Heterogeneous and Homogeneous Electron...
14 downloads
118 Views
1MB Size
Saceant et al.
/ Redox Catalysis in Halobenzenes and -pyridines
343 1
Heterogeneous and Homogeneous Electron Transfers to Aromatic Halides. An Electrochemical Redox Catalysis Study in the Halobenzene and Halopyridine Series Claude P. Andrieux, Claude Blocman, Jean-Michel Dumas-Bouchiat, and Jean-Michel Saveant* Contribution froni thr L.aboratoire d'Elrctrochiniir de I'UnirersitP de Paris V I I , 2 Place Jirssieir, 75221, Priris Cedes OS, France. Kecrired October 18, I978
Abstract: Halobenzenes and halopyridines undcrgo a t ~ o - e l e c t r o nirreversible reductive cleavage. I t involves fast dccomposition of the assumed anion radical intermediate. ARX-e ( k , , , order >> IOJ s - ' ) . and tranhient formation of a neutral radical. Ar. easier to reduce than the starting molecule. M ' i t h the exception of Phl. t h e clectrocheniical reaction is under the kinetic control of the initial electron transfer step. Standard application of the electrochemical techniques provides the transfer coefficients and the forward electron transfer rate constants but no information about the standard potcntials and the standard rate constants. T h e latter quantities can be determined through the kinetic analysis of homogcncous redox catallsis of the electrocheniical reduction. T h e following points a r e discussed: activation vs. diffusion control of the homogeneous electron transfer: existence of the anion radical as a n actual reaction intermediate vs. simultaneous electron transfer and breaking o f the carbonhalogen bond: correlation betueen homogeneous and heterogeneous electron transfers: determination of the standard potcnt i a Is and the s t a nd a rd rat e cons t a nt s: r el a t ions h i ps bet ween structure a nd electron t r a n s fe r the r mod yn a m ics and k i net ics . T h e remarkable slowness of the electron transfer as compared to aromatic and heteroaromatic compounds of similar molecular size implies, besides solvent reorganization. a significant internal reorganimtion due to stretching of the carbon-halogen bond when passing from the starting molecule to the anidn radical
Introduction The electrochemical reduction of aromatic halides in organic nonaqueous solvents (see ref 1 and references cited therein) generally involves, in the cases of CI, Br, and I , the cleavage of the carbon-halogen bond of the initially generated anion radical. ArX le e ArX-. ( I el)
+
Ar. ArX-.
Ar-
+
1
+ le
+ Ar-
H l 0 (residual) R3N+-CHrR' Ar'
+ SH +
-
+
Ar-
(3)
ArX
+ Ar-
ArH
+
ArH
+ S.
(4)
Is-
1
OH(5) R~N+-CH-R' (6)
+
ArX-. S. ArX S(8) The neutral radical thus formed may then undergo an electrode (3) or a solution reduction (4) leading, after protonation by the solvent, the residual water, or the quaternarb ammonium cation of the supporting electrolyte (5), to an overall hydrogenolysis of the carbon-halogen bond. Ar. may also abstract an hydrogen atom from the solvent ( 6 ) ,the solvent radical thus formed being further reduced to the solvent conjugated base at the electrode ( 7 ) or in the solution (8). In such conditions, the main reduction process will then appear ax a two-electron irreversible wave provided that the lifetime of the ArX-. anion radical is short within the time scale of the experiment. The yield of ArH upon electrolysis at the potential of this wave will then be 100%. This is what is observed in a number of cases, e.g., with chloro- and br0moben7ene,~chlorobiphenyl,4 and 1 -chloro-, bromo-, and iodobenzonitriles.h A second wave is observed when ArH is reducible before the background discharge, its reversibility depending upon the stability of the ArH-. anion radical. 0 0 0 2 - 7 8 6 3 / 7 9 / 1 5 0 l - 3 4 3 1$01.OO/O
In some cases, however (see ref I and references cited therein), the apparent number of faradays per mole at the first wave is less than 2 and the yield in ArH significantly less than 100%. This phenomenon has been analyzed in detail in the case of 4-bromobenzophenone and has been shown to be caused by an S u ~ substitution l of the halogen by the solvent conjugated base S-.'.' The substitution products give rise themselves to reduction waves that cannot practically be distinguished from that of ArH. That such a process occurs to a much lesser extent in the present ArX series can be explained in this context by the weaker electrophilic character of the AP radical. It could, however, be more effective at potentials located a t the foot of the second wave as revealed by the frequent occurrence of a current dip in this region.'Such dips are indeed observed when the electron-induced substitution is carried out with purposely added nucleophiles? and can be interpreted as an acceleration of the substitution process in this potential r e g i ~ n . ~ A detailed discussion of these phenomena is beyond the scope of the present paper and will be the object of forthcoming publications. We will here focus our attention upon halo compounds that give rise to a two-electron process with 100% .4rH yield at the first wave, the reduction mechanism of which can be pictured by the above scheme (reactions 1-8) without further interference of the solvent conjugated base. When reaction 2 is fast, i.e., for rate constants larger than 1 04- 1 Oi s-I, no chemical reversibility can be detected using the clectrochemical kinetic techniques even at the lowest extremity of their time scale (e.g., cyclic voltanimetry with sweep rates in the kilovolt per second range). With the exception of compounds bearing strong electron-withdrawing groups, such as C O R and NO?, this is a very common situation among the chloro, bromo, and iodo aromatics. The kinetic control of the reduction process is then by the initial electron transfer ( I ) and the cleavage reaction (2). I f the initial electron transfer is slow and/or the cleavage reaction is fast the reduction process tends to be kinetically controlled by the electron transfer.x When such a situation prevails the standard application of the usual electrochemical techniques gives no information not only about the lifetime of ArX-. but also about the standard potential, E", and the electrochemical standard rate constant, ks", of the ArX1ArX-e couple. The only available data are then the
0 I979 American Chemical Society
3432
Journal of the American Chemical Society
transfer coefficient constant :
CY
krCl
and the electrochemical forward rate
= ksCl exp( CKFE O/R T )
of the electron transfer. I n other words, the kinetics and the thermodynamics of the electrode electron transfer cannot be characterized separately. It has been recently shown9 that the kinetic analysis of homogeneous redox catalysisI0 of such electrochemical reactions by a series of reversible redox couples may provide a much more complete characterization of the electron-transfer process allowing the determination of the standard heterogeneous rate constant. It provides in addition the value of the standard rate constant of the homogeneous electron transfer. The method was illustrated by the experimental study of the reduction of chlorobenzene using classical p ~ l a r o g r a p h y . ~ One purpose of the present paper is to describe the application of the method to a series of aromatic halides using cyclic voltammetry for which the theory of homogeneous catalysis has been recently worked out.’ I Cyclic voltammetry indeed leads to a better accuracy than polarography and to an extension of the potential range in which the electron-transfer kinetics can be determined. Time needed to record a cyclic voltammogram is much shorter than for a polarogram. This leads to a very significant time saving since a detailed redox catalysis study requires a large number of experiments involving a series of redox couples. The compounds selected for this study-chloro-, bromo-, and iodobenzene and 2- and 3-chloro- and -bromopyridinefall, with the exception of iodobenzene, in the category defined above, i.e., give rise to an irreversible 2e reduction kinetically controlled by the forward electron transfer at the electrode as well as in the solution. The thermodynamic and kinetic data obtained from the redox catalysis study of these compounds will then serve to discuss the following points: activation vs. diffusion control of the homogeneous electron transfer; existence of the anion radical as an actual reaction intermediate vs. simultaneous electron transfer and breaking of the carbon-halogen bond which is a problem of general interest for both aliphatic and aromatic halo compounds (see, e.g., ref 12- 14); correlation between homogeneous and heterogeneous electron-transfer kinetics; relationships between structure and electron-transfer thermodynamics and kinetics. Redox Catalysis as a Method for Determining Standard Potentials and Standard Electron Transfer Rate Constants The method consists in introducing into the solution the oxidized form P of a chemically and electrochemically reversible couple P/Q, the redox catalyst, the standard potential of which, EOpq, is positive to the reduction potential ArX. The reduced form Q is then generated at the electrode and transfers an electron to the substrate ArX within the electrode diffusion layer giving rise to the following reaction sequence:
Q + ArX
F==
P + ArX’
(1sol)
followed by reactions (2, 4,5, 6, 8 ) and also by
(9)
Q + S’
-
p+s-
(10)
P is regenerated through reactions ( 1 sol), (9), and ( I O ) resulting in an enhancement of its reduction current as com-
/
101:13
/
June 20, 1979
pared to that observed in absence of substrate. This “catalytic” effect is only a function of k I in the case where the forward homogeneous electron transfer is the rate-determining step. It can therefore be used to determine k I according to an appropriate treatment of the kinetics of this p h e n ~ m e n o nThe .~ experiment and treatment of data are then repeated with other catalyst couples so as to explore the largest possible range of standard potentials. The kinetics of the cross electron exchange between the ArX/ArX-- and P / Q couples, as reflected by the values of k I and k2, is under the control of both activation and diffusion factors and is a function of the intrinsic characteristics of the electron transfer to each couple.Is However, when using as catalysts aromatic or heteroaromatic molecules, both the diffusion coefficient and the isotopic activation free energy of the catalyst can be considered as approximately constant in the series.Ih It is also noted that the isotopic free energies of activation are small,16 which reinforces the validity of the approximation. I n this context, the variations of k l from one catalyst to the other reflects how the intrinsic kinetics of electron transfer i n the ArX/ArX-- varies with potential. I n order to describe the relative interference of activation and diffusion factors, let us consider reaction ( 1 sol) as composed of three successive steps: ! ,, + Q 3(ArX, Q) e(ArX-., dcl
ArX
dil
I, 2”cl
P)
Add
ArX-.
+P
Adif
(ArX, Q) and (ArX-a, P) represent the reactants in their reaction sites, whereas ArX Q and ArX-. P represent the reactants beyond the average diffusion distance. k d , f is the diffusion-limited rate constant, and k lac‘ and kyic‘ are the activation-controlled rate constants. Writing the stationarystate assumption for both (ArX, Q) and (ArX-., P) one readily obtains the expression of k l as
+
+
(Ilk11 = ( I / k i a c t )
+ ( I / k d i f ) [ l + exp(F/RT)(EoPQ - E o ) ]
( 1 1)
with
log klac‘ = log ksso’- C K ~ ~ I ( E-O E0)/0.058 ~Q (at 22 “C) (12) where aSo1 and ksbolare the transfer coefficient and the standard rate constant of the homogeneous electron transfer to the ArX/ArX-- couple. cxSo0lis close to 0.5 when the potential is not too far from the Eo. According to the magnitude of E 0 p q - Eo two zones of diffusion control with 0 and - I / % mV slopes, respectively, and an intermediate zone of activation control with a - 1/ 1 16 mV slope are obtained on the log k I E0pq plots as shown in Figure I . The extension of the activation zone is larger the lower ksSol. The possibility of determining Eo is based on the experimental observation of the diffusion zone at E O ~ Q> EO: Eo is then obtained by extrapolation of the corresponding straight line up to the value of the diffusion-limited rate constant. The existence of this diffusion-controlled zone is not significantly affected by reaction 2 as long as it is not so fast as to occur within the average diffusion distance. The latter quantity being on the order of molecular sizes, this condition will be fulfilled as long as k is smaller than about 1 O l o s-l. It is noted incidentally that beyond this limit the very existence of the ArX-. anion radical becomes ambiguous. The homogeneous electron transfer process can be considered as an electrolysis with an extremely small diffusion layer thickness, on the order of a few Angstroms, as opposed to electrolysis in the context of the usual electrochemical techniques where the diffusion layer thickness is at best 1000 times larger. In electrochemical conditions reaction 2 occurs within a small portion of the diffusion layer whereas in homogeneous
Saveant et al.
/ Redox Catalysis in Halobenzenes and -pyridines
3433
log k (1) DIFFUSION CONTROL
I EO
Figure 1. Predicted variations of the homogeneous forward electron tr:insfer rate constant with potential showing activation and diffusion control. ( 1 ) Diffusion control: log k , = log k d i f . ( 2 ) Activation control: log X I = log ks"O1 - W " ~ ( E ~-~E0)/0.058. Q ( 3 ) Diffusion control: log X I = log X d , f - (E"PQ - E0)/0.058. Figure 2. Cyclic voltammetry of halobcnzenes and holopqridines: (a) Phl, ( b ) PhBr. (c) PhCI. (d) 3-pyCI. (e) 2-pyCI. (g) 2-pyBr ( h ) 3-pyBr at 0.33
conditions it occurs outside the diffusion layer. These are fundamental reasons why in the present case quantities such as E o are inaccessible through electrochemical techniques while they can be obtained through redox catalysis which thus amounts to be a "molecular sized electrode electrolysis." On the other hand, reaction 2 has been assumed to be fast enough so as the forward reaction ( 1 sol) is the rate-determining step. The condition to be fulfilled then is k > k2C0pq. (C0pg is the bulk concentration of the catalyst.) If k2 is at the diffusion limit and with standard experimental condition this amounts to k > 10h-I O7 s-l. This point can anyway be checked experimentally since the catalytic increase of the current has a characteristic dependence upon CO~Qwhen ( 1 sol) is the rate-determining step whereas there is no concentration dependence when ( 2 ) is the rate-determining step. It is indeed clear that the overall process is second order toward Q i n the former case and first order i n the latter. Once E o is determined using the above procedure, the standard rate constant of the homogeneous electron transfer ks5*01is obtained as shown in Figure I . This requires the activation control domain not to be too small i n order to be able to locate the straight line with the - 1 / 1 16 mV slope. The activation free energy of the cross exchange reaction is given by
AG*,,I/F = 0.058 log (ZSol/kS"')
(1 3)
(Z\"l: collision frequency in the solution) and the intrinsic kinetic characteristics of the homogeneous electron transfer to ArX can then be obtained from I
+
AG*sol = ( A G A r ~ * J b 0 AGp*.'\" )/2
(14)
under the form of the activation free energy of the isotopic transfer to ArX if the corresponding quantity is known for the catalysts. When employing aromatic or heteroaromatic molecules as catalysts an average valuei6
ACp*J"/F = 0.159 V can be used with a good approximation. Knowing Eo, the heterogeneous electron transfer kinetics can also be characterized. Using, e.g., cyclic voltammetry, the electrochemical transfer coefficient, a , and standard rate constant, ksel, can be derived from the measurement of the peak potential as a function of sweep rate u:I7
(dE,/d log V ) = -0.058/(~ V
(15)
V
6-l;
( f ) 3-pyCI. ( i ) 2-pyBr ai 330 V s-l. Substrate concentration 2 X L - I , Vertical scale: PA. Horizontal scale: V vs. SCE.
M
and
E , = Eo
+ (0.058/a)log kSel,ap
- (1 / ~ ) [ 0 . 0 6+ 7 0.029 log CUOD]
a t 22 " C . k ~ ~ l iss the ~ p electrochemical apparent standard rate constant, Le., uncorrected from double-layer effects. The standard rate constant is then obtained through the Frumkin correction: log kse' = log k ~ ~ '-%(~q52/0.058) ~ p
(16)
where 42 is the potential difference between the outer Helmholtz plane and the solution. The activation free energy of the heterogeneous electron transfer is finally given by
AG*,I/F = 0.058 log (Zcl/ksel)
(17)
(ZcI:heterogeneous collision frequency).
Experimental Section Electrochemical instrumentation and Procedures. The experiments were carried out in D M F with 0.1 M BudNI as supporting electrolyte. The temperature was 22 O C . The working electrode used in cyclic voltammetry was a mercury drop hanging on a gold disk of about 0.8 mm2 surface area. The reference electrode was an aqueous saturated calomel electrode. The solution resistance was compensated using a positive feedback device down to a residual value on the order of I O R. The procedure for extracting the values of k l for each catalyst couple was similar to that used in polarography9 involving the determination of the catalytic efficiency for a series of values of the substrate/catalyst concentration ratio in each case. Working curves corresponding to cyclic voltammetry' were used for deriving k I from the catalytic efficiency. When possible, Le., for not too slow catalytic processes, several sweep rates were employed in order to obtain a better accuracy than in polarography. For high-efficiency catalysis corresponding to the wave of the catalyst being close to the wave of the substrate, sweep rates up to 1000 V S K I were used, allowing the investigation of a more extended potential range than in polarography. A more detailed description of the procedure and discussion of these points is given elsewhere." Chemicals. The aromatic halides were obtained from commercial sources (chloro-, bromo-, and iodobenzene, Prolabo; 2- and 3-chloroand -bromopyridine, EGA Chemie) and were used as received.
Results Electrochemical Behavior. Figure 2 shows the cyclic voltammograms obtained with the seven aromatic halides at
3434
Journal of the American Chemical Society
/
101:13
/
June 20, 1979
l a b l e I . Electrochemical Characteristics of Haloben7encs and Halopyridines"
PhCl ip1.-l/? V-l/? :iv bc'tween 0.33 and 100 V s-I -dE,/d log I , , mV/unit, from 0. I to
IO0 v s - I ALP. m V (av), between 10 and 100 V CY ( n v ) (from -?)E,/?) log r )
x (.Cl..IP
2-pqCI
3-p)Rr
2-plBr
26 f 2
23 f 2
28 f 2
22 f 2
25 f 2
48 f 2
56 f 2
57 f 2
60 f 2
58 f 2
Phl
2h
22f 2
55rt3h
64f2
23 f
s-I
3-pjC1
PhBr
85 f 5 95 f 5 15 f 5 83 f 5 90 f 5 84 rt 5 88 f 5 0.53 rt 0.03 0.45 f 0.02 0.60 f 0.02 0.52 f 0.02 0 . 5 1 f 0.02 0.48 f 0.02 0.50 f 0.02 7.5 X IO-'' 2.2 X IO-?? 3.0 X I O - l 9 1.5 x IO-?? 8.5 X 10-13 5.0 X IO-?' 2.8 X IO-??
'' i , = peak current: E , = peak potential: LE, = peak width: 1' = sweep rate; a = transfer coefficient: kfl.ep = apparent for\+ardrate constant of elcctron transfer. Determined between 0.1 and 2 V s-I only, since for higher sweep rates the wave merges with the discharge current of the supporting electrolyte.
'
Table 11. Redox Catalvsis of Bromobenzene Reduction
a n t h r accne
phcnanthridine
bcnzoV]quinoline
- I .89
20 50 50
0.033 0.033 0.33
1.21
I .46 I .06
0.4 0.4 0.5
0.4
20 50 50
0.33 0.33 3.3
I .37 2.07 1.10
1.7 I .9 1.8
1.8
5 20 5 50 50
0.33 0.33 3.3 3.3 33
2.7 5.2 1.28 2.85 1 28
3.1 3 3.2 3. I 3.1
3.1
5 20 5 20 50
0.33 0.33 3.3 3.3 33
2.8 5.8 I .3 2.3 I .48
3. I 3.2 3.3 3.3 3.4
3.3
-2.20
I
0.33 0.33 3.3 3.3 33 33
I .87 5.2 1.4 25
3.6 3.8 3.7 3.6 3.6 3.7
-2.24
5 2 IO 20 50 2
0.33 0.33 3.3 3.3 33
3 90 11.4
-2.00
-2.08
-2.12
chrjscne
bcnzonitrile
10 1 5 20
I,/-toluonitrile
-2.27
1 j
20 IO p-toluonitrile
p h c n a n t h re ne
-2.34
-2.41
0.5 5 2 5 2 5 I
-7
moderate sweep rates. The cyclic voltammograms remain irreversible up to 2000 V s-l, showing that, if the anion radical is actually the first reaction intermediate, its decomposition is fast (lifetime smaller than I O p 4 s). With the halopyridines the first irreversible wave is followed by a wave located a t the same potential as that obtained with pyridine itself, showing the formation of pyridine by reductive cleavage of the halogen at the first wave. The cathodic peak current is proportional to the square root of sweep rate within experimental uncertainty (Table I ) . The value of i , , ~ - l /is~almost the same for the seven compounds.
3.3 3.3 3.3 33 3.3 3.3 33 33 10
10 100 I00
I .3 1.9
1.6
3.8 2.8 1.75
4.0 9.4 2. I
4.2 4.3 4.3 4.3 3.4 4.4 4.5
4.5
4.5
3.2 3.8 6.7 1.4 1.8
5,6 5.6 5.5 5.6
-1
4.3
4.6 5.0 5.4 5.2 5.2
1 .8 8
3.1
5.2
5.6
The slight differences observed passing from one compound to the other are more likely to be due to some irreproducibility in electrode surface area than to actual variations in the apparent number of electrons. Since it is known that chloro- and bromobenzene' are reduced along a two-electron process leading quantitatively to ArH, it can be concluded that this is also the case with the other compounds, the difference in diffusion coefficients being negligible. To further check this point, coulometric experiments were carried out in the case of iodobenzene and 2-chloro- and 2-bromopyridine; they all gave an apparent number of electrons between I .8 and 2 and a quan-
Saceant et ai.
/ Redox Catalysis in Halobenzenes and -pyridines
3435
Table 111. Redox Catalysis of Chlorobenzene Reduction
typical values of y and L', V s - I
EOPQ,
catalyst bcnzonitrile
phenanthrene
V vs. S C E
-2.24
-2.41
0.033 0.033 0.33
1.25 I .50 I .06
0.5 0.5 0.5
2 10
0.033 0.033 0.33 0.33 3.3 3.3 33
3.2 8.0 I .45 4.9 1.3
2.8
50
-2.54
log k I. M-I L s-I
20 50 50
2 20 IO 50 diphenyl
(;p)c/(;p)o
2 2 IO 1
5
5
titative yield of pyridine as checked by cyclic voltammetry of the electrolyzed solution. A separate cyclic voltammetric study of pyridine in the same medium shows that a two-electron irreversible behavior is observed at low sweep rates. Chemical reversibility appears upon raising the sweep rate and is practically complete at 2000 V s-I, resulting then in a one-electron quasi-reversible behavior due to charge-transfer slowness. The second wave of halopyridines also shows increased reversibility and tendency toward one-electron behavior upon raising the sweep rate. This is apparently obtained somewhat more easily than in the case of pyridine itself. This is in fact the anodic counterpart of the dip observed on the cathodic trace at the foot of the second wave. The dip is more apparent i n polarography than in cyclic voltammetry and its magnitude decreases upon raising the sweep rate. This is to be related to the interference of substitution by S- in the framework of an ECE mechanism as discussed in the Introduction. The fact that this phenomenon is more apparent than in the case of, e.g., halonaphthalenes'.' indicates that the electrophilic character of the Ar. radical is enhanced by the presence of the ring nitrogen. Pyridine may indeed be considered as the vinylog of an imine which points to the possibility of localizing a second electron pair and a negative charge on the nitrogen. All these observations are consistent with a two-electron reductive cleavage of the halogen which can therefore be represented by the reaction sequence (reactions 1 el-8) given above. The peak potentials were observed to shift linearly with the logarithm of sweep rate in the range 0.1 - 100 V s - I (Table I). The values of the corresponding slopes dE / b log e as well as those of the peak width A E , (Table I) showf7 that the electrode reaction is kinetically controlled by the initial charge transfer with the exception of iodobenzene. In the latter case the observed slope is compatible either with a rate-determining charge transfer with a transfer coefficient significantly larger than 0.5 (-0.60) or with a mixed kinetic control by charge transfer and by a follow-up chemical reaction.x In all the other cases the values of the transfer coefficient do not differ significantly from 0.5. That the charge transfer is rate controlling in the case of, e.g., bromobenzene and 2-bromopyridine is confirmed by the negative shift in peak potential when passing from Et,N+ to Bu4N+ (0.1 M ) as supporting cation (1 20 mV for chlorobenzene and 1 I O mV for 2-bromopyridine) resulting from a change in the double-layer characteristics.l* With iodobenzene, the shift is only 40 mV, indicating that mixed charge transfer chemical reaction control is more likely than pure charge transfer control with a large a. The complete ir-
0.033 0.33 0.33 3.3 3.3 33
2.8 I .2 5.2 4.4 14.4 1.65 4. I 1.50
2.8 2.8
aV 0.5
2.9
2.8 2.9 2.9 3.0
4.1 4.4
4.5 4.3 4.5 4.5
4.3
reversibility of the CV wave observed for all compounds prevents the determination of the standard potential and of the standard rate constant. The measurement of the peak potential as a function of sweep rates provides, however, the apparent forward rate constant of electron transfer kp'4',17the values of which are given in Table I. Homogeneous Redox Catalysis. In most cases the forward rate constant of electron transfer, k 1, was determined for the experimental ratio (iJc/(iP)o (peak currents of the catalyst wave in the presence and absence of substrate, respectively).' For very high catalytic efficiencies, this procedure cannot be used safely since (i& is too close to its limiting value. In such conditions k1 was derived from the peak potential differences dE, between the peak potentials for the catalyst wave in presence and in absence of substrate." The results are shown in Tables 11-VII, which give, for each halo compound with the exception of iodobenzene, the list of the catalysts with their standard potential ( E o p ~ typical ), values of the excess factor (y: ratio of the concentrations of substrate and catalyst), and sweep rate c with the resulting values of either (i&/(i,+ or dE,. The k I values figured in the last column were obtained as an average corresponding to a larger number of y and c values than shown in the tables. Figures 3-8 represent the final results as plots of the logarithm of the homogeneous forward rate constant, k 1, vs. the standard potential of the catalyst couple E O ~ Q . In the case of chlorobenzene, the results obtained previously using classical polarography9 were used together with those of Table 111 for the diagram in Figure 4 . Such detailed and systematic analyses were not carried out for iodobenzene since it was not possible in this case to ascertain the nature of the rate-determining steps. An important qualitative observation was, however, made: for a given distance between the catalyst and the substrate waves, catalytic efficiency is always markedly lower for iodobenzene than for either chloro- or bromobenzene.
Discussion Activation vs. Diffusion Control of the Homogeneous Electron Transfer. The log k I-EO~Qplots exhibit two distinct regions. For the most positive potentials, the experimental points are located on a straight line with a slope of about - 1 / 5 8 mV per decade, whereas for the most negative potentials they are situated on a straight line having a slope of about - 1/ 1 16 mV per decade. These two straight lines are joined by a curve in the intermediate potential regions. Actually, these two straight lines are unequally extended depending upon each particular halo compound. The first one just begins to appear with bro-
Journal of the American Chemical Society ,I101:13 ,I June 20, I979
3436 'l'ahle 11'.Redox Catalysis of 2-Broniopyridine Reduction
typical values of y and I ' . V $ - I
E'PQ.
c;I til I>s1
V vs. SCE
(ip)c/(;ph~
log k 1. c1-1 L s-1
av
n a p h t h o n i t ri IC
- I .78
5 50
0.03: 0.033
I .30 3.7
1.3 I .3
9. IO-diphen) Ian t h riiccnc
- I .80
5 50 20
0.033 0.033 0.033
I .40
3.70 I .20
1.4 I .3 1.4
anthraccnc
- I .89
50 2 20 IO 50 50
0.033 0.33 0.33 3.3 3.3 33
24.2 I .65 5.42 1.34 2.53 1.20
3.0 2.9 3.1 3.0 2.9 3 .O
2 20 1 IO 5
0.33 0.33 3.3 3.3 33
4.2 19.2 2.04 6.5 I .52
4.3 4.4 4.6 4.6 4.5
4.5
20 5 50 I 5
0.33 3.3 3.3 33 33
30.8 7.6 39 1.71 4.10
5.1 5.3 5.4 5.4 5.5
5.3
20 1 5
3.3 33 33
31.6 2.5 6.8
6. I 6.0 6.2
6. I
phcna n t h ridinc
-2.00
-2.08
bcn7o[/i]quinoline
ni C I 11 4 I bcn 702 t c
bcnronitrilc
-2.12
-2.17
I 5 0.5 I I
-2.24
I
I
1.3 I .4
3.0
33 33 1000 1000 0.033
2.6 8 .O 1.19 I .43 615,= 105 niV
6. I 6.4 6. I 6.2 6. I
0.033 0.33
6Ep = 140 niV 6 E , = 110 mV
7.3 7.3
7.3
log k , . M-I L s-I
av
6.2
Table 1'. Redox Catalcsis of 3-Bromopyridine Reduction catalyst
n a p h t honi t ri l e
Y. I 0-diphenyl-
E'PQ, V
VS.
- I .78
- I .80
anthracene ;I
n t h r ;ice ne
p h c n a n t h r id i n e
bcn7o[ h]quinoline
mcthq I bcn70,ite
-1.89
-2.00
-2.12
-2.17
typical values of y and c , V s-l
SCE
(iPLl(ip)o
2 IO 20
0.033 0.033 0.033
I .3 2.2 2.6
10
20 50
0.033 0.033 0.33
2.0 2.6 1.7
1.5 I .4 1.4 1.4 1.5
I 2 2 I0 20 2 10 I 10 IO
0.033 0.033 0.33 0.33 3.3 0.33 0.33 3.3 3.3 33
2.2 3. I 1.6 2.8 1.80 4.35 10.5 1.65 5.2 2.4
2.8 2.8 2.9 2.8 3 .O 4.4 4.1 4.3 4.4 4.5
10 2 20
3.3 33 33
19.2 4.0 19.0
6. I 6.2 6.4
6.2
2.8 19.2
6.4 6.2 6.3
6.3
1 IO 0.5
mobenzene and is somewhat more apparent with chlorobenzcne: with the bromopyridines the first straight line is dominant. This trend is even more pronounced with the chloropyridines, the second line being almost absent. The comparison
33 33 I000
1.1
I .4 I .4
I .4
2.9
4.3
between the six diagrams makes, however, clearer the existence of these two modes of variation of k i with E O ~ Qnoted previously in the case of chlorobenzene.' They correspond, as discussed above, to ( I ) activation control of the electron
Saceant et ai.
/ Redox Catalysis in Halobenzenes and -pyridines
Table VI. Redox Catalysis of 2-Chloropyridine Reduction cat;dyst
3431
__
..
EOPQ, V vs. SCE
typical values of y and c, V s - -~ -I -..
.
-~
.-
log k I , .
(I
.-
/ ( i,)o
M-I L
av
SKI
9. I 0-diphenylanthracene
-1.80
20 50
0.033 0.033
1.06 I .20
0.0 0.0
anthracene
- I .89
10 50 50
0.033 0.033 0.33
I .45 2.90 I .29
1.1
I .0 1.2 3. I 3.0 3.2 3.0 3.2 5.4 5.2 5.3 5.2 5.4 5.7
5.5
6.0
6.0
phenanthridine
bcn7o[h]quinoline
methyl benzoate
chrysene
-2.00
-2.12
2 20 IO 50 50
0.33 0.33 3.3 3.3 33
2 20 2 20
3.3 3.3 33 33
I .78 5.24 1.6 2.75 I .30 4.4 22.1 2.2 7.6
2 2
3.3 33
4.3 2.9
-2.17 -2.20
0.033
1
Table VII. Redox Catalysis of 3-Chloropyridine Reduction catalyst 9. I O-diphenylanthracene
E'PQ, V vs. SCE
.-. _.
typical values of Y and c. V s-I
ALP = I00 niV
____-
0.0
1.1
3. I
5.2
. _ _ . _ . . _ I
- .. ....
log ki, .
(Ip)
