Kinetics and mechanism of oxime formation from methyl benzoylformate

Article abstract of DOI:10.1002/poc.932

Rate and equilibrium constants for methyl benzoylformate oxime formation were determined as a function of pH over the range from about 0 to 6. The reaction occurs with rate-determining carbinolamine dehydration over the entire range of pH investigated. Specific acid catalysis is dominant at pH < 4. Above that value, a pH-independent reaction becomes apparent. This early appearance of an uncatalyzed reaction is interpreted as intramolecular assistance of the ester moiety in carbinolamine dehydration. Copyright

Full text of DOI:10.1002/poc.932

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2005; 18: 945–949
Published online 19 April 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.932
Kinetics and mechanism of oxime formation
from methyl benzoylformate
A. Malpica* and M. Calzadilla
Escuela de Quimica, Facultad de Ciencias, Universidad Central de Venezuela, 47102 Caracas, Venezuela
Received 19 November 2004; revised 18 February 2005; accepted 25 February 2005
ABSTRACT: Rate and equilibrium constants for methyl benzoylformate oxime formation were determined as a
function of pH over the range from about 0 to 6. The reaction occurs with rate-determining carbinolamine dehydration
over the entire range of pH investigated. Specific acid catalysis is dominant at pH < 4. Above that value, a pH-
independent reaction becomes apparent. This early appearance of an uncatalyzed reaction is interpreted as
intramolecular assistance of the ester moiety in carbinolamine dehydration. Copyright # 2005 John Wiley &
Sons, Ltd.
KEYWORDS: methyl benzoylformate; oxime formation; kinetics; carbinolamine dehydration
INTRODUCTION
range from about 0 to 5. In contrast, the same constant
observed for oxime formation from methyl pyruvate has a
value of 200 Mꢀ1 sꢀ1 and a contribution of the uncata-
lyzed reaction is noted since from a pH of about 5.3 Since
the nucleophile is common, their different behavior must
be due to the substrate structures. Apparently, the pre-
sence of an ester group directly bonded to the carbonyl
group increases the rate constant of the uncatalyzed
dehydration. Since to our knowledge there are no studies
that relate the values of pH at which spontaneous reac-
tions are important to acid-catalyzed dehydrations, it was
of interest to examine the behavior of the same reaction
using methyl benzoylformate as substrate. In comparison
with methyl pyruvate it should be less disposed to accept
catalysis by the hydrated proton because of the electron-
withdrawing effect of the benzene ring directly bonded to
the central carbon atom of the tetrahedral carbinolamine
intermediate, an argument that will lead to the observa-
tion of the appearance of the uncatalyzed reaction at a
lower pH value.
Investigations of the kinetics and mechanism of imine
formation from the most basic amines and/or the most
activated carbonyl compounds exhibits a pH–rate profile
of type C, as results from plotting pH values versus
logarithms of second-order rate constants.1 The figures
obtained from these plots show only one break, indicative
of a transition in the rate-determining step from uncata-
lyzed carbinolamine formation to hydronium ion cata-
lyzed dehydration with increasing pH. The change of
rate-determining step occurs at lower pH values when the
basicity of the amine and/or the activity of the carbonyl
compound are sufficiently increased, e.g. no break is
observed in the pH–rate profile in the pH range from
about 1 to 5 when oxime and phenylhydrazone are
formed from the very activated 2-, 3- and 4-formyl-1-
methylpyridinium ions.2 The justification for this beha-
vior was basically attributed to two factors: (i) very
activated substrates for addition and (ii) the difficulty of
the acid-catalyzed dehydration of the carbinolamines
owing to an unfavorable electrostatic situation. Accord-
ing to this last argument, an uncatalyzed carbinolamine
dehydration has a better chance of existence when the
acid-catalyzed process is more difficult. This require-
ment, although indispensable, is not sufficient. In fact, the
acid-catalyzed dehydration rate constants for oxime for-
mation from 2-, 3- and 4-formyl-1-methylpyridinium
ions have values of 4.11, 175 and 117 Mꢀ1 sꢀ1, respec-
tively, and do not show spontaneous reactions2 in the pH
EXPERIMENTAL
Materials
Hydroxylamine and semicarbazide hydrochlorides, hy-
drochloric acid, benzoylformic acid, formic acid, chlor-
oacetic acid, acetic acid, potassium hydrogenphosphate
and potassium chloride were obtained commercially.
Solutions of these reagents were prepared just prior to
their use to minimize the possibility of decomposition.
Buffer solutions from hydrochloric acid, formic acid,
chloroacetic acid, acetic acid and potassium hydrogen-
phosphate were employed according to the pH
investigated. Glass-distilled water was used throughout.
*Correspondence to: A. Malpica, Escuela de Quimica, Facultad
de Ciencias, Universidad Central de Venezuela, 47102 Caracas,
Venezuela.
E-mail: amalpica@yahoo.com
Contract/grant sponsor: Consejo de Desarrollo Cientifico y Humanis-
tico de la Universidad Central de Venezuela.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 945–949

946
A. MALPICA AND M. CALZADILLA
Methyl benzoylformate was prepared by mixing concen-
trated sulfuric acid (99.99%; Aldrich) with a solution of
benzoylformic acid (97%; Aldrich) in methanol (analy-
meters. Oxime formation was followed by observing the
appearance of the product at 240 nm. Second-order rate
constants, kobs/(amine)fb, were obtained from the slopes
of plots of the first-order rate constants against the
concentration of amine free base and were corrected for
accumulation using kobs. ¼ kexp[1 þ Kadd (amine)fb] In
cases where the reactions were very slow (pH 5 4.70),
the progress of the reaction was followed by the initial
rate method: with pH and hydroxylamine concentration
constant, using varying ester concentrations (3.74 ꢂ 10ꢀ4
to 9.35 ꢂ 10ꢀ5 M) the absorbance changes of the product
were measured at 10 s intervals until ꢃ100 s. Plots of
absorbance versus time under these conditions give ex-
cellent lines. Since oxime solutions obey Beer’s law, the
slope of the line represents the value of the initial rate. A
second plot of the slopes obtained versus ester concen-
tration permitted the first-order rate constant (kobs) to be
evaluated and finally the second-order rate constant was
obtained from the ratio kobs/(amine)f.
¨
tical reagent grade; Riedel-de Haen). The isolation and
purification of the product were carried out following the
technique described for the synthesis of ethyl benzoyl-
formate.4 The product was isolated by distillation (b.p.
90 ꢁC at 8 mmHg; lit. b.p. 246–248 ꢁC and 137 ꢁC at
14 mmHg5). Its purity is indicated by the 13C NMR
spectra recorded on a JEOL 270 MHz instrument in
CDCl3: ꢀ 52.84, 128.96, 130.11, 132.43, 135.07, 164.11
and 186.16. Methyl benzolylformate oxime was prepared
by mixing water–methanol solutions of methyl benzoyl-
formate and hydroxylamine hydrochloride. The crystals
obtained were recrystallized from methanol–water (m.p.
135–139 ꢁC; lit. m.p. 138–139 ꢁC5).
Equilibrium constants
At constant pH values of 2.94, 3.00, 3.10, 3.35 and 3.65
(AcOH/AcOꢀ buffer), the second-order rate constants
were measured at different buffer concentrations (0.2–
0.5 M). Plots of rate constant values against buffer con-
centration gave horizontal parallel lines. A second plot of
the intercept values of the lines obtained versus hydro-
nium ion concentration yielded an excellent straight line
(r ¼ 0.997) whose slope provided the rate constant for the
specific acid component (KaddkH) and an intercept that
provided the rate constant for the pH-independent route
(Kaddk0) (kH and k0 are the rate constants of the acid-
catalyzed and spontaneous dehydration of the carbinola-
mine respectively).
Equilibrium constants for the addition of hydroxylamine
and semicarbazide to methyl benzoylformate (Kadd),
were determined at 30 ꢁC, ꢁ ¼ 0.5 (KCl), by monitoring
spectrophotometrically the disappearance of the chromo-
phore of the substrate at ꢂ ¼ 260 nm. The values of Kadd
were obtained from the negative intercepts of plots of
1/Á Aequi. vs 1/(amine)fb. In the case of hydroxylamine
addition, pH 7.02 was maintained constant by the use of
0.92 M potassium phosphate–potassium hydrogenpho-
sphate buffer. The concentration of the amine free base
was varied from 2.75 ꢂ 10ꢀ2 to 9.10 ꢂ 10ꢀ2 M. An aver-
age of 20 determinations were made and the value of Kadd
ꢀ1
obtained was 80.00 M
(r ¼ 0.94, standard deviation
ꢃ5%). When the amine used was semicarbazide, Kadd
was estimated by the same procedure, under the follow-
ing experimental conditions: pH 5.00 (AcOH/ACOꢀ,
0.47 M), (semic)fb ¼ 0.187–0.426 M. An average of 25
determinations gave Kadd ¼ 18.90 Mꢀ1 (r ¼ 0.97, standard
deviation ꢃ5%), ((semic)fb is the concentration of semi-
carbazide free base).
RESULTS AND DISCUSSION
Second-order rate constants for methyl benzoylformate
oxime formation were determined under pseudo-
first-order conditions over the pH range 0.54–5.92 at
30 ꢁC in aqueous solutions of ionic strength 0.5 (KCl).
The logarithms of the values obtained are plotted as a
function of pH in Fig. 1. The values decreased linearly
with increasing concentration of the hydrated proton up
to pH ꢄ4.2. Above this limiting pH value the logarithms
show positive deviations, more pronounced as the pH
increases, reaching a near constant value between pH 5.5
and 6.0. This pH–rate profile is interpreted as a process in
which the sole rate-determining step over the entire range
of pH investigated is the dehydration of the carbinola-
mine, which occurs with specific acid and spontaneous
catalysis. The rate law is:
Kinetic measurements
UV spectra of methyl benzoylformate oxime obtained at
pH from about 0 to 5 indicate quantitative kinetic yields.
Above pH 5.0 it was necesassary to correct the values of
the experimental rate constants owing to the presence of
the lateral reaction of ester hydrolysis, as indicated in the
Results and discussion section. All rate measurements
were carried out spectrophotometrically employing a
Zeiss PMQ II spectrophotometer equipped with a ther-
mostated cell holder. Rate constants were measured in
water at 30 ꢁC and ionic strength 0.5 (KCl) under pseudo-
first-order conditions. The pH was maintained constant
through the use of buffers with hydrochloric, formic,
chloroacetic and acetic acid and potassium hydrogenpho-
sphate. Values of pH were measured with Radiometer pH
Kobs=ðamineÞfb ¼ Kadd½kHðH3OþÞ þ k0ꢅ
ð1Þ
where Kadd, kH and k0 are the equilibrium constant for
addition and the acid-catalyzed and spontaneous rate
constants of carbinolamine dehydration, respectively.
The solid line in Fig. 1 is a theoretical line based on
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 945–949

KINETICS OF OXIME FORMATION FROM METHYL BENZOYLFORMATE
947
4
3
mic acid and its anion have appreciable absorbances at
the same wavelength at which measurements of oxime
formation were made, the possibility that the positive
deviations from linearity were the result of a contribution
of ester hydrolysis to the experimental values measured,
it was necessary to correct them for the extent of this
contribution. The substrate shows similar behavior to
methyl pyruvate,6 its hydrolysis is insensitive towards
acid catalysis, as one would expect with a substrate
strongly activated by the carbonyl group, and the rate
constants of specific base catalysis and spontaneous
2
1
0
0
1
2
3
4
5
6
7
-1
-2
catalysis are 1.94 ꢂ 103 Mꢀ1 sꢀ1 and 5.98 ꢂ 10ꢀ6 sꢀ1
,
pH
respectively (A. Malpica and M. Calzadilla, unpublished
data). Although these values represent very small con-
tributions to the experimental constants obtained here at
the higher pH values used, they were corrected for its
contributions. It was assumed that the complete law at
high pH values is
Figure 1. Logarithms of second-order rate constants for
methyl benzoylformate oxime formation plotted as a func-
tion of pH. &, Experimental points. The solid line was
calculated based on the rate law in Eqn (1) and the constants
in Table 1
kobs ¼ ðamineÞfbKadd½kHðH3OþÞ þ k0ꢅ þ kOHðOHꢀÞ þ k0ꢆ
ð2Þ
Eqn (1) and on values of Kadd, kH and k0 evaluated
experimentally and summarized in Table 1. The points
in the figure are the experimental values of the second-
order rate constants corrected for accumulation and for
ester hydrolysis. The positive deviations shown in Fig. 1
are not a consequence of a general acid catalysis because
plots of second-order rate constants against the concen-
tration of acetic acid buffer at different pH values show
that at constant pH the rate constants are independent of
the buffer concentration. This suggests that the ꢃ value
for the dehydration of carbinolamine should be close to
unity and, therefore, difficult to determine. A plot of the
intercepts of the parallel lines obtained against the con-
centration of hydronium ion yields an excellent straight
line (r ¼ 0.997) whose slope provides the rate constant for
the specific acid-catalyzed component and an intercept
that provides the rate constant for the pH-independent
route. The observed positive deviations are not the result
of a contribution of rate constants for ester hydrolysis, a
reaction that, without doubt, accompanied the process,
because the experimental values of the second-order rate
constants used to construct Fig. 1 have been corrected for
the extent of this additional reaction. Since benzoylfor-
kobs ꢀ ½kOHðOHꢀÞ þ k0ꢆꢅ ¼ ðamineÞfbKadd½kHðH3OþÞ þ k0ꢅ
ð3Þ
where kOH and k0* are the specific basic and spontaneous
rate constants for the ester hydrolysis, respectively. The
values obtained for the first term of Eqn (3) were
corrected for accumulation and the logarithms of the
corr
ratio k /(amine)fb were inserted in Fig. 1.
obs
Note that the theoretical limit of Eqn (3) will be
reached when the proton concentration is small enough
to make the acid-catalyzed process negligible; in such
a case the limit is kobs/(amine)fb ¼ Kaddk0, since
K
addk0 ¼ 5.33 ꢂ 10ꢀ2
M
ꢀ1 sꢀ1, the corrected value of the
experimental measure at pH 5.92 (4.78 ꢂ 10ꢀ2
M )
ꢀ1 sꢀ1
indicates that the process of oxime formation at this pH is
close to attaining the limit.
Experimental values of the second-order rate constants
show accumulation with increasing pH, indicating that it
is difficult for the intermediate to render the imine, hence
their values were corrected for the extent of carbinola-
mine accumulation using
Table 1. Summary of rate and equilibrium constants for
methyl benzoylformate oxime formation in aqueous solu-
tions at 30 ꢁC and ionic strength 0.5
kobs ¼ kexp½1 þ KaddðamineÞfbꢅ
Parameter
Value
Kadda(M
Kaddcorra(M
)
80.0
83.0
34.62
6.66 ꢂ 10ꢀ4
0.078
18.9
ꢀ1
The value of Kadd was determined at pH 7.02 using
potassium phosphate–potassium hydrogenphosphate and
varying the concentration of the nucleophile. The value
obtained is included in Table 1.
The proposed mechanism is outlined in Scheme 1. This
scheme does not include the conversion of the hydrate of
methyl benzoylformate to the reactive keto form, there-
fore the experimentally found equilibrium constant for
ꢀ1
)
kHb(Mꢀ1 sꢀ1
)
k0b (sꢀ1
KHyd
)
Kadda,c (M
)
ꢀ1
a Standard deviation was ꢃ5%.
b In most cases the data agree within 3%.
c In semicarbazone formation.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 945–949

948
A. MALPICA AND M. CALZADILLA
compared with acetaldehyde as 4 ꢇ 0.4 kcal molꢀ1
O
OH
K
add
(1 kcal ¼ 4.184 kJ).
COOMe
OMe + NH OH
C
2
It is interesting that a w value, defined as w ¼
logKHyd*/KHyd, where KHyd* and KHyd are the equili-
brium constant of water addition to a given compound
and of benzaldehyde, respectively, is obtained as 0.85.
Use of this w value in estimating the equilibrium constant
NHOH
O
+
(H
k
)
kH
o
N
OH
+ H O
of hydroxylamine to the keto form of the substrate, from
ketoform
another structure–reactivity relationship, logK
¼
C
2
add
Sw þ B,10 where S is a measure of the reactivity of the
reaction examined (carbinolamine formation in this case)
compared with the model of hydration equilibrium con-
stant and B is a constant for a given reaction series
COOMe
+
k
kH (H ) +
0
(1)
k
/ (amine) =K
fb
obs
add
(B ¼ 1.06 when the nucleophile is hydroxylamine), gives
¼ 86 Mꢀ1, in reasonable accord with the previous
ketoform
add
Scheme 1
K
estimated value of 83.00 M
ꢀ1
.
There are numerous reported cases in which sponta-
neous carbinolamine dehydrations occur, and in general
they are observed at pH values > 8.11–13 This is inter-
preted as a consequence of their higher competition with
an acid-catalyzed process carried out in solutions with
low proton concentration. This study, as for oxime for-
mation from methyl pyruvate,3 shows an important con-
tribution of the uncatalyzed carbinolamine dehydration
noted at relatively low pH values. Since this observation
applies also to oxime formation of methyl pyruvate, a
possible explanation is that an intramolecular general
basic catalysis exerted by the carbonyl oxygen atom on
the hydrogen atom bonded to nitrogen is helping the
expulsion of the leaving group:
addition of hydroxylamine to the keto ester should be
corrected for the hydration of the substrate using
exp
¼ Kadd ð1 þ KHyd
corr
add
K
Þ
ð4Þ
where KHyd represents hydrate/keto ester. To our knowl-
edge, the equilibrium hydration constant, KHyd, has not
been reported. A reliable value of the keto ester absor-
bance prior to its reaction could not be obtained because
the equilibrium is reached in less than 10 s. On the other
hand, use of an NMR method for evaluating this thermo-
dynamic parameter requires a considerably higher con-
centration of the substrate than in the corresponding
spectrophotometric method, restricting its use to water-
soluble substrates, and in addition giving erroneous
results unless account is taken of the sensitivity of such
a process to water concentration.7 Therefore, the value of
this equilibrium constant was estimated using the struc-
ture–reactivity correlationship of Sander and Jencks8 for
the addition of nucleophiles to aromatic carbonyl com-
pounds. In this linear free energy relationship, log
K0 ¼ Áꢄ þ A, Á is a measure of the sensitivity of the
carbonyl compound to the affinity of nucleophilic re-
agents and ꢄ is a measure of this affinity for a given
nucleophile. Based on the reported values of ꢄ ¼ 1.24,
0.46 and ꢀ 3.58 for hydroxylamine, semicarbazide and
water, respectively, and on values of Kadd ¼ 80.00 and
18.90 Mꢀ1 for addition of hydroxylamine and semicarba-
zide, respectively, to methyl benzoylformate, a value of
KHyd ¼ 0.078 was found from the relationship Álog
KHyd ¼ Áꢄ. With this value in hand, it is possible to
return to the original data and correct the apparent
OH
OMe
C
C
N
O
H
HO
Although this type of catalysis should be more im-
portant in the dehydration of carbinolamine derived from
the addition of hydroxylamine to the conjugated base of
pyruvic acid, it has not been reported.14 In that case, the
favorable elecrostatic attraction between the intermediate
and the hydrated proton, reflected in the high value of kH
(5.75 ꢂ 104 Mꢀ1 sꢀ1) does not permit any contribution,
even when the spontaneous reaction is carried out by
intramolecular catalysis.
Future study of the behavior of imine formation from
methyl benzoylformate and N-methylhydroxylamine
could possibly clarify the detailed nature of the transition
state of the reaction.
addition equilibrium constant for the extent of hydrate
corr
formation [Eqn (4)]. The corrected value, Kadd
¼
83.00 Mꢀ1, has been included in Table 1. In comparison
with the value KHyd ¼ 2.8 exhibited by methyl pyruvate,3
methyl benzoylformate is ꢃ36 times less hydrated,
which can be interpreted as a consequence of its reso-
nance stabilization. In fact, Fersht9 estimated that the
extra stabilization of the carbonyl in benzaldehyde
Acknowledgements
We are indebted to Professor Alirica Suarez for her
assistance with purification and NMR spectra of the
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 945–949

KINETICS OF OXIME FORMATION FROM METHYL BENZOYLFORMATE
949
5. Heilbron I, Cook AH, Bunbury HM, Hey DH. Dictionary of
Organic Compounds (4th edn), vol. 1. Eyre and Spottiswood:
London, 1965; 359.
6. Pocker Y, Meany JE, Jones CR. J. Am. Chem. Soc. 1982; 104:
4885–4889.
substrate. This work was supported in part by the Consejo
de Desarrollo Cientifico y Humanistico de la Universidad
Central de Venezuela.
7. Pocker Y, Meany JE, Zadorojny C. J. Phys. Chem. 1971; 75: 792–
799.
8. Sander EG, Jencks WP. J. Am. Chem. Soc. 1968; 90:
6154–6162.
9. Fersht AR. J. Am. Chem. Soc. 1971; 93: 3504–3515.
10. Malpica A, Calzadilla M. J. Phys. Org. Chem. 2003; 16: 1–4.
11. Do Amaral L, Bastos MP. J. Org. Chem. 1971; 36:
3412–3417.
REFERENCES
1. Sayer JM, Pinsky B, Schonbrun A, Wasthien W. J. Am. Chem.
Soc. 1974; 96: 7098–8009.
2. Sojo P, Viloria F, Malave L, Possamai R, Calzadilla M,
Baumrucker J, Malpica A, Moscovici R, do Amaral L. J. Am.
Chem. Soc. 1976; 98: 4519–4525.
3. Cordova T, Peraza AJ, Calzadilla M, Malpica A. J. Phys. Org.
Chem. 2002; 15: 48–51.
4. Gilman R, Blatt AH. Org. Synthe. Coll. Vol. 1941; 1: 241–245.
12. Fett R, Simionato SL, Yunes RA. J. Phys. Org. Chem. 1970; 3:
620–626.
13. Costa Brigmente M, Votero LR, Terezani AJ, Yunes RA. Bull.
Chem. Soc. Jpn. 1993; 66: 2289–2293.
14. Malpica A, Calzadilla M, Cordova T. J. Phys. Org. Chem. 2000;
13: 162–166.
Copyright # 2005 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 945–949