General scheme of adsorption, electroreduction and catalytic hydrogenation of nitro-compounds on platinum—II. Mechanism of electroreduction and catalytic hydrogenation

General scheme of adsorption, electroreduction and catalytic hydrogenation of nitro-compounds on platinum—II. Mechanism of electroreduction and catalytic hydrogenation

CQ15-4686/E I jQW563- 15 502.00/o Pusamen PressLtd. GENERAL SCHEME OF ADSORPTION, ELECTROREDUCTION AND CATALYTIC HYDROGENATION OF NITRO-COMPOUNDS ON ...

1MB Sizes 0 Downloads 10 Views

CQ15-4686/E I jQW563- 15 502.00/o Pusamen PressLtd.

GENERAL SCHEME OF ADSORPTION, ELECTROREDUCTION AND CATALYTIC HYDROGENATION OF NITRO-COMPOUNDS ON PLATINUM--II. MECHANISM OF ELECTROREDUCTION AND CATALYTIC HYDROGENATION Yu. B. VASSILIEV,

V. S. BAGOTZKY, 0. A. KHAZOVA, T. N. YASTREBOVA and T. A. SERGEEVA

Institute of Electrochemistry,

U.S.S.R.

Academy

of Sciences,

Moscow,

U.S.S.R.

(Received 9 July 1980) Abstract-The effect of basic kinetic parameters (potential, the nature and concentration of nitrocompounds solution pH, coverage of the surface with organic particles and hydrogen) on electroreduction and catalytic hydrogenation of nitro-compounds on platinum has been investigated. Direct comparison of kinetic and adsorption data obtained under the same conditions enabled the mechanism of these processes to be clarified. Kinetics of electroreduction of preliminarily chemisorbed semireduced nitro-compounds particles and kinetics of their hydrogenation by molecular and atomic hydrogen diffuused through a palladium membrane have been investigated. Kinetics of interaction of hydrogen preliminaryly adsorbed on platinum with nitrocompounds has also been investigated. One and the same stage of interaction between a semireduced chemisorbed particle, formed in the preceding rapid stages of nitro-compound chemisorption, and adsorbed atomic hydrogen, formed in the preceding rapid electrochemical stage

H++e

+Heds

or in the preceding rapid stage of molecular hydrogen adsorption

HZ = =&.ds has been shown to remesent the slow staae of electroreduction nitrocompounds odplatinum. INTRODUCTION

EXPERIMENTAL investigations

were

mainly

performed

of

smooth platinum electrodes according to the technique similar to that previously described[1,2]. Polarization measurements were conducted as follows. A cleaning pulse was applied to the electrode before recording each point of the polarization curve and after the electrode having been cleaned the potential was changed jumpwise to the investigated value and the i-t curve was recorded. The steady-state current value was measured after some time r when the steady-state coverage of the surface with chemisorbed particles is reached. All the current values are normalized on the unit true electrode surface[3]. Electroreduction of nitro-compounds chemisorbed particles was investigated after their adsorption at cd* for the time 7ti = 34 min necessary for the coverage to reach a steady-state value and after jumpwise change of the potential up to p (electroreduction potential), at which the surface coverage with these particles was determined after different electroreductlon times, zd_. Hydrogenation of chemisorbed particles with mols cular hydrogen was investigated by means of a diskring electrode (platinum or palladium disk, platinum ring, o = 4000 rev/min). The platinum ring was used as a working electrode on which at E, = 0.2 V a nitrocompound was adsorbed for 5-6 min, then the ring circuit was opened, the dc cathodic current was fed to the disk and hydrogen formed on the disk was

Without reliable data on the regularities of adsorption, on the nature of chemisorbed particles and on their participation in hydrogenation and electroreduction no unambiguous conclusions on the mechanism of electroreduction and catalytic hydrogenation can be made for metals, like platinum, capable ofconsiderable adsorption of both, hydrogen and the reaction organic component. In the first Part of the Paper we investigated the kinetics andmechanism of adsorption of nitro-compounds (nitromethane, nitroethane and nitrohenzene) on platinum[ 11. In this paper the effect of basic kinetic parameters (potential, the nature and concentration of nitro-compounds, solution pH, coverage of surface with organic particles and hydrogen) on electroreduction and catalytic hydrogenation of nitro-compounds on platinum is systematically investigated_ Direct comparison of kinetic and adsorp tion data obtained under the same conditions enabled one to have a deeper understanding of the process mechanism and suggest a general scheme of adsorption, electroreduction and catalytic hydrogenation of nitro-compounds on platinum.

The

and catalytic liquid-phase hydrogenation

transferred

on

563

by

the liquid

flow

to the

ring

where

it

564

Yu. B.

VASSILIEV,

V.

S.

BAWTZKY,

0. A.

KHAZOVA,

hydrogenated the nitro-compound chemisorbed particles. After different hydrogenation times a pulsed potentiostat was switched on and a measuring pulse was applied to the ring, thereby enabling the remaining coverage with chemisorbed particles to be determined. The magnitude of the cathodic current through the disk was specified by the required hydrogen pressure[4]. The potential shift curves technique was used to investigate the interaction of nitro-compounds with hydrogen preliminaryly adsorbed on platinum. Two versions of this technique were employed. In the first one the platinum electrode potential was stabilized potentiostatically at E, = 0 V in 0.5 M HzS04 solution so that it practically did not change with time when the potentiostat was switched off. Then a nitrocompound was introduced into the solution under intensive stirring and the potential change with time was recorded. In some experiments the potentiostat was switched on at certain moments of the potential drop and a measuring pulse was applied to determine the electrode surface coverage with hydrogen and chemisorbed particles. In the second version a negative potential ( -0.02 V) was applied, after the preparatory pulse, to the electrode immersed into the nitro-compound solution, so that all the platinum surface proved to be covered with hydrogen. The adsorption of nitro-compounds did not occur due to their hydrogenation. Then the circuit was opened and the E-7 curve was recorded. The results obtained by both the techniques were the same. Interaction of nitro-compounds chemisorbed particles with atomic hydrogen was studied by a palladium membrane cell similar in principle, to that described in[3-51. The palladium membrane was platinized on the contact side of the cell containing the nitro-compound solution (that is electroreduction proceeded on platinum). A nitro-compound was chemisorbed at edr = 0.2 V on platinized palladium from 0.5 M H,SOn solution containing the nitrocompound until the steady-state coverage was reached. Then the potentiostat on the cel1 contact side was switched off and the cathodic polarization was switched on in the diffusion part of the cell containing pure 0.5M HzS04 background solution. Hydrogen formed on the diffusion side diffused through the palladium membrant onto the contact side (that was easily observed by the change in the contact side potential), where it hydrogenated nitrocompound chemisorbcd particles. After some hydrogenation period the pulsed potentiostat was energized on the contact side and a measuring pulse was applied to determine the remaining surface coverage with chemisorbed particles. Electroreduction products were analyzed qualitatively and quantitatively after the electrolysis at different controlled potentials (potentiostatic electrolysis). Gas-liquid chromatography, titration, polarograbby, ir spectroscopy and elemental analysis were used for this Duruose. In most of the cases the yield of methylhydroxylamine was determined iodomdtrically of a slightly modified de Ga&ne by means technique[6, 73 in which a part of measured alkaline copper solution (the Felling solution) is reduced in boiling with methylhydroxylamine solution and the copper excess is iodometrically titrated by sodium

T.

N.

YASTREBOVA

AND

T. A.

SERGEEVA

thiosulfate. The methylamine yield was determined after its separation in the form of hydrochloride salt. After the electrolysis the solution was treated with 40% KOH solution introduced from the dropping funnel. Liberated pure methylamine was driven off under heating into the receiver containing 70 ml of 15 % HCl. To determine the amount of CHINHZHCl the solution was titrated with KOH by an indicatorthymolphthalein. The reaction mixture was analysed quantitatively under the nitrobenzene electroreduction by chromatographic and polarographic techniques. A Vyzukhzom chromatograph with the flameionization detector, Model A-I, was employed for investigations. The analysis was performed in the column 5 m long and N-AW” diameter. “Chromaton 4mm in (0.2XXl.315 mm) treated with “apiezon” (15 wt. %) was used as the immovable phase. The column temperature during the analysis was 210°C; the velocity of carrier-gas (argon) - 70 f 5 ml/min. The quantitative composition of the reaction mixture was determined by the direct calibration technique, and the chromatographic curves were calculated the peak hights. The data obtained show that under the conditions employed chromatographic analysis enables the content of aniline and the total content of nitrobenzene and phenylhydroxylamine to be determined. The concentration of aniline in the reaction mixture was determined by the chromatographic technique. The polarographic analysis was performed on a HfIT-I polarograph. The buffer solution of HCl-KC1 (pH = 1.1) with 10 “/, of ethyl alcohol was used as the background solution. The measurements have shown that under these conditions nitrobenzene is reduced on the dropping mercury electrode at two stages-its poiarogram exhibits two waves, whose half-wave potentials are equal to 0.15 and 0.61 V, respectively. According to literary data[8], in acid medium nitrobenzene is first reduced to phenylhydroxylamine and only in the potential range of the second wave the latter is reduced to aniline. The polarograms of the mixtures analysed also exhibit two waves with the same half-wave potentials as in the case of nitrobeuzene, however the ratio of the second wave height to the first wave height is much more than in reduction of pure nitrobenzene. Unchanged E, ,* value of the second wave under its increase makes it possible to suggest that phenylhydroxylamine is present in the solutions analyzed. Polarographic reduction of pure phenylhydroxylamine has shown the potential of its half-wave to be close to E, 12 of the second wave on the nitrobenzene polarogram, and its addition to the solution of the latter increases the height of this wave proportionally to the amount added. It was also found experimentally that some other possible intermidiate products of nitrobenzene reduction, such as azo-, azoxy- and hydrazobenzene in concentrations corresponding to their saturated solution in the polarographic background, give, under the conditions investigated, very insignificant reduction steps. According to the literary data, nitrobenzene in extremely various media is reduced at potentials much more positive than that of hydrazobenzene. Therefore, phenylhydroxylamine, formed under electroreduction of nitrobenzene on platinum, is likely to be responsible for the increase in

General scheme of adsorption, electroreduction and catalytic hydrogenation of nitrocompounds

the second wave. The former may be detected qualitatively in the mixtures in question by other techniques. The polarograms of the mixtures under investigation were calculated from the calibration curves of nitrobenzene and phenylhydroxylamine reduction recorded in the concentration range 10-4-5.10-4, where the waves height depends practically linearly on the concentration of these compounds in the solution.

RESULTS Kinetics

process. The maximal adsorption rate and electroreduction rate become comparable ar E, c 0.1 V and the surface coverage with chemisorbed particles drops (Figs I, 2). At these potentials the adsorption stage proceeds rather slow. If nitromethane and nitrobenzene are electroreduced only through the intermidiate formation of a semireduced chemisorbed particle, then under steadystate conditions the adsorption rate must be equal to electroreduction rate at the steady-state coverage value. Using the adsorption kinetics data and dependence of the steady-state coverage on the potential, we calculated polarization curves of ,nitromethane and nitrobenzene electroreduction, when the reaction proceeds through the intermidiate formation of a semireduced chemisorbed particle. Figures 1 and 2 reveal complete coincidence between these curves and the experimental polarization curves. This shows that nitromethane electroreduction cannot proceed without the chemisorption stage. A complicated dependence of nitromethane and nitrobenzene electroreduction rates on their volume concentration also indicates that chemisorbed particles participate in the electroreduction process. As can be seen from Fig. 3, the electroreduction rate first increases in proportion to the volume concentration raised to a fractional power

AND DISCUSSION

of nitro-compounds

electroreduction

Figure 1 and 2 show typical polarization curves of nitromethane and nitrobenzene electroreduction in solutions of a different concentration. Under the cathodic shift of the potential the electroreduction rate first increases with a Tafel slope of 60 mV for nitromethane and 75-85 mV for nitrobenzene, reaches a maximum and remains practically unchanged for nitrobenzene, but for nitromethane it drops rapidly with a Tafel slope of 80 mV under the further cathodic shift of the potential. Direct comparison of the data on the nitrobenzene and nitromethane adsorption kinetics in this potential range with those of their electroreduction kinetics shows that the maximum (as & -P 0) adsorption rate at E, > 0.1 V exceeds greatly the electroreduction rate, that is the adsorption is not the limiting stage of the

i=k.&’

I 01

I 02

E/V

I

C

I 02

I 01

I 0

E,/v

E,/v

maximal adsorption rate (4”) and surfacecoverage with se&reduced chemisorbed particles (I-II) on the smooth platinum electrode potential in MM HISO, solutions with different nitromethane concentrations (h4) l-5.10-‘; 2-10-l; 3-5.10-2;4-10-2: 5-3.10m3; 6Fig. 1. Dependence of electroreduction rate (l-lo),

10m3; 7-3.10-‘;

8-10m4; 9-5.10-‘;

(1)

where B varies from 0.5 to I with an increase in the

-2 I

565

lo-3.10w5; 11-1O-5 curve 4’ calculated for steady-state condition using kinetic adsorption data.

Yu.

B. VASSILIEV, V.

S. BAGOTZKY, 0. A. KHAZOVA, T. N. YASTREBOVA AND T. A. SERGEEVA

cathodic potential, then lt remains constant or even slightly drops with increasing the volume concentration, that is characteristic of processes in whose slow stage a chemisorbed particle participates. Figure 4a shows the dependence of the nitromethane electroreduction rate on the surface concentration of chemisorbed (semireduced) particles. At a constant potential the electroreduction rate increases exponentially with coverage i = kexp (BJ’&)

04

02

(2)

where Bf’ = 69. At E, -z 0.1 V when the surface is significantly covered with hydrogen, a large coverage effect is observed, that is the electroreduction rate first increases with coverage, passes through a maximum and then drops. This indicates that besides a semireduced organic particle a second chemisorbed particle participates in the slow stage, and the former particle competes with the latter one for the surface. In this potential range only adsorbed hydrogen may be such an organic particle. The polarization curves recorded at a constant volume concentration do not reflect the true influence of the potential on the electroreduction rate, because in this potential range the surface coverage with chemisorbed organic particles changes. Therefore, to find the true influence of the potential on the reaction rate the polarization curves were recorded at WR= const (Fig. 4b). It follows from these curves that

0

E/V

Fig. 2. [email protected] of cleetroreduetionrate (l-3), maximal adsorption rate at &a = O(l’-2’). adsorption rate at ORRftion = 0 (1”-2”), hydrogenationrateof particlesprechemisorbed (3) at e = 0.2V (2-3) and surfacecoveragewith chemisorbed particles (4,s) on the platinum electrode potential in where a = 1.5, and the dependence is linear up to E 0.5M H&O4 solutions with differentnitro-benzeneconant= 0.05-0.075 V but at larger coverage values it deflects rations (M): 1,4, 5-1.55lo-‘; 2-1.55lo-*; 3-1.55lo-’ M; C from the linear one. It can be Seen from Figs 5 and 6 that in acid and neutral solutions the form of the polarization curves of nitromethane and nitrobenzene electroreduction does

(b)

-4

-2 log k/M)

0

-4

-2 Log k/Ml

Fig. 3. Dependence of clectroraiuction rate of nitromethane (a) and nitrobenzene (b) on their volume concentrationat different electrode potential (V): l-0.025; Z-0.05; 3-0.075; 4-0.1; 5-0.125; 6-0.15; 7-0.175: S-0.2; 9-0.25; 10-0.275; 11-0.3.

General scheme of adsorption,

electroreduction

and catalytic hydrogenation

567

of nitro-compounds Ib)

(0)

P

a

04

02

06

0

02

8;

E/V

Fig. 4. Dependence of nitromethane electroreductian rate on surface coverage with semireduced chemisorbed particles (a) in OSM H,SO, at different platinum electrode potentials (V): l-0.025; 2-0.05; 3-0.075;4-0.1; S-0.125;6-O-15;7-0.2 and on the potential (b) at different surface coverage with chemisorbed particles, 0;: 80.55; 9-0.5; 10-0.45; 1l-0.4; i2-0.35; 13-0.3; N-0.25; 15-0.2~16-0.15. not change; as the solution pH varies, they shift, in the first approximation, in the same manner as the equilihrium hydrogen electrode potential, and also as the region of hydrogen adsorption on platinum. As can be Seen from Fig. Sb, the nitromethane electroreduction rate, at a constant potential relative to the hydrogen electrode in the same solution, is practically unchanged in the range of pH = O-4, and then drops sharply at pH 6. This drop is synchronous with the reduction in the platinum electrode surface coverage with semireduced chemisorbed particles due to the transformation of nitromethane into the aciform. For nitrobenzene, which is not transformed into the a&form, the electroreduction rate at a constant E, is unchanged up to

pH = 8. At a constant E, and t& = const the nitromet-

hane and nitrobenzene

electroreduction

rate in acid

and neutral solutions does not depend on the solution pH, that is

X0

Then the general kinetic equation for nitromethane electroreduction on platinum may be written in the form

i=..exp(,,.,).exp(-BE,)

I I

4 -a

(b)

3.0

4

-I

02

Ed?

0

Fig. 5. Polarization curves of ekctroreduction of lo-‘M nitromethsne solution in phosphate buffer solutions with duffermt pH values (a)r l-0.4 (0.5M H,SO.+); 2-0.75: 3-2.7: 4-3.9; 5-5.5 and dependenas of electroreduction rate (6) and surfaae coverage with chcmisorbed particles (6’) on the solution pH at E, 3 0.2 (b).

FA

26.4

r/

(4)

(5)

YLJ. B. VASSILIEV, V. S.

568

BAOGTZKY,

0. A.

KHA~OVA,

I

I -04

04

T. N.

I

YASTREBOVA AND

/ -0 6 E/V

I

T. A.

SERGEEVA

I -0 6

Fig.

6. Polarization curves of electroreductioa of 1.55.10-‘M nitrobenzene solution in solutions with different pH: I-0.4; 2-1.8; 3-2.9; 4-3.9; 5-5.3; 6-6.2; 7-7.08; 8-8.38; g-10.3;

phosphate buffer 10-l 1.36; 1 l-12.6; 12-13.56 and the effect of solution of pH on the electroreduction rate (13-14) and on the surface coverage with semireduced chemisorbcd particles (13’) at E, = 0.1 V (13, 13’) and E, = 0.3 V (14) and with type I particles (12’).

The character of the solution pH and electrode potential influence on the rate of nitromethane and nitrobenzene electroreduction on platinum in acid and neutral solutions shows that besides an adsorbed semireduced particle a second particle, formed in the preceding rapid electrochemical stage, participates in the slow stage. Adsorbed hydrogen is such a particle in a given potential range; it is formed in the electrochemical stage H+ +e sHldr

(6)

In an alkaline medium nitromethane is not electroreduced due to intramolecular rearrangement of nitromethane and its transformation into the unreduced aciform. In alkaline solutions nitrobenzene is electroreduced, however at pH ICIthe character of the pH influence on nitrobenzene electroreduction changes. In solutions with pH ranging from 10.3 to 13.5 the polarization curves plotted in coordinates relative to the normal hydrogen electrode coincide (Fig. 6), that is the solution pH does not affect the nitrobenzene electroreduction rate and the Tafel slope is 60 mV as compared to 75-85 mV in acid and neutral solutions. All this indicates that in the case of nitrobenzene the mechanism of electrode processes changes when passing to alkaline medium. The measurements performed in 1M KOH show that in nitrobenzene alkaline solutions at potentials ranging from 0.0 to 0.7V the platinum surface is not covered with semireduced type II particles, decreasing the total amount ofadsorbed hydrogen, but is covered only with anion type particles, that is type I particles[l]. The study of nitrobenzene adsorption kinetics in alkaline solutions shows that at E, L- 0.35 V the maximal adsorption rate (as @;, -+ 0) is much higher than the steady-state electroreduction rate and adsorp tion is not the slow stage of the process. The adsorp

tion stage rate becomes slower at more cathodic potentials, which results in the drop of the surface coverage with chemisorbed particles (Fig. 6). The experimental data obtained show that under nitrobenzene electroreduction on platinum in alkaline solutions the slow stage is the addition of two electrons to a chemisorbed type I particle with the formation of dianion (C,H,NO,),d,

+ 2e-(C,HSN0,):G

(7)

which then is protonated (C,H,NO,&+

2H + + C6H,N0

+ H,O

(8)

Thus, nitrobenzene is electroreduced in alkaline solutions via the direct electrochemical mechanism, that is direct transition of an electron from the electrode to the organic molecule in the slow stage (cf- with the mechanism for metals with a high hydrogen overvoltage[9, lo]. We investigated the effect of the platinum electrode platinization on nitrobenzene electroreduction and showed that the character of polarization curves remains practically unchanged when passing from a smooth platinum electrode to a platinized one (*pa’ = 0.2 V), the Tafel slope slightly increasing (see Fig. 7). However, in transition from a smooth to a platinized platinum electrode at a constant potential the specific (that is normalized on the unit true electrode surface) elcctroreduction rate decreases, first rapidly and then slower, with an increase in the roughness coefficient, y = S,,,/S,,, . This is likely to be related to an increase in the fraction of valenceunsaturated or defect platinum atoms, on which the nitromethane electroreduction rate is much lower than on platinum atoms on an ideal single crystal face[l 11. When passing from smooth platinum electrodes to

General scheme of adsorption,

electroreduction

and catalytic

of t&o-compounds

hydrogenation

(a)

569

(b)

E/V

Y

Fig. 7. Polarization curves (a) of electroreduction of 10-‘M CHsN02 +0.5M H,SO, on smooth and platinized platinum electrodes with different roughness factors y (l-1.4; 2-380; 3-980; 4-2100) and dependences of ektroreduction rate on the electrode roughness factor (b) at different potentials (V): S-0.05: 6-0.1; 7-0.15; 8-0.2.

well platinized ones the specific nitromethane electroreduction rate drops by a factor of almost 50. The yield of various products in nitromethane and nitrobenzene electroreduction on a platinum electrode was investigated in acid solutions at different potentials. It can be seen from Table 1 that in nitromethane electroreduction the yield, with respect to the current, of methylhydroxylamine drops with an increase in the cathodic potential from almost 100 per cent at 0.15 V to 20-25 per cent at 0 V, whereas the methylamine yield increases from0 to almost 80 per cent. Moreover, the methylamine yield should also be noted to increase with the nitromethane solution exhaustion. This is likely to be due to the fact that the methylhydroxylamine formed participates in further reduction. The nitrobenzene electroreduction proceeds generally in a similar way (Table 2), though much less pronounced, mainly due to the possibility of forming some other products which manifests itself most of ail in long electrolysis. It is also interesting to note that in nitrobenzene electrolysis on platinum at potentials more cathodic than 150 mV, when a limiting kinetic Table 1. Results

current is observed on the polarization aniline yield drops with a further cathodic potential.

Kinetics of elect&reduction of particles preliminarily ckemisorbed from nitro-compounds solutions If a nitro-compound is adsorbed on an electrode in its high adsorption potential range and then the potential is changed jumpwise to a value near the hydrogen zero, one may observe the kinetics of removing chemisorbed organic particles from the electrode surface due to their electroreduction up to a steady-state value corresponding to a new potential (Figs. 8, 9). The surface coverage with nitrocompounds chemisorbed particles drops linearly with an increase in the desorption time logarythm 1 Pa = B--Blflnr where

j3,f”

of potentiostatic electrolysis of losolution on a platinized platinum

= 10.2 at E, = -0.025

’ h-i CHaNO,

Electrolysis

Nitromethane

Yield with respect to

consumption,

the current %

W)

methylhydroxyl amine

methylamine

0

50 85

25 19

75

0.1

20 45 5s 63 65

44 30 :: 40

56 70 68 65 60

10 7

95 107

5 0

0.15

+ 0.05 V for nitro-

in 0.5 M H,SO,

electrode

potential

E,

curves, the shift of the

81

570

Yu. 0.VASSILIEV. V. S.BAGOTZKY, 0. Table 2. Results

of potentiostatic

A. KHAZOVA, T. N. YASTREBOVAAND T. A. SERGEEVA

electrolysis

solution on a platinized Electrolysis potential E,,

w

methane benzene;

Nitrobenzene consumption

of 2 10m3 M C,H,NO, platinum electrode

in 0.5 M H,SOI

Yield with respect to the substance

%

phenylhydroxyl amine

aniline

0.03

70 87

48.6 47.1

34.3 35.6

0.1

5s 82 95.5

43.6 47.5 SO.3

46.4

0.2

45.5 64.5 83.5

24.2 24.8 28.14

51.6 58.1 58.2

0.3

39 50 80

39.7 36.0 27.5

32.0 39.0 62.0

and 11-l 7 at E, = - 0.025 + 0.05 V for nitrothe factor @If decreases significantly with

cathodic shift of the potential and increases with its anodic shift. Nitro-compounds adsorbed particles are removed from the platinum electrode surface independent of the adsorption potential and of what type of particles (I or II) is formed in adsorption. In fact, if the adsorption is carried out at different potentials and desorption -at one and the same potential, the curves of the total surface coverage with all the chemisorbed particles coincide in time domain (Fig. 9). In this case the rate of chemisorbed particles electroreduction does not depend on whether the nitro-compound is present in the solution, or desorption is carried out after washing in the background solution (Fig. IO). This indicates that the bond of nitro-compounds chemisorbed particles with the surface is not strengthened during washing or changing their nature. For the same surface coverage the rate of chemisorthe

48.8 45.5

bed particles electroreduction is practically independent of the nature of the nitro-compound (Fig. 8) from which they are formed (nitromethane, nitroethaw and nitrobenzene). Only for HNOs solutions the rate of chemisorbed particles electroreduction differs considerably from the rate of particles electroreduction from organic nitro-compounds solutions. It is noteworthy that chemisorbed particles formed in adsorption with dehydrogenation of hydroxylamine (the product of HNO, partial reduction) are electroreduced much slower than the particles formed from the nitrous acid itself. This means that NHsOH is the main product of HN02 reduction on platinum. The adsorbed particles electroreduction rate grows sharply with the cathodic shift of the potential. At a constant surface coverage with adsorbed particles the electroreduction rate logarythm at E, 0 increases linearly with the cathodic shift of the potential (Fig. 11). Investigation of adsorbed particles electroreduction in phosphate buffer solutions with different pH

Fig. 8. Kinetics of ekctroreduction at E, = 0 V of partides chemisorbed at E, = 0.2 V on a smooth platinum 3-1.55-lo-‘M C,H,HO,; 4electrode from solutions: 1-1O-3 M HNO*; 2-10-4 M NH,OH; 10V2 M C,H,NO,;

5-lo-’

M CHaN02

in 0.5 M HISO*.

General scheme of adsorption, electroreduction and catalytic hydrogenation

of nitro-compounds

571

values (Figs. 10,12) shows that at a constant potential relative to the hydrogen electrode in the same solution, the particles electroreduction rate practically does not vary with the change in the solution pH. The general kinetic equation for electroreduction of nitrocompounds particles preliminarily. chemisorbed on a platinum electrode is of the form -~=k,exp(/31f’t&)exp

(-+)

(10)

where ai = 1.2-1.3 for nitromethane and 2.1 for nitrobenzene. This equation is completely analogous to kinetic equation (5) for electroreduction of a nitrocompound from the solution volume. Moreover, direct comparison (for the same conditions) between the nitro-compounds electroreduction currents and the removal rates of preliminarily chemisorbed particles shows that they are equal to each other (Fig. 12), that is in both cases the processes have the same slow stage: interaction of a semireduced chemisorhed particle with adsorbed hydrogen formed in the preceding rapid electrochemical stage.

E,cO

(El Fig. 9. Kinetics of electroreduction (l-8) and catalytic hydrogenation with molecular hydrogen (9) of particles chemisorbed from solutions lo-‘M CHSNOL f 0.5 M H,SO, at I$* = 0.2 V (a) for potentials: 1- -0.075,2- -0.05, 3- -0.025,40; S-0~02~6-O.~S; 74.075; g-0.1 V and electroreduction at E, = 0 V of type I (4) and type II (1,3) particles chemisorbed at different potentials (b) (l-0.2; 240.4 V), and also of mixture )_ of particles of both types adsorbed at E, = 0.4 V(

Kinetics of liquid-phase catalytic hydrogenation of prechemisorbed on platinum semireduced nitrocompounds particles with molecular hydrogen Kinetics of liquid-phase catalytic hydrogenation with molecular hydrogen was investigated by the diskring electrode technique_ As can be seen from Figs 9,10, the curves of prechemisorbed nitromethane and nitrobenzene with molecular hydrogen coincide completely with the eiectroreduction curves, within experimental accuracy. The rate of pre-chemisorbed nitromethane and nitrobenzene hydrogenation at a constant hydrogen pressure may be written as

06

Fig. 10. Kinetics of electroreduction (l-9, 11-U) and hydrogenation with molecular hydrogen (10) of particles chemisorhed at E, = 0.2 V from solutions of 1.55 lo- M (l-lo), 0.47 10-&M (11) and 1.55 lo-“M (12-13) nitrobenzene in OSM HISO, (I-6, 11-13) in phosphate but&r solutions p (7 -pH = 1.5 and 32; 8 -pH = 6.2) and in 1M KOH (9) at electroreduction potentials (V): l--0.05; 2--0.025; 3-6, 4.6-9, ll-130.25; 5-0.05 in the same solution (l-12) or after washing in the background solution (13).

Yu. B. VASSILIEV,V. S.

572

0. A. KHAZOVA, T. N. YASTREBOQAAND

BAGOTZKY,

(b)

(a)

I +o 05

T. A. SERGEEVA

I

E,/o”

I

I

0

-005

I

-005

‘005

EJV

Fig. 11. Dependence of electroreduction rate of pre-adsorbed at Efdd”= 0.2V nitromethane (a) and nitrobenzene (b) on the potentialat differentsurfacecoverage withchemisorhedparticles (ok): l-0.1 5; 2-0.2; 3-0.3; 4-0.4; 54.45; 64.6.



hydr =

Kl

exp(h.f;Bk)

(11)

Coincidence between the rates of electroreduction and hydrogenation of prcchemisorbed nitro-compounds particles with molecular hydrogen shows that both the processes proceed via the same mechanism and with the same rate. Comparison between the experimental rates of nitro-compounds hydrogenation and molecular hydrogen adsorption[4] indicates, even with due account for the effect of surface coverage with a chemisorbed nitrocompound, that the adsorption of hydrogen is always the more rapid stage. Thus, hydrogenation of a nitro-compound prechemisorhed on platinum with molecular hydrogen does not proceed with a noticeable rate via a “shock” mechanism, but (a)

like electroreduction proceeds through the interaction of a prechemisorbed semireduced particle with adsorbed hydrogen.

Kinetics of nitro-compounds catalytic interaction hydrogen pre-chemisorbed on platinum surface

Figure 13 shows the curves of the potential drop forL smooth and platinized platinum electrodes with an adsorbed hydrogen monolayer after they were brought into contact with nitromethane and nitrobenzene solutions of different concentrations. As can be. seen from the figure, unlike the curves of the potential shift in introducing maleic acid[2], the curves for nitromethane and nitrobcnzene plotted in E-lnr coordinates have two sections with different slopes E, = n+glnr

(12)

1

,.

with

Direct measurement of the electrode surface coverage with adsorbed hydrogen and chemisorbed organic particles under the potential shift (Figs 14, 15) shows that the potential shift is due to a decrease in the surface coverage with adsorbed hydrogen and the potential shift rate is determined by the rate of removing pre-adsorbed hydrogen from the platinum surface because of hydrogen interaction with the nitrocompound, ie

3 0

aE _K.aBH -_=

(b)

Fig. 12. Kinetics of electroreduction at E, = 0.0 V of nitramethane pre-chemisorhed at E:ds = 0.2 V in phosphate btiffer solutions with different pH vaiuer l-0.4; 2-1-O; 3-1.65; 4-2.65; S-3.15; 6-5.15; 7-6.45; 8-7.25; 9-8.35; lo-lo.1 (a) and kinetics of ekctroreduction of solely hydrogenated particles (b).

a?



ar

(13)

The potential shift rate (or the rate of removing preadsorbed hydrogen from the piatinum surface) grows as a fractional power of the nitro-compound volume concentration and exponentially decreases with an increase in the anodic potential (Fig. 16) aE

-

ar

= IWexp(--g,,)

(14)

where y varies from 1 to 0.5 for nitromethane and from

General

scheme of adsorption,

electroreduction

Id

04

(cl Fig. 13. Variation with time of the potential of platinized (a, c) and smooth (b, d) platinum electrodes wilb preadsorbed hydrogen after bringing them into contact with nitromethane (a, b) and nitrobenzene (c, d) solutions in 0.5 M H,SO, (l-8). in phosphate buffer with pH = 2.9 (Q), pH = 6.8 (1O)andin 1MKOH (11): l-5-lo-*M;2-lo-*M; 4-0.47 lo-‘M; 3-3 10-s M; 5 9-11-l 55 10-sM. 60.47 1O-3 M; 7-1.55 10-4ti; S-.l.Sj lo-$ M. ’

1 to 0 for

nitrobenzene

with

the potential

changing

from 0.05 to 0.3 V. Direct measurements of the potential shift curves in phosphate buffer solutions show the potential shift

and catalytic

hydrogenation

of nitro-compounds

573

rate to he practically independent of solution pH at a constant potential relative to the hydrogen electrode in the same solution, because E, = const means that the surface is almost constantly covered with adsorbed hydrogen. Thus, the kinetic equation for nitromethane and nitrobenzene catalytic hydrogenation proves to be completely analogous to that for electroreduction of these substances. A proportional change in the potential shift rate at E, -z 0.1 V with an increase in the nitrosompound volume concentration shows that in this section the potential shift rate and the rate of catalytic interaction of nitromethane and nitrobenzene are determined by their adsorption rate. This becomes especially evident from direct comparison between the maximal adsorption rate, as & -+ 0 (dashed line in Fig. 16d) and the potential shift rate, which prove to be equal in this section. In this section the nitromethane and nitrobenzene chemisorbed particles are not accumulated on the surface (Figs 14,15), because all the adsorbed substance is at once hydrogenated to the product moving away into the solution volume. At potentials more positive than 0.1 V the adsorg tion rate exceeds that of hydrogenation which results in the accumulation of chemisorbed particles on the surface. It is interesting to note (Fig. 17) that under formation of a chemisorbed particle, occupying one adsorption centre, first one adsorbed hydrogen atom is removed from the surface and then two adsorbed hydrogen atoms. It can be seen from Fig. 17 that there is a linear relationship between the platinum catalyst surface coverage with a chemisorbed nitro-compound and adsorbed hydrogen, ie ek = C-n&

(15)

where ni = 1 and nz = 0.5. Two sections in the curve E,-lnr, lnu - E, and & - B, tell us that unlike maleic acid[2] the removal of hydrogen from the electrode surface proceeds in this case in two stages. The first stage is the rapid interaction between an adsorbed nitrobenzene or nitromethane molecule and adsorbed hydrogen with the formation of a semireduced chemisorbed particle,

LC

0.0

4 0.6 0.4

0.2

Fig. 14. Dependence of coverage of smooth (1) and pkttinized (2,3) platinum electrode with chemisorbed nitrobenzme (1,2), nitromethane (3) and hydrogen (1’. 2’) on the time of being in contact: l-l.55 10-&M; 21.55 10-3M; 3-10-2M.

Yu. B. VASSILIEY,V. S. BAGOTZKY. 0. A. KHAZOVA,T. N.

‘8

YASTREBOVA

AND

T. A.

SERGEEVA

stage of electroreduction and catalytic hydrogenation of nitro-compounds on platinum is the stage of interaction between a semireduced chemisorbed particle, formed in the preceding rapid stages of nitrocompound chemisorption

-

and semireduced

particle

formation

R-NN~+~H.~,-,RN-O+H~O xx xxx X

and adsorbed atomic hydrogen, preceding rapid electrochemical

(17)

x

formed stage

either in the

H*+e=Hndg or in the preceding adsorption

.6 -

(18)

rapid stage of molecular H2

+

2H&

hydrogen (19)

that is the stage R -

N - 0 + mH& + products xxx x For the slow stage (20) we may write Fig. 15. Coverage of smooth (1) and platinized (2, 3) platinum electrode with chemisorbed nitrobenzene (1,2), nitromethane (3) and hydrogen (l’, 2’) as a function of the electrode potential: l-l.55 1Om4M; 2-1.55 1Om3M; 3-10ea M. which is indicated by the fractional exponent in the dependence of the potential shift rate on the volume concentration. The second stage is much slower further addition of hydrogen to the semireduced particle with the formation of a product desorbed into the solution volume. Catalytic interaction of chemisorbed nitro-compounds particles with atomic hydrogen diflised through a palladium membrane Kinetics of catalytic hydrogenation of semireduced chemisorbed nitrobenzene particles by atomic hydrogen was studied by means of a platinized palladium membrane. As can bc seen from Fig. 18, the curves of removal of pre-chemisorbed at Ep = 0.2 V semireduced nitrobenzene particles under their hydrogenation by atomic hydrogen diffused through a palladium membrane and under their electroreduction completely coincide, within experimental accuracy. This is one more unambiguous indication that electroreduction and catalytic hydrogenation proceed via the same mechanism and that in both the cases the interaction of a semireduced chemisorbed particle with adsorbed atomic hydrogen is the slow stage of these processes. Mechanism of electroreduction and catalytic nation of nitro-compounds on platinum The experimental

data available

hydroge-

show that the slow

u = [email protected]&exp(-/?AQ)

(20)

(21)

where the exponent takes into account the effect of the change in the adsorption energy of both chemisorbed particles on the reactions activation energy. Since in the potential range O.&O.3 V the surface coverage with chemisorbed semireduced particles and with adsorbed hydrogen has medium values, we obtain analogously to[1,12-151 u = KtIEt?k exp[ /lrn(j”_” +B(&-R

+&H)&l

+ ji;-a)OH (22)

where fijare the terms taking into account the chemisorbed Particles’ mutual effect induced through the surface layer[l2]. As can be seen from Figs 19-20, equation (22) describes well the experimental data for nitromethane and nitrobenzene electroreduction on platinum. Participation of two chemisorbed particles (organic particle and hydrogen) in the slow stage and their mutual effect[13] explain some other features of nitrocompounds reduction on platinum, for instance the extremal character of the lni+ dependence at E, = cons:, or practically horizontal delay section on the charging curve in electroreduction of nitrocompound chemisorbed particles. Stopping of electroreduction and hydrogenation of aliphatic nitro-compounds when passing to alkaline solutions is due to the transformation of aliphatic nitro-compounds in alkaline solutions into aciform, which is adsorbed because of opening of double carbon-nitrogen bond with the formation of new C-Pt and N-Pt bonds, among which the C-Pt bond is practically not hydrogenated. Therefore, nonhydrogenated chemisorbed particles are formed on the surface. In electroreduction and hydrogenation of nitrobenzene in aikaline solutions the electrode potential is

General scheme of adsorption, electroreduction

and catalytic hydrogenation of nitro-eompounds

(b) -2

J ? -3.

-5.

-1

-2 ‘4

-3

-4

>

Fig. 16. Dependence of the potential shift rate of platinized (1-19) and smooth (l’-18’) platinum electrodes on the nitromethane (a) and nitrobenzene (c) volume concentration at different potentials: I-0.05; 2-0.1; 30.125; 4-0.1s; 5-0.175; 6-0.0225; 7-0.250; 8-0.275; 9-0.3; 10-0.325; 1l-0.350; 12-0.375V and on the catalyst potential in nitromethane (b) and nitrobenzene (d) solutions with different concentration: 13-S IO-*; 14 10-s; 15-5 lo-‘; 16-1.55 10-s; 17-0.47 10-s; 18-1.55lO-4; 19-0.47 lo-sM.

Fig. 17. Coverage of the surface of smooth (1) and platinized (2) platinum catalyst with chemisorbed nitrobenzene particles as a function of coverage of platinum catalyst surface with pm-adsorbed hydrogen in its interaction with nitrobenzene solution: l-l.55 fOm4M; 2-1.55 IO-sM.

57s

576

Yu. B. VASSILIEV, V. S. BAGOTZKY, 0. A.

KHAZOVA,

T. N.

YASTREEWVA

AND

T. A.

SERGEEVA

Fig. 18. Variation of the coverage of the platinized palladium membrane surface with pre-chemisorbedat E, = 0.2 V nitrobenzene particles due to electroreduction at E, = 0.02 V (1) and due to hydrogenation with atomic hydrogen diffused through the palladium membrane (2). _

tion and catalytic hydrogenation of nitro-compounds in acid and ueutral solutions the processes of nitrocompound adsorption, formation of semireduced makes it possible simultaneous transition of two chemisorbed particles and hydrogen adsorption proelectrons in the slow stage. In this case reduction on ceed simultaneously. Hydrogenation of a chemisorbed platinum proceeds via the rare electrochemical nitrocompound may proceed either through 0-Pt bond or through N-Pt bond, which results in the mechanism. desorption of products of an incompletely reduced nitro-compound into the solution volume. Therefore, all the process of nitro-compounds electroreduction and catalytic hydrogenation on platinum may be General scheme of adsorption, electroreduction and represented by the scheme shown in Fig. 21. The yield catalytic hydrogenation of nitro-compounds on of various products under. steady-state conditions is pfatinum determined by the relationship between the rates of different stages. Under the steady-state conditions of electroreducso negative that

electrons can pass directly from the electrode to an adsorbed nitrobenzene molecule. Conjugation of bonds in a nitrobenzene molecule

(a)

/

(b)

Fig. 19. Dependence of elcctrorcduction rate of nitromethane (a), pre-adsorbed semireduced nitromethane particles (b) and nitrobenzene (a) on the platinum electrode surface coverage with adsorbed hydrogen at a cOnsWantcoverage with organic substance t& : 1- - 0.6; 2-0.55; 3-0.5; 4-0.45; 5-0.4; 6-0.35; 7-0.3; S-0.25; 9-0.2; lo-O.1 5; 11-O.

General

scheme

of adsorption,

electroreduction

and catalytic

hydrogenation

of nitro-compounds

577

(b)

a4

06

ek

Fig. 20. Dependence of electroreduction rate of nitromethane (a) and pre-adsorbed semireduced nitrobenzene particles (b) on the electrode surface coverage with the organic particles at a constant surface coverage with hydrogen, 6,: t-0.5; 2-0.45; 3-0.4; 4-0.35; 5-0.3; 6-0.25; l-0.2; 8-0.15; 9-0.8; 10-0.75; 1 I-0.7; 12-0.65.

Hydrogenation

Fig. 21. General

scheme

of adsorption,

2.

3. 4.

5. 6.

N-O

bond

electroreduction and catalytic on platinum.

REFERENCES 1. Yu. B. Vassiliev.

thrwgh

V. S. Bdgotzky, 0. A. Khazova, N. N. Krasnova and T. A. Sergeeva, Electrochim. Acto 26, 545 (1981). Yu. B. Vassiliev, V.S. Bagotzky, 0. A. Khazova, V. V. Cherny and A. M. Meretsky, J. elecrroanal. Chem. 98, 2533273 (1979). V. S. Bagotzky and Yu. B. Vassiliev, Electrorhim. Acfa 11, 1439 (1966); 12, 1323 (1967). 1. Gonz, Yu. B. Vassiliev and V. S. Bagotzky. Elektrokhimiya 6,325 (1970); V. A. Gromyko, Yu. B. Vassiliev, V. S. Bagotzky, Elektrokhimiya 8, 914 (1972). A. G. Polyak, Yu. B. Vassiliev and V. S. Bagotzky. Hekrrokhimiya 1, 968 (1965). F. Treadwell. Anolyrical Chemisrry, Vol. II, book 2, p. 144. Inostrannaya Literatura Publishers, Moscow (1931). (Russian translation).

hydrogenation

of nitro-compounds

7. N. A. Izgaryshev and A. A. Petrova. Zh. Fiz. Khim. 24, 745 ( 1950). 8. Ya. P. Stradyn’. Polarography of Organic Compounds, p. 49. Acad. Sci. Latvian SSR, Riga (1961). 9. W. R. Riti in N. L. Weinberg (Editors), Technique ofElectroorganic Synthesis, Part II, p. 137. WileyInterscience, New York (1975). 10. B. Kastening, Narurwiss 47, 443 (1960). 1 I. 0. A. Khazova, Yu. B. Vassiliev and V. S. Bagotzky. Elektrokhimiya 3, 1020 (1967); 6, 1367 (1970). 12. M. I. Temkin. Zh. Fiz. Khim. 15, 296 (1941). 13. 0. A. Khazova, Yu. B. Vassiliev and V. S. Bagotzky, Elekrrokhimiya 10, 606 (1974). 14. V. V. Cherny and Yu. B. Vassiliev. Elekfrokhimiya 11, 79s (1975). 15. T. N. Yastrebova. A. A. Sutyagina, C. D. Vovchenko and Yu. B. Vassiliev, Elekrrokhimiya 13, 1778

(1977).