Hydrogen-bonding and protonation effects in electrochemistry of quinones in aprotic solvents

Article abstract of DOI:10.1021/ja970028j

Hydrogen-bonding and protonation are fundamental factors controlling potentials and mechanisms in the reduction of quinones. These are studied systematically in benzonitrile, acetonitrile, and dimethylsulfoxide solutions by cyclic voltammetry of a series of quinones of increasing basicity (chloranil to duroquinone in the presence of hydroxylic additives of increasing hydrogen-bonding power (tert-butyl alcohol to hexafluoro-2- propanol) or acidity (trifluoroacetic acid). Electrochemical effects are demonstrated over the complete interaction range, from hydrogen bonding of reduced dianions to protonation of unreduced quinones. With increasing concentrations of additives, three clearly different types of electrochemical behavior are observed for weakly (I), moderately (II) and strongly (III) interacting quinone-additive pairs, as follows: (I) Two well-separated reduction waves, corresponding to formation of quinone mono- and dianions, shift positively, with no loss of reversibility. The second wave is smaller, shifts more strongly, and finally merges with the first. The relative heights of the waves remain constant. (II) The positive shift is accompanied by increasing height of the first peak and broadening and irreversibility of the second wave. (III) One or even two, more positively shifted, new prior waves appear, together with a new anodic wave. In interpreting these phenomena, the role of hydrogen-bonding is clearly distinguished from protonation on the basis of pK(a) values of relevant species, effects of solvent variation, magnitude of potential shifts, and the onset of irreversibility. Type I behavior is attributed to stabilization by hydrogen-bonding of mono- and dianion reduction products; the number of bonds per quinone ion and bonding equilibrium constants are estimated from the shifts in peak potentials with additive concentration. These results are supported by simulating the experimental cyclic voltammograms using these parameters. Type III behavior is assigned to initial hydrogen-bonding or protonation of the quinones. Type II is attributed to a reduction mechanism involving disproportionation of primary radicals, assisted by hydrogenbonding or protonation of the dianion.

Full text of DOI:10.1021/ja970028j

6384
J. Am. Chem. Soc. 1997, 119, 6384-6391
Hydrogen-Bonding and Protonation Effects in Electrochemistry
of Quinones in Aprotic Solvents
Neeraj Gupta and Henry Linschitz*
Contribution from the Department of Chemistry, Brandeis UniVersity,
Waltham, Massachusetts 02254-9110
ReceiVed January 3, 1997. ReVised Manuscript ReceiVed May 2, 1997X
Abstract: Hydrogen-bonding and protonation are fundamental factors controlling potentials and mechanisms in the
reduction of quinones. These are studied systematically in benzonitrile, acetonitrile, and dimethylsulfoxide solutions
by cyclic voltammetry of a series of quinones of increasing basicity (chloranil to duroquinone), in the presence of
hydroxylic additives of increasing hydrogen-bonding power (tert-butyl alcohol to hexafluoro-2-propanol) or acidity
(trifluoroacetic acid). Electrochemical effects are demonstrated over the complete interaction range, from hydrogen
bonding of reduced dianions to protonation of unreduced quinones. With increasing concentrations of additives,
three clearly different types of electrochemical behavior are observed for weakly (I), moderately (II), and strongly
(III) interacting quinone-additive pairs, as follows: (I) Two well-separated reduction waves, corresponding to formation
of quinone mono- and dianions, shift positively, with no loss of reversibility. The second wave is smaller, shifts
more strongly, and finally merges with the first. The relative heights of the waves remain constant. (II) The positive
shift is accompanied by increasing height of the first peak and broadening and irreversibility of the second wave.
(III) One or even two, more positively shifted, new prior waves appear, together with a new anodic wave. In
interpreting these phenomena, the role of hydrogen-bonding is clearly distinguished from protonation on the basis of
pKa values of relevant species, effects of solvent variation, magnitude of potential shifts, and the onset of irreversibility.
Type I behavior is attributed to stabilization by hydrogen-bonding of mono- and dianion reduction products; the
number of bonds per quinone ion and bonding equilibrium constants are estimated from the shifts in peak potentials
with additive concentration. These results are supported by simulating the experimental cyclic voltammograms using
these parameters. Type III behavior is assigned to initial hydrogen-bonding or protonation of the quinones. Type
II is attributed to a reduction mechanism involving disproportionation of primary radicals, assisted by hydrogen-
bonding or protonation of the dianion.
Introduction
pathways of the various species which appear in these quinone-
hydroquinone systems, both in Vitro and in mitochondrial or
photosynthetic membranes. Among these factors, we are
particularly concerned in this paper with hydrogen-bonding,
which has been implicated in the biological function of the
quinone systems11 and on which surprisingly little systematic
research has been done.1,2
In well-buffered aqueous media (including mixtures with
ethanol, dioxane, etc.), quinone-hydroquinone couples provide
familiar, reversible two-electron redox systems in which po-
tentiometric or polarographic potentials vary with pH in a
straightforward Nernstian manner.3 This behavior is conve-
niently summarized in E-pH (Pourbaix) diagrams, showing
regions of existence of various redox and protonated species
and their respective pKa values.1,14 In the strongly basic region
(pH > pKa of QH-) only the unprotonated forms, Q, Q-•, and
Q2- appear.1,14
Quinone-hydroquinone couples have been studied over many
decades as the prototypical examples of organic redox systems.1-3
In this regard, their electrochemical behavior, involving the
kinetics and equilibria of coupled electron and proton transfer
reactions, has given much information concerning the effect of
molecular structure1,2,4 and environment5-8 on these basic
processes. In addition to their intrinsic chemical interest, these
studies are particularly important in view of the key biological
functions of quinone-based couples as electron-proton transfer
agents in oxidative phosphorylation or photosynthesis.9-13
Thus, it is necessary to understand, as far as possible, the
environmental factors which regulate the potentials and reaction
X Abstract published in AdVance ACS Abstracts, June 15, 1997.
(1) Chambers, J. Q. In The Chemistry of the Quinonoid Compounds;
Patai, S., Rappoport, Z., Eds.; Wiley: New York, 1988; Vol. II, Chapter
12, pp 719-757; 1974, Vol. I, Chapter 14, pp 737-791.
(2) (a) Peover, M. E. In Electroanalytical Chemistry; Bard, A. J., Ed.;
Dekker: New York, 1967; pp 1-51. (b) Peover, M. E. J. Chem. Soc. 1962,
4540-4549.
(3) Kolthoff, I. M.; Lingane, J. J. Polarography, 2nd ed.; Interscience:
New York, 1952; Vol. I, Chapter XIV and XV; Vol. II, Chapter XL.
(4) Zuman, P. Substituent Effects on Organic Polarography; Plenum
Press: New York, 1967; Chapter VIII.
(5) Russel, C.; Jaenicke, J. J. Electroanal. Chem. 1986, 199, 139-151.
(6) Laviron, E. J. Electroanal. Chem. 1986, 208, 357-372.
(7) Wilford, J. H.; Archer, M. D. J. Electroanal. Chem. 1985, 190, 271-
277.
(8) Jaworski, J. S.; Lesniewska, E.; Kolinowski, M. K. J. Electroanal.
Chem. 1979, 105, 329.
Similarly, in dry, neutral, aprotic media, quinones show
typically two cathodic polarographic waves, E1 and E2, separated
by about 0.7 V, which correspond to the formation of Q-• and
Q
2-, respectively.1 In these reductions, the first step is generally
reversible and the second is at least quasi-reversible at customary
scan speeds. The potentials of these reductions depend on the
polarity of the solvent,2,7,8 the nature of the supporting
(11) Okamura, M. Y.; Feher, G. Annu. ReV. Biochem. 1992, 61, 861-
896.
(12) Crofts, A. R.; Wraight, C. A. Biochem. Biophys. Acta 1983, 726,
149-185.
(13) Rich, P. R. Biochem. Biophys. Acta 1984, 768, 53-79.
(14) Bailey, S. I.; Ritchie, I. M. Electrochim. Acta 1985, 30, 3-12; J.
Chem. Soc., Perkin Trans. 2, 1983, 645-562.
(9) Swallow, A. J. In Function of Quinones in Energy ConserVing
Systems; Trumpower, B. L., Ed.; Academic Press: New York, 1982; Chapter
3, p 66.
(10) Trumpower, B. L. J. Biol. Chem. 1990, 265, 11409-11412.
S0002-7863(97)00028-0 CCC: $14.00 © 1997 American Chemical Society

Electrochemistry of Quinones in Aprotic SolVents
J. Am. Chem. Soc., Vol. 119, No. 27, 1997 6385
electrolyte1,5,15-17 and the presence of acidic additives, reflecting
respectively nonspecific solvation energies, ion-pairing, and
protonation equilibria. With regard to the last factor, numerous
studies of the effect of acid strength (extending into very strongly
acidic media), concentration, and quinone basicity1,2,6,18-23 have
been interpreted on the basis of the full 3 × 3 array6,24 of redox
and protonated species and reaction pathways in the Q-QH2
systems. For our purpose, we are concerned now mainly with
the effects of weak proton donors on the above two simple
reduction waves, E1 and E2. Two situations can be clearly
distinguished, depending on whether the donor can protonate
Q-• or only Q2-. As typified, for example, by reduction of
anthraquinone (AQ) in dimethylformamide (DMF),2,25,26 addi-
tion of increasing concentrations of a weak acid (phenol),
reacting only with Q2-, leads to a progressive positive displace-
ment of E2, due to protonation of the dianion,25 until this wave
merges with E1. A stronger donor (benzoic acid) causes the
first wave, E1, to increase at the expense of E2. This is generally
attributed to the well-known sequence, fast protonation of Q-•,
and reduction of the more easily reduced QH• radical at the
same potential.25,27 The polarographic effect of adding increas-
ing concentrations of water or acid to DMF or dimethylsulfoxide
(DMSO) quinone solutions has been interpreted in the same
way.25,28,29
However, quite similar polarograms are obtained in quinone
systems containing hydroxylic additives in which proton-coupled
reductions do not occur. For example, reversible waves of AQ
in DMF shift smoothly toward less-negative values upon
addition of ethanol (but not acetone).29a Nevertheless, at the
same time, the presence of unprotonated mono- and dianion
reduction products is directly demonstrated by their optical
spectra which also exhibit characteristic hydrogen-bonding
effects.29b Moreover, addition of water to DMF doubles the
height of the reversible E1 wave of AQ, at the expense of E2,
but ESR measurements establish the stability of the unprotonated
anion radical in the mixtures.30 The increase in wave height in
this case is explained by dismutation of AQ-• regenerating AQ,
which is favored by hydrogen-bonding of the AQ2- dianion.
Smooth shifts in quasi-reversible waves of p-benzoquinone,
caused by small increments of water in aprotic solvents, are
also found by Wilford and Archer, who state that these effects
“are not consistent simply with protonation of Q-• and Q2-”.7
The role of hydrogen-bonding in modifying the redox behavior
of quinones has been particularly emphasized by Peover.2 He
points out that in a series of methyl- and chloro-substituted
p-benzoquinones, the one-electron reduction potentials in strongly
alkaline aqueous ethanol are much more positive than in aprotic
solvents and that this discrepancy increases with the basicity
of Q-•.2,31 Accordingly, he has attributed this difference to
stabilization of anions by hydrogen-bonding in the aqueous
medium and has even estimated bonding equilibrium constants
as high as 1010, to account for the observed effects.2a In indirect
support of this is the positive shift in reduction potential of 9,10-
anthraquinones associated with R-OH substitution32,33 and the
interpretation of solvent-dependent ESR spectra of semiquinone
radical ions in terms of hydrogen-bonding.34,35
It is thus clear that neither a positive shift in potential nor a
change in wave height necessarily indicates involvement of
protonation by hydroxylic agents. Both these effects may result
from stabilization by hydrogen-bonding of the anionic products,
either of primary reduction or secondary disproportionation.
Work on this problem in the quinone system to date has been
rather scattered, involving few systems and over limited
concentration ranges. Indeed, references to the role of hydrogen-
bonding in quinone redox chemistry are remarkably sparse,36
in view of the enormous literature on the subject. To further
clarify the situation, we present here cyclic voltammetric studies
of quinone reduction in aprotic solvents, in which we systemati-
cally vary the acidity and hydrogen-bonding power of added
hydroxylic reagents (increasing from tert-butyl alcohol to
hexafluoro-2-propanol (HFIP)) as well as the basicity of
substituted p-benzoquinones (from chloranil to duroquinone).
The onset of irreversibility and the effect of solvent and an
unambiguous protonating agent (trifluoroacetic acid) are used
to help distinguish protonation from hydrogen-bonding. We
show that hydrogen-bonding interactions produce large positive
shifts in reduction potentials, particularly of the second reduction
step. From these shifts, we estimate the number of hydrogen
bonds per quinone anion and dianion in these various pairs and
their respective equilibrium constants.
Experimental Section
Materials. 1,4-Benzoquinones (BQs) were of the best available
grade (>97%) from Sigma, Aldrich, or Eastman and substituted as
follows: tetrachloro (TCBQ, chloranil); 2,5-dichloro (DCBQ); 2,5-
diphenyl (DPBQ); 2,5-dimethoxy (DMOBQ); 2,5-dimethyl (DMBQ);
tetramethyl (TMBQ, duroquinone). Benzonitrile (PhCN) (99.9%,
Sigma-Aldrich), acetonitrile (CH3CN) (HPLC reagent, J. T. Baker,
Inc.), and DMSO (99.9%, Sigma-Aldrich) were stored over molecular
sieves (4A, 8-12 mesh, Aldrich) preheated to 400 °C for 12 h, prior
to use. 1,1,1,3,3,3-Hexafluoro-2-propanol (99+%) (HFIP), 2,2,2-
trifluoroethanol (99.5+%) (TFE), trifluoroacetic acid (99+%) (TFA),
2-methyl-2-propanol (tert-butyl alcohol), and tetrabutylammonium
hexafluorophosphate (98%) (TBAPF6) were from Aldrich. Pure
ferrocene was kindly provided by Prof. M. Rosenblum. All chemicals
were used as received without further purification unless specified.
(15) Peover, M. E.; Davis, J. D. J. Electroanal. Chem. 1963, 6, 46-53.
(16) Nagaoka, T.; Okazaki, S.; Fujinaga, T. J. Electroanal. Chem. 1982,
133, 89.
(17) Eggins, B. R. Chem. Commun. 1969, 1267-1268.
(18) Jeftic, L.; Manning, G. J. Electroanal. Chem. 1970, 26, 195-200.
(19) Hanzlik, J.; Samec, Z. Collect. Czech. Chem. Commun. 1985, 50,
2821-2826.
Procedure. Cyclic voltammetry experiments were done using a
conventional three-electrode cell, with glassy carbon working electrode,
platinum wire counter electrode, and Ag/AgCl (containing aqueous
solution of 3 M NaCl and saturated with AgCl; EG&G) reference
electrode, separated from the solution by a Vycor plug. The supporting
electrolyte was 0.1 M TBAPF6 in all experiments. All potentials were
measured using ferrocene as internal reference and converted to aqueous
(20) Bessard, J.; Cauquis, G.; Serve, D. Electrochim. Acta 1980, 25,
1187-1197.
(21) Marcus, M. F.; Hawley, M. D. Biochim. Biophys. Acta 1970, 222,
163-173.
(22) Bauscher, M.; Mantele, W. J. Phys. Chem. 1992, 96, 11101-11108.
(23) Eggins, B. R.; Chambers, J. Q. J. Electrochem. Soc. 1970, 117,
186-192.
+
+
(Fc /Fc)
(Fc /Fc)
SCE taking E1/2
to be 0.56 V vs SCE. This value of E1/2
(C /C
60 60
was determined with respect to known C60 (E1/2
-•) ) -0.45 V vs
(24) Jacq, J. Electrochim. Acta 1967, 12, 1345-1361.
(25) Given, P. H.; Peover, M. E. J. Chem. Soc. 1960, 385-393.
(26) Wawzonek, S.; Berkey, R.; Blaha, E. W.; Runner, M. E. J.
Electrochem. Soc. 1956, 103, 456-459.
(31) Peover, M. E.; Davies, J. D. Trans. Farad. Soc. 1964, 60, 476-
478.
(27) Aten, A. C.; Buthker, C.; Hoijtink, G. J. Trans. Farad. Soc. 1959,
55, 324-330.
(32) Ashnagar, A,; Bruce, J. M.; Dutton, P. L.; Prince, R. C. Biochim.
Biophys. Acta 1984, 801, 351-359.
(28) Kolthoff, I. M.; Reddy, T. B. J. Electrochem. Soc. 1961, 108, 980-
(33) Jones, R.; Spotswood, T. M. Austr. J. Chem. 1962, 15, 492502.
(34) Gendell, J.; Freed, J. H.; Fraenkel, G. K. J. Chem. Phys. 1962, 37,
2832-2841.
985.
(29) (a) Hayano, S.; Fujihira, M. Bull. Chem. Soc. Jpn. 1971, 44, 2051-
2055. (b) Fujihira, M.; Hayano, S. Bull. Chem. Soc. Jpn. 1972, 45, 644-
645.
(35) Stone, E. W.; Maki, A. H. J. Am. Chem. Soc. 1965, 87, 454-458.
(36) Ge, Y.; Lilienthal, R.; Smith, D. K. J. Am. Chem. Soc. 1996, 118,
3976-3977.
(30) Umemoto, K. Bull. Chem. Soc. Jpn. 1967, 40, 1058-1065.

6386 J. Am. Chem. Soc., Vol. 119, No. 27, 1997
Gupta and Linschitz
Table 1. pKa Values and Hydrogen-Bonding Equilibrium
Constants of Alcohols and Acids
alcohols/acids
pKa (H2O)a
pKa (DMSO)b
log Kc
tert-butyl alcohol
2-propanol
ethanol
water
TFE
HFIP
19.0
17.1
15.9
15.7b
12.4
9.3
32.2
30.3
29.8d
32e
23.6
17.85f
3.45
0.78
0.91
1.21
2.0
2.83
3.55
TFA
0.52
a Reference 39. b Reference 58. c Hydrogen-bonding equilibrium
constant toward the common acceptor N-methylpyrrolidinone in 1,1,1-
trichloroethane.39 d Reference 59 e The high pKa of water in basic polar
solvents should be kept in mind in interpreting the role of traces of
moisture in these systems. f Bordwell, F. G., private communication.
SCE)37 in the same solvent-supporting electrolyte system, since the
+
(Fc /Fc)
E1/2
in this system is not available in the literature. Voltammo-
grams were run using an EG&G-PAR Versastat potentiostat, coupled
to a PC and EG&G M270 electrochemical software to record and
analyze the data. Solutions were purged with argon to remove oxygen,
and argon was passed over the solution during the measurement.
Quinone concentrations were in the range of 2-3 mM, and hydroxylic
reagents were added incrementally to the solution using a microsyringe.
To improve precision, shifts in potentials (∆E) following each addition
were read from a large-scale screen projection of the voltammogram.
Values of E1/2 were taken at 0.029 V, positive with respect to the peak
potential. (The calculations of hydrogen-bonding constants (Table 2,
below) involve only ∆E values.) Sweeps were initiated in the cathodic
direction. All experiments were done at room temperature (25 °C).
Simulation of the cyclic voltammograms was done using the Bioana-
lytical System’s Digisim, Version 2.1 electrochemical simulation
program.38
For later reference, Table 1 gives pKa values in water and DMSO
for the six hydroxylic reagents used in this study, as well as equilibrium
constants for hydrogen-bond formation, referred to a common acceptor,
N-methylpyrrolidinone in 1,1,1-trichloroethane.39 In this series, hydrogen-
bonding power parallels acidity, although over a much smaller range
of variation.
Results and Discussion
(I) Behavior in Pure Media. All quinones studied show
two reduction peaks in nonaqueous aprotic media, corresponding
to two single-electron reductions to give mono- and dianions
(Figure 1, “a” curves). The values of E1/2 vs SCE, for the first
and second reduction steps for all of the quinones, are given in
Table 2 and are in good agreement with the literature.2b,40 The
ratio of the cathodic peak current to the anodic peak current
for the first redox process is close to unity for all of the quinones,
throughout the sweep rate range from 20 to 500 mV/s.
However, at slower sweep rates (20 mV/s), the ratio of peak
currents for the second redox process was found to be 0.8-
0.9, indicating slight irreversibility. Cathodic to anodic peak
separations were typically 70-80 mV for the first reduction
wave and about 100 mV for the second wave.
Figure 1. Cyclic voltammograms of (a) TCBQ, (b) DCBQ, and (c)
DMOBQ at different concentrations of ethanol in PhCN. Sweep rate
) 100 mV/s.
experimental curves.41,42
Q + Q2- h [QQ]2-
(I)
It is important to note that while this suggested complexation
reaction accounts very well for the difference in peak heights,
it turns out to have no effect on the reduction potentials of the
(41) This reaction was suggested by Dr. Stephen W. Feldberg. Other
possible mechanisms were exhaustively tested by simulations, including a
change in diffusion coefficient of Q-• due to solvation or hydrogen-bonding
and dimerization of Q-• (see refs 25, 43, and 44). Neither of these gave
results which agree with the observations.
It is interesting that the height of the second peak is less than
that of the first in the voltammograms of all of the observed
quinones. The effect becomes more marked in passing from
TCBQ to TMBQ (Figure 1). Among several mechanisms
considered in attempted simulations, the assumption of a
complexation reaction between a quinone and its dianion
(reaction I) gave excellent agreement with this feature of the
(42) Cyclic voltammograms of quinones in which the second wave is
lower than the first have been observed by others, but this behavior seems
to have excited little comment.17,21 In one case, Marcus and Hawley21 show
explicitly that the first wave corresponds to one-electron reduction, so that
neither the disproportionation55 nor protonation followed by reduction
(ECE)27 of semiquinone can explain the decrease in the second wave. We
are of course aware that the assumption of reaction I poses various
difficulties, such as the apparent stability of the complex toward thermo-
dynamically favored dissociation to 2Q-•, etc. Nevertheless, as long as this,
or any other, mechanism does not modify the shifts in peak potentials, the
conclusions drawn here regarding hydrogen-bonding effects are still valid.
Further studies of this interesting problem are now in progress; results will
be published elsewhere.
(37) Dubois, D.; Moninot, G.; Kutner, W.; Jones, M. T.; Kadish, K. M.
J. Phys. Chem. 1992, 96, 7137-7145.
(38) Bioanalytical Systems Inc., 2701 Kent Ave., West Lafayette, IN
47906.
(39) Abraham, M. H.; Duce, P. P.; Prior, P. P.; Barratt, D. G.; Morris,
J. J.; Taylor, P. J. J. Chem. Soc., Perkin Trans. 2 1989, 1355-1375.
(40) Sasaki, K.; Kashimura, T.; Ohura, M.; Oshaki, Y.; Ohta, N. J.
Electrochem. Soc. 1990, 137, 2437-2443.

Electrochemistry of Quinones in Aprotic SolVents
J. Am. Chem. Soc., Vol. 119, No. 27, 1997 6387
Table 2. Electrochemical Parameters: Reduction of Quinones in Presence of Hydrogen-Bonding Agents
(1) a
(2) a
(1) b
(2) b
quinones
TCBQ
solvent
-E1/2
-E1/2
reagent
n
Keq
m - n
m
Keq
PhCN
PhCN
PhCN
CH3CN
CH3CN
CH3CN
PhCN
PhCN
PhCN
CH3CN
CH3CN
CH3CN
DMSO
DMSO
DMSO
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
PhCN
0.05
0.04
0.20
0.20
0.18
0.83
0.81
0.95
0.93
0.98
-
-
c
c
-
c
c
-
0.75
1.0
-
-
c
c
-
c
c
-
15
30
-
c
c
-
c
c
-
50
-
30
-
50
-
370
10
10
-
-
-
-
EtOH
TFE
-
EtOH
TFE
-
EtOH
TFE
-
EtOH
TFE
-
EtOH
TFE
-
EtOH
-
EtOH
-
EtOH
-
2.9
3.5
-
3.1
3.4
-
3.0
4.6
-
3.0
3.7
-
3.1
3.0
-
3.9
-
3.3
-
3.7
-
2.9
3.5
-
3.1
3.4
-
3.75
5.6
-
2.6 × 104
3 × 1010
-
6.3 ×103
5 × 107
-
DCBQ
2.5 × 106
1.8 × 1014
-
c
3.0
1.2 × 104
2 × 1011
-
∼0.5
∼4.2
-
-
c
c
-
1.2
-
1.4
-
1.1
-
2.03
0.87
0.76
-
3.1
3.0
-
5.2
-
4.7
-
4.8
-
6.5
3.5
2.6
-
1.9 × 103
8 × 106
-
BQ
0.53
0.7
1.21
1.38
1.23
1.42
1.0 × 109
-
DMBQ
DPBQ
DMOBQ
∼108
0.52
0.74
-
2.0 × 108
-
EtOH
2-PrOH
t-BuOH
-
4.5
2.6
1.8
-
2.2 × 1011
2.0 × 106
3.0 × 104
-
TMBQ
0.9
1.56
EtOH
1.3
85
4.2
5.5
8.5 × 109
a Potentials vs aqueous SCE b Values of equilibrium constants are estimated by taking n and m - n values (0.5 to obtain best fit to data. Units
(1)
(2)
(1)
of Keq ) M-n and Keq ) M-m. Accuracy (30%.60 c Potential shifts are very small. We take n and Keq to be negligible in these cases.
two waves and therefore does not modify the hydrogen-bonding
equilibrium constants derived therefrom (see below). Thus, in
of BQ and TMBQ are 4.0 and 5.1, respectively, in aqueous
medium,9,46 which are so much lower than the pKa of ethanol
(see Table 1) that we may safely assume a correspondingly large
difference in benzonitrile. Water-based pKa values are used
here, since related data in aprotic solvents are not available for
the reduced quinones. However, we note that the charge on
unprotonated quinone reduction products is more delocalized
than that on the alcohol anion. Thus, in passing from water to
benzonitrile or DMSO, the alcohols will become even weaker
acids (Table 1) relative to the protonated quinone monoanions,
since the loss of hydration energy will be greater for alcohol
anions.47 This applies also to the dianions of BQ and the other
more basic quinones (Figure 1). Changes in energies of
quinones on reduction, caused by dispersion forces, should be
small compared with energies of hydrogen-bonding and pro-
tonation.48 No complications due to ion-pairing appear in this
work using TBAPF6 as supporting electrolyte.15
(2)
simulations, a plot of E1/2 calculated without reaction I vs
(2)
E′1/2 with reaction I was linear with slope 1.0 for all values
(2) 42
of E1/2
.
A small reduction peak was observed between the
two main reductions for many quinones (DMBQ, BQ, and
DMOBQ) and has been seen also in reduction of analogs of
ubiquinone.22 This small peak disappeared at high sweep rate
and may be attributed to the reduction of a dimer formed by
the reaction between two semiquinone molecules.25,43,44
(II) Effect of Weak Hydrogen-Bonding Agents: Ethanol,
2-Propanol, and tert-Butyl Alcohol. Addition of ethanol to
benzonitrile solutions of all quinones studied results in positive
shifts of both reduction steps, which increase smoothly with
ethanol concentration and with no further loss of reversibility
(up to 2.5 M ethanol).45 Representative cyclic voltammograms
are shown in Figure 1. For given ethanol concentration, the
shift is much larger for the second wave and, for both waves,
increases with quinone basicity, as in passing from TCBQ to
DMOBQ or TMBQ (Figure 1). These marked effects are
apparent even at ethanol concentrations less than 0.05 M, in an
already somewhat polar medium (benzonitrile ꢀ ) 25.2). Thus,
they cannot be due to changes in bulk polarity, but must
evidently be ascribed to quite specific ethanol-quinone interac-
tions (i.e., hydrogen-bonding) which increase with both the
degree of reduction (i.e., anionic charge) and the basicity of
the quinone. Earlier observations of similar shifts in the
potential of different quinones caused by the addition of water
to aprotic solvents have been ascribed to fast protonation of
the dianion.28,29 We may rule out the possibility of protonation
of quinone monoanions simply on the basis of the unfavorable
pKa values of these semiquinones. The pKa of the semiquinones
Similarly, the pKa of BQ dianion is 11-121,14 and, hence,
should not favor protonation by ethanol in PhCN. While, pKa
values for the dianions of more basic TMBQ and DMOBQ are
not available in the literature, they are presumed to be higher
than that for BQ.49 To check the possibility of protonation of
these dianions by ethanol, we compare the effects observed with
the weaker acids 2-propanol and tert-butyl alcohol for which
such protonation is less likely. Qualitatively similar but smaller,
continuous positive shifts of the second peak were observed
for these alcohols even in reduction of DMOBQ. We thus rule
out the possibility of protonation of DMOBQ and TMBQ
dianions by these alcohols in PhCN, at the experimental
concentrations of Figure 1.45 Quite different, unambiguous
protonation effects, such as much larger potential shifts and
irreversible and merging waves, are caused by hydroxylic
additives more strongly acidic than ethanol and will be described
(43) Baizer, M. M. Organic Electrochemistry; Marcel Dekker: New
York, 1973; p 408.
(44) Linschitz, H.; Rennert, J.; Korn, T. M. J. Am. Chem. Soc. 1954,
76, 5839-5842.
(45) In neat ethanol, the electrochemistry of TCBQ is similar to DMOBQ
with TFE in PhCN (Figure 5b), and TMBQ behaves similarly to DMOBQ
at lower concentrations of HFIP in PhCN (Figure 7b).
(46) Patel, K. B.; Willson, R. L. J. Chem. Soc., Faraday Trans. 1 1973,
69, 814-825.
(47) Bordwell, F. G. Acc. Chem. Res. 1993, 26, 510-517.
(48) Grunwald, E.; Price, E. J. Am. Chem. Soc. 1964, 86, 4517-4525.
(49) The pKa of the 2,3-dimethoxy-5-methyl-6-isoprenyl-1,4-benzo-
quinone dianion in acetonitrile is 11.3.22

6388 J. Am. Chem. Soc., Vol. 119, No. 27, 1997
Gupta and Linschitz
later. We therefore assign the potential shifts typically illustrated
in Figure 1 to changes in fast hydrogen-bonding equilibria which
are closely coupled to reduction.
In order to treat the situation more quantitatively, we take
the hydrogen-bonding constant of the quinone to be negligible
compared to that with mono- and dianion (no change is observed
in absorption spectra of duroquinone on addition of ethanol in
PhCN). Accordingly, we write the following:
Q + e h Q-•
Q
-• + nEtOH h Q-•(EtOH)n
Q-•(EtOH)n + e h Q2-(EtOH)n
(2)
Figure 2. Change in -E1/2 for different quinones vs log[EtOH] in
Q2-(EtOH)n + (m - n)EtOH h Q2-(EtOH)m
PhCN.
Peover and Davis15 have noted the dependence of quinone
reduction potentials on the concentration and properties of
cations and have interpreted these reactions in terms of cation-
anion association equilibria.50-52 By analogy with their treat-
ment, we write for the first reduction step
RT
F
E1/2 ) E°1/2
+
ln(1 + Keq(1)[EtOH]n)
(1)
(
)
where E°1/2 is the half-wave potential in the absence of ethanol.
If Keq(1)[EtOH]n . 1, a plot of E1/2 vs log[EtOH] should give
a straight line with slope 2.3nRT/F, from which the value of n
can be estimated. This procedure can be reasonably applied
only to the more strongly bonded basic quinones where we find
larger shifts in E1/2(1). Values of n thus obtained are listed in
Table 2 and are less than 1 for quinones of low basicity and
close to 2 for the more basic quinones.
Figure 3. Simulated curves of Figure 1a using the parameters given
in Table 2. Other parameters used for simulations are ke (heterogeneous
rate constant) ) 1 cm s-1 for both redox processes, K[QQ] (equilibrium
constant for reaction 1 ) 105 M-1, and kf (rate constant for the formation
of [QQ]2-) ) 105 s-1 M-1. A wide range of values of K[QQ] and kf
produce good fits to the data provided reaction 1 remains to the right
and reversible.
The equilibrium constant for the first reduction step can be
estimated by rearranging eq 1 to obtain
exp(f∆E1/2) ) 1 + Keq(1)[EtOH]n
(2)
for the monoanion, the value of Keq(2) increases from TCBQ to
TMBQ in accord with the basicity of these quinones.
(1)
where f ) F/RT and ∆E1/2 ) E1/2 - E°1/2. Values of Keq
,
Figure 3 shows the results of the simulations performed for
TCBQ using the parameters given in Table 2 and assuming a
estimated for different quinones, are listed in Table 2. For
binding of ethanol, Keq is highest for DMOBQ, lowest for
(1)
complexation between Q and Q2- 53
Similar shifts in potential
.
DCBQ, and negligible for TCBQ, consistent with their respec-
tive basicities.
Using the same analogy between cation association and
hydrogen-bonding equilibrium for the second reduction step,52
we have
were also obtained when complexation was ignored. As noted
earlier, since the complexation reaction affects only the peak
height of the second wave and not the peak potential, the values
of Keq, n, and m determined on the basis of shifts in potential
apply with or without the complexation reaction. The good
agreement between the experimental (Figure 1a) and the
simulated curves (Figure 3) convincingly supports the hydrogen-
bonding model and above treatment of data.
1 + Keq(2)[EtOH]m
1 + Keq(1)[EtOH]n
exp(f∆E1/2) )
(3)
We note however, that n, m, and Keq values for DMOBQ are
higher than those for the presumably more basic TMBQ. This
could be due to greater steric hindrance toward bonding for
TMBQ. This aspect is further examined in Figure 4, which
shows the variation of E1/2 for the second reduction step of
DMOBQ with log[alcohol] in PhCN for the three weakly acidic
alcohols. It is apparent from the figure that the slope and
therefore the value of m decrease with increasing size of the
alcohol molecules, from 6.5 for ethanol to 3.5 and 2.6 for
(2)
where m and Keq are the number of molecules of ethanol
(2)
hydrogen-bonded to Q2- and Keq is the corresponding equi-
librium constant, respectively. For strong hydrogen-bonding,
we may neglect 1 in the denominator and numerator and obtain
the value of m - n by plotting E1/2 for the second step vs the
log[EtOH], as shown in Figure 2. Taking n from the first
reduction step, values of m are estimated for all of the quinones
and are listed in Table 2. Having established m, n, and Keq
(1)
,
(2)
eq 3 then permits evaluation of Keq (Table 2). As indicated
(53) Simulations of Figure 3 are based on the square scheme of redox
and chemical reactions (i.e., Q + e h Q-•, Q-• + e h Q2-, Q-• + nX h
QXn-•, Q2- + mX h QXm2-, and QXn-• + e + (m - n)X h QXm2-). For
chloranil, only the second reduction need be considered. For more basic
(50) Galus, G. Fundamentals of Electrochemical Analysis; Ellis Har-
wood: Chichester, 1976; Chapter 14.
(51) Chauhan, B. G.; Fawcett, W. R.; Lasia, A. A. J. Phys. Chem. 1977,
81, 1476-1481.
(52) Fawcett, W. R.; Opallo, M.; Fedurco, M.; Lee, J. W. J. Am. Chem.
Soc. 1993, 115, 196-200.
(1)
(2)
quinones, in which both E1/2 and E1/2 shift, two square schemes are
required. The values of E1/2 and the equilibrium constants for the
homogenous reactions were taken from Table 2.

Electrochemistry of Quinones in Aprotic SolVents
J. Am. Chem. Soc., Vol. 119, No. 27, 1997 6389
(2)
Figure 6. Dependence of -E1/2(2) on log[alcohol] for TCBQ with two
alcohols in acetonitrile.
Figure 4. Plot of -E1/2 for DMOBQ vs log[alcohol] for different
alcohols in PhCN.
(2)
Figure 7. Plot of -E1/2 for DCBQ vs log[TFE] in three solvents.
situation probably also applies to the dianions. Thus, the shifts
in E1/2 for both reduction steps are attributed primarily to
hydrogen-bonding. Figure 6 shows the changes in E1/2 of the
second peak of TCBQ, caused by the addition of TFE and
ethanol in acetonitrile. The slopes of the lines for the two
reagents are very similar, but the displacement in E1/2 at a given
additive concentration is much larger for TFE, indicating a large
(2)
difference in Keq for similar values of m - n (Table 2).
Estimated values of m and Keq(2) for all of the quinones, as given
in Table 2, show that, in PhCN, 5-6 molecules of TFE
hydrogen-bond with DCBQ dianion compared to 4 for ethanol,
while the even less basic TCBQ dianion associates with 3 or 4
TFE molecules. In addition to these differences in the number
Figure 5. Cyclic voltammograms of (a) DCBQ and (b) TMBQ at
different concentrations of TFE in PhCN. Sweep rate ) 100 mV/s.
(2)
of associated molecules, Keq with TFE is 106 to 108 times
2-propanol and tert-butyl alcohol respectively (Table 2). The
reduction in m from ∼6 for ethanol to about ∼3 for the hindered
alcohols suggests a steric effect, but also agrees with a possibly
sterically related influence on hydrogen-bonding power of these
alcohols, as measured with respect to a common acceptor (Table
1). In either case, the assignment of the shift in potential to a
hydrogen-bonded interaction is supported.
(III) Effect of Stronger Hydrogen-Bonding Agents. A.
Trifluoroethanol. Addition of TFE to quinone solutions in
benzonitrile leads to two very different types of redox behavior
for different quinones. For quinones with low basicity, TCBQ
and DCBQ, the effect as shown in Figure 5a is similar to that
of ethanol. We again observe continuous positive shifts for both
reduction waves, but this now occurs at much lower concentra-
tions for TFE. Again the pKa values for TCBQ and DCBQ
semiquinones, which should be lower than that of BQ,9 do not
favor protonation of their monoanions by TFE (see Table 1),
and judging from the voltammograms of Figure 5a, the same
higher than that with ethanol for TCBQ and DCBQ in PhCN.
(i) Solvent Effects. These interpretations are further sup-
ported by the effects of solvent variation, which are clearly
displayed in the second reduction step of DCBQ. Figure 7
(2)
shows that the effect of TFE on E1/2 is largest in PhCN,
intermediate in CH3CN, and smallest in DMSO, as indicated
by both the slopes of the lines and their relative displacements.
The values of m for TFE-binding (Table 2) decrease from 5.6
in PhCN and 4.2 in CH3CN to 3.0 in DMSO. Correspondingly,
(2)
Keq (M-m), decreases from 1.8 × 1014 and 2 × 1011 to 8 ×
106. This order agrees with the sequence of polarity in moving
from PhCN (ꢀ ) 25.2, ET(30) ) 41.5) and CH3CN (ꢀ ) 35.9,
ET(30) ) 45.6) to DMSO (ꢀ ) 46.5, ET(30) ) 45.1).54a
Increasing solvent polarity favors proton transfer over hydrogen-
bonding,54b which, in addition, is further suppressed in DMSO
(54) (a) Reichardt, C. SolVents and SolVent effects in Organic Chemistry;
VCH Publishers: New York, 1988; p 370. (b) Ibid.; p 105.

6390 J. Am. Chem. Soc., Vol. 119, No. 27, 1997
Gupta and Linschitz
Figure 8. Cyclic voltammograms of DCBQ in DMSO at different
concentrations of TFE. Sweep rate ) 100 mV/s.
since this basic solvent itself forms competitive hydrogen-bonds.
Thus, this sequence of solvent effects provides further compel-
ling evidence that the positiVe shift in potential is caused by
hydrogen-bonding of the dianion, not protonation. This applies
a fortiori to the more weakly basic monoanion as well. In
DMSO, addition of TFE, even up to 0.29 M, has no effect at
all on the first reduction wave of DCBQ (Figure 8), although a
marked positive shift of this wave occurs in benzonitrile (Figure
5a). Moreover, protonation of the dianion in polar DMSO is
indicated by the onset of irreversibility at high TFE concentra-
tions and appearance of a new anodic peak at TFE ) 0.29 M.
Thus, the effect of TFE on DCBQ in DMSO falls between that
of ethanol in DMSO (no irreversibility) and that of TFE with
basic quinones in benzonitrile, as shown below.
With more basic quinones, the character of the voltammogram
changes sharply, as seen by comparing Figures 5a and 5b.
Addition of TFE to benzonitrile solutions of DMBQ, DPBQ,
DMOBQ, and TMBQ results in complex electrochemical
behavior. The second wave shifts positively, becomes very
broad and irreversible, and finally disappears at higher [TFE]
(Figure 5b), quite unlike the case of DCBQ or TCBQ. At TFE
) 0.03 M (Figure 5b), the first reduction peak becomes nearly
1.5 times higher than the original peak, with a decrease in the
corresponding anodic peak. A new broad anodic peak at ∼-0.4
V is also seen at this concentration. In view of the pKa values
for TMBQ semiquinone9 and TFE (Table 1), protonation of the
monoanion seems very unlikely. Therefore, the increase in peak
current is attributed to disproportionation of the semiquinone,
facilitated by the shift in reduction potential of Q-• due to strong
hydrogen-bonding of Q2-, to give quinone and the dianion.55
The hydrogen-bonded dianion, thus formed, has a higher pKa
(compare pKa (BQ2-) ) 11-12) than TFE and is likely to be
protonated. The irreversibility of the first peak in Figure 5b in
presence of TFE is attributed to slow deprotonation of the
dianion. Our attempts to simulate these experimental observa-
tions did not give an acceptable fit to the curves because of the
complex scheme of possible chemical and redox processes and
the large number of unknown parameters, although gross
features of the voltammogram could be reproduced.
Figure 9. Cyclic voltammograms of (a) DCBQ and (b) TMBQ at
different concentrations of HFIP in PhCN. Sweep rate ) 100 mV/s.
hydrogen-bonding agent or acid (HFIP) is similar to a relatively
more basic quinone (TMBQ) in the presence of a weaker
hydrogen-bonding agent or acid (TFE). However, with HFIP,
the electrochemistry of TMBQ is sharply different from that of
DCBQ, as shown clearly in Figure 9b. An additional reduction
peak prior to the first reduction of TMBQ now appears, grows
in height, and shifts positively with increasing concentration of
HFIP. Also, the first peak becomes irreversible and a new
anodic peak at 0.27 V increases at higher concentrations of
HFIP. Similar behavior is also shown by other basic quinones
such as DMOBQ and DMBQ (figure not shown).
This new reduction peak prior to the original first reduction
observed for TMBQ (Figure 9b) suggests the formation of
another easily reducible species in the medium before the first
reduction step. This species can either be protonated or
hydrogen-bonded quinone. Since TMBQ is a very weak base
as indicated by its pKa ()-11) in aqueous medium, its
protonation by HFIP is ruled out. However, the UV-vis
spectrum of TMBQ in PhCN shifts slightly to the red upon
addition of HFIP. This effect is further enhanced in nonpolar
methylene chloride and is ascribed to hydrogen-bonding between
TMBQ and HFIP. (The hydrogen-bonding equilibrium constant
for this system in CH2Cl2 is estimated from the shift in spectrum
to be ∼2-3 M-1.) Therefore, the prior peak in Figure 9b, which
increases in height with addition of HFIP, is attributed to the
reduction of hydrogen-bonded TMBQ.
B. Effect of Hexafluoro-2-propanol. HFIP is a stronger
acid and hydrogen-bonding reagent than TFE and ethanol (Table
1). Figure 9 shows the cyclic voltammograms of DCBQ and
TMBQ in benzonitrile at different concentrations of HFIP. We
note that the electrochemical characteristics of DCBQ in the
presence of HFIP are very similar to TMBQ with TFE (compare
Figures 5b and 9a). That is, as seen above, the electrochemistry
of a less-basic quinone (DCBQ) in the presence of a stronger
Since protonation of TMBQ monoanion (pKa ) 5.1) by HFIP
(pKa ) 9.3) is thermodynamically unfavorable, the increase in
peak current of the first wave (Figure 9b) is again assigned to
the fast disproportionation mechanism. However, on the basis
of the pKa of the BQ dianion, HFIP is sufficiently acidic to
protonate the more basic dianions of TMBQ and DMOBQ. Slow
deprotonation of these dianions would explain the irreversibility
of the first peak and the appearance of the new anodic peak at
0.27 V, which is assigned to the oxidation of hydrogen-bonded
TMBQH-.23
(55) Mastragostino, M.; Nadjo, L.; Saveant, J. M. Electrochim. Acta 1968,
13, 721-749.

Electrochemistry of Quinones in Aprotic SolVents
J. Am. Chem. Soc., Vol. 119, No. 27, 1997 6391
the oxidation of TCBQH2 formed by the two-electron reduction
and protonation of chloranil at 0.2 V.23
Addition of TFA to the more basic TMBQ brings out yet
another process in that the voltammogram now shows two new
prior reduction peaks at -0.24 and -0.62 V, shifted positively
by 0.66 and 0.28 V from the original peak (Figure 10b). At
0.0065 M TFA, these two peaks grow in height and the original
quinone peaks disappear. At TFA ) 0.03 M, the peak at -0.62
V also disappears and only one peak is finally observed. These
two peaks are attributed respectively to the reduction of
hydrogen-bonded (-0.62 V peak) and protonated (-0.24 V
peak) TMBQ, which are consistent with both the pKa of TMBQ
and the size of their respective positive displacements.1,2a At a
higher concentration of acid, most of the quinone is finally
protonated, and hence, only the peak at -0.24 V is observed.
This is supported by the change in UV-vis spectrum of TMBQ
caused by addition of TFA (up to 0.3 M). Similar spectral
changes were obtained with methylsulfonic acid (pKa ) -0.6)58
and are attributed to the protonation of TMBQ. The anodic
peak at 0.62 V (Figure 10b) shifts to more positive potential
with an increase in acid concentration, stabilizes at 0.72 V at
0.1 M TFA, and is assigned to the oxidation of TMBQH2.23
Conclusions
1. In the electrochemistry of quinones in neutral aprotic
solvents, hydroxylic additives cause such large effects at such
low concentrations that specific interactions between additives
and quinone species must occur.
Figure 10. Cyclic voltammograms of (a) TCBQ and (b) TMBQ at
different concentrations of TFA in PhCN. Sweep rate ) 100 mV/s.
2. The effect of these agents depends essentially on the
degree of hydrogen-bonding interactions. Weakly basic quino-
nes paired with strongly bonding additives behave similarly to
strongly basic quinones with weakly bonding agents.
(IV) Effect of Strong Acid. Trifluoroacetic Acid. Figures
10a and 10b show the effect of the strong acid TFA on the
electrochemistry of the representative weakly and strongly basic
quinones TCBQ and TMBQ in PhCN. The sequence of
voltammograms of TCBQ with added TFA is generally similar
to that observed for TMBQ in presence of HFIP (compare
Figures 9b and 10a). A cathodic peak prior to the first original
reduction peak grows in height at the expense of the original
peak, suggesting the presence of two reducible species in
equilibrium. At higher TFA concentrations (0.015 M), the prior
peak increases to 1.6 times the height of the original peak and
become fully irreversible, while an anodic peak develops around
0.7 V shifting slightly positive with increasing TFA. The
relative pKa values of BQ (-7)1,56 and TFA (0.52) show that
protonation of TCBQ by TFA is highly unfavorable. This is
supported by the lack of significant change in absorption of
chloranil in benzonitrile upon addition of TFA, up to 0.5 M.
Moreover, the positive shift of the first peak from -0.06 to 0.2
V is too small to be caused by protonation.1,2a Furthermore, in
polar DMSO no prior peak and only a small (smaller than
observed for PhCN) continuous positive shift in potential of
the first reduction step is observed on addition of TFA to
chloranil, unlike the result in PhCN. We therefore attribute the
prior peak, as for the TMBQ-HFIP system, to the reduction
of hydrogen-bonded chloranil. This is consistent with the strong
hydrogen-bonding power of TFA. Since TFA is significantly
more acidic than the semiquinone of TCBQ,57 the increase in
peak current at 0.2 V is now assigned to immediate reduction
of the protonated semiquinone (ECE process)27 rather than to
disproportionation. The anodic peak at 0.75 V is attributed to
3. The effect of hydrogen-bonding can be clearly distin-
guished from that of protonation by consideration of appropriate
pKa values and the characteristics of the cyclic voltammogram
itself. Continuous shifts in potential with no change in wave
height, reversibility, or appearance of new waves indicate
hydrogen-bonding of reduction products. Even if the first wave
increases in height at the expense of the second, hydrogen-
bonding is not excluded, since a disproportionation mechanism
assisted by hydrogen-bonding may be responsible.
4. Analysis of shifts in potential as a function of additive
concentration gives trends in the values of the hydrogen-bonding
parameters n, m, and Keq in agreement with the expected strength
of hydrogen-bonding interaction in a given quinone-additive
pair.
5. Simulation studies of the relative heights of the first and
second reduction waves suggest the formation of a quinone-
quinone dianion complex.
Acknowledgment. It is a pleasure to thank Dr. Stephen W.
Feldberg for helpful discussions regarding digital simulations.
Work done at Brookhaven National Laboratory was supported
under U.S. Department of Energy contract no. DE-AC02-
76CH00016. We much appreciate support of this work by the
Division of Chemical Sciences, Office of Basic Sciences, U.S.
Department of Energy (grant no. FG02-89EB 14072 to Brandeis
University).
JA970028J
(58) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463.
(59) Olmstead, W. N.; Margolin, Z.; Bordwell, F. G. J. Org. Chem. 1980,
45, 3295-3298.
(56) Laviron, E. J. Electroanal. Chem. 1984, 164, 213-224.
(57) The change in pKa of the BQ semiquinone on addition of methyl
groups is only 0.25 units per group;9 thus, the change from BQH• (pKa )
4.0) to TCBQH• should certainly not reduce the latter pKa to that of TFA
(pKa ) 0.52).
(60) Errors in measurement of the small changes in E1/2 of the first peak
(1)
lead to large uncertainties in the values of Keq and contribute thereby to
(2)
errors also in Keq
.