Mechanism and reactivity in perborate oxidation of anilines in acetic acid

Article abstract of DOI:10.1039/b208199g

Perborate but not percarbonate in acetic acid generates peracetic acid on standing and the peracetic acid oxidation of anilines is fast. The oxidation with a fresh solution of perborate in acetic acid is smooth and second order but the specific oxidation rate increases with increasing [perborate]₀ or [boric acid]. Perborate on dissolution affords hydrogen peroxide and a borate; the latter assists the former in the oxidation. The oxidation rates of anilines under identical conditions do not conform to any of the linear free energy relationships but the reaction rates of molecular anilines do. Perborate oxidation proceeds via two reaction paths but the overall oxidation rates of molecular anilines conform to structure reactivity relationships; the transition states do not differ significantly. Analysis of the oxidation rates of perborate and percarbonate reveals that while perborate oxidation is faster than percarbonate it is at least as selective as the latter.

Full text of DOI:10.1039/b208199g

View Article Online / Journal Homepage / Table of Contents for this issue
Mechanism and reactivity in perborate oxidation of anilines in
acetic acid
Chockalingam Karunakaran* and Ramasamy Kamalam
2
Department of Chemistry, Annamalai University, Annamalainagar, 608002, India.
E-mail: karunakaranc@rediffmail.com
Received (in Cambridge, UK) 21st August 2002, Accepted 13th September 2002
First published as an Advance Article on the web 15th October 2002
Perborate but not percarbonate in acetic acid generates peracetic acid on standing and the peracetic acid oxidation
of anilines is fast. The oxidation with a fresh solution of perborate in acetic acid is smooth and second order but the
specific oxidation rate increases with increasing [perborate]0 or [boric acid]. Perborate on dissolution affords hydrogen
peroxide and a borate; the latter assists the former in the oxidation. The oxidation rates of anilines under identical
conditions do not conform to any of the linear free energy relationships but the reaction rates of molecular anilines
do. Perborate oxidation proceeds via two reaction paths but the overall oxidation rates of molecular anilines conform
to structure–reactivity relationships; the transition states do not differ significantly. Analysis of the oxidation rates of
perborate and percarbonate reveals that while perborate oxidation is faster than percarbonate it is at least as selective
as the latter.
acid. On mixing the aged solution of perborate in glacial
Introduction
acetic acid with aniline in acetic acid, there is a rapid drop of
Sodium perborate, a peroxo salt of anionic formula: B2(O2)2-
titre with simultaneous formation of azobenzene followed
(OH)42Ϫ, is a cheap, non-toxic, stable, easily handled, large-scale
industrial chemical, extensively used in detergents as a bleach-
by a smooth slow oxidation. The UV–visible spectra of the
reaction solution confirm the formation of azobenzene on
the addition of the aged solution of perborate in acetic acid
to solution of aniline in acetic acid, and also during the course
ing agent. In glacial acetic acid it is a highly effective reagent,
the oxidant of choice,1–3 and oxidizes anilines to azobenzenes,
used in the manufacture of azo dyes, in good yield.4,5 The liter-
of the smooth reaction. With aging of perborate solution
ature lacks reports on the mechanism of perborate oxidation in
in acetic acid, the fraction of the oxidation that occurs rapidly
glacial acetic acid and hence this study. Perborate in aqueous
solution yields hydrogen peroxide and the kinetic studies in
aqueous and partly aqueous acidic media confirm perborate
on mixing the reactants increases but the specific rate of
the remaining part of the oxidation is constant: a unique
observation (Fig. 1). For example, the approximate percen-
oxidation as hydrogen peroxide oxidation.6–9 At pH 8.4–10.5
tages of the oxidation that occurs rapidly on mixing are
peroxoborate anion also acts as active oxidant.10 Operation of
20, 45 and 70 at 0.67, 1.33 and 2.0 h of aging, respectively; the
linear free energy relationships in the oxidation of anilines by a
oxidation with perborate solution aged 6 h or more is
variety of reagents is well known but the perborate oxidation
instantaneous.
rates of anilines under identical conditions do not conform to
any of the structure-reactivity relationships. However, the vari-
ation of the reaction rates of molecular anilines is in agreement
with the usual Hammett equation. Like sodium perborate,
sodium percarbonate is also an inexpensive, innocuous and
versatile oxidizing agent being used in functional group
oxidations.1–3 Although these reagents are being employed
interchangeably in detergent industry the chemistry of these
two peroxygen compounds is distinct in several aspects and so is
in organic synthesis.1–3 Examination of the reactivity and select-
ivity of these reagents under identical conditions show perbo-
rate as more reactive than percarbonate and its selectivity is not
less than that of the latter.
Results and discussion
Aging effect of perborate in acetic acid
The kinetics of the oxidation, studied in glacial acetic acid
under the condition [aniline] ӷ [perborate], were followed by
iodometric estimation of the unconsumed oxidizing agent. The
oxidation is sluggish at room temperature but is smooth at
45–65 ЊC. The decomposition of the oxidant is negligible up to
60 ЊC and the specific rate of decomposition at 65 ЊC, measured
as 1.2 × 10Ϫ5 sϪ1, is also negligible in comparison with the oxid-
ation rate. Studies on the perborate oxidation of aniline in
acetic acid reveal the aging effect of perborate in acetic
Fig. 1 Aging effect of perborate in acetic acid. Oxidation of aniline
(102[perborate]0 = 1.0 mol dmϪ3, [aniline]0 = 0.50 mol dmϪ3, medium:
acetic acid, 60 ЊC).
DOI: 10.1039/b208199g
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018
This journal is © The Royal Society of Chemistry 2002
2011

View Article Online
Reaction order
dmϪ3, 60 ЊC). The oxidation occurs under anhydrous condition
but not in aqueous acetic acid.
The oxidation, under the condition: [aniline] ӷ [perborate] is
first order with respect to the oxidant. The kinetics of the
reactions, using fresh solution of perborate in acetic acid, was
followed from 15 to 80% consumption of the oxidizing agent
and the plots of log titre versus reaction time are linear. The
least squares slopes of the linear plots afford the pseudo-first-
order rate constants, kЈ and the rate constants are reproducible
to 4%. The reaction under the condition: [aniline] ӷ [per-
borate] was also followed spectrophotometrically at 429 nm, the
λmax of azobenzene, and the product formation also conforms
to pseudo-first-order kinetics (Fig. 2: number of data points,
It is interesting and also surprising to note that although the
reaction is first order in the oxidizing agent, the pseudo-
first-order rate constant increases with increasing initial
concentration of perborate, [perborate]0; a plot of kЈ versus
[perborate]0 is a straight line with a finite y-intercept (Fig. 3). In
Fig. 3 Variation of kЈ with [perborate]0 ([Aniline]0 = 0.50 mol dmϪ3
medium: acetic acid, ᭺ 60 ЊC, ᭹ 65 ЊC).
,
an attempt to unravel the complexity of the reaction order with
respect to perborate, the influence of boric acid on the oxidation
rate was examined (102[perborate]0 = 1.0 mol dmϪ3, [aniline]0 =
0.5 mol dmϪ3, 60 ЊC). Boric acid enhances the oxidation;
initial addition of orthoboric acid increases the oxidation rate
and a plot of kЈ versus the sum of the concentrations of per-
borate and boric acid ([perborate]0 ϩ [boric acid]) is a straight
line with a definite y-intercept (Fig. 4). The oxidation is first
order with respect to aniline. At fixed [perborate]0, the oxid-
ation rate increases with increasing [aniline]0 and the plot of
kЈ versus [aniline]0 is a straight line almost passing through the
origin (Fig. 5). The reaction is not by a radical pathway; addi-
tion of vinyl monomer, acrylonitrile to the reaction solution
(0.10 mol dmϪ3) does not suppress the reaction. Also, the added
vinyl monomer fails to polymerize during the course of the
reaction. These results lead to the experimental rate law:
Fig. 2 Spectrophotometric study of oxidation of aniline with fresh
solution of perborate in acetic acid (103[perborate]0 = 1.25 mol dmϪ3
[aniline]0 = 0.50 mol dmϪ3, medium: acetic acid, 50 ЊC).
,
n = 175, correlation coefficient, r = 0.9997, standard error,
s = 3.5 × 10Ϫ3). The reaction was allowed to go to completion,
the absorbance was measured at different time intervals as well
as at the completion of the reaction and the concentrations
of the unconsumed oxidizing agent at different reaction times
obtained. The spectrophotometric study was made at a very low
concentration of perborate, a concentration that is too low for
iodometric estimation, to ensure that the absorbance of the
product formed is within the Beer–Lambert law limit. Hence
the reaction rates obtained by the two methods could not be
compared.
Ϫd[oxidizing agent]/dt =
(A ϩ B[borate])[aniline][oxidizing agent]
The solubility of sodium perborate in a variety of organic
solvents was tested; it is insoluble in methanol, ethanol,
2-propanol, t-butanol, dimethylformamide, dioxane, aceto-
nitrile, 2-ethoxyethanol, 2-butoxyethanol, 2-methoxypentan-
2,4-diol and glycerol but dissolves readily in glacial acetic acid
and less so in 1,2-diols like ethylene glycol and propylene glycol.
Use of ethylene glycol or propylene glycol along with acetic
acid is neither convenient nor cost effective; addition of high
boiling ethylene glycol or propylene glycol to acetic acid slows
down the oxidation, reduces the yield and makes the isolation
of the product difficult. Also, experiments show glacial acetic
acid as the best solvent for the oxidation. The reaction rate
decreases with decreasing content of acetic acid; each kinetic
run was made with fresh solution of perborate in acetic acid.
For example, on decreasing the content of acetic acid in the
reaction medium to 70% v/v, with the addition of methanol, the
specific oxidation rate (kЈ) decreases from 6.3 × 10Ϫ4 to 0.89 ×
10Ϫ4 sϪ1 (102[perborate]0 = 1.0 mol dmϪ3, [aniline]0 = 0.5 mol
where [borate] = [perborate]0 or [boric acid], and A and B are
constants.
Oxidation by hydrogen peroxide and peracetic acid
The oxidation of aniline with anhydrous hydrogen peroxide,
using sodium percarbonate (Na2CO3ؒ1.5H2O2) as the source of
dry hydrogen peroxide,11–14 under the conditions of perborate
oxidation, is slower than the perborate oxidation;15 the presence
of hydrogen peroxide in a solution of percarbonate in glacial
acetic acid was also confirmed by chemical tests. The oxidation
of aniline by anhydrous hydrogen peroxide in glacial acetic
acid, under the condition [aniline]0 ӷ [H2O2]0 is first order with
respect to hydrogen peroxide; plots of log [oxidant] versus reac-
tion time are linear. The least squares slope of the linear plot
yields the pseudo-first-order rate constant that remains con-
stant (2.3 × 10Ϫ4 sϪ1 at [aniline]0 = 0.50 mol dmϪ3 and 60 ЊC) at
2012
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018

View Article Online
absence of peracetic acid in fresh solutions of perborate in
acetic acid. Kinetic experiments with fresh solutions of per-
acetic acid in acetic acid show that peracetic acid oxidation of
aniline, under the reported conditions of perborate oxidation, is
almost instantaneous. Hence, the aging effect is the slow form-
ation of peracetic acid from perborate in acetic acid; on mixing,
aniline is oxidized almost instantaneously by the peracetic acid
formed. Experiments show that the generation of peracetic acid
on aging of perborate solution in glacial acetic acid is very
much slower than the oxidation of aniline with a fresh solution
of perborate in acetic acid.
The selective solubility of sodium perborate in acetic acid,
ethylene glycol and propylene glycol suggests breakdown of the
dimeric structure, B2(O2)2(OH)42Ϫ, on dissolution. In acetic acid
perborate is unlikely to exist as the peroxoborate anion,
(HO)3B(O2H)Ϫ. With 1,2-diols, boric acid in aqueous solution
forms glycol borates. Similarly, formation of a glycol borate
along with the generation of hydrogen peroxide on dissolution
of perborate in ethylene glycol and propylene glycol is possible.
Chemical tests confirm the formation of hydrogen peroxide on
dissolution of perborate in acetic acid, ethylene glycol and pro-
pylene glycol; but attempts to identify the boron species in the
reaction medium by GC-mass spectrometry and cyclic voltam-
metry were unsuccessful. Analytical experiments show that on
dissolution of perborate in glacial acetic acid, perborate is
quantitatively converted into hydrogen peroxide. But the kinetic
studies reveal that the perborate oxidation is faster than the
oxidation by anhydrous hydrogen peroxide (percarbonate). The
experimental rate law points out that the perborate oxidation
proceeds via two paths; one is independent of the initial concen-
tration of perborate or the concentration of added boric acid
and the other exhibits first order dependence on [perborate]0 or
added [H3BO3]. An examination of the rates of oxidation by
perborate and anhydrous hydrogen peroxide (percarbonate)
reveals that the perborate oxidation that is independent of the
initial concentration of perborate is hydrogen peroxide oxid-
ation. The extrapolated specific rate of the perborate oxidation
(kЈ) to zero initial concentration of perborate ([perborate]0 = 0)
is the same as the specific rate of the hydrogen peroxide (per-
carbonate) oxidation. The oxidation that shows first order
dependence on [perborate]0 or added [H3BO3] is also hydrogen
peroxide oxidation but mediated by borate (Scheme 1). Ogata
Fig. 4 Influence of boric acid on kЈ (102[perborate]0 = 1.0 mol dmϪ3
[aniline]0 = 0.50 mol dmϪ3, medium: acetic acid, 60 ЊC).
,
Fig. 5 First order dependence of rate on [aniline] (102[perborate]0 =
1.0 mol dmϪ3, medium: acetic acid).
Scheme 1
different initial concentrations of hydrogen peroxide (0.40 ×
10Ϫ2 to 1.8 × 10Ϫ2 mol dmϪ3). It is interesting to note that
although sodium perborate and sodium percarbonate are used
interchangeably in the detergent industry as well as in organic
synthesis, unlike the former the latter in glacial acetic acid does
not generate peracetic acid on aging. The product of hydrogen
peroxide oxidation is also azobenzene, identified by its mp,
mixed mp, IR, UV-visible and GC-mass spectra.15 Experi-
ments with peracetic acid as the oxidant, under the conditions
of perborate and hydrogen peroxide oxidations, show that per-
acetic acid oxidation of aniline is very fast; too fast to follow by
titrimetry. The product of the peracetic acid oxidation is also
azobenzene, identified by its mp, mixed mp, IR, UV–visible and
GC-mass spectra.
and Shimizu4 proffered that the oxidizing species of perborate
in glacial acetic acid is hydrogen peroxide coordinated with
boric acid but failed to observe an increase of kЈ with increasing
[perborate]0 and the generation of peracetic acid on standing of
perborate solution in acetic acid. Hydrogen peroxide associated
with borate is a better electrophile than molecular hydrogen
peroxide. The mechanism of oxidation of anilines by hydrogen
peroxide is classical. The electrophilic attack of the peroxide
oxygen on the amine nitrogen leading to formation of phenyl-
hydroxylamine is slow and rate determining.4 Phenylhydroxy-
lamine is further oxidized to nitrosobenzene which in turn
couples with aniline, present in a large excess, to yield azo-
benzene; the latter two reactions are fast. Scheme 1 differs from
the mechanism suggested by Ogata and Shimizu4 as it includes
oxidation by molecular hydrogen peroxide; kinetic studies on
the perborate oxidation of organic sulfides in glacial acetic acid
reveal perborate oxidation as hydrogen peroxide oxidation, the
rates of oxidation with perborate and dry hydrogen peroxide
Mechanism
Perborate in glacial acetic acid generates peracetic acid. Chem-
ical tests confirm the formation of peracetic acid on aging of
perborate solution in acetic acid. Also, chemical tests reveal the
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018
2013

View Article Online
(percarbonate) under identical conditions are identical.16
Methanol is a nucleophile, and association of methanol with
the electrophilic peroxide oxygen atom may be a reason for the
suppression of the oxidation with the addition of methanol.
The oxidation is sluggish in aqueous and partly aqueous media
and a possible reason is the reduced electrophilicity of hydrogen
peroxide due to the association of water molecules. Scheme 1
leads to the rate law:
conform to Charton’s LDS (localized, delocalized, steric)
equations; the carboxy, nitro and methoxycarbonyl substituents
individually take either planar or orthogonal orientations
requiring appropriate steric substituent constants. The acetyl
substituent was excluded due to the non-availability of the
steric substituent constant. The values of the substituent
constants used are cited elsewhere.15
Reaction rates of molecular anilines
Ϫd [oxidizing agent]/dt =
Reports on the kinetics of oxidation of anilines are numerous;
the reaction rates in basic, neutral, acidic and non-aqueous
media conform to the Hammett equation or to one of its modi-
fied forms. Anilines are present as free bases in basic and neu-
tral media but exist in two forms in acidic solutions, as free
bases and as conjugate acids. The ratio of the concentrations of
the free base to the conjugate acid ([XC6H4NH2]/[XC6H4NH3ϩ])
depends on the pKa of the aniline and the acidity of the
medium. The hitherto reported oxidations of anilines were
studied under pseudo-first-order conditions of [anilines] ӷ
[oxidant], and [oxidants] at different reaction times were esti-
mated by titrimetry or spectrophotometry. The pseudo-first-
order rate constants (kЈ) were obtained from the least squares
slopes of log[oxidant] versus time plots and the second order
rate constant = kЈ/[aniline]T where [aniline]T is the total concen-
tration of aniline. Since the pKa varies from 5.36 (p-OCH3) to
Ϫ0.28 (o-NO2) and molecular anilines are the easily oxidizable
species (nucleophiles) the reported kЈ/[aniline]T values are not
the rate constants of the oxidant–molecular aniline reactions.
Furthermore, the analysis of structure-reactivity relationships
using kЈ/[aniline]T is erroneous. In the reactions of anilines in
acid medium the specific reaction rates of anilines are to be
obtained using the concentrations of the free bases and not the
total concentrations of anilines. The concentrations of the free
bases may be deduced from the acid strength of the medium
and the pKa values of the anilines.15 Although the pKa values
used correspond to those in aqueous solutions, a detailed exam-
ination reveals that they may be used to obtain the concen-
trations of the free bases even in glacial acetic acid. The ratio of
the ionization constant of acetic acid to that of anilinium ion
2(k1 ϩ Kk2[perborate]0)[aniline][oxidizing agent]
with a pseudo-first-order rate constant of kЈ = 2(k1 ϩ Kk2-
[perborate]0)[aniline].
The rate constants obtained from the slopes and intercepts of
the kinetic plots are 104k1 = 2.4, 2.8 dm3 molϪ1 sϪ1; 102Kk2 = 3.7,
4.8 dm6 molϪ2 sϪ1 at 60 and 65 ЊC, respectively.
Hammett correlation of reactivity
Application of the Hammett correlation to reactions proceed-
ing through two competing mechanisms results in non-linear
Hammett plots; one such case is the acetolysis of threo-3-aryl-2-
butyl brosylates (brosylate = 4-bromobenzene sulfonate) which
proceeds via two reaction paths, competition between form-
ation of phenonium ions and the solvent assisted process.17 The
perborate oxidation of anilines proceeds via two paths but in
both cases the oxidizing species is hydrogen peroxide. Hence it
is of interest to examine the structure-reactivity relationships in
this reaction and 31 anilines were used for this study. The per-
borate oxidation rates of para- and meta-substituted anilines in
acetic acid do not conform to the usual Hammett equation at
any of the temperatures measured (eg., Fig. 6). In anilines, the
results in the equilibrium constant of the reaction PhNH2 ϩ
PhNH3 ϩ OAcϪ. In the absence of water auto-
ϩ
HOAc
ionization of acetic acid is unlikely and hence [PhNH3ϩ] =
[OAcϪ]; [PhNH3ϩ] ϩ [PhNH2] = [PhNH2]T . Solving the quad-
ratic expression on [PhNH2] yields [PhNH2]. Since the pKa
values of anilines at the experimental temperatures are unavail-
able, as an approximation, the pKa values at 25 ЊC were used in
the calculation; the pKa values employed are given elsewhere.15
The specific reaction rates of molecular anilines in glacial acetic
acid {k = kЈ/[XC6H4NH2] = 2(k1 ϩ Kk2[perborate]0); [per-
borate]0 = 1.00 × 10Ϫ2 mol dmϪ3 } thus obtained conform to
the usual Hammett equation at all the temperatures studied;
Fig. 7 is the Hammett plot at 45 ЊC (r = 0.98, s = 0.21, n = 17,
ρ = Ϫ2.9 at 45–65 ЊC). The large deviation of p-amino-
benzoic acid is likely due to complication of zwitterion form-
ation not allowed for in the calculation of the concentration
of the molecular form of the aniline. The meta-acid does not
deviate significantly and
a possible explanation is that
Fig. 6 The Hammett plot with the rate constants obtained under
identical conditions at 45 ЊC.
zwitterion formation is insignificant. The reaction constant is
close to that of the hydrogen peroxide (percarbonate) oxid-
ation of molecular anilines in glacial acetic acid under identical
conditions (ρ = Ϫ2.7 0.1 at 45–65 ЊC).15
The specific oxidation rates of ortho-substituted anilines
were analyzed in terms of Charton’s LDS equations. The oxid-
ation of o-mercaptoaniline is instantaneous. The UV–visible
spectra of the reaction solution after the completion of
the reaction show absence of formation of azo compound; the
reaction is not the oxidation of the amino group but that of
the thiol. The reaction of o-aminophenol is also too fast to
measure; o-aminophenol enters into intramolecular hydrogen
reaction center is likely to cross-conjugate with the para-
substituents and correlations of the oxidation rates of
para- and meta-substituted anilines separately using the usual
Hammett equation (σp or σm) or the Brown–Okamoto equation
(σpϩ or σmϩ) or the modified Hammett equation (σpϪ) were also
unsuccessful. The oxidation rates measured at 45, 55 and 65 ЊC
were also analyzed in terms of the dual substituent parameter
(DSP) equations (para: σI, σR; σI, σRϩ; σI, σRϪ; F, R; meta: σI, σR;
σI, σRϩ; σI, σRϪ; F, R)18 but this analysis was unsuccessful.
The oxidation rates of ortho-substituted anilines also fail to
2014
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018

View Article Online
where the temperature was 45 ЊC; n = 8, 100R2 = 98.6,
s = 0.156.15
Separation of reaction rates
The satisfactory linear Hammett correlation observed in the
perborate oxidation of anilines in glacial acetic acid that occurs
through two competing reaction pathways is surprising.17
A
possible explanation is that the transition states of both the
competing mechanisms do not differ significantly and so are the
reaction constants, ρ. This could be confirmed by analyzing
individually the specific rates of hydrogen peroxide (percar-
bonate) oxidation of molecular anilines15 and the specific rates
of borate assisted hydrogen peroxide oxidation of molecular
anilines, the latter may be obtained from the rates of oxidation
of molecular anilines by perborate (this work) and by hydrogen
peroxide (percarbonate);15 k(H2O2 ؒ ؒ ؒ borate) = k(perborate) Ϫ
k(percarbonate). The specific rates of borate assisted hydrogen
peroxide oxidation of molecular anilines obtained thus con-
form to the usual Hammett equation at all the temperatures
studied; Fig. 8 is the Hammett plot at 55 ЊC (r = 0.98, s = 0.23,
Fig. 7 The Hammett plot using the rate constants of molecular
anilines at 45 ЊC.
bonding. o-Phenylenediamine in glacial acetic acid may exist in
ϩ
three forms, viz., o-ϩNH3C6H4NH3ϩ, o-NH2C6H4NH3 and
o-NH2C6H4NH2. Calculation of the individual concentrations,
using the pKa values, shows that o-phenylenediamine exists
predominantly as o-NH2C6H4NH3ϩ, the other two forms are
present in only trace amounts. The protonated amino group
(NH3ϩ-) is highly electron-demanding, the activation energy
for the oxidation of the conjugated acid is likely to be high and
hence the free base is the probable reactive species. The k values
reported in Table 1 correspond to this species. The oxidation of
molecular o-phenylenediamine encounters two identical reac-
tion centers; one acts as the substituent and the other as the
reaction site. Hence only half of the rate is to be taken into
account for the correlation analysis. Because of these compli-
cations and also that the reported reaction rates are only the
extrapolated values o-phenylenediamine was excluded from
the regression. Correlation analysis of the specific rates of
the ortho-substituted molecular anilines in terms of the LDS
equations gave results ranging from fairly satisfactory to rather
unsatisfactory. The values of the substituent constants used
are given elsewhere.15 Each data set involved 8 points (methyl,
methoxy, ethoxy, chloro, nitro, carboxy, methoxycarbonyl and
hydrogen), acetyl being omitted due to the unavailability of the
Fig. 8 The Hammett plot for the borate assisted hydrogen peroxide
oxidation of molecular anilines at 55 ЊC.
n = 17, ρ = Ϫ3.0). The reaction constant is close to that of the
hydrogen peroxide (percarbonate) oxidation of molecular ani-
lines in glacial acetic acid, obtained under identical conditions
Ϫ
steric parameter. Table 2 reveals that σR is less satisfactory in
explaining the variation of the rate with the substituent. This is
in accordance with the literature on the oxidation of anilines;
most of the satisfactory correlations involve either σ or σpϩ. The
results obtained by using σI and σRϩ were fairly better and still
so when it was assumed that the carboxy and nitro groups
were in the orthogonal and methoxycarbonyl substituent in the
planar orientations. One of the most satisfactory correlation
equations was as follows:
(ρ = Ϫ2.7
0.1).15 The correlation of the rates of borate
assisted oxidation in terms of Charton’s LDS equations is
less satisfactory and is explained as follows. The ortho-series
includes strongly electron-withdrawing substituents like nitro,
carboxyl and methoxycarbonyl, and the oxidations of these
substrates by percarbonate and perborate are slow. The error in
the rates of borate assisted oxidation that is obtained from the
difference in the measured rates of perborate and percarbonate
oxidations is likely to be significant. This error could be a
reason for the less satisfactory correlation.
log k = Ϫ1.99( 0.18) Ϫ
3.14( 0.38)σI Ϫ 0.91( 0.17)σRϩ Ϫ 0.48( 0.20)υ
where the temperature was 45 ЊC; n = 8, 100R2 = 97.1, s = 0.222.
The regression coefficients yield the percentages of inductive
(σI), resonance (σRϩ) and steric (υ) terms as 69, 20 and 11,
respectively at 45 ЊC. These results are comparable to those of
the hydrogen peroxide (percarbonate) oxidation of anilines in
glacial acetic acid under identical conditions:
Reactivity and selectivity of perborate and percarbonate
oxidations
Sodium perborate (NaBO3ؒ4H2O) and sodium percarbonate
(Na2CO3ؒ1.5H2O2), two cheap, non-toxic, stable, safe, large-
scale industrial chemicals primarily employed in the detergent
industry, now find extensive usage in organic synthesis.1–3 The
oxidations of anilines by perborate (this work) and per-
carbonate15 in glacial acetic acid involve molecular anilines and
logk = Ϫ2.28( 0.15) Ϫ
2.90( 0.28)σI Ϫ 1.15( 0.12)σRϩ Ϫ 1.18( 0.35)υ
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018
2015

View Article Online
Table 1 Pseudo-first-order rate constants (kЈ) and specific oxidation rates of molecular anilines (k) in glacial acetic acida
104kЈ/sϪ1
45 ЊC
103k/dm3 molϪ1 sϪ1
Entry
Substituent
55 ЊC
65 ЊC
45 ЊC
55 ЊC
65 ЊC
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
H
m-CH3
m-OH
1.95
2.97
4.54
1.40
0.592
1.97
1.73
0.486
1.85
2.10
2.30
3.98
7.42
9.30
2.58
1.17
4.38
3.80
1.56
4.40
5.00
6.29
3.41
3.59
0.732
1.22
1.08
3.57
0.903
4.28
5.37
5.09
1.48
0.548
5.44
1.26
0.686
373c
21.1
4.95
7.78
13.7
17.0
5.99
2.16
9.26
8.08
3.04
9.30
11.8
16.5
6.15
6.49
2.25
2.19
2.99
10.4
1.81
7.90
10.7
10.2
3.02
1.59
10.2
2.42
1.26
1510c
55.7
9.37
11.3
21.8
14.3
1.12
0.184
0.936
35.4
0.209
30.9
66.0
56.3
2.48
2.17
0.0498
0.153
0.484
8.39
0.100
9.51
23.0
54.5
29.3
2.06
0.363
2.08
77.6
0.672
73.4
157
154
5.80
45.0
101
53.6
4.78
0.670
4.40
165
1.31
155
371
404
10.5
9.08
0.489
0.663
1.87
60.2
0.441
33.2
53.3
44.8
m-Cl
m-NO2
m-COOH
m,p-(CH3)2
m,p-Cl2
p-CH3
p-OCH3
p-OC2H5
p-Cl
1.46
p-Br
p-NO2
1.55
5.02
0.229b
0.507
0.774
1.45
0.159
0.369
0.676
20.7
0.220
18.0
26.8
22.3
0.507
0.112
1.45
0.373
0.192
p-COOH
p-COOC2H5
p-NHCOCH3
o,m-Cl2
o-CH3
o-OCH3
o-OC2H5
o-Cl
0.412
2.26
2.65
13.2
8.34
1.90
0.580
0.260
2.14
0.666
0.332
84.5c
8.87
2.35
Fast
Very fast
0.199
0.0530
0.571
0.197
0.0927
545
88.6
45.9
1.03
0.324
2.72
0.716
0.352
o-NO2
o-COOH
o-COOCH3
o-COCH3
o-NH2
-NHCH3
-N(CH3)2
o-OH
2400
211
96.7
9730
556
183
o-SH
a [perborate]0 = 0.010 mol dmϪ3; [aniline]0 = 0.50 mol dmϪ3
1.32, 7.62, and 38.7 at 20, 30, and 40 ЊC, respectively.
.
b Extrapolated using Arrhenius plot: 1.46 at 60 ЊC. c Extrapolated using Arrhenius plot:
the oxidizing species in the slow and rate-determining
steps. Also, the reactions conform to the Hammett equations.
Examination of the oxidation rates of perborate (Table 1) and
percarbonate15 reveals that perborate oxidation is faster than
percarbonate. Association of the electron-deficient borate
species with hydrogen peroxide leads to increased electro-
philicity. Highly reactive systems are often less selective and, if
amenable to correlation, show lower susceptibility than similar
less reactive systems. The oxidations of anilines by perborate
and percarbonate do not conform to the original reactivity–
selectivity relationship; although the reactivity of perborate is
higher than that of percarbonate the former does not show
lower selectivity, the reaction constants are Ϫ3.0 and Ϫ2.7,
respectively. Exner19 redefined and reexamined the reactivity–
selectivity principle (RSP) on an experimental basis, showing
that the original narrow definition of RSP is invalid, and
formulated the principle in terms of simple mathematical
expressions, involving only rate constants and avoiding any σ
constants. The rates of oxidation of molecular anilines by per-
borate and percarbonate at all the temperatures studied were
analyzed by the Exner method; the double logarithmic plots of
k(perborate) versus k(percarbonate) are straight lines (n = 28;
r = 0.992, 0.993, 0.992; s = 0.138, 0.130, 0.142; slope, b = 1.04,
1.01, 0.994; intercept, a = 0.438, 0.345, 0.311; mean difference,
∆ = 0.322, 0.328, 0.324, at 45, 55 and 65 ЊC, respectively). Fig. 9
is the RSP plot at 55 ЊC. Since the slope b is not significantly
different from unity, the RSP correlation shows indifferent
behavior, i.e., no change in selectivity. While perborate oxid-
ation is faster than percarbonate oxidation it is not less
selective than the latter. Also, perborate is of longer shelf life
and cheaper than percarbonate; its decomposition is less than
that of percarbonate. The experimentally measured reaction
rates (kЈ) of perborate and percarbonate oxidations under
Fig.
9
RSP plot of perborate and percarbonate oxidations of
molecular anilines at 55 ЊC.
identical conditions were also analyzed by the Exner method
but the results are less satisfactory (logkЈ(perborate)–logkЈ-
(percarbonate) correlations: n = 28; r = 0.933, 0.944, 0.934;
s = 0.141, 0.130, 0.140; slope, b = 1.08, 1.00, 0.934; intercept,
a = 0.669, 0.326, 0.093; ∆ = 0.322, 0.328, 0.324, at 45, 55 and
65 ЊC, respectively). This is likely to be a consequence of the
lack of Hammett correlation using kЈ.
2016
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018

View Article Online
Table 2 Charton’s LDS correlation of the rate constants of perborate oxidation of molecular ortho-substituted anilines.a
45 ЊC
55 ЊC
65 ЊC
Explanatory
variables
100R2
s
100R2
s
100R2
s
b
σI, σR
88.9
92.5
93.8
93.4
96.0
93.8
92.6
92.4
95.4
0.382
0.353
0.323
0.333
0.260
0.323
0.353
0.355
0.277
88.4
91.9
94.0
92.6
95.9
93.6
91.8
91.9
94.5
0.391
0.365
0.315
0.351
0.261
0.325
0.367
0.365
0.302
86.0
90.3
92.2
91.6
95.4
91.7
90.1
89.9
93.8
0.420
0.387
0.345
0.360
0.265
0.357
0.389
0.393
0.310
σI, σR, υc
σI, σR, υd
σI, σR, υe
σI, σR, υf
σI, σR, υg
σI, σR, υh
σI, σR, υi
σI, σR, υj
Ϫ b
σI, σR
85.6
87.1
89.3
87.7
91.0
89.3
87.0
87.2
89.6
0.436
0.464
0.424
0.453
0.387
0.423
0.466
0.463
0.416
85.3
86.5
89.7
86.9
91.0
89.6
86.4
87.0
88.8
0.440
0.472
0.412
0.465
0.385
0.415
0.473
0.463
0.430
83.1
84.7
87.8
85.7
90.5
87.6
84.5
84.6
87.9
0.460
0.484
0.432
0.469
0.382
0.436
0.488
0.486
0.431
σI, σRϪ, υc
σI, σRϪ, υd
σI, σRϪ, υe
σI, σRϪ, υf
σI, σRϪ, υg
σI, σRϪ, υh
σI, σRϪ, υi
σI, σRϪ, υj
ϩ b
σI, σR
88.6
93.3
94.8
94.3
97.1
93.4
92.9
92.9
96.8
0.389
0.333
0.294
0.309
0.222
0.331
0.345
0.345
0.232
88.3
93.1
95.5
93.8
97.4
93.5
92.6
92.6
96.2
0.392
0.338
0.272
0.320
0.208
0.329
0.350
0.350
0.250
85.9
91.7
93.9
93.1
97.1
91.7
90.8
90.8
95.8
0.421
0.357
0.307
0.326
0.210
0.358
0.376
0.376
0.253
σI, σRϩ, υc
σI, σRϩ, υd
σI, σRϩ, υe
σI, σRϩ, υf
σI, σRϩ, υg
σI, σRϩ, υh
σI, σRϩ, υi
σI, σRϩ, υj
F, Rb
92.7
93.3
94.0
94.4
96.2
94.0
93.6
93.3
96.4
0.311
0.334
0.318
0.305
0.252
0.316
0.328
0.334
0.245
92.5
93.2
94.5
94.1
96.6
94.3
93.2
93.1
96.0
0.314
0.335
0.300
0.313
0.237
0.307
0.334
0.338
0.258
90.3
91.5
92.7
93.0
96.0
92.3
91.5
91.2
95.2
0.348
0.362
0.334
0.327
0.247
0.344
0.361
0.367
0.270
F, R, υc
F, R, υd
F, R, υe
F, R, υf
F, R, υg
F, R, υh
F, R, υi
F, R, υj
a n = 8 (o-NH2 and o-COCH3 excluded); 100R2: the percentage of variation. b n = 9, o-COCH3 included. c COOH: planar, NO2: planar, COOCH3:
planar; d COOH: orthogonal, NO2: planar, COOCH3: planar; e COOH: planar, NO2: orthogonal, COOCH3: planar; f COOH: orthogonal, NO2:
orthogonal, COOCH3: planar; g COOH: planar, NO2: planar, COOCH3: orthogonal; h COOH: orthogonal, NO2: planar, COOCH3: orthogonal;
i COOH: planar, NO2: orthogonal, COOCH3: orthogonal; j COOH: orthogonal, NO2: orthogonal, COOCH3: orthogonal.
over chromium() oxide for 6 h and distilled through a column.
Conclusions
Other chemicals used were also of AR or LR grade. Perborate
Perborate but not percarbonate in glacial acetic acid generates
and percarbonate were dissolved in glacial acetic acid, stand-
peracetic acid on standing. The oxidation of anilines with a
ardized iodometrically and used immediately. For the experi-
fresh solution of perborate is faster than with percarbonate.
ments with peracetic acid, a solution of peracetic acid in glacial
The Hammett correlation of the perborate oxidation rates of
acetic acid was prepared by the standard method, estimated
anilines in acetic acid under identical conditions is unsatisfac-
iodometrically and also used immediately. The kinetics of
tory but the specific reaction rates of molecular anilines con-
the oxidation at fixed temperature were followed by mixing
form to the linear free energy relationships. The perborate
required volumes of the oxidant and aniline in glacial acetic
oxidation occurs via two reaction paths but the over all reaction
acid and estimating the unreacted oxidizing agent at different
rates of molecular anilines conform to the structure–reactivity
relationships, and this is because the transition states do not
reaction times iodometrically. Perborate (0.01 mol) was added
to aniline (0.01 mol) in acetic acid at 65 ЊC, diluted with water
differ significantly. While the perborate oxidation is faster than
after 4 h; the separated solid was filtered, recrystallized from
the percarbonate oxidation the selectivity of the former is not
less than that of the latter.
petroleum ether (40–60) and identified as azobenzene by its
mp, mixed mp, IR, UV–visible and GC-mass spectra. The
spectra were identical with those of an authentic sample and
the GC-mass spectral study showed azobenzene as the only
product; the yield was not less than 90%.5 The UV-visible
spectra of the reaction solutions, during and after the comple-
tion of the reaction, of all the anilines studied except
p-mercaptoaniline, N-methylaniline and N,N-dimethylaniline
revealed the formation of azobenzenes. Hence the reaction is
represented as:
Experimental
General procedure
Sodium perborate, NaBO3.4H2O and sodium percarbonate,
Na2CO3. 1.5H2O2 (Fluka) were used as such. Anilines (AR or
LR) were redistilled or recrystallized prior to use and the
identity of the anilines used was confirmed by their physical
constants as reported earlier.15 Acetic acid, AR was refluxed
2PhNH2 ϩ 2NaBO3
PhN = NPh ϩ 2NaBO2 ϩ 2H2O
2017
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018

View Article Online
11 M. A. A. F. de C. T. Carrondo, W. P. Griffith, D. P. Jones and
A. C. Skapski, J. Chem. Soc., Dalton. Trans, 1977, 2323.
12 J. M. Adams and R. G. Pritchard, Acta Crystallogr., Sect. B, 1977,
B33, 3650.
13 J. M. Adams, R. G. Pritchard and A. W. Hewat, Acta Crystallogr.,
Sect. B, 1979, B35, 1759.
References
1 A. McKillop and W. R. Sanderson, J. Chem. Soc., Perkin Trans. 1,
2000, 471.
2 A. McKillop and W. R. Sanderson, Tetrahedron, 1995, 51, 6145.
3 J. Muzart, Synthesis, 1995, 1325.
4 Y. Ogata and H. Shimizu, Bull. Chem. Soc. Jpn., 1979, 52, 635.
5 S. M. Mehta and M. V. Vakilwala, J. Am. Chem. Soc., 1952, 74, 563.
6 C. Karunakaran, V. Ramachandran and P. N. Palanisamy, Int. J.
Chem. Kinet., 1999, 31, 675.
7 C. Karunakaran and P. N. Palanisamy, Synth. React. Inorg.
Met.-Org. Chem., 1998, 28, 1115.
14 D. P. Jones and W. P. Griffith, J. Chem. Soc., Dalton. Trans., 1980,
2526.
15 C. Karunakaran and R. Kamalam, J. Org. Chem., 2002, 67,
1118.
16 C. Karunakaran and R. Kamalam, Eur. J. Org. Chem., 2000,
3261.
8 C. Karunakaran and B. Muthukumaran, React. Kinet. Catal. Lett.,
1997, 60, 387.
9 C. Karunakaran and B. Muthukumaran, Transition Met. Chem.
(London), 1995, 20, 460.
10 C. Karunakaran and R. Kamalam, Synth. React. Inorg. Met.-Org.
Chem., 1999, 29, 1463.
17 H. C. Brown, C. J. Kim, C. J. Lancelot and P. v. R. Schleyer,
J. Am. Chem. Soc., 1970, 92, 5244.
18 J. Shorter, in Similarity Models in Organic Chemistry, Biochemistry
and Related Fields, ed. R. I. Zalewski, T. M. Krygowski and
J. Shorter, Elsevier, Amsterdam, 1991, pp. 77–147.
19 O. Exner, J. Chem. Soc., Perkin Trans. 2, 1993, 973.
2018
J. Chem. Soc., Perkin Trans. 2, 2002, 2011–2018