The ortho effect on the acidic and alkaline hydrolysis of substituted formanilides

Article abstract of DOI:10.1002/kin.20925

The kinetics of formanilides hydrolysis were determined under first-order conditions in hydrochloric acid (0.01-8 M, 20-60°C) and in hydroxide solutions (0.01-3 M, 25 and 40°C). Under acidic conditions, second-order specific acid catalytic constants were used to construct Hammett plots. The ortho effect was analyzed using the Fujita-Nishioka method. In alkaline solutions, hydrolysis displayed both first- and second-order dependence in the hydroxide concentration. The specific base catalytic constants were used to construct Hammett plots. Ortho effects were evaluated for the first-order dependence on the hydroxide concentration. Formanilide hydrolyzes in acidic solutions by specific acid catalysis, and the kinetic study results were consistent with the A_AC2 mechanism. Ortho substitution led to a decrease in the rates of reaction due to steric inhibition of resonance, retardation due to steric bulk, and through space interactions. The primary hydrolytic pathway in alkaline solutions was consistent with a modified B_AC2 mechanism. The Hammett plots for hydrolysis of meta- and para-substituted formanilides in 0.10 M sodium hydroxide solutions did not show substituent effects; however, ortho substitution led to a decrease in rate constants proportional to the steric bulk of the substituent.

Full text of DOI:10.1002/kin.20925

The Ortho Effect on the
Acidic and Alkaline
Hydrolysis of Substituted
Formanilides
SALIL DILEEP DESAI, LEE E. KIRSCH
Division of Pharmaceutics and Translational Therapeutics, College of Pharmacy, The Dale E. Wurster Center for
Pharmaceutical Technology, The University of Iowa, Iowa City, IA 52242
Received 4 February 2015; revised 17 April 2015; accepted 17 April 2015
DOI 10.1002/kin.20925
Published online 9 June 2015 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: The kinetics of formanilides hydrolysis were determined under first-order condi-
tions in hydrochloric acid (0.01–8 M, 20–60°C) and in hydroxide solutions (0.01–3 M, 25 and
40°C). Under acidic conditions, second-order specific acid catalytic constants were used to
construct Hammett plots. The ortho effect was analyzed using the Fujita–Nishioka method. In
alkaline solutions, hydrolysis displayed both first- and second-order dependence in the hydrox-
ide concentration. The specific base catalytic constants were used to construct Hammett plots.
Ortho effects were evaluated for the first-order dependence on the hydroxide concentration. For-
manilide hydrolyzes in acidic solutions by specific acid catalysis, and the kinetic study results
were consistent with the AAC2 mechanism. Ortho substitution led to a decrease in the rates
of reaction due to steric inhibition of resonance, retardation due to steric bulk, and through
space interactions. The primary hydrolytic pathway in alkaline solutions was consistent with a
modified BAC2 mechanism. The Hammett plots for hydrolysis of meta- and para-substituted
formanilides in 0.10 M sodium hydroxide solutions did not show substituent effects; however,
ortho substituꢀtCion led to a decrease in rate constants proportional to the steric bulk of the
substituent.
2015 Wiley Periodicals, Inc. Int J Chem Kinet 47: 471–488, 2015
In strongly acidic solutions, formanilide has been
INTRODUCTION
reported to degrade via the acid-catalyzed acyl cleav-
age bimolecular (AAC2) mechanism [6]. At high con-
centrations of acid, a linear Hammett relationship was
observed for para and meta substituents. However, at
lower acid concentrations a nonlinear Hammett rela-
tionship was observed due to a change in the rate-
determining step for the reaction. [7,8].
The hydrolytic instability of formanilides is important
to the efficacy of drug molecules [1–5]. A survey of
the literature reveals that the effects of ortho substi-
tutions on formanilide hydrolysis has not been exten-
sively studied although the effects of para and meta
substitutions are well documented.
In alkaline solutions, the pH–rate profile for for-
manilide displayed nonlinear first- and second-order
dependence on the hydroxide concentration [9]. Ham-
mett plots showed no difference in reactivity with
Correspondence to: L. E. Kirsch; e-mail: lee-kirsch@uiowa.edu.
2015 Wiley Periodicals, Inc.
ꢀC

472
DESAI AND KIRSCH
change in substituent except for strongly electron-
withdrawing compounds such as cyano and nitro
groups. Kvalek and Sterba interpreted the alkaline hy-
drolysis kinetics based on a combination of substrate
ionization and the second-order hydroxide-dependent
pathway [9]. The Hammett plots were a concave up-
ward curve, suggesting a change in a mechanism or
rate-determining step. At lower hydroxide concentra-
tions, kinetics were first order in hydroxide [7,8]. Ef-
fects of nucleophilic catalysis, ionic strength, and tem-
perature studies have been studied; however, the results
were not mechanistically interpreted [7, 8, 10–12].
More recently, Cheshemdzhieva et al. have inves-
tigated the alkaline amide hydrolysis of acetanilides
including a series of para-substituted acetanilides
wherein the reactivity was correlated to the electro-
static potential of the carbon and nitrogen amide atoms
[13,14].
In the present study, the acidic and alkaline hydroly-
sis kinetics of para- and meta-substituted formanilides
are compared to the kinetics of ortho-substituted com-
pounds. Substituents were chosen such that they cover
the wide range of electronic and steric effects. The
substituent constants of the selected substituents are
listed in Table I [15,16]. Substrates were synthesized,
purified, and characterized and then hydrolyzed under
constant pH conditions in acidic and alkaline solutions.
Linear-free energy relationships (LFER) and pH-rate
profiles were used to evaluate effects of the electronic,
steric, and anchimeric factors on acidic and alkaline
hydrolysis kinetics and mechanisms.
constants (pKa) were determined for the carboxyl-
and hydroxyl-substituted formanilides using poten-
tiometry (p-hydroxy (9.58)) and UV spectroscopy (p-
carboxyl (4.02), m-carboxyl (3.86), o-carboxyl (3.37),
m-hydroxyl (9.59), and o-hydroxyl(8.98)) [21,22]. The
compounds had melting points and NMR spectra in
good agreement with literature values [6, 9, 23–39].
The spectra of the compounds not previously syn-
thesized are reported below. The purities of the com-
pounds synthesized were >97%.
N-(3-Methyl phenyl)formamide: δH (300 MHz;
CDCl3, Me4Si): 9.6 (br s, ½H), 9.2 (s, ½H), 8.7 (d,
½H), 8.25 (s, ½H), 7.4 (d, 1H, Ar-H), 7.2 (m, 1H,
Ar-H), 6.8 (m, 2H, Ar-H), 2.25 (m, 3H, CH3). δC
(300 MHz; CDCl3): 163.20, 160.07,139.39, 138.59,
136.98, 136.77, 129.19, 128.57, 125.66, 125.29,
120.76, 119.10, 117.27, 115.39, 21.12, 21.05.
N-(3-Ethylphenyl)formamide: δH (300 MHz;
CDCl3, Me4Si): 9.5 (br d, ½H,), 8.75 (d, ½H), 8.60
(br d, ½H,), 8.25 (s, ½H,), 7.40 (m, 1H, Ar-H), 7.2 (m,
1H, Ar-H), 6.9 (m, 2H, Ar-H), 2.60 (m, 2H, CH2), 1.20
(m, 3H, CH3). δC (300 MHz; CDCl3)163.26, 161.11,
145.93, 145.11, 137.16, 136.88, 129.42, 128.74,
125.59, 124.10, 119.63, 118.16, 117.51, 115.80, 28.64,
28.57, 15.31, 15.29.
N-(3-Isopropylphenyl)formamide: δH (300 MHz;
CDCl3, Me4Si): δ 9.45 (br d, ½H), 9.76 (br m, 1H,),
8.35 (s, ½H), 7.50 (m, 1H, Ar-H), 7.25 (m, 1H, Ar-
H), 7.0 (m, 2H, Ar-H), 2.8 (m, H, CH), 1.29 (m, 6H,
2CH3). δC (300 MHz; CDCl3)163.64, 160.18, 151.08,
150.21, 137.47, 137.17, 129.86, 129.17, 123.61,
123.02,118.67, 117.38, 116.44, 34.32, 24.17, 24.13,
24.08.
EXPERIMENTAL
Synthesis
Kinetic Studies
Hydrolysis reactions were carried out under pseudo–
first-order conditions. Reactions in concentrated hy-
drochloric acid solutions (8.15–0.50 M) and sodium
hydroxide solutions (5.0–1.0 M) were followed by us-
ing UV spectroscopy. Reactions in dilute hydrochloric
acid solutions (0.50–0.010 M) and dilute sodium hy-
droxide solutions were followed using HPLC.
For studies conducted using HPLC analysis, the fol-
lowing protocol was used. A 10 mM stock solution
of the formanilide substrate was prepared in absolute
ethanol. A 10-mL of aliquot of dilute hydrochloric acid
solution or sodium hydroxide solution was equilibrated
to the reaction temperature. Reactions were initiated by
adding 0.1 mL of the stock solution of the substrate to
the acidic or alkaline solution to give an initial substrate
concentration equal to 0.1 mM formanilide and 1% v/v
ethanol. Aliquots were withdrawn and quenched with
either acetate buffer (1.0 M, pH 4) or tris buffer (1.0 M,
Solvents and chemicals were obtained commercially
and used without further purifications. The substituted
formanilides were prepared from the unsubstituted ani-
lines by formylation with formic acid using diisopropy-
lether as the solvent (azeotropic synthesis) [17]. The
p-nitro formanilide was prepared by formylating the
p-nitro aniline using acetyl formyl anhydride prepared
in situ [18]. The volatile formanilides were purified
using vacuum distillation, whereas the formanilides
with high melting points were purified using recrystal-
ization or extraction. Melting points and purity were
determined using a Perkin Elmer DSC7 differential
scanning calorimeter [19]. H and 13C NMR spectra
1
were recorded at room temperature on a Bruker Avance
300 spectrometer. 1H NMR and 13C NMR spectra
were referenced using tetramethylsilane or residual
solvent signals as internal standards [20]. Ionization
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
473
Table I Compilations of the substituent parameters; second order rate constants (40° C) and activation parameters
for acidic hydrolysis; and first order rate constants for alkaline hydrolysis (0.1 M sodium hydroxide at 40° C) of
substituted formanilides used in this study
Acidic
Alkaline
kH SE
(M–1 h–1
ꢀH‡ SE –ꢀ S‡ SE
kobs
(kJ mol–1
)
(J mol–1 K–1) (h−1 × 101)
+
+
–
a
Substituent σm
σp
σm
σp
σp
Es
F
)
H
0
7.39
5.06
6.58
0.08 71.7
0.11 66.5
0.18 66.5
0.04 68.0
0.07 66.7
0.02 65.0
0.03 65.0
0.7 60.3
0.2 64.4
0.2 61.2
2.8 69.3
0.5 88.7
1.0 85.8
0.9 89.7
1.1 84.6
0.9 87.4
0.8 89.7
0.7 93.5
0.9 87.9
0.7 91.8
9.1
1.6
3.3
2.8
3.5
3.0
2.6
2.1
2.8
2.2
4.08
3.51
3.24
0.99
3.56
3.2
2.05
1510
4.13
2950
p-CH3
m-CH3
o-CH3
p-Cl
m-Cl
o-Cl
p-NO2
m-NO2
o-NO2
–0.17
–0.31
0.11
–0.07
–0.07
0.4
–1.24 –0.04 2.57
0.23
0.78
7.04
9.65
0.41 7.24
27.3
11.3
0.67 24.9
0.37
0.71
–0.97
0.79 1.24
0.15
0.67
–1.01b,
–2.52c
p-Br
m-Br
o-Br
p-C2H5
m-C2H5
o-C2H5
p-isoC3H7
m-isoC3H7 –0.07
o-isoC3H7
0.23
–0.15
–0.15
–0.27
0.45
8.32
10.8
0.44 2.57
4.60
0.19 67.3
0.1 65.1
0.04 63.4
0.10 66.6
0.18 63.0
0.01 73.1
0.10 68.0
0.13 66.4
0.02 71.4
0.03 67.7
0.03
0.21 65.3
0.5 61.9
0.25 63.6
0.3 59.9
0.08 68.4
0.02 61.0
0.03 64.8
0.18 70.4
0.02 63.8
0.6 81.9
0.8 87.1
1.1 95.5
2.0
2.7
3.4
3.39
2.71
3.3
0.39
0.41
–1.16
–0.3
0.5 87.9 1.7
3.58
3.01
0.75
3.43
2.84
0.675
3.61
3.09
2.54
–0.07
–0.06
6.21
1.2 96.8
2.4 75.4
1.0 83.1
0.5 86.4
3.1 83.9
1.6 88.3
3.7
7.3
3.2
1.6
9.7
4.9
–1.31 –0.05 2.07
–0.28
5.42
6.05
–1.71 –0.05 1.39
p-OCH3
–0.79
2.95
8.59
0.26 5.23
18.2
8.21
24.6
0.29 2.67
7.80
m-OCH3
o-OCH3
p-COOH
m-COOH
o-COOH
p-OH
m-OH
o-OH
p-Phenyl
o-Phenyl
0.12
0.36
0.12
0.05
0.32
–0.55
0.8 91.1
0.3 91.4
0.3 92.6
0.3 95.6
4.3 87.6
1.8 101
4.3 95.3
1.8 73.6
1.8 100
2.6
1.1
1.0
1.0
13.4
6
13.2
5.6
6
0.42 0.73
–0.92
–0.37
–0.01
4.13
3.81
7.53
0.08 3.17
2.8
2.95
0.75
–0.18
–1.01b,
–3.82c
aPoint of reference of Es values is shifted to H. bMinimum perpendicular dimension. cMaximum coplanar dimension. Substituent constants
were taken from the compilations by Fujita or Exner.
pH 8), and the samples were stored at 4°C until HPLC
analysis. The pH value was measured at the start and
end of the reaction at the reaction temperature. The
degradation reactions were followed to at least three
half lives. Concentration time profiles were generated
using standards of authentic formanilides and their cor-
responding anilines in the appropriate concentration
range (0.1–0.05 mM). HPLC analyses were performed
on a thermo-separations RP-HPLC system consisting
of a UV6000LP diode array detector, AS3000 autoin-
jector, P4000 pump, and an online membrane degasser.
Chromatograms were integrated, and data stored using
ChromquestꢀR chromatography data system software
(Version 4.0). HPLC conditions and mobile phase com-
positions are provided in Table II.
For studies conducted using UV analysis (Hewlett
Packard HP8453 UV–vis diode array spectrophotome-
ter equipped with a Peltier thermostatted cell holder), a
2.5-mL aliquot of hydrochloric acid or sodium hydrox-
ide solution was equilibrated to the reaction tempera-
ture in a closed glass cuvette in the thermostatted cell
holder of the HP 8453 spectrophotometer. Reactions
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

474
DESAI AND KIRSCH
Table II Chromatographic conditions used in the analysis of substituted formanilides
Retention Times (min)
Flow Rate
(mL/min)
Wavelength
(λ) (nm)
Substituent
H
Column
C
Substrate
2.8
Aniline
3.6
Mobile Phase
0.8
235
30:70 ACN:10 mM phosphate
buffer pH 7
p-Cl
m-Cl
1
1
C
C
239
240
2.6
2.6
3.8
4.0
38:62 ACN: water
38:62 ACN:10 mM phosphate
buffer pH 7.2
o-Cl
1
D
236
1.3
2.3
38:62 ACN:10 mM phosphate
buffer pH 7.2
p-Br
m-Br
1
1
B
D
242
242
3.6
1.8
4.2
2.7
55:45 ACN: water
35:65 ACN:10 mM phosphate
buffer pH 7.2
o-Br
1
1.2
1
C
C
C
C
D
D
D
D
D
D
D
D
D
D
A
A
A
A
A
B
230
232
240
228
230
230
231
231
231
340
236
233
270
220
260
238
235
230
210
230
2.7
2.9
3.3
2.4
3.9
2.0
4.2
4.2
2.7
6.9
3.3
4.0
4.1
3.2
2.8
2.8
4.3
5.7
6.8
6.5
5.5
3.7
4.5
4.0
5.0
3.6
5.6
5.9
5.3
5.6
4.3
2.2
5.6
6.8
2.3
2.3
2.7
3.7
4.0
3.6
38:62 ACN:10 mM phosphate
buffer pH 7
30:70 ACN:10 mM phosphate
buffer pH 7
30:70 ACN:10 mM phosphate
buffer pH 7
28:62 ACN:23 mM phosphate
buffer pH 7
30:70 ACN:10 mM phosphate
buffer pH 7.2
30:70 ACN:10 mM phosphate
buffer pH 7.2
32:68 ACN:10 mM phosphate
buffer pH 7.2
32:68 ACN:10 mM phosphate
buffer pH 7.2
32:68 ACN:10 mM phosphate
buffer pH 7.2
11:89 ACN:10 mM phosphate
buffer pH 7.2
20:80 ACN:10 mM phosphate
buffer pH 7.2
25:75 ACN:10 mM phosphate
buffer pH 7.2
35:65 ACN:10 mM phosphate
buffer pH 7.2
35:65 ACN:10 mM phosphate
buffer pH 7.2
10:90 ACN:100 mM phosphate
buffer pH 7
10:90 ACN:100 mM phosphate
buffer pH 7
p-CH3
m-CH3
o-CH3
1.2
1
m-C2H5
o-C2H5
p-isoC3H7
m-isoC3H7
o-isoC3H7
p-NO2
1
1
1
1
1.2
1
m-NO2
o-NO2
1
p-Phenyl
o-Phenyl
p-COOH
m-COOH
o-COOH
p-OH
1
1
0.8
0.8
0.8
1
10:90 ACN:100 mM phosphate
buffer pH 7
05:95 ACN:100 mM phosphate
buffer pH 7
05:95 ACN:100 mM phosphate
buffer pH 7
05:95 ACN:10 mM phosphate
buffer pH 7
m-OH
1
o-OH
1.2
Continued
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
475
Table II Continued
Retention Times (min)
Wavelength
Flow Rate
Substituent
(mL/min)
Column
A
(λ) (nm)
Substrate
Aniline
Mobile Phase
p-OCH3
1
235
5.2
4.3
20:80 ACN:100 mM phosphate
buffer pH 7
m-OCH3
o-OCH3
0.5
0.5
D
D
235
235
1.9
2.2
2.9
2.8
40:60 ACN:100 mM phosphate
buffer pH 7
40:60 ACN:100 mM phosphate
buffer pH 7
Column: A = Zorbax SB-CN, 5 μ, 4.6 × 150 mm: B = Zorbax SB-C8, 5 μ, 4.6 × 150 mm: C = C12 HydroRP Phenomenex, 4 μ, 4.6 ×
75 mm; D = Metasil ODS, 5 μ, 4.6 × 500 mm; and ACN = acetonitrile.
were initiated by adding 25 μL of stock solution
(10 mM substrate in ethanol) to the reaction mixture,
so that the final concentration of ethyl alcohol was 1%
v/v and the final concentration of substrate in solution
was 0.1 mM. Spectra were collected as a function of
time using the kinetics module in the ChemstationꢀR
software. Reactions were followed until there was no
significant change in spectra with time. Reactions fol-
lowed to at least three half lives. The pH value was
determined at the beginning and end of the reaction
at the reaction temperature. The rate constants deter-
mined using UV were reproducible within 5% and
also matched those determined from the HPLC studies
( 10%).
Identity of the hydrolysis products for substituted
formanilides was confirmed by comparing the spec-
tra associated with the final reaction mixture to an
authentic sample of the predicted end product at the
same concentration (i.e., the corresponding aniline and
formic acid). Formic acid does not have appreciable
absorbance at these concentrations and cannot be de-
tected; however, aniline has a stronger chromophore
and could be detected. The spectrum was collected and
compared to the spectrum of the final products using
the peak purity compare function in the ChemstationꢀR
software. The procedures were all identical for all for-
manilides except for o-hydroxyformanilide, which is
insoluble in ethanol, and the stock solution was pre-
pared in dimethyl sulfoxide.
hydrochloric acid in water. Dilute hydrochloric acid so-
lutions below 1.0 M were prepared by diluting certified
hydrochloric acid solutions (1.0 or 0.10 M) in water
and adjusting the ionic strength to 1.0 using potas-
sium chloride. The pH–rate profiles for the remaining
substituted formanilides were determined in aqueous
hydrochloric acid solutions in the range of 1–0.01 M.
The pH–rate profiles for selected formanilides (for-
manilide, p-chloroformanilide, p-methylformanilide,
p-methoxyformanilide, and m-nitroformanilide) at
25°C were determined by using aqueous sodium hy-
droxide solutions in the concentration range between
0.010 and 5.0 M. Sodium hydroxide solutions in the
range of 1.0–5.0 M were prepared by diluting 10.0 M
certified sodium hydroxide solution in carbonate-free
water. Dilute sodium hydroxide solutions (<1 M) were
prepared by diluting certified sodium hydroxide acid
solutions (1.0 or 0.10 M) in carbonate-free water. Ionic
strength was not adjusted for these studies.
The effect of structure on the rate of alkaline hy-
drolysis of substituted formanilides was evaluated by
hydrolyzing them in 0.10 M sodium hydroxide solu-
tions at 40°C. The ionic strengths of these solutions
were not adjusted, and reactions were followed using
HPLC. The p-nitroformanilide and o-nitro formanilide
degrade very rapidly even at low hydroxide concen-
trations; hence, the rate of degradation of these com-
pounds was studied using UV spectroscopy.
Studies to determine the effects of temperature on
formanilide hydrolysis kinetics were carried out in
1 M hydrochloric acid at 20, 30, 40, 50, and 60°C
using UV spectroscopy.
Effects of pH and Temperature on
Formanilide Hydrolysis Kinetics
The pH–rate profile for the unsubstituted formanilide
at 40°C was determined by using aqueous hydrochloric
acid solutions in the concentration range between 8.15
and 0.010 M. Hydrochloric acid solutions in the range
of 8.15–1.01 M were prepared by diluting concentrated
RESULTS
Reaction schemes were determined by mass balance
and kinetic analysis for all substrates; the general ap-
proach is illustrated using unsubstituted formanilide.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

476
DESAI AND KIRSCH
Figure 1 (a) Concentration–time profiles for hydrolysis of formanilide in 0.075 M hydrochloric acid at 1.0 ionic strength and
60°C. (b) Concentration–time profiles for hydrolysis of formanilide in 0.075 M sodium hydroxide solution at 1.0 ionic strength
and 60°C. Solid circles (•) represent loss of formanilide, solid squares (ꢀ) represent appearance of aniline, and solid triangles
(ꢁ) represent mass balance for the reaction. The solid line is a first-order fitting for formanilide loss, and the broken lines are
interpolations.
HPLC chromatograms for the acidic and alkaline
degradation of formanilide indicated the presence of
two compounds identified as the substrate and corre-
sponding aniline. In both cases, the concentration of
formanilide decreases as a function of time with a con-
current increase in the concentration of aniline (Fig. 1).
The mass balance of the reaction remained constant,
which indicated that formanilide quantitatively con-
verted to aniline. Similarly, UV analysis confirmed the
presence of a simple mixture composed of substrate
and aniline. In acidic solutions the hydrolysis of for-
manilide produces the anilium ion and formic acid,
whereas in alkaline media it forms the aniline and for-
mate ion (Scheme 1).
All of the substituted formanilides hydrolyzed
irreversibly to the corresponding substituted aniline,
except for m-methoxyformanilide (acidic) and p-
and o-hydroxylformanilide. In acidic solutions,
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
477
DISCUSSION
Acid Hydrolysis
The logarithms of the observed rate constants (kobs
)
were plotted as a function of pH or HA to construct pH–
rate profiles (Fig. 2). For reactions in dilute hydrochlo-
ric acid solutions (<1.0 M HCL), the pH values were
measured and at high concentrations of hydrochloric
acid the HA scale was used without temperature cor-
rections [41].
A typical bell-shaped pH–rate profile for amide hy-
drolysis was observed [42]. Formanilide is known to
undergo hydrolysis by the AAC2 mechanism [6], which
predicts a simple pH–rate profile based on the proto-
nation of the carbonyl oxygen at very low pH (pKa
< –2.5) followed by solvent attack on the protonated
substrate to form the rate-limiting tetrahedral interme-
diate (Scheme 3). The expected pH rate profile is a
downward bend occurring in the pH region of the sub-
strate pKa and a profile slope of –1 in the pH regions
above the pKa. The observed rate constants decrease
as a function of decreasing pH in concentrated acid
(below the pKa) has been attributed to a decrease in
water activity [42].
The rate law for the AAC2 mechanism can be given
by Eq. (1), where [H+] is the concentration or activity
of the hydronium ion; STotal stands for the total substrate
concentration; Ka is the ionization constant; ks is the
rate constant for solvolytic hydrolysis of the ionized
substrate; and fSH is the fraction of ionized species.
Scheme 1 Reaction scheme for the acidic and alkaline hy-
drolysis of formanilides.
m-methoxyformanilide
degraded
to
m-
methoxyaniline; however, mass balance was not
maintained. Since m-methoxyaniline was found to
be stable under identical conditions, parallel loss of
m-methoxyaniline occurred and the observed rate
constant for m-methoxyaniline loss (kobs) was the
sum of the rate constant (kꢁ) for the formation of
m-methoxyaniline, and the rate constant (kunk) for the
parallel loss (Scheme 2). The values for kꢁ and kunk
were estimated by nonlinear regression (WinNonlinꢀR ).
In alkaline solutions, the concentration–time pro-
files for ortho- and para-hydroxyformanilides showed
an absence of mass balance due to the rapid degrada-
tion of the aniline in alkaline solutions [40]. Thus the
estimated kobs value for the loss of the hydroxylfor-
manilides was attributable to the formanilide to aniline
conversion.
For all substrates, log-linear substrate concentration
–time profiles were used to estimate the observed first-
order rate constants (kobs), which were then used to
construct pH–rate profiles and to estimate second-order
rate constants associated with specific acid or base
catalysis. LFER plots were prepared using either first-,
second-, or higher order rate constants.
Rate = kSfSHSTotal
(1)
[H+]
Ka + [H+]
kobs = kS
(2)
In dilute hydrochloric acid concentrations (1.0–
0.010 M) formanilide is largely unionized (pKBH+
= –2.71 at 25°C), and, therefore, the hydrolysis rate
is first order in the hydronium ion concentration
(Eq. (3)) [6]. The second-order rate constant (kH) as
defined in Eqs. (4) and (5) were estimated by lin-
ear least-squares fitting of kobs as a function of the
Arrhenius plots were obtained for the temperature
dependence of specific acid catalytic constants for the
deformylation of substituted formanilides and used to
calculate the enthalpies and entropies of activation over
the temperature range 20–60°C (Table I).
Scheme 2 Reaction scheme for the acid-catalyzed hydrolysis of m-methoxyformanilide.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

478
DESAI AND KIRSCH
Figure 2 pH–rate profile for formanilide hydrolysis at 40°C in hydrochloric acid solutions. The broken line represents the
specific acid-catalyzed region where the slope is –1.
Scheme 3 Acid-catalyzed acyl cleavage bimolecular (AAC2) mechanism for hydrolysis of amides.
hydrochloric acid concentration (Table I).
plot is nonlinear and displays downward curvature
(Fig. 3A). For meta-substituted formanilides, electron-
withdrawing substituents (e.g., nitro-, chloro-, bromo-,
hydroxyl-, methoxy-, and carboxyl-) displayed an in-
creased rate constant value with increasing electroneg-
ativity, whereas electron-donating substituents (e.g.,
methyl-, isopropyl-, -phenyl, and ethyl-) displayed a
decrease with increasing electron-donating capability.
Similarly for para substituents electronegative sub-
stituents (e.g., chloro- and bromo-) exhibited an in-
crease in the kH, whereas the electron-donating groups
(e.g., methyl-, ethyl-, isopropyl-, phenyl-, methoxy-,
and hydroxyl-) showed a decrease. These two types
of substituents appeared to fall on two separate lines,
which intersect at formanilide. Strongly electron-
withdrawing substituents capable of resonance inter-
actions (p-nitro and p-carboxyl) displayed positive de-
viation from the rest of the compounds.
[H+]
kobs = kS
Ka
(3)
kS
kH
=
(4)
(5)
Ka
kobs = kH [H+]
The pH–rate profiles for the substituted for-
manilides were similar to formanilide and treated in
an analogous manner. For the acidic hydrolysis of m-
methoxyformanilide the kꢁ value was used to estimate
kH for that reaction.
The kH values for para- and meta-substituted for-
manilides were used to generate Hammett plots. Un-
der the conditions studied, all of the formanilides
were essentially unionized substrates. The Hammett
This type of nonlinear Hammett plot is consis-
tent with a complex mechanism wherein sequential
reaction steps possess opposing electronic substituent
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
479
Figure 3 Hammett plots for the acidic hydrolysis of formanilides at 40°C. (a) Hammett sigma values. Broken lines are not
fitted to the data (b) selected sigma values. The solid line is a linear fitting to the Hammett equation.
effects that are compensatory, and the second step is
rate determining [43]. These results are consistent with
the two-step AAC2 reaction mechanism in which the
second-order rate constant is the ratio of ks and Ka
(Eq. (4)). Thus the overall Hammett reaction constant
(ρ) is a function of difference between the reaction
constant values for ionization (ρpKa) and solvent attack
steps (ρs) (Eq. (6)).
of the tetrahedral intermediate is accelerated by the
electron-withdrawing substituents (ρs = positive). At
the high hydronium concentration, the first step (ion-
ization of the carbonyl oxygen) is not kinetically im-
portant since most of the substrate is ionized and thus
overall ρ (1.75) is a function of the second step (wa-
ter attack) [6]. At lower acid concentrations, the first
step becomes kinetically important since the ionized
fraction of the substrate is significantly reduced and
the ρ is a function of both the steps. The ionization
Protonation is enhanced by the electron-donating
substituents (ρpKa = negative), whereas the formation
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

480
DESAI AND KIRSCH
Figure 4 Enthalpy–entropy compensation plot for substituted formanilides. The line is a linear fit to the substituents, which
display pure inductive effects. The values in parenthesis are the Es values.
of formanilides is controlled mainly by inductive ef-
fects (ρpKa = –0.48), but the water attack is influenced
strongly by resonance effects [6]. Substituents capable
of both inductive and resonance effects affect the both
steps, which leads to the curvature in the Hammett
plot even though the rate-determining step remains
solvolytic attack.
Phenyl rings in the para position are capable of reso-
nance interactions only if the phenyl group is planar to
the formanilide benzene group. Nonplanar phenyl sub-
stituents can interact only through inductive effects.
Substituting the appropriate σ values, a Hammett plot
with selected sigma values (σ+, σ, or σ−) for the para
and meta substituents resulted in the LFER relation-
ship displayed in Fig. 3B. This plot was linear with
a slope (ρ) of 0.457 0.022 and a correlation coef-
ficient (R2) value of 0.957. The intercept was 0.843
0.011, which corresponded well to the value for
the substituted formanilide (0.869). The value of ρ is
positive and indicates that the reaction is accelerated
by electron-withdrawing groups. The value of ρ is a
composite value and corresponds well to the published
ρ value for the acidic hydrolysis of acetanilide (ρ =
0.293, 50% ethanol, 70°C) [46].
Substituent effects can be reflected in activation en-
tropies and enthalpies of the reaction. The enthalpy–
entropy plot for meta-, para-, and ortho-substituted
formanilides was generated by using activation param-
eters ꢀH‡ and ꢀS‡ (Fig. 4). A relationship between the
para- and meta-substituted formanilides was obtained
when the substituents were classified into three groups:
(1) substituents capable of mainly inductive effects in-
cluding all the meta-substituted formanilides and some
para substituents (phenyl, chloro, and bromo), (2) sub-
stituents that show strong electron-withdrawing capa-
bilities through resonance (–R effects) including the p-
carboxyl and p-nitro substituents, and (3) substituents
that show strong electron-donating capabilities
log kH = (ρS + ρpKa)σ + c = ρσ + c
(6)
An attempt was made to linearize the Hammett plot
by selecting the most appropriate substituent constant
consistent with the AAC2 mechanism and chemical in-
tuition. The Hammett substituent constants (σ) were
developed from benzoic acid ionization and repre-
sent inductive effects. The substituents with electron-
donating capabilities were assigned electrophilic sub-
stitution constants (σ+), whereas those displaying
electron-withdrawing capabilities were assigned en-
hanced sigma constants (σ−). The σ+ scale is used
in reactions that occur with the development of an
electron deficiency, such as aromatic electrophilic sub-
stitutions [44]. The σ+ values are present for most
of the para substituents and some of the meta sub-
stituents. If the σ+ value was unavailable, the σm
m
value was substituted. The enhanced sigma constants
(σ−) were available for the p-nitro and p-carboxy
groups [45].
This reconstructed Hammett plots displayed linear
behavior except for the p-phenyl substituent, which
deviated from the correlation line using the σ+ value.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
481
through resonance (+R) including the rest of the para-
substituted formanilides.
The Taft–Kutter–Hansch steric substituent values
(Es) are noted in Fig. 5A. These values have been de-
rived from the van der Waals dimensions and are in-
dicative of the steric bulk of the substituent. In the ortho
position, the inductive and resonance effects are higher
in the absence of steric bulk effects and the hydrolysis
rate of both o-carboxyl and o-nitro substituent groups
should increase. However, the rate constant for the o-
carboxyl group was nominally higher, and the rate con-
stant for the o-nitro group decreased as compared to
their para-substituted counterparts. This behavior is
likely attributable to steric inhibition of resonance for
carboxyl and nitro groups.
To investigate the presence of steric inhibition of
resonance, the ortho substituents were plotted using
Hammett sigma (σp) values instead of the enhanced
sigma values (Fig. 5B). The o-hydroxyl and o-methoxy
substituents fell closer to the correlation line. The
methoxy substituent showed a steric substituent effect
consistent with its steric bulk. The nitro and phenyl
groups have two values associated with them since
they are planar groups, and their cross-sectional areas
change with the rotation of the group. The reduction in
the rate of hydrolysis for the nitro group and the phenyl
group are consistent with the value corresponding to
the lower Es value, i.e. the plane of the substituent is
perpendicular to the plane of the formanilide benzene
ring. Since these substituents are not planar to the ring,
the resonance effects of the nitro and phenyl groups
were minimal, leaving steric and inductive effects as
primary contributors.
The reaction constants of the remaining sub-
stituents (chloro, bromo, methyl, ethyl, and iso-
propyl) all decrease in proportion to their Es values.
The o-carboxyformanilide degrades faster than the p-
carboxyformanilide. This acceleration could be due to
the increase in the inductive effect and/or a weak in-
tramolecular catalytic effect.
The Fujita–Nishioka equation (Eq. (8)) was used to
describe the steric effects [15]. The equation assumes
that the overall rate constant for ortho-substituted com-
pounds is a linear function of three components, i.e.
electronic effects, steric effects, and the proximity po-
lar effects. The first term (ρσ) corresponds to the elec-
tronic effects using various substituent constants (σm,
σp, σ+, σ−). The second term (δEs) corresponds to the
steric substituent constants (Esortho). The third term (f
F) corresponds to the proximity polar effects, which
are approximated by the Swain Lupton (F) constant.
The reaction constants (ρ, δ, and f) correspond to the
electronic, steric, and proximity reaction constants, re-
spectively. The proximity effects and steric terms are
only used for ortho-substituted compounds. The last
term (c) is a constant and corresponds to the logarithm
The substituents associated with predominant in-
ductive effects (Group 1) showed a linear correlation
(R2 = 0.974), which suggests a common interaction
mechanism [47]. This linear relationship is referred to
the as the entropy–enthalpy compensation and can be
represented by Eq. (7) where β is the compensation
temperature and α is a constant. The squared corre-
lation coefficient R2 value was somewhat less than
unity; hence an isokinetic relationship could not be
assigned [48]. This observation is consistent with the
complex rate law [49]. The compensation tempera-
ture (β) is 334 17 K (61.6 17.2°C). At the com-
pensation temperature, the differences in reactivities
due to free energy changes become negligible and
the substituent effects are muted [45]. The compen-
sation temperature for these substituents is very close
to the reaction temperature (40°C) and could account
for the small values for the Hammett reaction constant.
Groups 2 and 3 substituents show significant devia-
tions from this correlation probably due to their strong
electron-donating/accepting capabilities, which affect
the interaction mechanisms involved in the hydrolysis.
ꢀH ‡ = βꢀS‡ + α
(7)
In addition to inductive and resonance effects, or-
tho substituents are also capable of steric interactions.
Steric interactions include the ability to accelerate or
decelerate the reactions due to steric bulk, proximity
effects, and anchimeric assistance. A Hammett plot
for the ortho-substituted formanilides was prepared by
assuming that the electronic effects (inductive and res-
onance) in the ortho position were similar to those
for corresponding para substituents. The Hammett plot
for ortho-substituted formanilides has been overlaid on
the Hammett plots for the para and meta formanilides
(Fig. 5). The Hammett correlation line based on
the para and meta substituents divides the ortho-
substituted formanilides into two groups: Those
containing oxygen substituents (i.e., o-methoxy, o-
hydroxyl, and o-carboxyl) fall above the line, and
those bulky substituents without hydrogen-bonding ca-
pability fall below the line. An increased rate of hy-
drolysis for the o-hydroxyl group due to anchimeric
assistance is possible but fails to explain the rate en-
hancement observed for the methoxy analog; therefore,
electronic field effects may be responsible for the ob-
served rate increases for both analogs. Steric hindrance
is the likely cause of the slower hydrolysis rates for
ortho-substituted methyl, ethyl, nitro, chloro, bromo,
isopropyl, and phenyl substrates.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

482
DESAI AND KIRSCH
Figure 5 (a) Hammett plots for para-, meta-, and ortho-substituted formanilides using (a) selected sigma values. (b) Hammett
plots for para-, meta-, and ortho-substituted formanilides using Hammett sigma values for ortho substituents (assuming steric
inhibition of resonance). The values in parenthesis are the Taft–Kutter–Hansch values for the steric substituent Es.
of the unsubstituted compound.
hibition of resonance hence the Hammett sigma (σp)
values were used. The Es values and F values used
for the ortho-substituted formanilides are given in Ta-
ble I. The o-hydroxyl and o-carboxyl substituents were
not included in the analysis since Es and F values are
not available for these substituents. Multiple linear re-
gression (JMPꢀR , SAS Institute, Cary, NC) was used to
estimate the parameters associated with Eq. (8). The
log kH = ρσo,m,p + δESortho + f F + c
(8)
The substituent constants used for the para and
meta substituents were the same ones used to fit the
linear Hammett plot in Fig. 3B. In the ortho posi-
tion, the resonance effects are muted due to the in-
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
483
Figure 6 pH–rate profile for alkaline hydrolysis of formanilide using the observed rate constants (kobs) at 25°C. The solid line
was simulated using Eq. (9) and estimated from Table II.
adjusted squared correlation constant (R2 adjusted) value
was 0.974. All four terms (ρ, δ, c, and f) were found
to be statistically significant by the t test (p < 0.001).
The F test for the model was significant (p < 0.001).
The ρ value, 0.469 0.021, was similar to the value
obtained from the Hammett plot for the para- and meta-
substituted formanilides (0.457). The δ parameter had
a value of 0.327 0.019. The positive sign indicates
steric inhibition, and the magnitude of the steric con-
stant is consistent with minimal crowding at the rate-
determining reaction site [15]. The estimated value is
consistent with similar reactions, i.e. acidic hydrolysis
of phenyl acetates [50].
tent with minimal steric effects and significant induc-
tive effects. The other ortho substituents lie above the
correlation line associated with inductive effects, indi-
cating the presence of significant ortho effects on the
hydrolysis reaction [49]. The substituents appear to be
distributed randomly without a clear dependence on
Es. Therefore, the ortho effect for these substituents
appears to be a function of both the steric and prox-
imity effects in agreement with the Fujita–Nishioka
analysis.
Alkaline Hydrolysis
The alkaline hydrolysis of formanilide was studied in
sodium hydroxide (0.010–3.0 M) solutions at 25°C. At
high concentrations of sodium hydroxide, the acidity
function (H-, based on the structurally similar thioac-
etamide) was used. These H-values (obtained from the
literature [51,52]) were used to construct the pH–rate
profiles (e.g., Fig. 6) wherein the kobs values displayed
a dependence on the hydroxide concentration that was
higher than first order but less than second order. This
behavior has been previously reported for anilides, and
the modified base- catalyzed acyl cleavage bimolecular
(BAC2) mechanism (Scheme 4) involving both a higher
order dependence on the hydroxide concentration and
reduction in reactive substrate due to ionization has
been used for these reactions [53–55]. The unionized
substrate can deprotonate to form a conjugate base,
which is unreactive, or it can be attacked by the hy-
droxide ion to form a tetrahedral intermediate that
The f value was 0.697
0.062, which is similar
to values seen for other acid-catalyzed hydrolysis re-
actions [15]. This value was expected to be small as
it is the sum of the proximity polar effects of two
steps, i.e. ionization of the substrate (–f value), and
rate-determining nucleophilic attack (+f value). The
intercept value (c) was 0.836
0.011, which corre-
sponds to the value of log kH for the unsubstituted
formanilide (0.869).
The activation enthalpy–entropy plot for the ortho-
substituted formanilides was also examined (Fig. 4).
The ortho-substituted formanilides appear to be scat-
tered around the para- and meta-substituted for-
manilides. On overlaying the ortho effects against the
relationship developed for the meta and para relation-
ships, the ortho substituents are split into two groups.
The o-nitro and o-carboxyl substituted formanilides
lie close to their para counterparts, which is consis-
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

484
DESAI AND KIRSCH
Scheme 4 Modified base catalyzed acyl cleavage bimolecular (BAC2) mechanism for hydrolysis of anilides.
subsequently degrades to starting materials or prod-
ucts via water-catalyzed and/or hydroxide-catalyzed
pathways.
The pH–rate profiles for selected substituted
formanilides were similar (p-chloro, p-methyl, p-
methoxy, and m-nitro) to that of formanilide. The pKa
values were taken from the literature (formanilide and
m-nitroformanilide) or calculated using a Hammett re-
lationship (Eq. (12)) for the ionization constants for for-
manilides at 20°C using σ0 (Taft substituent constants
for substituents that do not show resonance delocaliza-
tion interactions) and are given in Table III [9,56].
In Scheme 4, ka is the rate constant for the addi-
tion of the hydroxide ion to the unionized substrate,
k-a is the rate constant for the degradation of the in-
termediate to reactants, kꢁ1 is the rate constant for the
conversion of the intermediate to the products via the
water-catalyzed pathway, kꢁ2 is the rate constant for
the conversion of the intermediate to the products via
the hydroxide-catalyzed pathway, and Ka is the ioniza-
tion constant for the formanilide. The rate law is given
by Eq. (9), where k1 is the complex rate constant for
the conversion of reactants to products via the water-
catalyzed pathway (Eq. (10)); k2 is the complex rate
constant for the conversion of reactant to products via
the hydroxide-catalyzed pathway (Eq. (11)); and Kw is
the dissociation constant for water.
pKa = −1.52 × σ0 + 13.8
(12)
Nonlinear regression (JMPꢀR , SAS Institute, Cary,
NC) was used to estimate k1, k2, and ka values. Rate
constant values were weighed using the square of re-
ciprocal of the rate constant [57]. Good agreement be-
tween model estimated and observed rate constants as
a function of pH (or H-) is depicted in Fig. 6. The
estimated rate constants are presented in Table III.
Hammett plots constructed for each pathway were
linear with R2 > 0.91. The reaction constant value for
ka (ρka) was 1.40 0.09, indicating that the reaction is
accelerated by electron-withdrawing groups (Fig. 7).
The reaction constant for steps k1 and k2 was (ρk1)
–0.23 0.04 and (ρk2) of 0.63 0.12, respectively,
indicating that the reaction process for k1 was acceler-
ated by electron-donating groups whereas the reaction
process for k2 was accelerated by electron-withdrawing
groups.
ꢀ
ꢁ
k1 + k2[OH]
Kw
kobs
=
ka[OH]
ka + k1 + k2[OH]
[OH]Ka + Kw
(9)
ka
k1 = kꢁ
(10)
(11)
1 k
−a
ka
k2 = kꢁ
2 k
−a
Table III Rate constants and ionization constants for the base catalyzed hydrolysis of formanilides at 25° C
pKa
ka SE (M–1 h–1
)
k1 SE (M–1 h–1
)
k2 SE (M–2 h–1
)
p-Methoxyc
p-Methylc
Hc
p-Chloroc
m-Nitrod
14.1b
14.0b
13.9a
13.4b
12.8a
30.7 3.9
40.0 4.7
51.1 5.0
121 26
0.684 0.117
0.629 0.079
0.514 0.054
0.527 0.113
0.389 0.231
12.8 1.95
11.2 1.25
12.6 1.00
18.2 2.27
50.8 7.70
760 366
a pKa values from the literature. b pKa values calculated using the Hammett relationship (Eq. (12)). c95% confidence intervals were generated
for all parameters. d90% confidence intervals were generated for all parameters.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
485
Figure 7 Hammett plots for the rate constants of alkaline hydrolysis of formanilides at 25°C. The lines are linear regression
fits of the data to the Hammett equation.
A modified BAC2 mechanism is consistent with the
Hammett plots. The formation of the tetrahedral in-
termediate (monoanion) occurs by the addition of hy-
droxide to the carbonyl carbon. The strong electron-
withdrawing groups enhance the susceptibility of the
carbonyl carbon to nucleophilic hydroxide attack. The
ρka value of 1.34 is in agreement with reaction con-
stants for similar reactions such as alkaline hydroly-
sis of acetanilide (1.0), and 2,2,2-trifluoroacetanlide
(1.18) [54,58].
The values for k1 and k2 can be explained by a
Scheme 5 Hypothetical structures for intermediates in the
two-step process involving (1) the formation of the
base-catalyzed hydrolysis of anilides.
intermediate and (2) partitioning of the intermediate
to products or reactants. The formation of the inter-
mediate is given by ka, whereas the partition of the
to be the rate-limiting intermediate [58]. The dipolar
ion is formed by the rate-limiting proton transfer from
the hydroxyl group to the amine group by water. The
formation of the protonated amine is dependent on the
basicity of the nitrogen atom of the anilide. The ba-
sicity of the amine is enhanced by electron-donating
groups; therefore, the anilines with higher pKa values
will enhance the formation of the dipolar ion. The sign
and magnitude of the reaction constant for the amine
protonation can be approximated by the reaction con-
stant for the ionization of anilines (ρ = –2.77) [46].
intermediate via k1 and k2 processes is given by the
partition ratios k1ꢁ /k–a and k2ꢁ /k–a, respectively. The par-
tition ratios were calculated by dividing the complex
rate constant k1 and k2 with ka (Eqs. (10) and (11)).
In Fig. 7, the Hammett plot for the rate constants (ka,
k2ꢁ /k–a and k2ꢁ /k–a) are shown. The value of the reac-
tion constant ρka for the addition step is 1.40, whereas
the value of reaction constant for the partition ratio
ꢁ
(ρk1/k–a) is –1.63 0.07. The overall reaction constant
for the first-order process (ρk1), –0.23, is the sum of
the reaction constants for the two steps.
The magnitude of the substituent effect for the parti-
tioning process predominates the net reaction constant.
ꢁ
The significant negative value for ρk1/k–a is consistent
with this process.
The reaction constant for the pathway involving
second-order dependence on hydroxide (described by
the rate constant k2) was estimated to be +0.63, and
ꢁ
This value for ρk1/k–a is comparable to the value of –
1.1 obtained in the alkaline hydrolysis of acetanilide
where a dipolar ion (Scheme 5) has been hypothesized
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

486
DESAI AND KIRSCH
Figure 8 Hammett plots for the hydrolysis of substituted formanilides in 0.10 M sodium hydroxide solutions at 40°C.
the value of the reaction constant for the partition ratio
used for p-hydroxyformanilide in the Hammett plot
(Fig. 8).
ꢁ
(ρk2/k–a) was –0.76 0.05. This value is significantly
smaller than the reaction constant for the addition pro-
cess and hence the substituent effect for the formation
of the intermediate predominates in the net reaction
constant (ρk2 = 0.63). A likely mechanism for the
second-order hydroxide pathway involves formation
of a dianion intermediate (Scheme 5) and breakdown
to products via rate-determining formation of a dipo-
lar ionic intermediate [54,59]. The formation of the
dipolar ion is likely the rate limiting. The rate con-
stant for the second-order reaction (k2ꢁ ) contains the
rate constant for the formation of the dianion and the
rate-determining formation of the dipolar ion. The ki-
netic process for the formation of the dianion is very
similar to the kinetic process for the formation of the
monoanion, and hence the reaction constant for the for-
mation of the dianion is also expected to be positive.
Nearly all of the observed first-order rate constants
values for the para- and meta-substituted substrates
were within the range of 0.2–0.7 h–1, and therefore ap-
peared to be independent of the Hammett sigma values.
These results agree with studies on the alkaline hydrol-
ysis of anilides [58,60]. The estimated kobs value for
the p-nitro substrate was about 350-fold greater than
the other para- and ortho-substituted substrates, which
suggests a different mechanism of alkaline hydroly-
sis [60–63].
Results from the alkaline hydrolysis of ortho-
substituted substrates were compared to their para-
substituted counterparts (Fig. 8). With the exception of
the o-nitroformaniline, the ortho compounds typically
degraded more slowly than para or meta substrates.
The degree of stabilization varied from <10% to
>80% for the bromine- and isopropyl-substituted com-
pounds, respectively. In general, stabilization could
be attributed to steric effects using the Taft–Kutter–
Hansch treatment (Eq. (13), Fig. 9) wherein a lin-
ear correlation was observed between the logarithm
of kobs and the Taft–Kutter–Hansch steric substituent
ꢁ
The value of ρk2 is a sum of the positive reaction con-
stant for the dianion formation and negative reaction
constant for the formation of the dipolar ion (amine
ꢁ
basicity); and hence ρk2/k–a is only slightly negative.
Owing to the complexity of the kinetic scheme, the
effects of ortho substitution on each elementary step
in a pathway could not be determined. Hence sub-
stituent effects were evaluated using the observed first-
order rate constant (kobs) estimated from a series of
studies conducted under identical experimental condi-
tions (0.10 M sodium hydroxide and 40.0°C). Since
the hydroxyl-substituted substrates were primarily de-
protonated under the experimental conditions (pKa =
9.6), the phenoxide-substitutent constant (–0.81) was
value (Es) with a δapp of 0.505
0.094 and a cor-
relation coefficient (R2) of 0.8787. The positive sign
for the value indicates that the steric bulk inhibits the
reaction, whereas the value indicates that the steric
bulk inhibition of reaction is significant. Thus the first-
order process is inhibited strongly by steric bulk. The
rate-limiting states for both the first- and second-order
pathways for alkaline hydrolysis of formanilides are
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

THE ORTHO EFFECT ON THE ACIDIC AND ALKALINE HYDROLYSIS
487
Figure 9 Plot of the logarithm observed rate constants for ortho-substituted formanilides versus the steric substituent constant
(Es) determined in 0.10 M sodium hydroxide at 40°C. The solid line indicates the linear correlation line to Eq (13), whereas the
broken lines indicate 95% confidence intervals. The substituents included in the regression analysis are denoted by solid circles
(•) and those not included are denoted by open circles (◦).
similar, and, therefore, it is possible that the steric bulk
inhibition effect will be similar for both the pathways.
The alkaline hydrolyses showed first- and second-
order dependence on hydroxide, which be explained
adequately by the modified BAC2 mechanism. The
Hammett plots for each step were found to be con-
sistent with the BAC2 mechanism. Ortho effects in al-
kaline were found to be primarily dependent on steric
bulk.
log kobs = δappEs + c
(13)
The relative instability of the o-nitro substituent
is similar to p-nitroformanilide, which may degrade
by a different pathway then the other substituted for-
manilides. The o-bromo and o-chloro substituents are
strongly electron withdrawing, and thus the second-
order hydroxide pathway may play a predominant role
in their alkaline hydrolysis. These substrates were not
considered in evaluating steric effects using Eq. (13).
BIBLIOGRAPHY
1. Yoshioka, S.; Stella, V. J. Stability of Drugs and Dosage
Forms; Kluwer: New York, 2000; pp. 3–137.
2. Wang, W. Int J Pharm 1999, 185(2), 129–188.
3. Ma, M.; Sun G. Dyes Pigm 2004, 63(1), 39–49.
4. Brown, R. S. In The Amide Linkage: Selected Struc-
tural Aspects in Chemistry Biochemistry and Material
Science; Greenberg, A.; Breneman, C. M.; Liebman,
J. F., Eds.; Wiley: New York, 2000; pp. 85–114.
5. Challis, B. C.; Challis J. In The Chemistry of
Amides; Zabicky, J., Ed.; Interscience: New York, 1970;
pp. 731–858.
6. Kavalek, J.; Sterba V. Collect Czech Chem Commun
1975, 40(6), 1924–1231.
7. Bergstrand, B. Acta Pharm Suec 1983, 20(2), 81–98.
8. Bergstrand, B. Acta Pharm Suec 1984, 21(1), 43–54.
9. Kavalek, J.; Sterba V. Collect Czech Chem Commun
1975, 40(4), 1176–1182.
CONCLUSION
The acidic hydrolyses of formanilides occurs by the
AAC2 mechanism and showed nonlinear Hammett
plots. The nonlinear Hammett plots were explained
using the two-step AAC2 mechanism. Ortho substitu-
tion led to inhibition of resonance between the sub-
stituent and reaction site by disrupting the planarity of
the amide group with the phenyl ring. The analysis of
the ortho effect using Fujita–Nishioka analysis showed
the presence of steric bulk and through space effects.
No evidence of anchimeric assistance was observed for
ortho-substituted substrates.
10. Bergstrand, B. Acta Pharm Suec 1984, 21(1), 55–66.
11. Bergstrand, B. Acta Pharm Suec 1985, 22(1), 1–16.
12. Bergstrand, B.; Sundelof L. O. Acta Pharm Suec 1981,
18(5), 283–298.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925

488
DESAI AND KIRSCH
13. Cheshmedzhieva, D.; Ilieva, S.; Hajieva, B.; Galabov,
B. J Phys Org Chem 2009, 22(6),619–631.
14. Cheshmedzhieva, D.; Ilieva, S.; Hajieva, B.; Trayanova,
T.; Galabov, B. Mol Phys 2009, 107, 1187–1192.
15. Fujita, T.; Nishioka, T. In Progress in Physical Organic
Chemistry; Taft, R. W., Ed.; Wiley: New York, 1976;
pp. 49–89.
16. Exner, O. In Correlation Analysis in Chemistry: Recent
Advances; Chapman, N. B.; Shorter, J., Eds. Plenum
Press: New York, 1978; pp. 439–540.
39. Kamer, P. C. J.; Nolte, R. J. M.; Drenth, W. J Am Chem
Soc 1988, 110(20), 6818–6825.
40. Koshy, K. T.; Lach, J. L. J Pharm Sci 1961, 50(2), 113–
118.
41. Yates, K.; Riordan, J. C. Can J Chem 1965, 43(8), 2328–
2335.
42. Bruice, T. C.; Benkovic, S. Bioorganic Mechansims;
W. A. Benjamin: New York, 1966; pp. 1–211.
43. Jencks, W. P. Catalysis in Chemistry and Enzymology;
Dover: New York, 1987; pp. 480–483.
17. Werner, W. J Chem Res, Synop 1981(4), 120–120.
18. Huffman, C. W. J Org Chem 1958, 23(5),727–729.
19. Plato, C.; Glasgow, Jr A. R. Analyt Chem 1969, 41(2),
330–336.
20. Gottlieb, H. E.; Kotlyar, V.; Nudelman, V. J Org Chem
1997, 62(21), 7512–7515.
21. Gran, G. Analyst 1952, 77, 661–671.
22. Albert, A.; Serjeant E. P. The Determination of Ion-
ization Constants: A Laboratory Manual; Chapman and
Hall: New York, 1984; pp. 70–73.
23. Hosseini-Sarvari, M.; Sharghi H. J Org Chem 2006,
71(17), 6652–6654.
24. Yale, H. L. J Org Chem 1971, 36(21), 3238–3240.
25. Hill, D. R.; Hsiao, C. N.; Kurukulasuriya, R.; Witten-
berger, S. J. Org Lett 2002, 4(1), 111–113.
26. Di Nunno, L.; Scilimati, A. Tetrahedron 1986, 42(14),
3913–3920.
44. Brown, H. C.; Okamoto, Y. J Am Chem Soc 1958, 80,
4979–4987.
45. Connors, K. A. Chemical Kinetics: The Study of Re-
action Rates in Solution; VCH: New York, 1990;
pp. 311–384.
46. Jaffe, H. H. Chem Rev 1953, 53, 191–261.
47. Leffler, J. E.; Grunwald, E. Rates and Equilibria of Or-
ganic Reactions. Dover: New York, 1989; pp. 128–170.
48. Liu, L.; Guo, Q.-X. Chem Rev 2001, 101, 673–695.
49. Leffler, J. E.; Grunwald, E. Rates and Equilibria of Or-
ganic Reactions; Dover: New York, 1989; pp. 315–402.
50. Nishioka, T.; et al. Fujira, T.; Kitamura, K.; Nakajima,
M. J Org Chem. 1975, 40(17), 2520–2525.
51. Edward, J. T.; Wang, I. C. Can J Chem 1962, 40, 399–
408.
52. Bowden, K.; Bromley, K. J Chem Soc, Perkin Trans 2:
Phys Org Chem 1990(12), 2103–2109.
27. Pettit, G. R.; Kalnins, M. Liu, T.; Thomas, E.; Parent,
K. J Org Chem 1961, 26, 2563–2566.
53. Biechler, S. S.; Taft, R. W. J Am Chem Soc 1957, 79(18),
4927–4935.
28. Berrier, C.; Jacquesy, J. C.; Jouannetaud, M. P.; Morel-
let, G. Bull Soc Chim Fr 1987(1), 158–164.
29. Farley, T. M.; Strong, F. M.; Bydalek, T. J. J Am Chem
Soc 1965, 87(15), 3501–3504.
30. Pettit, G. R.; Thomas, E. G. J Org Chem 1959, 24, 895–
896.
31. Exner, O.; Lakomy, J. Collect Czech Chem Comm 1970,
35(5), 1371–1386.
32. Buss, C. E.; Mann, K. R. J Am Chem Soc 2002, 124(6),
1031–1039.
33. Wiberg, K. B.; Shryne, T. M.; Kintner, R. R. J Am Chem
Soc 1957, 79, 3160–3164.
34. Lesiak, T. Chem Stosowana, Ser A, 1963, 7(4), 593–
601.
35. Gim, H. J.; Kang, B. M.; Jeon, R. Arch Pharm Res 2007,
30(9), 1055–1061.
36. Arient, J.; Dovorak, J. Chem Listy Vedu Prum 1956, 50,
1636–1638.
54. Kershner, L. D.; Schowen, R. L. J Am Chem Soc 1971,
93(8), 2014–2024.
55. Eriksson, S. O. Acta Pharm Suec 1969, 6(2), 121–138.
56. Yukawa, Y.; Tsuno, Y.; Sawada, M. Bull Chem Soc Jpn
1972, 45(4), 1198–1205.
57. Motulsky, H.; Christopoulos, A. Fitting Models to Bi-
ological Data Using Linear and NonLinear Regression.
A Practical Guide to Curve Fitting; GraphPad Software:
San Diego CA, 2003; pp. 84–89.
58. Bender, M. L.; Thomas, R. J. J Am Chem Soc 1961,
83(20), 4183–4189.
59. Schowen, R. L.; Jayaraman, H. Kershner, L.; Zuorick,
G. W. J Am Chem Soc 1966, 88(17), 4008–4012.
60. DeWolfe, R. H.; Newcomb, R. C. J Org Chem 1971,
36(25), 3870–3878.
61. Sorci, J. J.; Macalady, D. L. J Agric Food Chem 1993,
41(10), 1760–1766.
62. Pollack, R. M.; Bender, M. L. J Am Chem Soc 1970,
92(24), 7190–7194.
63. Broxton, T. J.; Deady, L. W. J Org Chem 1975, 40(20),
2906–2910.
37. Zimmermann, J. Ber Dtsch Chem Ges 1880, 13(2),
1963–1969.
38. Kisfaludy, L.; Otvos, L., Jr. Synthesis 1987(5), 510.
International Journal of Chemical Kinetics DOI 10.1002/kin.20925