Abnormal solvent effects on hydrogen atom abstraction. 3. Novel kinetics in sequential proton loss electron transfer chemistry

Article abstract of DOI:10.1021/jo051474p

A prolonged search involving several dozen phenols, each in numerous solvents, for an ArOH/2,2-diphenyl-1-picrylhydrazyl (dpph^.) reaction that is first-order in ArOH but zero-order in dpph^. has reached a successful conclusion. These unusual kinetics are followed by 2,2′-methylene-bis(4-methyl-6-tert-butylphenol), BIS, in five solvents (acetonitrile, benzonitrile, acetone, cyclohexanone, and DMSO). In 15 other solvents the reactions were first-order in both BIS and dpph^. (i.e., the reactions followed "normal" kinetics). The zero-order kinetics indicate that in the five named solvents the BIS/dpph^. reaction occurs by sequential proton loss electron transfer (SPLET). This mechanism is not uncommon for ArOH/dpph^. reactions in solvents that support ionization, and normal kinetics have always been observed previously (see Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2003, 68, 3433 and Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69, 5888). The zero-order kinetics found for the BIS/dpph^. reaction in five solvents, S, imply that BIS ionization has become the rate-determining step (rds, rate constants 0.20-3.3 s ^-1) in the SPLET reaction sequence: S + HOAr ? S- HOAr →^rds SH⁺ + ^-OAr →^dpph. SH⁺ + ^.OAr + dpph^- → S + ^.OAr + dpph-H, where ArOH = BIS. Some properties specific to BIS that may be relevant to its relatively slow ionization in the five solvents are considered.

Full text of DOI:10.1021/jo051474p

Abnormal Solvent Effects on Hydrogen Atom Abstraction. 3. Novel
Kinetics in Sequential Proton Loss Electron Transfer Chemistry
Grzegorz Litwinienko*,† and K. U. Ingold
National Research Council, Ottawa, Ontario, Canada K1A 0R6, and Department of Chemistry,
Warsaw University, Pasteura 1, 02-093 Warsaw, Poland
litwin@chem.uw.edu.pl
Received July 15, 2005
A prolonged search involving several dozen phenols, each in numerous solvents, for an ArOH/2,2-
diphenyl-1-picrylhydrazyl (dpph•) reaction that is first-order in ArOH but zero-order in dpph• has
reached a successful conclusion. These unusual kinetics are followed by 2,2′-methylene-bis(4-methyl-
6-tert-butylphenol), BIS, in five solvents (acetonitrile, benzonitrile, acetone, cyclohexanone, and
DMSO). In 15 other solvents the reactions were first-order in both BIS and dpph• (i.e., the reactions
followed “normal” kinetics). The zero-order kinetics indicate that in the five named solvents the
BIS/dpph• reaction occurs by sequential proton loss electron transfer (SPLET). This mechanism is
not uncommon for ArOH/dpph• reactions in solvents that support ionization, and normal kinetics
have always been observed previously (see Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2003, 68,
3433 and Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69, 5888). The zero-order kinetics
found for the BIS/dpph• reaction in five solvents, S, imply that BIS ionization has become the
rate-determining step (rds, rate constants 0.20-3.3 s-1) in the SPLET reaction sequence: S + HOAr
•
dpph
rds
•
•
h S- HOAr
SH+
+
-OAr
SH+ + OAr + dpph- f S + OAr + dpph-H, where ArOH )
f
f
BIS. Some properties specific to BIS that may be relevant to its relatively slow ionization in the
five solvents are considered.
Introduction
peroxyl, and alkyl) have revealed that reaction 1 exhibits
large kinetic solvent effects (KSEs) when XH is a
hydrogen bond donor (HBD) and the reaction is carried
out in a hydrogen bond acceptor (HBA) solvent, S. These
very extensive kinetic results have been rationalized by
the simple chemistry shown in Scheme 1.3,5
The rates of hydrogen atom transfer (HAT) from a wide
variety of substrates, XH (hydrocarbons,1 aniline,2 tert-
butyl hydroperoxide,3 phenol,2-5 ring-substituted phen-
ols,4-8 and aromatic diols8,9) by a number of radicals, Y•
(tert-alkoxyl, 2,2-diphenyl-1-picrylhydrazyl (dpph•), alkyl-
kHAT
Y• + XH HAT8 YH + X•
(1)
* Corresponding author. Phone: 48-22-8225996. Fax: 48-22-
8220211 ext. 335.
According to Scheme 1, there is no HAT from the
hydrogen-bonded XH (XH-S) and the reaction involves
† Warsaw University.
(1) (a) Avila, D. V.; Brown, C. E.; Ingold, K. U.; Lusztyk, J. J. Am.
Chem. Soc. 1993, 115, 466-470. (b) Wayner D. D. M.; Lusztyk, E.;
Page´, D.; Ingold, K. U.; Mulder, P.; Laarhoven, L. J. J.; Aldrich, H. S.
J. Am. Chem. Soc. 1995, 117, 8737-8744.
(6) Valgimigli, L.; Banks, J. T.; Lusztyk, J.; Ingold, K. U. J. Org.
Chem. 1999, 64, 3381-3383.
(2) MacFaul, P. A.; Ingold, K. U.; Lusztyk, J. J. Org. Chem. 1996,
61, 1316-1321.
(7) de Heer, M. I.; Mulder, P.; Korth, H.-G.; Ingold, K. U.; Lusztyk,
J. J. Am. Chem. Soc. 2000, 122, 2355-2360.
(3) Avila, D. V.; Ingold, K. U.; Lusztyk, J.; Green, W. H.; Procopio,
D. R. J. Am. Chem. Soc. 1995, 117, 2929-2930.
(4) Valgimigli, L.; Banks, J. T.; Ingold, K. U.; Lusztyk, J. J. Am.
Chem. Soc. 1995, 117, 9966-9971.
(8) Barclay, L. R. C.; Edwards, C. E.; Vinqvist, M. R. J. Am. Chem.
Soc. 1999, 121, 6226-6231.
(9) (a) Foti, M. C.; Barclay, L. R. C.; Ingold, K. U. J. Am. Chem.
Soc. 2002, 124, 12881-12888. (b) Foti, M. C.; Johnson, E. R.; Vinqvist,
M. R.; Wright, J. S.; Barclay, L. R. C.; Ingold, K. U. J. Org. Chem.
2002, 67, 5190-5196.
(5) Snelgrove, D. W.; Lusztyk, J.; Banks, J. T.; Mulder, P.; Ingold,
K. U. J. Am. Chem. Soc. 2001, 123, 469-477.
10.1021/jo051474p CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/01/2005
8982
J. Org. Chem. 2005, 70, 8982-8990

Novel Kinetics in SPLET Chemistry
SCHEME 1. Cause of a KSE in a HAT Reaction
SCHEME 2. Origin of KSEs in a Mixed HAT and
SPLET Reactiona
only that fraction of XH that is not H-bonded. This “free”
XH reacts with Y• by HAT that occurs with a character-
istic, solvent-independent rate constant, k0. This char-
acteristic rate constant is the same as the rate constant
measured in non-HBA solvents, notably saturated hy-
drocarbons. Kinetic analysis of Scheme 1 relates the
experimentally measured rate constant for the reaction
of XH with Y• by a HAT mechanism, kHAT, in any solvent,
S, to the characteristic rate constant and the equilibrium
constant for H-bond formation as shown in eq 2:3,5
a The SPLET reaction is shown in red.
The SPLET mechanism, which was independently
discovered by Foti et al.,14 is consistent with the high
reactivity of a phenolate anion toward a dialkyl-substi-
tuted dpph• reported by Nakanishi et al.15 SPLET is
shown in Scheme 2 (in red) as an extension of the HAT
mechanism (in black) from Scheme 1.
In Scheme 2, the rate constants for proton and electron
transfer are indicated by kPT and kET, respectively.
Although all the reactions in this scheme will be revers-
ible (e.g., k0PhOH/dpph• < kd0pph-H/PhO•), this complexity has
been ignored for three of the reactions in favor of
simplicity.
Consistent with Scheme 2, the SPLET mechanism has
been shown experimentally to be favored with respect to
the underlying, and always present, HAT process in
solvents that support XH ionization (notably methanol
among organic solvents), with XH having low pKa’s and
with electron-deficient Y• (e.g., dpph•).12 In many sol-
vents, the rates of ArOH/dpph• reactions are increased
by the addition of base (e.g., sodium methoxide) and
decreased by the addition of acid (e.g., acetic acid).12,13
These changes in rate reflect a base-induced increase and
an acid-induced decrease in the degree of phenol ioniza-
tion and hence in the importance of the SPLET mecha-
nistic pathway.
k
HAT ) k0/(1 + KS[S])
(2)
Room-temperature KSEs for the many XH/Y• reactions
investigated have been shown to be extremely well
correlated by empirical eq 3:5
log(kHAT/M-1 s-1) ) log (k0/M-1 s-1) - 8.3R2H â2H (3)
In this equation, R2H is Abraham et al.’s10 parameter
quantifying the relative HBD activities of XH (range 0.00
for alkanes to ∼1 for strong organic acids), and âH2 is
Abraham et al.’s11 parameter quantifying the relative
HBA activities of S (range 0.00 for alkanes to 1.00 for
hexamethylphosphoramide, the strongest organic HBA).
More than 500 RH2 values and more than 500 âH2 values
are known.10,11
In our recent work on KSEs, the kinetics of ArOH/
Equation 3 has proven to be a very useful mechanistic
probe for XH/Y• reactions. If the kinetic data in several
solvents fit eq 3 or, equivalently, yield a straight line
when logarithms of the experimental rate constants are
plotted against âH2 , there is a HAT mechanism in all the
solvents examined. Under these circumstances, the XH/
Y• reaction is said to exhibit a normal KSE. The reactions
of phenols and ring-substituted phenols with, for ex-
ample, tert-alkoxyl radicals all exhibit normal KSEs.2,5,7
However, very early in this work,4 it was discovered that
although the KSEs were normal with XH ) phenol and
R-tocopherol and Y• ) dpph• in most solvents, they were
abnormal in tert-butyl alcohol. Specifically, for both
substrates the rate constants in tert-butyl alcohol were
about five times greater than would have been expected
on the basis of the HBA activity (i.e., âH2 value) of this
alcohol. Enhanced rate constants (sometimes very much
enhanced) were subsequently observed in the reactions
of various phenols with dpph• in many alcohols.12 These
abnormal KSEs were attributed to the occurrence in
these systems of a new reaction path leading to the same
products.12 This new mechanism was later named se-
quential proton loss electron transfer (SPLET).13
dpph• have been monitored by following the loss of dpph•
spectrophotometrically at 517 nm (ꢀ ≈ 11 000 M-1 cm-1 16
)
A
using a rapid-mixing, stopped-flow apparatus.9a,b,12,13
large excess of ArOH was generally employed, and the
complete loss of dpph• occurred with “clean” first-order
kinetics, with just the four exceptions noted below. Rates
of ArOH/dpph• reactions were always directly propor-
tional to the ArOH concentration. Therefore, the vast
majority of the ArOH/dpph• reactions examined follow
“clean” bimolecular kinetics (eq 4).
-d[dpph•]/dt ) kexptl[ArOH][dpph•]
(4)
This was the case for “HAT-only” conditions (nonionizing
solvents and ionizing solvents containing sufficient acetic
acid to completely suppress the SPLET process), for
conditions where SPLET contributed g90% to kexptl, and
(12) Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2003, 68, 3433-
3438.
(13) Litwinienko, G.; Ingold, K. U. J. Org. Chem. 2004, 69, 5888-
5896.
(14) Foti, M. C.; Daquino, C.; Geraci, C. J. Org. Chem. 2004, 69,
2309-2314.
(15) Nakanishi, I.; Miyazaki, K.; Shimada, T.; Iizuka, Y.; Inami, K.;
Mochizuki, M.; Urano, S.; Okuda, H.; Ozawa, T.; Fukuzumi, S.; Ikota,
N.; Fukuhara, K. Org. Biomol. Chem. 2003, 1, 4085-4088.
(16) Incorrectly given as ∼50 000 in ref 13. We thank an anonymous
reviewer for bringing this matter to our attention.
(10) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Duce, P. P.; Morris,
J. J.; Taylor, P. J. J. Chem. Soc., Perkin Trans. 2 1989, 699-711.
(11) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Morris, J. J.;
Taylor, P. J. J. Chem. Soc., Perkin Trans. 2 1990, 521-529.
J. Org. Chem, Vol. 70, No. 22, 2005 8983

Litwinienko and Ingold
TABLE 1. Bimolecular Rate Constants, kS/M-1 s-1, for
for conditions where the HAT and SPLET processes were
more evenly matched.
the BIS/dpph• Reaction in 20 Solventsa
solventb
â2H
kS
solventb
â2H
kS
c
c
The four exceptions mentioned above would not have
been discovered without the stopped-flow equipment.
This is because the anomalous kinetics occurred within
<100 ms of mixing methanol and ethanol solutions of
curcumin and dehydrozingerone (sometimes called “half
curcumin”, pKa 9.1 in water/methanol, 1:1 v/v) with
solutions of dpph• in the same solvents.13 These very
initial reactions were >1000 times as rapid as the same
reactions carried out in the presence of 5 mM acetic acid,
despite the fact that the rates with 5 mM acid were still
1 order of magnitude greater than those for the “HAT
only” reactions that were observed only at much higher
concentrations of acetic acid.13 The four short duration,
fast reactions were followed by slower (but still rapid)
first-order loss of dpph•. The initial, fast processes were
attributed to reactions of dpph• with the anions, X-, of
the substrates, XH, that would have been present at low
equilibrium concentrations in the alcohols (see reaction
labeled kET in Scheme 2). These “preformed” anions are
quickly depleted, and ionization of XH (the proton loss
1
2
3
CH3(CH2)5CH3 0.00 57
11 1,4-dioxane
12 EtCOEt
0.47 0.31
0.48 6.5
0.48 87
CCl4
0.05d 25
0.20e 7.0
PhH + 1 M
13 PhCOMe
Ph2CO
HC(O)OMe
MeOH
PhCN
MeCN
4
5
6
7
0.38 1.0
14 t-BuCOMe
15 MeCOMe
16 THF
0.49 0.64
0.50 460f
0.51 1.4
0.41 2100
0.42 620f
0.44 15000f 17
0.52 240f
(CH2)5CO
18 c-(C3H5)2CO
19 γ-valerolactone 0.55 1300
20 Me2SO
8
9
EtOH
t-BuCN
0.44 1300
0.44 13
0.47 12
0.53 1500
10 MeC(O)OEt
0.78 160f
aFull data with errors are given in the Supporting Information.
b Solvents are listed in order of increasing â2H values. The sol-
vents are identified in parts A and B of Figure 1 by the numbers
in this column. c Reference 11. d Reference 5. e Estimated from
âH2 (benzene) ) 0.14 and âH2 (benzophenone) ) 0.47 and the mole
fractions of the two components of this solvent mixture. f This rate
constant refers only to very initial stages of the reaction (0 to <100
ms) where the reaction is first-order in both BIS and dpph•.
3-8 × 10-5 M) was monitored to completion17 in the
presence of excess BIS at five or more different BIS
concentrations in 20 solvents (see Supporting Informa-
tion). In all solvents the rate of loss of dpph• was directly
proportional to the concentration of BIS. In 15 of the
solvents the dpph• loss followed pseudo-first-order kinet-
ics for the entire reaction that can be represented as:
portion of SPLET) then becomes partly rate-determining
PT
XH/S
(see reaction labeled k
in Scheme 2).
Substrate ionization in an ArOH/dpph• reaction oc-
curring by the SPLET mechanism that is truly the rate-
determining step (rds) would have an interesting and
readily observed kinetic consequence. Specifically, in the
second (slower) stage of the reaction after the “preformed”
ArO- has been rapidly consumed, the loss of dpph•
should follow zero-order kinetics. That is, the reaction
should obey eq 5, rather than eq 4.
kexptl
dpph• + ArOH
8 dpph-H + ArO•
(ArOH ) BIS) (6)
-d[dpph•]/dt ) kexptl[ArOH]
(5)
Bimolecular rate constants in each solvent, kS/M-1 s-1
,
were calculated from the linear least-squares slopes18
Extensive kinetic studies on ArOH/dpph• reactions
involving numerous monophenols and many solvents
failed to reveal a single example of eq 5 kinetics.
However, our efforts were rewarded when we extended
our studies to bisphenols and discovered that 2,2′-
methylene-bis(4-methyl-6-tert-butylphenol):
derived from plots of kexptl versus [BIS], eq 7.
kexptl ) const + kS[BIS]
(7)
The kS values in these 15 solvents are recorded in Table
1 in normal type. The solvents have been listed in order
of increasing HBA activity, that is, increasing âH2 which,
for a “clean” HAT process, would mean also in order of
decresing kS, with the value in heptane, viz., k0 ) 57 M-1
s-1, being the largest (see eq 3). It is obvious that in
several of these 15 solvents the reaction being monitored
is not a HAT process because kS > k0.
For the other five solvents (listed in bold face in Table
1), the same kinetics (i.e., first-order in both [dpph•] and
[BIS]) applied only in the period immediately (<100 ms)
following the rapid (∼1-5 ms) mixing of the dpph• and
BIS solutions. The “initial” kS values (eq 7) are given in
Table 1 in bold face (the zero-order kinetics that follow
this initial (fast) reaction phase are described later). In
these five solvents the kS values are also considerably
greater than k0 (Table 1). Indeed, a plot of log kS for all
20 solvents against âH2 provides a rather fine “shotgun
pattern” (Figure 1A) rather than the straight line (slope
designated hereafter as BIS, sometimes exhibited the
sought-after kinetics. In a 20-solvent study, eq 4 was
obeyed throughout the entire BIS/dpph• reaction in 15
of the solvents, but only in the very initial stages in the
remaining five solvents. In these five solvents, after the
initial, very fast reaction with the “preformed” BIS-
anion, the kinetics changed to become “clean” zero-order
in dpph• (i.e., the rate law changed from eq 4 to eq 5).
These kinetic solvent effects on what, at first sight, might
be thought to be a simple hydrogen atom abstraction are
described herein.
(17) This is never all the way to zero absorbance at 517 nm because
dpph-H absorbs weakly at this wavelength (ꢀ ≈ 400 M-1 cm-1) and
because the BIS oxidation product(s) also absorb at this wavelength;
see various figures in the text and Supporting Information.
Results
Kinetics of BIS/dpph• Reaction in Different Sol-
(18) In all the kinetic work reported herein the overwhelming
vents. Overview. Decay of dpph• (initial concentration
majority of the R2 values were 0.99 (see Supporting Information).
8984 J. Org. Chem., Vol. 70, No. 22, 2005

Novel Kinetics in SPLET Chemistry
FIGURE 1. Plots of log kS for the BIS/dpph• reaction in different solvents against âH2 . (A) All 20 solvents, data from Table 1. (B)
Five solvents selected as being unlikely to support BIS ionization (4, Table 1) and seven solvents containing 1 M acetic acid (red
2, Table 2). The least-squares straight line in (B) has been drawn ignoring the points for solvents 6, 15, and 20.
TABLE 2. Bimolecular Rate Constants, kS/M-1 s-1, for
the BIS/dpph• Reaction in Seven Solvents Containing
Acetic Acida
In all our earlier ArOH/dpph• KSE work,12,13 10 mM
acetic acid either had no effects on kS (i.e., no SPLET
occurred in the solvent) or it reduced kS relative to the
rate constant in the nonacidified solvent (i.e., there was
partial or complete suppression of SPLET and kS f kHAT).
For the BIS/dpph• reaction, in contrast, the addition of
10 mM acetic acid causes the values of kS to increase
substantially in acetone and DMSO and to increase
slightly in benzonitrile (see Table 2). In acetone and
benzonitrile, kS decreases on increasing the acetic acid
concentration to 100 mM and decreases still further with
1 M acid. In DMSO, kS reaches a maximum with 100 mM
acetic acid and even with the addition of 1 M acid kS
remains larger than in the nonacidified solvent.
Earlier work12,13 has shown that for ArOH/dpph•
reactions the SPLET mechanism is either completely or
nearly completely suppressed by 1 M acetic acid, meaning
that the measured rate constant, kS, now reflects the
underlying HAT process, kHAT. The logarithms of the 1
M acetic acid rate constants have therefore been plotted
(in red) against âH2 in Figure 1B together with log kS (in
black) for those neat solvents that are unlikely to support
a kinetically significant SPLET process (viz., solvents 1,
2, 3, 11, and 16). A rather good straight line is obtained
provided the benzonitrile (6), acetone (15), and DMSO
(20) data points are ignored. The deviations of these three
data points from the line suggest that in these three
solvents the ionization of BIS is incompletely suppressed
by 1 M acetic acid.
kS
[acetic acid]/
mM solventb
10
100
mM
1000
mM
d
Ac
0.75 32.7 2100
25.2
620e
ꢀr
0
mM
5
MeOH
PhCN
MeCN
EtOH
MeC(O)OEt 0.21 6.0
MeCOMe
Me2SO
9.1
750
3.6
23
274
3.1
1.6
1550
4900
1.4
15
6
-
7
0.37 35.9 15000e 5700
2.0
8
0.66 24.6 1300
12
0.25 20.6
0.34 46.5
36
12
1.7
10
15
20
0.9
460e 13000
4.9
160e
890
520
a Full data with errors are given in the Supporting Information.
b See footnote b to Table 1. c For solvent acity, see ref 44. d For
permittivity (dielectric constant), see: Reichardt, C. Solvents and
Solvent Effects in Organic Chemistry, 3rd ed.; Wiley: Weinheim,
Germany, 2003. e See footnote f to Table 1.
) -8.3RH2 ) expected (eq 3) if there is a HAT mechanism
in all solvents. This suggests that the scatter in Figure
1A is largely19 due to contributions from the SPLET
mechanism (Scheme 2).
The HAT-Only Reactions and the Determination
of r2H for BIS by Kinetics and by IR Spectroscopy.
The effects of acetic acid on the BIS/dpph• KSEs were
explored using seven solvents chosen because they were
judged to be capable of supporting the ionization of BIS
and had structural features representative of other
solvents that were also likely to support BIS ionization.
Bimolecular rate constants, kS, were measured in the
presence of 10, 100, and 1000 mM acetic acid in the seven
selected solvents, and the results are presented in Table
2. Those solvents shown in bold face in Table 1 are also
shown in bold face in Table 2, though it should be noted
that with all concentrations of acetic acid the reactions
were not zero-order in [dpph•] (i.e., they did not follow
the kinetics observed in the nonacidified solvents; see
below).
The line in Figure 1B has a slope of -3.5, indicating
that the “best” RH2 value for BIS is -3.5/-8.3 ) 0.42 for
this array of 11 solvents.20 Since an RH2 value for BIS (or
any related diol) has not been previously reported,
“solvent-specific” RH2 values were determined by the
usual IR method5,10,12,13 using two of Abraham’s10 “cali-
brated” HBA’s: DMSO gave RH2 ) 0.49 and acetonitrile
gave RH2 ) 0.40 (for full details see the Supporting
Information). These two “solvent-specific” IR-derived R2H
(19) For sterically nonhindered phenols, scatter in such plots
appears to arise solely from the SPLET mechanism. However, for
sterically hindered phenols (e.g., BIS, see later) scatter also arises from
the fact that the strength of the hydrogen bond between the phenol
and the HBA depends on the HBA’s size and shape as well as on its
HBA activity toward nonhindered phenols.12
(20) The word “best” has been used because sterically congested
phenols do not possess a single, unique RH2 value; see footnote 19 and
ref 12.
J. Org. Chem, Vol. 70, No. 22, 2005 8985

Litwinienko and Ingold
FIGURE 2. Decay of dpph• during the reaction with BIS in acetone (A and B), in DMSO (C), and in benzonitrile (D). In the
reactions shown in (A), the initial concentrations of dpph• were always the same and the initial concentrations of BIS were
varied. In the reactions shown in (B, C and D), the initial concentrations of [BIS] were always the same in each panel and the
initial concentrations of dpph• were varied. All concentrations are given in the legends within each panel.
values for BIS nicely bracket the “best” kinetics-derived
Discussion
RH2 value of 0.42.
The KSEs observed in BIS/dpph• reactions are famil-
iar, for the most part. That is, in all solvents the rates of
the initial reactions are directly proportional to the
concentration of BIS and to the concentration of dpph•.
In the majority of the 20 solvents examined these
bimolecular kinetics are followed until complete con-
sumption of dpph•, although the very initial stages of
the reaction are often much faster than the later stages.
As with other ArOH/dpph• reactions,12-14 SPLET (Scheme
2) is the dominant reaction mechanism in solvents that
can support the ionization of ArOH. Intially, the dpph•
will react with the low concentration (if any) of “pre-
formed” phenoxide anion present in equilibrium with the
phenol. This involves a very rapid electron transfer and
occurs with bimolecular kinetics (eq 4). Once all the
preformed anion has been consumed, the reaction rate
slows and the SPLET mechanism takes control with
kinetics again described by eq 4. The pre-ionization of
ArOH and the SPLET process can generally be sup-
pressed, or even eliminated, leaving just the “back-
ground” HAT mechanism, by the addition of acetic acid.
All the foregoing aspects of BIS/dpph• reaction kinetics
are familiar, in view of our earlier work.12,13 What is
unfamiliar and has not previously been observed in
ArOH/dpph• reactions involving monophenols,4,5,8,12-14
catechols,8,9 or naphthalene diols9 is a process that is first-
order in [ArOH] but zero-order in [dpph•]. The following
The Reactions that are Zero-Order in [dpph•].
These kinetics (i.e., reactions following the rate law
described by eq 5, with the rds being deprotonation of
phenol) were observed to commence some 100 ms after
mixing BIS and dpph• solutions in five solvents: ben-
zonitrile, acetonitrile, acetone, cyclohexanone, and DMSO.
They do not occur immediately after the reagent solutions
are mixed because during the first few tens of mil-
liseconds the dpph• is reacting with, and hence destroy-
ing, the equilibrium concentration of “preformed” BIS
anion (see Introduction). The dpph• solutions are stable
for days in four of these five solvents. However, in
cyclohexanone the dpph• slowly decays. This we attribute
to its reaction with the 1.2% enol present in neat cyclo-
hexanone21 by analogy with other ketone(enol)/dpph•
reactions we have studied.13 Because of this side reaction,
consideration will henceforth only be given to the other
four solvents. Typical kinetic traces are shown in Figure
2 (acetone (15) panels A and B; DMSO (20), panel C;
benzonitrile (6), panel D). These traces and data in the
Supporting Information demonstrate that these reactions
are first-order in [BIS] (e.g., Figure 2A) and zero-order
in [dpph•] (e.g., parts B, C, and D in Figure 2).
(21) Solomons, T. W G. Organic Chemistry, 6th ed.; Wiley: New
York, 1996.
8986 J. Org. Chem., Vol. 70, No. 22, 2005

Novel Kinetics in SPLET Chemistry
discussion will therefore focus on the zero-order kinetics
found for BIS/dpph• reactions in certain solvents.
Kinetic Requirements for an XH/Y• Reaction To
Be Zero-Order in [Y•]. Two conditions must be met: (i)
the SPLET mechanism must completely dominate the
HAT process and (ii) proton transfer from ArOH to the
solvent, S, must be the rds in the SPLET pathway.
Application of the steady-state approximation to Scheme
2 yields:
KSkPXTH/S[XH][S]
[X-] )
(8)
PT
k
[HS+] + kET[Y•]
HS+/X-
and hence:
KSkPXTH/SkET[XH][S][Y•]
d[Y•]
dt
-
)
PT
(9)
k
[HS+] + kET[Y•]
HS+/X-
FIGURE 3. OH stretching region of the IR spectrum of 13.4
mM BIS in CCl4 containing various concentrations of DMSO
(in mM): (a) 0.0, (b) 9.1, (c) 13.6, (d) 18.2, (e) 24.2, (f) 32.3, (g)
43.1, (h) 57.4, (i) 76.5, (j) 98.4. The free OH band at 3632 cm-1
was used for calculation of the concentration of nonintermo-
lecularly hydrogen-bonded BIS.
The reaction will follow zero-order kinetics provided:
PT
k
ET [Y•] . k
[HS+]
(10)
HS+/X-
in which case:
-d[Y•]/dt ) KSkXPTH/S [XH][S]
SCHEME 3. νOH/cm-1 for BIS Conformers in CCl4
(11)
Intramolecular H-Bonding and Structures of BIS
in Solution. The secret of the BIS/dpph• reaction’s
unique zero-order kinetics must lie in the solution
properties and structures of BIS. In 1950, an early IR
study of BIS revealed two O-H stretching bands.22 In
1965, IR measurements with better resolution showed
that BIS actually has four O-H stretching bands, two
of which are barely resolved.23 It was suggested that BIS
adopted a number of conformations that “may be in
equilibrium.”23 The O-H bands were assigned to “free”
OH, to OH hydrogen bonded to the “other” aromatic ring,
OH-Ar, and to OH hydrogen bonded to the “other” OH
group, OH-OH.23 The presence of four O-H stretching
hydrogen bonded to the oxygen atom of the adjacent
phenol. However, we suggest that in CCl4 all conformers
present in reasonable abundance will have one intramo-
lecular hydrogen bond. The electron-withdrawing (EW)
effect of these intramolecular hydrogen bonds will cause
the band due to the remaining “free” OH to shift to lower
wavenumbers than those for the OH bands in 2-tert-
butyl-4,6-dimethylphenol and 2,4,6-trimethylphenol. For
meta- and para-substituted phenols the OH fundamental
stretching frequencies, νOH, correlate very well with the
substituents’ Hammett σ-constants with EW substituents
shifting νOH to smaller wavenumbers.30,31 The EW effect
of the intramolecular hydrogen bonds on the neighboring
“free” OH groups will be greater for the OH-OH hydro-
gen bond than for the OH-Ar hydrogen bond. Our
assignment of the four OH stretching bands of BIS in
CCl4 is shown in Scheme 3. For simplicity, this repre-
sentation is two-dimensional but it is certain that the
two aromatic rings of BIS will not be coplanar in any
stable conformer.28,29,33
bands has been confirmed in three later publications,24-26
27
though their reported frequencies in CCl4
and the
conformers suggested were not always in agreement.28
We have confirmed that BIS has four O-H stretching
bands (Figure 3) and agree broadly with their original
assignments by Cairns and Eglinton.23 That is, the bands
at 3632 and 3610 cm-1 are assigned to “free” OH groups,
that at 3510 cm-1 to OH that is H-bonded to the adjacent
aromatic ring, and that at 3450 cm-1 to OH that is
(22) Coggeshall, N. D. J. Am. Chem. Soc. 1950, 72, 2836-2844.
(23) Cairns, T.; Eglinton, G. J. Chem. Soc. 1965, 5906-5913.
(24) Kovac, S.; Solcaniova, E.; Baxra, J. Tetrahedron 1971, 27, 2823-
2830.
(25) Pivcova, H.; Jirackova, L.; Pospisil, J. J. Polym. Sci., Polym.
Symp. 1973, 40, 283-295.
(26) Arzamanova, I. G.; Naiman, M. I.; Romm, I. P.; Tovbin, U. K.;
Guryanova, E. N.; Gurvich, Y. A.; Logvinenko, R M. Zh. Fiz. Khim.
1981, 55, 1554-1555.
(29) Chetkina, L. A.; Zavodnik, V. E.; Bel’skii, V. K.; Arzmanova, I.
G.; Naiman, M. I.; Gurvich, Ya. A. Zh. Struct. Khim. 1984, 25, 109-
113. English translation: J. Struct. Chem. 1985, 25, 935-939.
(30) Ingold, K. U. Can. J. Chem. 1960, 38, 1092-1098.
(31) νOH values for para-substituted 2-tert-butyl, 2,6-di-tert-butyl,
2-methyl, and 2,6-dimethylphenols also correlate with σp.30,32
(32) (a) Ingold, K. U.; Taylor, D. R. Can. J. Chem. 1961, 39, 471-
480. (b) Ingold, K. U. Can. J. Chem. 1962, 40, 111-121.
(33) According to DFT calculations, the conformers of bis(2-hydroxy-
phenyl)methane also have noncoplanar aromatic rings; see: Katsyuba,
S.; Chernova, A.; Schmutzler, R.; Gronenberg, J. J. Chem. Soc., Perkin
Trans. 2 2002, 67-71.
(27) Band maximum frequencies (in cm-1): 3639, 3620, 3516, 3450
(ref 23); 3648, 3623, 3516, 3444 (ref 24); 3627, 3610, 3506, 3436 (ref
25); 3651, 3635, 3516, 3454 (ref 26). Our work (Figure 3 and Scheme
3) yields: 3632, 3610, 3510, 3450.
(28) Most notably, any OH-OH hydrogen-bonded conformers of BIS
were ignored in ref 24. The crystal structure of BIS reveals an
intramolecular OH-OH hydrogen bond with the “free” OH (on the
right in structure 1, Scheme 3) twisted out of the plane of the aromatic
ring to which it is attached.29
J. Org. Chem, Vol. 70, No. 22, 2005 8987

Litwinienko and Ingold
kinetic RH2 value for BIS (0.42) is consistent with the two
values obtained by IR (0.49 and 0.40), vide supra. Indeed,
from the viewpoint of the BIS/dpph• KSEs and reaction
rates, BIS generally behaves as a “normal”, moderately
acidic,38 moderately sterically crowded phenolic HBD.
The generally “normal” behavior of BIS is also indicated
by by Pedulli and co-workers’40 study of the BIS/ROO•
reaction:
The two intramolecular hydrogen bonds formed by BIS
in CCl4 (Scheme 3) do not exist in the presence of >100
mM DMSO (see Figure 3). Similarly, the BIS analogue
in which the two 4-methyl groups were replaced by two
ethyl groups showed only a single broad band in diethyl
ether (3350 cm-1 23 and in pyridine (3185 cm-1).24 Thus,
)
in strong and even in moderately strong HBA solvents,
BIS adopts conformation 3.
kinh
ROO• + ArOH
8 ROOH + ArO• (ArOH ) BIS) (12)
(13)
ROO• + ArO• fast8 nonradical products
These workers found that the autoxidation of styrene (4.3
M in chlorobenzene at 30 °C) was retarded by BIS and
the duration of the induction period corresponded to an
overall stoichiometric factor, n, equal to 4 (i.e., 4 ROO•/
BIS) as befits a bisphenol. Initially, there was strong
inhibition during which 2 ROO• per BIS were consumed
This facile breaking of the BIS intramolecular H-bonds
(1 and 2) by HBA solvents indicates that they are
relatively weak and stands in contrast to the virtual
inability of such solvents to cleave the intramolecular
OH-OH hydrogen bonds in catechols and 1,8-naphtha-
lene diols.9a,34
and kinh was found to be 5.0 × 105 M-1 s-1 40
.
Weaker
inhibition followed for the remaining 2 ROO• per BIS with
kinh ) 2.0 × 103 M-1 s-1 41
. For comparison, under similar
The BIS/dpph• Reaction. The HAT process in non-
HBA solvents will involve only conformers 1 and 2. In
catechols (and naphthalene diols), the OH-OH intramo-
lecular hydrogen bond lowers the bond dissociation
enthalpy (BDE) of the “free” OH relative to, for example,
that of the OH BDE in hydroquinone.35,36 This is because
the fairly strong OH-OH hydrogen bond in these aro-
matic diols becomes stronger by several kcal/mol in the
radical OH-O• hydrogen bond.9b,35,36 By analogy, the most
reactive site for HAT in BIS is expected to be the “free”
OH in 1. Thus, in a saturated hydrocarbon and on a
molar basis, BIS would be expected to show a similar, or
slightly enhanced, reactivity toward dpph• compared
with 2-tert-butyl-4,6-dimethylphenol (ButMe2-phenol) and
2,4,6-trimethylphenol (Me3-phenol), but to be much less
reactive than catechol. The experimental rate constants
(M-1 s-1) are: BIS, 57 (Table 1); ButMe2-phenol, 17;37
Me3-phenol, 40,37 and catechol,9b 1800. Thus, the HAT
ability of BIS is consistent with its structure.
experimental conditions with ArOH ) Me3-phenol, kinh
) 8.5 × 104 M-1 s-1 and n ) 2.42 Thus, BIS is slightly
more reactive than Me3-phenol toward both ROO• and
dpph• in poor and non-HBA solvents, respectively. These
results are consistent with the slightly lower O-H BDE
for a BIS homologue,43 viz.40 81.2 kcal/mol, compared with
that for Me3-phenol, 82.7 kcal/mol.40
In two respects, however, the BIS/dpph• reactions
exhibit behavior that has not previously been encoun-
tered. The first relates to the effect on the reaction rates
in certain solvents of added acetic acid. In earlier ArOH/
dpph• kinetic work with many phenols in numerous
solvents,12,13 the addition of acetic acid either had no
effect on the reaction rate (indicating a HAT-only reac-
tion) or caused the rate to decline monotonically (to the
HAT limit) as the acid concentration increased (indicat-
ing a SPLET process). The rates of the BIS/dpph•
reactions decline monotonically with increasing acetic
acid concentrations in only three of the seven solvents
in which the effect of acid was explored (Table 2). In two
of the seven solvents, acetone (15) and DMSO (20), the
rates in the presence of 10 mM acetic acid are definitely
In HBA solvents, the HAT reactions of BIS will involve
the mono-desolvated form, 4, or, less probably, the di-
desolvated forms 1 and 2.
(37) Kinetic data on the reactions of ButMe2-phenol and Me3-phenol
with dpph• in various solvents are given in the Supporting Informa-
tion.
(38) The pKa of BIS has not been reported. Two much less sterically
congested homologues in which both of the tert-butyl groups of BIS
had been replaced either by two H-atoms or by two CH3 groups have
been titrated with base in isopropyl alcohol/benzene, 1:1 v/v.39 For both
compounds, one OH group was much more acidic than the other, a
result consistent with their monoanions having structures analogous
to 6.
The BIS/dpph• reaction would therefore be predicted to
exhibit a normal KSE provided any SPLET process was
suppressed because of the solvent’s low permittivity
(dielectric constant, ꢀr) or by the addition of acetic acid.
This normal KSE is shown in Figure 1B, and the derived
(39) Sprengling, G. R. J. Am. Chem. Soc. 1954, 76, 1190-1193.
(40) Amorati, R.; Lucarini, M.; Mugnaini, V.; Pedulli, G. F. J. Org.
Chem. 2003, 68, 5198-5204.
(41) Initially, one ring of BIS is oxidized to a peroxycyclohexadi-
enone, the carbonyl group of which promotes the formation of a strong
intramolecular hydrogen bond from the OH group on the second ring.
It is this hydrogen bond that makes the second phenolic hydrogen atom
more difficult to abstract.40
(42) Burton, G. W.; Doba, T.; Gabe, E. J.; Hughes, L.; Lee, F. L.;
Prasad, L.; Ingold, K. U. J. Am. Chem. Soc. 1985, 107, 7053-7065.
(43) One H-atom in the methylene bridge was replaced by a methyl
group because the BIS radical was not sufficiently persistent to allow
the O-H BDE in BIS to be determined.40
(44) Swain, C. G.; Swain, M. S.; Powell, A. L.; Alunni, S. J. Am.
Chem. Soc. 1983, 105, 502-513.
(34) Foti, M. C.; DiLabio, G. A.; Ingold, K. U. J. Am. Chem. Soc.
2003, 125, 14642-14647.
(35) Wright, J. S.; Johnson, E. R.; DiLabio, G. A. J. Am. Chem. Soc.
2001, 123, 1173-1183.
(36) (a) Lucarini, M.; Mugnaini, V.; Pedulli, G. F. J. Org. Chem.
2002, 167, 928-931. (b) Lucarini, M.; Pedulli, G. F.; Guerra, M.
Chem.-Eur. J. 2004, 10, 933-939.
8988 J. Org. Chem., Vol. 70, No. 22, 2005

Novel Kinetics in SPLET Chemistry
FIGURE 4. (A) Kinetic traces showing the loss of dpph• (initial concentration 9.5 × 10-5 M) during reaction with 2.5 × 10-4
M
BIS in dry acetone (a) and in acetone containing water at concentrations of 0.36 (b), 0.56 (c), 1.20 (d), and 1.44 M (e). (B) Plots of
the negative logarithms of normalized absorbance obtained from the kinetic curves in (A). The dashed straight lines show that
curve (a) does not but that curves (b-e) do follow first-order kinetics.
greater than the rates in the absence of acid. This acetic
acid-induced rate increase suggests that the concentra-
tion of the “preformed” BIS anion has been increased by
the added acid. Although this suggestion is counterin-
tuitive, it may be a consequence of the relative anion-
solvating abilities of acetic acid and the solvents. Relative
anion solvating abilities are described by Swain et al.’s44
solvent acity parameters, A (scales range from 0.0 for
alkanes to 1.0 for water). The A value for acetic acid is
0.93, which means that it will solvate the BIS anion much
more strongly than, for example, acetone (A ) 0.25) or
DMSO45 (A ) 0.34); see Table 2.
solvation of BIS or on its conformational equilibria in
acetone (âH2 ) 0.50).11 However, the added water (ꢀr )
78.4) would be expected to increase the rate of BIS
ionization above that in neat acetone (ꢀr ) 20.6). In the
event, the rate of the BIS/dpph• reaction in acetone was
increased by the addition of water (see Figure 4A), and
the reaction became first-order in [dpph•] (see Figure
4B). This result also suggests that ionization of 3 may
well be the rds in the five abnormal solvents.
Structure 5 is not intended to represent the most stable
conformer of the BIS monoanion. The stable conformer
of the monoanion is expected to have an intramolecular
OH-O- hydrogen bond (i.e., 6).
The second unusual aspect of BIS/dpph• reactions, and
the raison d’eˆtre for the present article, is those reactions
that are zero-order in [dpph•]. Such kinetics are consis-
tent with the ionization of BIS being the rds reaction 14:
The intramolecular H-bond in 6 is likely to retard the
PT
reverse proton-transfer step, 6 + HS+ f 3 (k
in
HS+/X-
Scheme 2). A slow reverse proton-transfer reaction is a
The five linear decay traces for the BIS/dpph• reaction
in acetone that are shown in Figure 2B have a mean slope
of -0.0214 absorbance units/s (see Supporting Infor-
mation). Conversion of absorbance into [dpph•] yields
kPXTH/S[BIS][acetone] ) 1.1 × 10-5 M s-1 and, since [BIS]
was 5.5 × 10-5 M in these five experiments, kPXTH/S[ace-
tone] ) 0.20 s-1, which is the rate constant for BIS
ionization in acetone. Rate constants for BIS ionization
in the other three solvents in which it could be measured
are: benzonitrile, 3.30 s-1; acetonitrile, 1.4 s-1; and
(45) Another possible reason for the unusual behavior of the BIS +
DMSO + CH3CO2H system (in comparison with other acidified
systems) lies in the fact that the acidities of carboxylic acids change
much more than the acidities of phenols on passing from water (or
alcohols) to DMSO. For example, the pKa’s for acetic acid in water and
in DMSO are 4.7 and 12.6, respectively (∆pKa ) 7.9). For 2,6-
dinitrophenol, the pKa’s in water and DMSO are 3.7 and 4.9, respec-
tively (∆pKa ) 1.2); see: Isutsu, K. Electrochemistry in Nonaqueous
Solutions; Wiley-VCH: Weinheim, Germany, 2002; Chapter 3, pp 66-
67. For some other carboxylic acids and phenols, the ∆pKa’s are: 6.8,
benzoic acid; 6.0, chloracetic acid; 1.44, 2,4-dinitrophenol; and -1.38,
picric acid.
(46) The rates of these zero-order reactions are, of course, much
greater than the rates of the HAT-only reactions in these four solvents.
The latter can be calculated from eq 2, using k0 ) 57 M-1 s-1 (Table
DMSO, 0.91 s-1 46,47
As a reviewer kindly pointed out,
.
1), RH2 ) 0.42 for BIS, and the appropriate âH2 value for the solvent. In
the protonation of the BIS anion will be close to diffusion-
acetone (âH2 ) 0.50), for example, this procedure yields kHAT) 1.0 M-1
s-1. The zero-order in [dpph•] traces in acetone shown in Figure 2B
were obtained with [dpph•] ) (2.2-9.6) × 10-5 M. Therefore, in this
PT
controlled (i.e., in Scheme 2, k
≈ 5 × 109 M-1 s-1).
HS+/X-
It follows that the dissociation constant for BIS, Ka(BIS),
in these four solvents ranges from 4.0 × 10-11 to 2.8 ×
10-10 M.49
solvent kHAT [dpph•] ) 1.0 × (2.2-9.6) × 10-5 ) (2.2-9.6) × 10-5 s-1
,
PT
XH/S
which is very much smaller than k
[acetone], viz. 0.20 s-1
.
(47) These rate constants are 3 or 4 orders of magnitude smaller
than the rate constants for the deprotonation of nitrophenols in water.48
(48) Eigen M.; Kustin, K. J. Am. Chem. Soc. 1960, 82, 5952-5953.
The addition of low concentrations of water (â2H
)
0.38)11 is not expected to have much effect on either the
J. Org. Chem, Vol. 70, No. 22, 2005 8989

Litwinienko and Ingold
Determination of RH2 by IR. The method has been de-
scribed previously.12,13 A Shimadzu FTIR 8201PC apparatus
with a 1.03 mm CaF2 cell was employed, and a baseline
correction was made using the same concentration of the HBA
in CCl4. Values of [BIS]free were determined from the intensi-
ties of the band at 3632 cm-1 (resolution 1 cm-1, 20 scans).
The detailed procedure and the tables and figures for BIS and
for ButMe2-phenol, OHfree ) 3618 cm-1 (OH group pointing
toward the 6-methyl substituent) and 3650 cm-1 (OH group
pointing toward the 2-tert-butyl substituent)31 are given in the
Supporting Information.
necessary condition for the overall kinetics to be zero-
order in [dpph•], that is, kHPTS+ must be ,kET[Y•] (see
/X-
eq 10 and Scheme 2). We therefore tentatively suggest
that the formation of an intramolecular H-bond in this
BIS monoanion is the reason that the BIS/dpph• reaction
exhibits zero-order kinetics in a few neat solvents.50
The discovery of ArOH/dpph• reactions that follow
zero-order kinetics in dpph• in certain solvents provides
strong support for the occurrence of the SPLET mecha-
nism in some of these systems. Whether such kinetics
are also obtained for any ArOH/ROO• reaction in biologi-
cal systems remains to be determined.
Acknowledgment. We are extremely grateful to
several anonymous reviewers for their thoughtful and
helpful comments and criticisms. G.L. thanks the
Department of Chemistry, Warsaw University, for
Grants BW-1637/06/04 and BW-1681/07/05.
Experimental Section
Materials. Because phenol/dpph• reaction kinetics in metha-
nol, ethanol, acetone, and acetonitrile are sensitive to traces
of acids and bases, these solvents were fractionally distilled
over a small amount of dpph• and a few beads of an ion-
exchange resin prior to use. All other solvents were of the
highest commercially available purity and were used as
received. Commercial BIS was recrystallized twice from n-
hexane, and the crystals were dried under vacuum.
Kinetic Measurements. These were made following the
procedure described previously.12,13 Decays of dpph• (initial
concentrations 3-8 × 10-5 M) in the presence of excess BIS
at known concentrations were monitored at 517 nm on an
Applied Photophysics stopped-flow spectrophotometer, SX 18
MV, equipped with a 150 W xenon lamp at ambient temper-
ature. The rate constants presented in Tables 1 and 2 are mean
values from at least two independent sets of measurements
of several pseudo-first-order rate constants, kexptl. Values of
kS were calculated using eq 7 in the Results. For additional
details, see the Supporting Information.
Supporting Information Available: Detailed kinetic
data for reactions of BIS with dpph• in 20 solvents and in
seven acidified solvents (Tables S1-S19), for the reaction of
ButMe2-phenol in heptane, ethanol, and acetone (Table S20),
for the reaction of Me3-phenol in 14 solvents (Table S21) and
plot of the logarithms of the Me3-phenol kinetic data against
âH2 (Figure S2), plots of dpph• decay in its reaction with BIS
in neat acetonitrile and in acidified acetonitrile (Figure S1),
plots and parameters used for calculation of HB equilibrium
constants, K, between BIS and DMSO and acetonitrile plus
calculated RH2 parameters (Figures S3, S4 and Tables S22-
S24), first-order plots of the dpph•/ButMe2-phenol reaction in
neat acetone (Figure S5), IR plots and parameters used for
calculation of HB equilibrium constants, K and RH2 for But-
Me2-phenol versus DMSO (Figures S6, S7 and Table S25),
plots of kinetic traces of dpph• decay in its reaction with BIS
in neat DMSO and DMSO containing 10, 100, and 1000 mM
acetic acid (Figure S8), plots of dpph• decay in its reaction
with BIS in acetone that were used for the determination of
the rates of BIS ionization (Figure S9), and plots of dpph•
decay in its reaction with BIS in acetonitrile (Figure S10). This
material is available free of charge via the Internet at
http://pubs.acs.org.
(49) For example, the first Ka for 2,2′-methylenebis(4-methyl)phenol
dissociation in methanol is 3.3 × 10-11 M (see: Takemura, H. J.
Inclusion Phenom. 2002, 42, 169-186.)
(50) The possibility that the zero-order kinetics arise from some
other conformational property that is unique to BIS or its monoanion
cannot be ruled out.
JO051474P
8990 J. Org. Chem., Vol. 70, No. 22, 2005