromethyl benzalde-yde oxime at constant ionic strength

Na2SO4] [NaCl] −E11/2 (V) −E2

1/2 (V)

.0 M H2SO4

.5 0.0 0.41 0.46

.4 0.3 0.395 0.46

.3 0.6 0.385 0.455

.2 0.9 0.38 0.455

.1 1.2 0.375 0.455

.0 1.5 0.37 0.455

Na2SO4] [NaBr] −E11/2 (V) −E2

1/2 (V)

.0 M H2SO4

.5 0.0 0.41 0.46

.4 0.3 0.38 0.46

.3 0.6 0.38 0.46

.2 0.9 0.38 0.46

.1 1.2 0.38 0.46

.0 1.5 0.38 0.465

NaNO3] [NaCl] −E11/2 (V) −E2

1/2 (V)

.0 M HNO3

.5 0.0 0.41 0.46

.2 0.3 0.405 0.46

.9 0.6 0.40 0.46

.6 0.9 0.395 0.46

.3 1.2 0.39 0.455

.0 1.5 0.385 0.45

otentials are reported vs. the SCE.

iroi

A

og

R

[[[[[

[[

cta 51 (2006) 5845–5852

ulfuric (E11/2 = 0.020 + 0.40[Cl−] with r2 = 0.97) and nitric

cid (E11/2 = 0.017 + 0.41[Cl−] with r2 = 0.99).

. Conclusions

Discrepancies, which led to an unsubstantiated speculation5] about the nature of electrode processes taking place in thewo-step reduction of some oximes in acidic media, have beenemonstrated to be caused by insufficient attention to the rolef ionic strength and the nature of the anion of the supportinglectrolyte. It has been shown that the half-wave potential of theecond wave (E2

1/2) corresponding to a reduction of the iminentermediate is independent of acidity, nature of the cation ofupporting electrolyte, and by variation in ionic strength. Onhe other hand, the potential of the reduction of the protonatedorm of the oxime (E2

1/2) depends at pH or H0 higher than pKoxcorresponding to a dissociation of the protonated form of thexime) on changes in acidity. The value of E1

1/2 is independentf the nature of the cation of the supporting electrolyte, andractically independent of the ionic strength, but depends onhe concentration of halide ions present. The possibility thatuch observed variations are due to a double layer effect wasxcluded. They are attributed to ion-pair formation facilitatinghe reduction. Separation of the two waves, observed practicallyn the presence of electron donating substituents on the aromaticing, is attributed to a smaller pKa-value of the protonated formf the oxime than the pKa-value of the protonated form of themine.

cknowledgement

The financial support of the Grant MSMT (Czech Ministryf Education, Youth and Sports) KONTAKT 1P 05 ME 785 isratefully acknowledged.

eferences

[1] P. Souchay, S. Ser, J. Chim. Phys. 49 (1952) C172.[2] J. Gardner, W.P. Georgans, J. Chem. Soc. (1956) 480.[3] U. Eisner, E. Kirowa-Eisner, in: A. Bard, H. Lund (Eds.), Encyclopedia

of Electrochemistry of the Elements, Vol. XIII, Mercel Dekker, NewYork, 1979, p. 219.

[4] T.V. Troepol’skaya, G.K. Budnikov, The Electrochemistry of the Azome-thines (in Russian), Nauka, Moscow, 1989, pp. 128.

[5] V.N. Leibzon, Russian J. Electrochem. 32 (1996) 11.[6] H. Lund, Acta Chem. Scand. 13 (1959) 249.[7] H. Celik, G. Ekmekci, J. Ludvık, J. Pıcha, P. Zuman, J. Phys. Chem.

B, in press.[8] M.S. Baymak, H. Celik, H. Lund, P. Zuman, J. Electroanal. Chem. 589

(2006) 7.[9] M.S. Baymak, H. Celik, H. Lund, P. Zuman, J. Electroanal. Chem. 558

(2005) 284.10] H. Lund, Acta Chem. Scand. 18 (1964) 563.11] K.V.K. Rao, S.B. Rao, J. Electroanal. Chem. 53 (1974) 159.12] V. Marek, Chem. Zvest. 28 (1974) 355.

13] I. Gurrappa, S.J. Reddy, Indian J. Environ. Protect. 9 (1989) 614.14] J. Heyrovsky, M. Kalousek, Collect. Czech. Chem. Commun. 11 (1939)

464.15] P. Zuman, O. Exner, Collect. Czech. Chem. Commun. 30 (1965) 1832.16] Referee’s comment.