Kinetics, mechanism, and products of the reaction of diphenylcarbonyl oxide with carboxylic acids

Article abstract of DOI:10.1023/A:1015851730981

The kinetics of the reactions of acetic, benzoic, formic, oxalic, malic, tartaric, trifluoroacetic, and hydrochloric acids with diphenylcarbonyl oxide Ph2COO was studied. The carbonyl oxide Ph2COO was generated by flash photolysis of diphenyldiazomethane Ph2CN2 in solutions of acetonitrile and benzene at 295 K. The apparent rate constants of the reaction range from 4.6 10⁸ for (COOH)2 in MeCN to 7.5 10⁹ L mol^-1 s^-1 for acetic acid in a benzene solution. The reaction mechanism was proposed, according to which at the first stage the carbonyl oxide is reversibly solvated by the solvent. Then the solvated carbonyl oxide reacts with the acid molecule by the mechanism of insertion at the O-H bond.

Full text of DOI:10.1023/A:1015851730981

608
Russian Chemical Bulletin, International Edition, Vol. 51, No. 4, pp. 608—615, April, 2002
Kinetics, mechanism, and products of the reaction
of diphenylcarbonyl oxide with carboxylic acids
A. M. Nazarov,aꢀ S. L. Khursan, E. M. Chainikova, and V. D. Komissarov
b
a
b
aInstitute of Organic Chemistry, Ufa Scientific Center of the Russian Academy of Sciences,
7
1 prosp. Oktyabrya, 450054 Ufa, Russian Federation.
Fax: (347 2) 35 6066
b
Bashkir State University,
3
2 ul. Frunze, 450074 Ufa, Russian Federation.
Tel: (347 2) 23 6727. Eꢀmail: KhursanSL@bsu.bashedu.ru
The kinetics of the reactions of acetic, benzoic, formic, oxalic, malic, tartaric,
trifluoroacetic, and hydrochloric acids with diphenylcarbonyl oxide Ph COO was studied.
2
The carbonyl oxide Ph COO was generated by flash photolysis of diphenyldiazomethane
2
Ph CN in solutions of acetonitrile and benzene at 295 К. The apparent rate constants of the
2
2
8
9
–1 –1
reaction range from 4.6•10 for (COOH) in MeCN to 7.5•10 L mol
s
for acetic acid in
2
a benzene solution. The reaction mechanism was proposed, according to which at the first
stage the carbonyl oxide is reversibly solvated by the solvent. Then the solvated carbonyl
oxide reacts with the acid molecule by the mechanism of insertion at the O—H bond.
Key words: carbonyl oxides, flash photolysis, carboxylic acids, kinetics of reactions.
Carbonyl oxides R COO are intermediates in several
oxidation processes, such as ozonolysis of unsaturated
by sublimation. Malic and tartaric (racemate) acids were reꢀ
crystallized from isopropyl alcohol. Acetonitrile was purified
2
1
3
according to a known procedure. Synthesis and purification
of Ph2CN2 were conducted according to a previously published
compounds and thermoꢀ and photoxidation of diazo comꢀ
pounds.1
—3
Unique electronic properties and high reacꢀ
1
4
method.
tivity of these intermediates attract permanent attention
to the chemistry of carbonyl oxides. We have previously
studied the kinetics of diphenylcarbonyl oxide recombiꢀ
Experiments were carried out at ∼20 °C. A solution of
Ph CN2 with a known concentration was subjected to flash
2
photolysis (FP) in a cell, and the kinetic curve of the absorꢀ
bance (A) decay of diphenylcarbonyl oxide at the maximum of
4
—7
nation in various solvents
and the kinetics of its reacꢀ
tions with organic substrates, viz., olefins, sulfoxides,
phenols, and amines.
In this work we studied by the combined flash phoꢀ
tolysis and timeꢀresolved spectrophotometry method
its absorption band was monitored (λ
= 410 nm, εmax =
max
8
—11
3
–1
–1 6
1.9•10 L mol cm ). Then a specified amount of an acid
was introduced into the solution, and the kinetics of carbonyl
oxide decay was monitored again. The procedure was repeated
several times varying the acid concentration. The obtained kiꢀ
netic curves were processed by nonlinear regression analysis.
The formation of the peroxide products of the reaction of
(
FP—TS) the regularities of Ph COO decay in the presꢀ
2
ence of acetic (1), benzoic (2), formic (3), oxalic (4),
malic (5), tartaric (6), trifluoroacetic (7), and hydroꢀ
chloric (8) acids.
Ph COO with AcOH was studied under the conditions of steadyꢀ
2
state diphenyldiazomethane oxidation photosensitized by meꢀ
thylene blue (MB). The MB dye was excited by the light with
λ > 560 nm (nitrogen incandescent lamp, power 500 W, OSꢀ13
light filter). The distance from the light source to the reactor
was 15 cm. A cylindrical reactor, whose temperature was mainꢀ
tained at 298 K, was charged with MeCN (15 mL), MB
(1•10–5 mol L ), and AcOH with a specified concentration.
Experimental
Kinetic studies were performed on an FP—TS setup with a
known design.12 The reactor was a quartz cell with the optical
length l = 10 cm and an inner diameter of ∼1 cm. Photolysis of
–1
diphenyldiazomethane Ph CN2 was carried out by the sepaꢀ
To prevent Ph COO consumption through the reaction with
2
2
rated light (UFSꢀ2 light filter, transmission region 270—380 nm).
diphenyldiazomethane, the latter was gradually added as it
To prevent Ph CN from decomposition under the light beam,
was consumed so that the Ph CN concentration in the soluꢀ
2 2
2
2
tion was at most 2•10–4 mol L . Air was bubbled through
the solution during photolysis. At the end of the reaction
(20—40 min) the reaction mixture was analyzed iodometrically
for the content of peroxide products.
–1
the absorption region of the diazo compound was attenuated
by an SSꢀ15 light filter (transmission region 300—520 nm).
Acetic, trifluoroacetic, and formic acids (reagent grade)
were used as received. Benzoic and oxalic acids were purified
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 4, pp. 565—571, April, 2002.
066ꢀ5285/02/5104ꢀ608 $27.00 © 2002 Plenum Publishing Corporation
1

Reaction of diphenylcarbonyl oxide with acids
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
A
609
Results and Discussion
ln(A0/A)
3.0
0
.02
During flash photolysis the following reactions ocꢀ
cur in a solution of Ph CN according to the known
2
2
.5
.0
2
2
2
4
—6,15,16
1
scheme
:
1
Ph CN → Ph C: + N ,
(1)
(2)
(3)
(4)
(5)
2
2
2
2
0
.01
1.5
1
3
Ph C: ↔ Ph C:,
2
2
1
0
.0
.5
3
Ph C: + O → Ph COO,
2
2
2
Ph COO + Ph CN → 2 Ph CO + N ,
2
2
2
2
2
0
500
1000
1500
t/s
Ph COO + Ph COO → 2 Ph CO + O .
2
2
2
2
Fig. 1. Typical kinetic curve (1) for Ph CN2 decomposition
during its irradiation with the light from a probing lamp and
2
When [Ph CN ] < 5•10 mol L–1, reaction (4) can
–
4
2
2 0
4
,6
be neglected, and the kinetics of the Ph COO absorꢀ
bance decay obeys the second order equation up to at
least 80%
the semilogarithmic anamorphosis of the kinetic curve (2). Solꢀ
2
–
4
–1
vent MeCN, [Ph CN ] = 2•10 mol L , [AcOH] = 0, 295 К,
2
2 0
0
light filter BSꢀ4, monitoring of the Ph CN concentration at
2
2
λ = 480 nm.
2
–
dA/dt = [2k /(εl)]A ,
(6)
5
shortꢀwavelength region of the ultraviolet spectrum was
attenuated by the BSꢀ4 light filter, which is transparent
in the region of the n → π* transition of the diazo group.
where ε is the molar absorption coefficient, and l is the
optical path length.
14
It is known that acids efficiently decompose diazo
The data in Table 1 indicate that in our experiments the
reaction rate of Ph CN with the acid is negligible. The
compounds.1
4
In addition, photolysis of the
2
2
Ph CN …...HOC(O)R complex can likely produce the
2
2
irradiation of the solution with the probing lamp light
substantially accelerates diphenyldiazomethane decomꢀ
position. However, its halfꢀdecay time remains rather
long compared to the duration of the pulse experiment
reaction products without the intermediate formation of
carbonyl oxide. The initial reactant concentrations and
the composition of the oxidation products can change
due to the indicated reactions, which impedes the interꢀ
pretation of kinetic experimental results. To estimate the
contribution of these reactions to the process, we studied
the kinetics of the reaction of diphenyldiazomethane with
acetic acid (Table 1) in the dark and under irradiation of
the solution with the light from a probing lamp. The
(
Fig. 1). In the presence of the acid, Ph CN photolysis
2 2
somewhat slows down. It is most likely that during the
photoexcitation of the diazo compound—acid complexes,
Ph CN can be deactivated by the excitation energy transꢀ
2
2
fer to the acid molecule to the vibrational levels of the
O—H oscillator, along with dissociation according to
reaction (1).
Table 1. Rate constants of diphenyldiazomethane deꢀ
Carbene Ph C formed by diphenyldiazomethane deꢀ
2
–
4
–1
composition ([Ph CN ] = 2•10 mol L ) in the
composition is either oxidized to the corresponding carꢀ
bonyl oxide (reaction (3)) or inserts at the O—H bond of
the acid molecule. The kinetic consequence of these
2
2 0
presence of AcOH during irradiation with the light
from a probing lamp (solvent MeCN, 295 К)
reactions is a decrease in the initial Ph COO concentraꢀ
2
3
–1
[
/
MeC(O)OH]
mol L
Light filter
k•10 /s
0
tion as the acid concentration increases (Fig. 2). The
reaction product, diphenylmethyl ester of the correspondꢀ
ing acid, is inert under experimental conditions. Thus,
side reactions initiated by the addition of the acid to a
solution of the diazo compound have no substantial efꢀ
fect on the kinetics and mechanism of the reactions of
diphenylcarbonyl oxide with acids.
–
1
.8•10–3
5
0
2
0
0
2
1
2
*
—
—
—
BSꢀ4
BSꢀ4
BSꢀ4
BSꢀ4
BSꢀ4
0.0055±0.0005
1.20±0.02
0.90±0.02
1.34±0.02
1.37±0.02
1.50±0.02
1.20±0.01
1.10±0.01
**
.2•10–4**
.2•10–
.1•10
.2•10
6
In the presence of even small amounts of the acid
–
–
5
4
∼10–5 mol L ) the rate of Ph COO consumption inꢀ
–1
(
2
creases sharply. At least three channels of interaction
can be proposed: the very fast, diffusionꢀcontrolled reꢀ
action of carbonyl oxide with acid (A) and the reactions
*
*
Without irradiation.
* Irradiation without light filters.

6
10
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
Nazarov et al.
5
–1
4
[
Ph COO] •10 /mol L
0.5
1.0
1.5
2.0
t•10 /s
2
0
A
1
1
0
0
0
0
.2
.0
.8
.6
.4
.2
0.12
0
0
.08
.04
1
2
–
5.6
–5.2
–4.8
logAcOH/mol L–1
Fig. 2. Initial absorbance of diphenylcarbonyl oxide as a funcꢀ
tion of the concentration of acetic acid in MeCN.
0
2
0
.4
0.8
1.2
t•10 /s
Fig. 3. Kinetic curves for the Ph COO decay absorbance
2
in the presence of benzoic acid in MeCN ([Ph CN ]
=
of the protonated Ph COO molecules with another carꢀ
bonyl oxide molecule (B) or with the initial diazo comꢀ
pound (C).
2
2 0
–1
2
–
4
–1
–7
1
•10 mol L ): 1, [PhC(O)OH] = 6.25•10 mol L ; and
0
–
5
–1
2
, [PhC(O)OH] = 1.4•10 mol L .
0
kinetic curve Ph COO recombination predominates, and
2
the presence of the fast region is caused by the reaction
of Ph COO with the acid characterized by the first order
2
with respect to the carbonyl oxide concentration
d[Ph COO]/dt =
2
=
k [RC(O)OH][Ph COO] + 2k [Ph COO],
(7)
b
2
5
2
where [RC(O)OH] is the acid concentration, and k is
b
the rate constant of the bimolecular reaction of Ph COO
2
with the acid. Going to absorbance, we obtain
2
(
–dA/dt)k [RC(O)OH]A + [2k /(εl)]A =
In the last two cases, the participation of a proton in
several steps of the reaction can explain the strong influꢀ
ence of the low acid concentration on the rate of carboꢀ
nyl oxide consumption. However, our finding that the
apparent rate constant of carbonyl oxide consumption
does not increase with an increase in the diazo comꢀ
pound concentration (see below) indicates a negligible
contribution of channel C to the reaction rate. The preꢀ
dominant occurrence of channel A is indicated by the
following experimental facts:
b
5
I
II 2
=
k A + k A ,
(8)
(9)
I
II
where k = k [RC(O)OH], k = 2k /(εl).
b
5
The analytical solution of Eq. (8) has the form
I
II
I
II
I
ln{(k + k A)A /[(k + k A )A]} = k t.
0
0
I
The k rate constant was calculated by Eq. (8) from
the initial fast region of the kinetic curve using k
II
=
–
1 4,6
2
k5/(εl) = 940 s
.
At rather high acid concentrations Ph COO is conꢀ
sumed according to a kinetic law of the first order (see
Fig. 3, curve 2), and k can be determined by the linearꢀ
ization of the kinetic curve in the coordinates of a firstꢀ
order equation
2
(
1) analysis of the products showed the accumulation
of peroxides in the reaction mixture, whereas both
Ph COO recombination and its reaction with Ph CN
I
2
2
2
barely produce peroxides;
(
2) when the acid (∼10–7—10–6 mol L–1) is introꢀ
duced into the system, an initial fast region at the curve
I
lnA = lnA – k t.
(10)
0
for the Ph COO absorbance decay appears (Fig. 3,
2
curve 1), and it disappears upon the next light pulse,
indicating acid consumption;
Influence of experimental conditions on the reaction
kinetics. The apparent rate constant for the reaction of
diphenylcarbonyl oxide with acids 1—6 depends variꢀ
ously on the experimental conditions. In solutions of
acetonitrile and benzene, when the initial acid concenꢀ
(
3) as the acid concentration increases, the degree of
Ph COO conversion corresponding to the fast region of
2
the kinetic curve increases.
I
Thus, Ph COO is consumed in two processes with
tration increases, the apparent k rate constant increases
2
noticeably different rates. At the final, slow region of the
according to a nonlinear law, approaching to a limiting

Reaction of diphenylcarbonyl oxide with acids
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
611
kI•10–4/s–1
kI•10–4/s–1
2.5
2.0
1.5
1.0
0.5
2.0
1.5
1.0
0.5
1
2
3
1
2
3
4
0
5
–1
1
2 [(CH(OH)C(O)OH) ]•10 /mol L
2
0
5
–1
1
2
3
[PhC(O)OH]•10 /mol L
Fig. 5. Apparent rate constant of Ph COO decay as a funcꢀ
2
Fig. 4. Apparent rate constant of Ph COO decay as a funcꢀ
2
tion of the of tartaric acid content at different concentraꢀ
tion of the benzoic acid content at different concentraꢀ
–1
–5
tions of Ph CN in MeCN; [Ph CN ]/mol L : 1, 5.0•10
2, 2.0•10 ; and 3, 5.0•10 .
;
2
2
2
2
–
1
–1
tions of Ph CN in MeCN; [Ph CN ]/mol L : 1, 1.6•10
;
–4
–4
2
2
2
2
–
1
–4
–4
2
, 5.0•10 ; 3, 2.0•10 ; and 4, 5.0•10 .
I
value. The curves for the k —[RC(O)OH] plot are monoꢀ
tonic for acids 1—4 (Fig. 4), whereas hydroxy acids 5
and 6 exhibit a pronounced, well reproducible Sꢀlike
shape of the plot of the apparent firstꢀorder rate constant
stant is of interest. This influence is most pronounced
for benzoic acid (see Fig. 4) and manifests itself as a
decrease in the slope of the initial linear region of the
I
k —[RC(O)OH] curve with an increase in the Ph CN
concentration. The limiting k value is virtually indepenꢀ
0
2
2
I
I
on the acid concentration (Fig. 5). The limiting k value is
4
–1
independent of the nature of acid and is (2.5±0.5)•10 s
dent of the initial concentration [Ph CN ] . A similar,
2
2 0
for acetonitrile. In a solution of nꢀpentane, the plot of kI
vs. acid concentration is almost linear and have no tenꢀ
dency to exceeding the limit in the studied concentraꢀ
although less pronounced influence of the concentration
of the diazo compound on the reaction kinetics is found
in the case of acetic and tartaric acids. The slope of the
–
5
–1
I
tion interval [MeC(O)OH] = (0—1)•10 mol L
.
k —[5] plot was determined in the inflection point of
0
The influence of the solvent on the reaction kinetics
the Sꢀlike curve.
was studied for the Ph COO—AcOH system. Unlike the
Kinetic scheme of the process. The found formal kiꢀ
netic regularities indicate an important role of complex
formation between the participants of the process: solꢀ
vent, diphenylcarbonyl oxide, its precursor diazo comꢀ
2
4
—11
regularity found earlier,
for the reactivity of diphenylꢀ
carbonyl oxide we found that the change in the medium
polarity in the series MeCN—PhOH—pentane has no
effect on the rate of Ph COO consumption. Taking into
pound, and, of course, acid RC(O)OH. The Ph COO
2
2
account the high value of the bimolecular rate constant
formed under flash photolysis conditions can react with
all components of the reaction mixture but, in our opinꢀ
ion, the most probable primary reaction of carbonyl oxꢀ
ide is its association with the solvent, in particular, acꢀ
etonitrile. The concentration of the diazo compound is
(
Table 2), this fact indicates, most likely, the diffusionꢀ
controlled regime of the reaction.
Assuming that carbonyl oxide interacts with the H+
proton, an increase in the degree of acid dissociation
should accelerate the reaction. However, it turned out
rather low, and in the absence of the acid Ph COO is
2
–
2
–1
that the addition of water (to 7•10 mol L ) has alꢀ
consumed according to a kinetic law of the second orꢀ
der. This indicates a negligible contribution of reaction (4)
to the overall process under experimental conditions.
The concentrations of the solvent and acid differ by at
least six orders, which enhances the probability of the
most no effect on the rate of the reaction of Ph COO
2
with AcOH. An increase in the H O concentration over
2
the indicated value results in diphenyldiazomethane preꢀ
cipitation on the cell walls and a sharp decrease in the
concentration of the generated carbonyl oxide.
In the series of weak acids 1—6, the nature of the
acid has almost no effect on the apparent rate constants
of their reaction with carbonyl oxide. It was difficult to
solvation of Ph COO by the solvent molecule. In addiꢀ
2
7
tion, quantumꢀchemical calculations showed that carꢀ
bonyl oxide H COO forms with ethylene and especially
2
with acetonitrile rather stable intermolecular complexes
of the type
study the reaction of Ph COO with trifluoroacetic and
2
hydrochloric acid because of the fast acidꢀcatalyzed deꢀ
composition of diphenyldiazomethane that is a precurꢀ
sor of diphenylcarbonyl oxide.
The influence of the initial concentration of the
I
∆H°298 = –5.6 kcal mol–1
.
Ph CN diazo compound on the apparent k rate conꢀ
2
2

6
12
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
Nazarov et al.
Table 2. Kinetic characteristics of the Ph COO decay rate in the presence of acids
2
5
–8
–4
b•105
Acid
Medium
MeCN
[Ph CN ]•10
kb•10
a•10
/s
pKa
2
2
–
1
/L mol s–1
–1
–1
/L mol–1 s–2
/
mol L
Acetic
1.6
.3
48.8
23.0
16.7
18.6
2.5±0.2
3.0±0.2
2.4±0.1
2.6±0.1
3.0±0.2
3.5±0.4
3.9±0.5
3.2±0.4
3.0±0.2
2.3±0.3
—
2.6±0.2
3.0±0.2
3.2±0.2
2.7±0.3
2.9±0.3
3.7±0.3
0.5±0.1
1.3±0.2
1.5±0.2
1.4±0.2
2.0±0.3
2.4±0.6
5.2±1.2
0.8±0.2
0.6±0.1
0.3±0.1
—
1.0±0.2
1.2±0.2
3.1±0.4
3.3±0.6
1.3±0.3
8.1±1.5
3
2
5
2
2
2
0
0
0
0
0
a
16.1
14.5
14.4
b
4.75
4.18
c
C H
4.9
42.3
51.9
74.5
29.6
24.9
24.5
10.2
8.4
6
6
2
4
0
9
nꢀC H
20
1.6
5
12
Benzoic
MeCN
5
.0
2
5
0
0
Formic
Oxalic
Tartaric
MeCN
MeCN
MeCN
20
20
5
22.6
4.6
3.75
1.27
d
9.9
8.8
8.5
8.7
d
2
5
0
0
2.98
3.40
d
d
Malic
MeCN
20
a
H O] = 6.9•10 mol L–1
–3
[
[
[
;
2
b
c
H O] = 6.9•10 mol L–1;
–2
2
Ph COO] ∼2•10 mol L–1, which is by ∼2 times higher than in all other cases;
–5
2
0
d
I
the rate constant was determined from the slope in the inflection point of the Sꢀlike plot of k vs.
acid concentration.
Then solvated Ph COO reacts with the acid to proꢀ
tion product of solvated Ph COO with acids 1—6 under
our experimental conditions (solvent molecule is released
in the reaction).
2
2
1
,17,18
duce molecular products. It is known
that the ozoꢀ
nolysis of alkenes in the presence of hydroxylꢀcontainꢀ
ing compounds (viz., H O, alcohols, hydroperoxides,
acids) produces αꢀsubstituted hydroperoxides through the
insertion of carbonyl oxide at the O—H bond.
This reaction explains the irreversible consumption
of the acid found in experiments with low initial conꢀ
centrations of RC(O)OH.
2
The formation of peroxides is confirmed by iodometric
titration of the reaction mixture after the complete conꢀ
sumption of diphenyldiazomethane under the conditions
of steadyꢀstate MB dyeꢀphotosensitized Ph CN decomꢀ
2
2
position in the presence of AcOH. The photoexcitation
of MB followed by its deactivation produces singlet
dioxygen, whose reaction with diphenyldiazomethane
affords Ph COO and N . Application of this method for
2
2
generation of carbonyl oxide instead of Ph CN phoꢀ
tolysis by the nearꢀUV irradiation (used in the kinetic
experiments) is due to the fact that photolysis is accomꢀ
2
2
By analogy to this reaction, we can assume that
αꢀhydroperoxydiphenylmethyl carboxylate is the reacꢀ
panied by a side reaction of carbene Ph C:, the product
2
of diazo compound photolysis, with the acid, decreasing
the yield of peroxide.
The amount of the peroxide formed depends on the
initial concentration of diphenyldiazomethane and is inꢀ
dependent of the acid concentration in the 0.012—0.082
–
1
mol L interval (Table 3). This indicates that the interꢀ
mediate product of photosensitized Ph CN decomposiꢀ
2
2

Reaction of diphenylcarbonyl oxide with acids
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
613
Table 3. Relative yield of peroxide upon the decomposition of Ph CN photosensiꢀ
2
2
tized by the MB dye in the presence of AcOH (solvent MeCN, 295 К)
3
[AcOH] •102
[ROOH] •103
Entry [Ph CN ] •10
[ROOH]/[Ph CN ]
2 2 0
2
2 0
0
0
mol L–1
1
2
3
4
5
6
1.60
1.17
1.17
3.51
5.85
8.19
1.17
0.9±0.2
1.7±0.2
1.2±0.2
1.6±0.4
1.3±0.4
2.1±0.4
0.56
0.50
0.35
0.47
0.38
0.47
3.42
3.42
3.42
3.42
4.48
tion reacts completely with the acid, and the peroxide is
formed from the diazo compound. The yield of peroxide
based on the initial diphenyldiazomethane is 46±10%.
Since the reaction of Ph CN with singlet dioxygen proꢀ
The rate of diphenylcarbonyl oxide decay according
to the first order kinetics is determined by the rate of
reaction (12), so that
2
2
1
9
I
duces diphenylcarbonyl oxide with 60% probability,
–d[PH COO]/dt = k [Ph COO] =
2
2
and the diazo compound decomposes partially by the
acid, we can conclude that hydroperoxide formation in
k12[Ph2COO...Solv][RC(O)OH].
(15)
the reaction of Ph COO with RC(O)OH is the main
Using the expression for the quasiꢀequilibrium conꢀ
2
reaction channel.
centration of the Ph COO complex with the solvent and
2
Finally, the complex formation between diphenylꢀ
diazomethane and carboxylic acid should be taken into
account. In organic media the catalytic decomposition
of diazo compounds by weak acid is rather slow. This is
indicated by the virtual constancy of the Ph CN conꢀ
taking into account Eq. (14), we obtain the equation for
the apparent rate constant k
I
.
(16)
2
2
centration during the photolytic experiment (see above).
In addition, the acid concentrations in our experiments
Equation (16) corresponds to the found kinetic reguꢀ
larities of the process. It explains the hyperbolic depenꢀ
dence of the apparent rate constant k on the acid conꢀ
–
7
–5
–1
are too low (∼10 —10 mol L ) for the efficient deꢀ
composition of the diazo compound. However, the reꢀ
action of the acid with Ph CN should be taken into
I
centration. At low [RC(O)OH]0 values, the first term of
the denominator prevails, which agrees with the presꢀ
2
2
account because the binding of RC(O)OH to a stable
intermolecular complex decreases the concentration of
free acid, which is active in reactions with solvated carꢀ
bonyl oxide.
I
ence of the linear region in the curve for the plot of k vs.
acid concentration
I
Thus, the kinetic scheme of the reaction of Ph COO
k = {k k /(k [Ph CN ] + 1)}[RC(O)OH].
(17)
2
11 12
13
2
2
with RC(O)OH under our experimental conditions has
the form
Equation (17) explains the decrease in the slope of
the initial region in the k —[RC(O)OH] curve with an
increase in the diphenyldiazomethane concentration (see
Fig. 4). In this case, the diazo compound plays a role of
the buffer decreasing the active concentration of the acid
and the rate of reaction (12).
I
Ph COO + Solv
Ph COO…Solv,
(11, –11)
2
2
Ph COO…Solv + RC(O)OH →
2
→
Ph C(OOH)OC(O)R + Solv,
(12)
(13)
2
RC(O)OH + Ph CN
Ph CN …HO(O)CR.
2 2
2
2
When the acid concentration increases, the rate of
irreversible Ph COO decay by reaction (12) begins to
2
The concentration of free acid [RC(O)OH] is reꢀ
lated to the total acid concentration by the relationship
а
exceed the rate of reverse reaction (–11), approaching to
I
the limiting klim value when (k [Ph CN ] + 1)k
<<
13
2
2
–11
<
< k [Ph C(O)OH)]
12 2
[
RC(O)OH] = [RC(O)OH]/(1 + k [Ph CN ]).
(14)
a
13
2
2
kIlim = k11[Solv].
(18)
In this expression, we neglected a change in the conꢀ
centration of diazo compound due to complex formaꢀ
tion because it is not substantial for qualitative analysis
of the kinetic scheme.
Equation (18) explains why the kI value is indeꢀ
pendent of the acid nature and diphenyldiazomethane
lim

6
14
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
Nazarov et al.
concentration. Equation (16) well reproduces the exꢀ
perimental plots. Presenting in the form
is determined by the nature of the R substituent and the
solvating ability of the solvent. Carbonyl oxides with
alkyl substituents are characterized by a substantial conꢀ
tribution of resonance structure C and, correspondingly,
the biradical properties resulting in their high reactivity.
The conjugation of the 4πꢀelectronic system of the carꢀ
bonyl oxide molecule with the aromatic system in
kI = ax/(b + x),
(19)
where x is the acid concentration, a = k [Solv] = kIlim,
1
1
b = (k [Ph CN ] + 1)k /k , we can calculate the a
1
3
2
2
–11 12
and b coefficients (see Table 2). It is seen that the a
value is independent of the experimental conditions beꢀ
cause it is determined only by the solvent nature. Coeffiꢀ
cient b and the concentration of diazo compound change
in parallel according to Eq. (16).
Ph COO stabilizes the zwitterionic component A and B
2
and decreases the reactivity of carbonyl oxide in the
reaction with another carbonyl oxide molecule and oxiꢀ
dized substrates. The polarization of the carbonyl oxide
fragment is responsible for the ambiphilic nature of
I
The Sꢀlike shape of the plots of k vs. concentrations
Ph COO and facilitates the molecular mechanism of its
2
of tartaric and malic acids can be due to the fact that the
concentration of active carboxylic groups in these acids
is decreased by the formation of an intramolecular hydroꢀ
gen bond.
reactions.
Thus, according to the above concepts, Me COO
2
reacts with hydroxylꢀcontaining compounds (viz.,
alcohols, water, and acids) completely or at least parꢀ
tially via the homolytic mechanism. The reaction rate is
determined by the strength of the O—H bond, i.e., deꢀ
1
8
creases in the series alcohol > water > acetic acid. The
insertion at the O—H bond of diphenylcarbonyl oxide,
which exhibits zwitterionic properties to a greater extent,
is facilitated when this bond is polarized (viz., in reacꢀ
tions with acids). It is most likely that the reaction of
carbonyl oxide with alcohols proceeds, in both cases, via
the insertion mechanism through transition states with
different polarities, which manifest in the limiting cases
exclusively the homolytic or molecular nature.
Since this equilibrium is shifted to the left, the reacꢀ
tion of carbonyl oxide with the acid is kinetically maniꢀ
fested at the tartaric acid concentrations > 10– mol L
5
–1
(
see Fig. 5).
The ratio of coefficients a/b, which, as can easily be
shown, represents the second order apparent rate conꢀ
stant k , can serve as a quantitative characteristic of the
b
reactivity of acids in the reaction studied. The k values
b
are presented in Table 2. It is seen that carbonyl oxide
molecules are inserted at the O—H bond of the acids
with almost diffusion constants. No relationship between
the strength of organic acid (pK ) and its reactivity was
a
observed. The very high reactivity of acids in this reacꢀ
tion is in dramatic contrast with the results1 demonꢀ
strating the low reactivity of AcOH in the reaction with
8
dimethylcarbonyl oxide and Me COO reacts with MeOH
2
most rapidly and with the highest yield of αꢀmethoxyꢀ
isopropyl hydroperoxide. The reason for this divergence,
most likely, is a distinction in the electronic nature of
carbonyl oxide intermediates. It is known that the strucꢀ
ture of carbonyl oxide cannot correctly be presented in
the framework of classical Lewis structures. Therefore, a
combination of several resonance structures is often used
to describe the electronic properties of carbonyl oxides.
References
1
. P. S. Bailey, Ozonation in Organic Chemistry, Academic Press,
New York—San Francisco—London, 1982, 2.
2. W. H. Bunnelle, Chem. Rev., 1991, 91, 335.
. K. Ishiguro, T. Nojima, and Y. Sawaki, J. Phys. Org. Chem.,
997, 10, 787.
3
1
4
. A. M. Nazarov, E. M. Chainikova, R. L. Safiullin, S. L.
Khursan, and V. D. Komissarov, React. Kinet. Catal. Lett.,
1
997, 61, 173.
5
. A. M. Nazarov, E. M. Chainikova, S. L. Khursan, I. A.
Kalinichenko, R. L. Safiullin, and V. D. Komissarov, React.
Kinet. Catal. Lett., 1998, 65, 311.
The contribution of one or another resonance strucꢀ
ture to the wave function of the carbonyl oxide molecule

Reaction of diphenylcarbonyl oxide with acids
Russ.Chem.Bull., Int.Ed., Vol. 51, No. 4, April, 2002
615
6
. A. M. Nazarov, E. M. Chainikova, S. L. Khursan, A. B.
Ryzhkov, R. L. Safiullin, and V. D. Komissarov, Izv. Akad.
Nauk, Ser. Khim., 1998, 1329 [Russ. Chem. Bull., 1998, 47,
12. S. I. Maslennikov, A. I. Nikolaev, and V. D. Komissarov,
Kinet. Katal., 1979, 20, 326 [Kinet. Catal., 1979, 20 (Engl.
Transl.)].
1
292 (Engl. Transl.)].
13. Technics of Organic Chemistry. Organic Solvents. Physical
Properties and Methods of Purification, Ed. A. Weissberger,
Interscience Publ. Inc., New York, 1955.
7
8
9
. S. L. Khursan, A. M. Nazarov, E. M. Chainikova, and
V. D. Komissarov, Izv. Akad. Nauk, Ser. Khim., 2001, 759
[
Russ. Chem. Bull., Int. Ed., 2001, 50, 793].
14. M. Regitz and G. Maas, Diazo Compounds. Properties and
Synthesis, Academic Press, Orlando, 1986.
15. M. Girard and D. Griller, J. Phys. Chem., 1986, 90, 6801.
16. J. C. Scaiano, W. G. McGimpsey, and H. L. Casal, J. Org.
Chem., 1989, 54, 1612.
17. E. S. Zelikman, Yu. N. Yur´ev, L. V. Berezova, and V. K.
Tsyskovskii, Zh. Org. Khim., 1971, 7, 633 [J. Org. Chem.,
1971, 7 (Engl. Transl.)].
. A. M. Nazarov, S. L. Khursan, P. V. Krupin, and V. D.
Komissarov, Zh. Fiz. Khim., 2000, 74, 2043 [Russ. J. Phys.
Chem., 2000, 74 (Engl. Transl.)].
. A. M. Nazarov, E. M. Chainikova, P. V. Krupin, S. L.
Khursan, and V. D. Komissarov, Izv. Akad. Nauk, Ser.
Khim., 2000, 1504 [Russ. Chem. Bull., Int. Ed., 2000,
4
9, 1496].
1
1
0. A. M. Nazarov, S. L. Khursan, E. M. Chainikova, and
V. D. Komissarov, Kinet. Katal., 2001, 42, 42 [Kinet. Catal.,
18. Y. Yamamoto, E. Niki, and Y. Kamiya, Bull. Chem. Soc.
Jpn., 1982, 55, 2677.
19. T. Nojima, K. Ishiguro, and Y. Sawaki, J. Org. Chem.,
1997, 62, 6911.
2
001, 42 (Engl. Transl.)].
1. A. M. Nazarov, S. L. Khursan, E. M. Chainikova, and
V. D. Komissarov, Zh. Fiz. Khim., 2001, 75, 245 [Russ.
J. Phys. Chem., 2001, 75 (Engl. Transl.)].
Received July 2, 2001