Equilibrium formation of anilides from carboxylic acids and anilinesin aqueous acidic media

Article abstract of DOI:10.1021/ja990104d

The formations of formanilide, p-methoxyformanilide, p-nitroformanilide, and acetanilide from their corresponding carboxylic acids and anilines in aqueous acidic media have been investigated at temperatures between 60 and 100 °C under a variety of conditions such as pH, D₂O, added phosphate, and added ethanol. In each case, the pseudo-first-order rate constants for the establishment of equilibrium (k_obs), from both the hydrolysis and formation directions, and the conditional equilibrium constant (K′ = [anilide]/[aniline]_total) were determined in excess formate. From K′, and knowledge of how the pK_a values of RCOOH and anilinium ion depend on the various conditions, is derived a corrected equilibrium constant, K′_eq, defined as [anilide]/ ([aniline][RCOOH]). In the case of formanilide, the K′ value is found to be invariant with temperature reductions, although the K′_eq value increases. In D₂O media, the K′ value drops slightly, but after correcting for the medium induced changes in [aniline] and [RCOOH], the K′_eq value is the same as in water. In the presence of added KH₂PO₄, the rate of establishment of equilibrium increases but the K′ and K′_eq values do not change relative to their values without phosphate. Added ethanol is found to increase both the rate of establishment of equilibrium and the K′ equilibrium constants, but reduces K′_eq. The mechanism of formation of anilides in water under acidic conditions is discussed.

Full text of DOI:10.1021/ja990104d

4598
J. Am. Chem. Soc. 1999, 121, 4598-4607
Equilibrium Formation of Anilides from Carboxylic Acids and
Anilinesin Aqueous Acidic Media
Ahmed M. Aman and R. S. Brown*
Contribution from the Department of Chemistry, Queen’s UniVersity, Kingston, Ontario, Canada K7L 3N6
ReceiVed January 11, 1999. ReVised Manuscript ReceiVed March 22, 1999
Abstract: The formations of formanilide, p-methoxyformanilide, p-nitroformanilide, and acetanilide from their
corresponding carboxylic acids and anilines in aqueous acidic media have been investigated at temperatures
between 60 and 100 °C under a variety of conditions such as pH, D2O, added phosphate, and added ethanol.
In each case, the pseudo-first-order rate constants for the establishment of equilibrium (kobs), from both the
hydrolysis and formation directions, and the conditional equilibrium constant (K′ ) [anilide]/[aniline]total) were
determined in excess formate. From K′, and knowledge of how the pKa values of RCOOH and anilinium ion
depend on the various conditions, is derived a corrected equilibrium constant, K′eq, defined as [anilide]/
([aniline][RCOOH]). In the case of formanilide, the K′ value is found to be invariant with temperature reductions,
although the K′eq value increases. In D2O media, the K′ value drops slightly, but after correcting for the medium
induced changes in [aniline] and [RCOOH], the K′eq value is the same as in water. In the presence of added
KH2PO4, the rate of establishment of equilibrium increases but the K′ and K′eq values do not change relative
to their values without phosphate. Added ethanol is found to increase both the rate of establishment of equilibrium
and the K′ equilibrium constants, but reduces K′eq. The mechanism of formation of anilides in water under
acidic conditions is discussed.
Introduction
bonds in aqueous media in the absence of enzymes is also of
interest since this is implicated in the origin of life. Attempts4-8
have been made to form peptides from amines and acids or by
coupling amino acids under conditions that resemble those of
primitive earth. In studies simulating so-called prebiotic condi-
tions in water, linear and cyclic polyphosphates,5 cyanamide,6
metal ions,7 silica, alumina and clay,8 and more recently iron
sulfide plus H2S9 have been used for promoting amide bond
formation from acids and amines.
The great bulk of studies concerning amide hydrolysis are
conducted under acidic or basic conditions where the reaction
proceeds essentially to hydrolyzed products.1 Consequently, it
is generally believed that amide bond formation in aqueous
solution is unfavorable because the experimental conditions
under which the reactions are conducted favor the hydrolysis
process (eq 1). However, from the biochemical perspective, it
Reformation of lactams from their amino acid hydrolysis
products is well-documented and, in many cases, is very
favorable due to the intramolecularity of the process.10 Studies
by Fersht and Requena,11 Morawetz and Otaki,12 Guthrie,13 and
has long been known that hydrolytic enzymes are also capable
of reforming peptides in solution.2,3 The formation of amide
(4) (a) Oie, T.; Loew, G. H.; Burt, S. K.; Binkley, J. S.; MacElroy, R.
D. J. Am. Chem. Soc. 1982, 104, 6169. (b) Oie, T.; Loew, G. H.; Burt, S.
K.; Binkley, J. S.; MacElroy, R. D. J. Am. Chem. Soc. 1983, 105, 2221.
(5) (a) Chung, N. M.; Lohrmann, R.; Orel, E.; Rabinowiz, J. Tetrahedron
1971, 27, 1210. (b) Yamanaka, J.; Inomata, K.; Yamagata, Y. Origins Life
1988, 18, 165. (c) Rabinowitz, J.; Flores, J.; Krebsbach, R.; Rogers, G.
Nature 1969, 224, 795. (d) Rabinowitz, J.; Hampai, A. J. Mol. EVol. 1985,
21, 199.
(6) (a) Steinman, G. D.; Kenyon, D. H.; Calvin, M. Biochem. Biophys
Acta 1966, 124, 339. (b) Ponnamperuma, C.; Peterson, E. Science 1965,
147, 1572. (c) Steinman, G.; Kenyon, D. H.; Calvin, M. Nature 1965, 206,
707.
(1) For reviews on this subject, see: (a) Bennett, A. J.; Brown, R. S.
Physical Organic Chemistry of Acyl Transfer Reactions. In ComprehensiVe
Biological Catalysis; Sinnott, M. L., Ed.; Academic Press Inc: London,
1997; Vol. 1, pp 293-326. (b) Brown, R. S. Studies of Amide Hydrolysis:
The Acid, Base, and Water Reactions. In Biochemical Significance of the
Amide Linkage; Greenberg, A., Breneman, C., Liebman, J., Eds.; Wiley-
Interscience: New York, 1999; Chapter 2, in press. (c) Jencks, W. P.
Catalysis in Chemistry and Enzymology; McGraw-Hill: New York, 1969;
pp 7-242, 463-554.
(2) (a) Fruton, J. S. In AdVances in Enzymology; Meister, A., Ed.; J.
Wiley and Sons: New York, 1982; pp 239-306. (b) Dobry, A.; Fruton, J.
S.; Sturtevant, J. M. J. Biol. Chem. 1952, 195, 149. Kullmenn, W. J. Biol.
Chem. 1980, 255, 8234. (c) Carpenter, F. H. J. Am. Chem. Soc. 1960, 82,
1111. (d) Inovyl, K.; Watanabe. K.; Morihara, K.; Tochino, Y.; Kanaya, T.
Emura, J.; Sekakibaras, S. J. Am. Chem. Soc. 1979, 101, 751. (e) Esowa,
Y.; Ohmori, M.; Lchikawa, T.; Kurite, H.; Sato, M.; Mori, K. Bull. Chem.
Soc. Jpn. 1977, 50, 2762. (f) Westeneys, H.; Borsook, H. Physiol. Re.V.
1930, 10, 110. (g) Morihara, K.; Oka, T.; Tsuzuki, H. Arch. Biochem.
Biophys. 1969, 132, 489. (h) Gawron, O.; Glaid, A. J.; Boyle, R. E.;
Odstrchel, G. Arch. Biochem. Biophys. 1961, 95, 203.
(7) Rishpon, J.; O’Hara, P.; Lahav, N.; Lawless, J. G. J. Mol. EVol. 1982,
18, 179.
(8) (a) Bujdak, J.; Rode, B. M. J. Mol. EVol. 1997, 45, 457. (b)
Schwendinger, G. M.; Rode, B. M. J. Mol. EVol. 1992, 22, 349.
(9) Keller, M.; Blochl, E.; Wachtershauser, G.; Stetter, K. O. Nature
1994, 368, 836.
(10) (a) Kirby, A. J.; Mujahid, T. G.; Camilleri, P. J. Chem. Soc., Perkin
Trans. 2 1979, 1610. (b) Camilleri, P.; Ellul, R.; Kirby, A.; Mujahid, T. G.
J. Chem. Soc., Perkin Trans. 2 1979, 1617. (c) Fife, T. H.; Duddy, N. W.
J. Am. Chem. Soc. 1983, 105, 74 and reference therein.
(11) Fersht, A. R.; Requena, Y. J. Am. Chem. Soc. 1971, 93, 3499.
(12) Morawetz, H.; Otaki, P. S. J. Am. Chem. Soc. 1963, 85, 436.
(13) (a) Guthrie, J. P. J. Am. Chem. Soc. 1974, 96, 3608. (b) Guthrie, J.
P.; Pike, D. C.; Lee, Y. Can. J. Chem. 1992, 70, 1671.
(3) (a) Homandberg, G. A.; Laskowski, M. Biochemistry 1979, 18, 586.
(b) Homandberg, G. A.; Mattis, J. A.; Laskowski, M. Biochemistry 1978,
17, 5220.
10.1021/ja990104d CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/01/1999

Equilibrium Formation of Anilides
J. Am. Chem. Soc., Vol. 121, No. 19, 1999 4599
us14 have shown that it is possible, under certain conditions, to
form substantial amounts of simple amides from intermolecular
reaction of their constituent acids and amines in water either
with or without catalysis. In view of the attractiveness of forming
the amide bond under “green” conditions that avoid the use of
acyl activating agents and dry, non-hydroxylic solvents, we have
embarked upon a program to investigate the scope and limita-
tions of the aqueous amide bond formation so as to define the
potential applications. The following reveals our findings for
the formation of simple anilides in aqueous acidic media under
acidic conditions.
obtained by injecting, and determining the peak areas of, a 1:1
formanilide and aniline mixture. For each run two rate constants could
be obtained, for example, in the hydrolysis direction, the pseudo-first-
order rate constant of disappearance of the formanilide as well as the
rate constant for the appearance of aniline.
Kinetic data for formation and hydrolysis of acetanilide were
obtained similarly at 98 ( 2 °C, pH 1.95 ( 0.03 and pH 3.72 ( 0.03
with (5.2-5.4) × 10-3 M solutions of acetanilide or aniline, µ ) 1.0
(KCl). At lower pH, HCl comprised the buffer and 1.0 M acetic acid
was added to the solutions. At higher pH, acetate buffer was used with
the total [buffer] being 1.0 M.
E. Kinetics by UV-Vis Spectrophotometry. The rates of formation
and hydrolysis of formanilide, p-nitroformanilide, and p-methoxyfor-
manilide were observed at 79 ( 1 °C using a Cary-219 UV-vis
spectrophotometer interfaced with an IBM 486 PC equipped with Olis
software (Online Instrument Systems, Jefferson GA, 1992). Kinetic data
were obtained by observing the rate of change in absorbance (increase
for formation and decrease for hydrolysis) of formanilide at 246 nm
and p-methoxyformanilide at 260 nm. For p-nitroformanilide, rates of
formation and hydrolysis were obtained by following the rates of
disappearance and appearance of p-nitroaniline, respectively, at 429
nm. Runs were initiated by injecting 5 µL of a stock solution in DME
((5.0-6.9) × 10-2 M) of either (p-H, p-NO2, or p-OCH3)-aniline or
(p-H or p-NO2 or p-OCH3)-formanilide into 3 mL of formate buffer
(pH 2.80-4.20, [buffer]total ) 0.001-1.0 M, µ ) 1.0 (KCl)), which
had been thermally equilibrated at 79 ( 1 °C in the instrument cell
holder for 30 min. The final pH of each of the runs was measured and
shown to agree with the initial pH. The pseudo-first-order rate constants
were obtained by NLLSQ fitting the absorbance vs time data to a
standard exponential model (At ) A∞ + (A0 - A∞) exp(-kt)).
Experimental Section
A. Materials and General Methods. Aniline, formic acid, and
p-nitroaniline were obtained from BDH. Acetanilide, formanilide, acetic
anhydride, and p-methoxyaniline were purchased from Aldrich and used
as supplied. Glacial acetic acid was obtained from Fisher Scientific.
Methanol (HPLC grade) was obtained from EM Science. Aniline and
acetic acid were distilled prior to use. All HPLC solvents were filtered
through a 0.45 µm filter before use. All melting points were obtained
using Fisher-Johns melting point apparatus and are uncorrected.
All buffers were prepared using purified deoxygenated water from
an Osmonic Aries water purification system. The pH was measured at
ambient temperature using a Radiometer Vit 90 video titrator equipped
with a GK2321 C combination electrode, standardized by Fisher
Certified pH 2, 4, 7, and 10 buffers.
B. Synthesis. p-Nitroformanilide and p-methoxyformanilide were
synthesized from p-nitroaniline and p-methoxyaniline and formic acetic
anhydride as described15 by Krishnamurthy. p-Nitroformanilide was
recrystalized 2 times from ethyl acetate before use and had a melting
point of 196-197 °C (lit.16 mp 196-198 °C). p-Methoxyformanilide
was purified by recrystalizing 3 times from a chloroform-hexane
mixture. The pure product had a melting point of 79-80 °C (lit.17 mp
78-80 °C).
The acid-catalyzed hydrolysis of formanilide was followed at 79 (
1 °C using 9.86 × 10-3 M HCl, µ ) 1.0 (KCl), with runs being initiated
as above. All the runs were followed for at least 5 half-lives.
F. D2O Studies. The rate of formation and hydrolysis of formanilide
at 79 ( 1 °C was determined in formate buffer ([buffer]total ) 0.25-
1.0 M) at pD 3.60 ( 0.03 where pD ) pHmeasured + 0.40.18 The rate of
hydrolysis of formanilide was also obtained in 1.10 × 10-2 M DCl, µ
) 1.0 (KCl). The kinetic data were obtained and analyzed as above.
C. Analytical HPLC Conditions. For HPLC analysis a Hewlett-
Packard 1050 series HPLC system, fitted with a variable-wavelength
UV-vis detector and autoinjector, was used. For separation, an
µ-Bonda-Pak C18 (Waters) cartridge column was used. A gradient
mixture of 0.005 M potassium phosphate buffer (pH 7.2) and methanol
was used to separate aniline from formanilide. For each injection the
initial solvent composition was 20% methanol:80% phosphate buffer,
which after 15 min was modified to 30% methanol:70% phosphate
buffer until 21 min, whereupon 100% methanol was used to wash the
column for 9 min before the next injection. For separation of aniline
and acetanilide, an isocratic mixture of 25% methanol:75% phosphate
buffer (0.005 M, pH 7.2) was used. In both cases, the flow rate was
1.2 mL/min and the detector was set at λ ) 231 nm.
D. Kinetics of Amide Formation and Hydrolysis by HPLC. For
kinetics studies at 98 ( 2 °C, (5.0-5.3) × 10-3 M formanilide or aniline
solutions were made in 10-20 mL of formate buffer (pH 3.2-4.2,
[buffer]total ) 0.1-1.0 M, µ ) 1.0 M (KCl)) and were degassed by
passing Ar through them for 30 min. These solutions were then divided
into 10-20 autosampler vials that were then sealed with Teflon-lined
septa. The vials were heated in a boiler containing boiling water, and
at certain intervals, the vials were withdrawn, cooled immediately in
ice water, and analyzed at ambient temperature by HPLC. The pseudo-
first-order rate constants for appearance and disappearance of aniline
and formanilide were obtained by NLLSQ fitting of peak area vs time
data to a standard exponential model (vide supra). The response factors
for formanilide and aniline under the experimental conditions were
G. Studies in the Presence of Phosphate. The rate of formation
and hydrolysis of formanilide was determined similarly at 79 ( 1 °C
in the presence of phosphate (0.10-0.50 M) in 1.0 M formate buffer
at pH 3.20 ( 0.03 and pH 3.60 ( 0.03, with the ionic strength again
being maintained at 1.0 (KCl).
H. Studies in Aqueous Ethanol. The rate of formation and
hydrolysis of formanilide was obtained in 1.0 M formate buffer in 20%
(v/v) ethanolic water, pH 3.59, µ ) 1.0 (KCl). The kinetic data were
obtained as above at 60 ( 0.3 °C. Kinetic data were also obtained in
80% (v/v) ethanolic water containing 1.0 M formate at pHs 3.60 and
4.92 at the same temperature. The pH was adjusted by adding a suitable
amount of concentrated NaOH or HCl and measured before and after
the reaction. The ionic strength in 80% ethanol-20%water was not
corrected. The rate of formation and hydrolysis of formanilide was also
obtained at 60 °C in 1.0 M aqueous formate buffer pH 3.60, µ ) 1.0
(KCl).
I. Determination of pKa. The pKa’s of aniline and formic acid were
determined by titration at 24 ( 1 °C. For each determination, 0.048-
0.051 mmol of aniline or formic acid was used in a 5 mL solution. In
the case of aniline enough HCl was added to convert all of it to
anilinium ion. The pH was measured using a Radiometer Vit 90 video
titrator equipped with a GK2321C combination electrode and interfaced
with an IBM PC. The pH was recorded as a function of added 0.0105
M NaOH, which was delivered by a Radiometer ABU 91 autoburet.
The ionic strengths of all of the solutions were maintained at 1.0 using
KCl. Data were analyzed by a computer version of Simms method.19
The pKa values reported are an average of three determinations.
Reported in Table 10 (as calculated in the Appendix) are pKa values
corrected for the temperature and medium effects.
(14) Keillor, J. W.; Neverov, A. A.; Brown, R. S. J. Am. Chem. Soc.
1994, 116, 4669.
(15) Krishnamurthy, S. Tetrahedron. Lett. 1982, 23, 3315.
(16) Perrin, C. L.; Thoburn, J. D.; Kresge, A. J. J. Am. Chem. Soc. 1992,
114, 8800.
(18) Glasoe, P. K.; Long, F. A. J. Phys. Chem. 1960, 64, 188.
(17) DeWolfe, R.; Newcomb, C. R. J. Org. Chem. 1971, 36, 3870.
(19) Huguet, J. Ph.D. Thesis, University of Alberta, 1980

4600 J. Am. Chem. Soc., Vol. 121, No. 19, 1999
Aman and Brown
constant, K′, is given in eq 3.
kobs ) kf + kr
K′ ) kf/kr
(2)
(3)
In practical terms, K′ is the ratio of [anilide]/[aniline]total at
equilibrium with a given excess of RCOOH. Given in Tables
1S-12S (Supporting Information) are the original kobs data for
appearance or disappearance of various the anilides as well as
the conditional equilibrium constants, K′, determined at equi-
librium by HPLC methods or from the residual absorbance in
the absorbance vs time UV kinetic plots. Given in Tables 1-8
are the pseudo-first-order rate constants for formation (kf) and
hydrolysis (kr) of the anilides as well as the corrected equilibrium
constant, K′eq, which is based on the concentrations of nonion-
ized aniline and formic acid with the concentration of water set
as unity as in eq 4.2b The acid and amine concentrations can be
Figure 1. Typical kinetic traces for formanilide equilibration at pH
3.60 at different [formate]total, 0.10-1.00 M, µ ) 1.0 (KCl), and T )
79 ( 1 °C. The lines are obtained from NLLSQ fitting of data to At )
A∞ + (A0 - A∞) exp(-kt). formation at 0.1 M (9), hydrolysis at 0.1 M
(0), formation at 0.5 M (2), hydrolysis at 0.5 M (1), formation at 1.0
M ([), and hydrolysis at 1.0 M (b).
Scheme 1
[formanilide]
K′eq
)
(4)
[aniline][formic acid]
determined from the pKa values for the various RCOOH and
Ar-NH3+, corrected for the conditions of temperature and
media as described in the Appendix, the corrected values being
listed in Table 10. We deal with the specific conditions of
temperature and structure below, followed by the effects of
[formate] and additives.
Results
The approach to equilibrium for the anilides given in Scheme
1 was followed both from the direction of formation and
hydrolysis under a variety of conditions such as temperature,
pH, [RCOOH]total, and the presence of the additives ethanol and
phosphate. Shown in Figure 1 are typical absorbance vs time
traces for the equilibration of formanilide with aniline and
formate at pH 3.60 and 79 °C in the presence of three different
[formate]total. The pseudo-first-order rate constant for the ap-
proach to equilibrium (kobs) can be obtained by NLLSQ fitting
of the absorbance vs time data to a standard exponential model,
At ) A∞+ (A0 - A∞) exp(-kt). The kobs can be expressed in
terms of the forward and reverse rate constants, kf and kr
(Scheme 1) as in eq 2, while the conditional equilibrium
A. Effect of Temperature and Structure on Anilide
Equilibrium Constants. Given in Tables 1-3 are the experi-
mental kf, kr, and K′eq values for equilibration of formanilide
and acetanilide at various pHs and [RCOOH]total at the indicated
temperatures. The data show that the K′eq for formanilide at 98
°C (12.6 M-1) is larger than that for acetanilide (3.4 M-1) by
a factor of about 3.7. As expected, equilibrium is reached faster
at higher temperature, but reducing the temperature from 98 to
79 °C leads to an increase in the K′eq for formanilide of roughly
1.6-fold, (12.6 vs 20.1 M-1). Given in Table 3 are data
determined in D2O at 79 °C that indicate that, at pH (pD) 3.60,
the overall solvent deuterium kinetic isotope effect (DKIE) on
Table 1. Pseudo-First-Order Rate Constantsa for Formation (kf) and Hydrolysis (kr) of Formanilide at 98 ( 2 °C and the Equilibrium
Constantsb (K′eq) Obtained by HPLC at Various pHs and Buffer Concentrations in Aqueous Formate Buffer
pH
[formate]totalc (M)
kr (s-1
)
kf (s-1
)
K′eq (M-1
)
4.17 ( 0.03
3.60 ( 0.03
3.18 ( 0.03
3.57 ( 0.03
3.57 ( 0.03
0.10
0.10
0.10
0.50
1.00
(1.22 ( 0.03) × 10-5
(4.15 ( 0.02) × 10-5
(1.29 ( 0.06) × 10-4
(9.72 ( 0.30) × 10-5
(1.27 ( 0.04) × 10-4
(0.23 ( 0.01) × 10-5
(1.13 ( 0.04) × 10-5
(2.56 ( 0.11) × 10-5
(1.20 ( 0.01) × 10-4
(3.36 ( 0.08) × 10-4
12.3 ( 0.2
13.1 ( 0.6
12.9 ( 0.2
11.9 ( 0.4
12.7 ( 0.2
a Calculated from the pseudo-first-order rate constants (kobs) for establishment of equilibrium, using kobs ) kr + kf and K′ ) kf/kr. Errors are
deviation from the mean for duplicate values and standard deviations for triplicate values. b Calculated using the nonionized concentrations at
equilibrium (eq 4), taking concentration of water as unity and accounting for changes in the pKa’s due to temperature and ionic strength variations.
c µ ) 1.0 (KCl).
Table 2. Pseudo-First-Order Rate Constantsa for Formation (kf) and Hydrolysis (kr) of Acetanilide at 98 ( 2 °C and the Equilibrium
Constantsb (K′eq) at Various pH Values in Aqueous HCl or Acetate Buffer
pH
buffer
kr (s-1
)
kf (s-1
)
K′eq (M-1
)
1.95 ( 0.05
3.75 ( 0.05
HClc,e,
(3.47 ( 0.04) × 10-5
(0.19 ( 0.02) × 10-5
3.6 ( 0.3
3.2 ( 0.3
acetated,e
(0.81 ( 0.02) × 10-6
(1.07 ( 0.02) × 10-6
a Calculated from the pseudo-first-order rate constants (kobs) for establishment of equilibrium, using kobs ) kr + kf and K′ ) kf/kr. Errors are
deviation from the mean for duplicate values and standard deviations for triplicate values. b Calculated using the nonionized concentrations at
equilibrium, taking concentration of water as unity and accounting for changes in the pKas due to temperature and ionic strength variations. c 1.0
M acetate added. d [buffer]total ) 1.0 M. e µ ) 1.0 (KCl).

Equilibrium Formation of Anilides
J. Am. Chem. Soc., Vol. 121, No. 19, 1999 4601
Table 3. Pseudo-First-Order Rate Constantsa for Formation (kf) and Hydrolysis (kr) of Formanilide at 79 ( 1 °C and the Equilibrium
Constantsb (K′eq) at Various pHs or pDs and at Various Buffer Concentrations in Aqueous Formate Buffer
pH or pD
[formate]totald (M)
Kr (s-1
)
kf (s-1
)
K′eq (M-1
)
2.80 ( 0.03
2.79 ( 0.03
2.81 ( 0.03
3.18 ( 0.03
3.20 ( 0.03
3.20 ( 0.03
3.56 ( 0.03
3.60 ( 0.03
3.61 ( 0.03
3.59 ( 0.03
4.01 ( 0.03
4.02 ( 0.03
4.03 ( 0.03
3.60 ( 0.03c
3.58 ( 0.03c
3.61 ( 0.03c
3.60 ( 0.03c
0.10
0.50
0.90
0.10
0.50
1.00
0.001
0.10
0.50
1.00
0.10
0.50
1.00
0.001
0.50
0.75
1.00
(7.02 ( 0.05) × 10-5
(7.60 ( 0.10) × 10-5
(7.69 ( 0.02) × 10-5
(3.32 ( 0.05) × 10-5
(3.66 ( 0.05) × 10-5
(4.09 ( 0.03) × 10-5
(1.36 ( 0.02) × 10-5
(1.42 ( 0.02) × 10-5
(1.76 ( 0.03) × 10-5
(2.15 ( 0.03) × 10-5
(5.93 ( 0.02) × 10-6
(7.90 ( 0.01) × 10-6
(1.06 ( 0.01) × 10-5
(1.34 ( 0.03) × 10-5
(1.93 ( 0.09) × 10-5
(2.01 ( 0.12) × 10-5
(2.18 ( 0.03) × 10-5
(0.77 ( 0.01) × 10-5
(4.29 ( 0.05) × 10-5
(8.22 ( 0.02) × 10-5
(0.66 ( 0.02) × 10-5
(3.60 ( 0.05) × 10-5
(8.20 ( 0.02) × 10-5
(0.42 ( 0.01) × 10-5
(2.64 ( 0.02) × 10-5
(5.97 ( 0.04) × 10-5
(1.83 ( 0.01) × 10-6
(1.25 ( 0.01) × 10-5
(2.94 ( 0.01) × 10-5
20.6
22.7
20.7
19.1
18.2
18.6
20.4
20.9
19.5
20.9
21.7
19.0
(1.77 ( 0.10) × 10-5
(2.76 ( 0.30) × 10-5
(3.96 ( 0.06) × 10-5
22.3
21.9
21.6
a Calculated from the pseudo-first-order rate constants (kobs) for establishment of equilibrium, using kobs ) kr + kf and K′ ) kf/kr. Errors are
deviation from the mean for duplicate values and standard deviations for triplicate values. b Calculated using the nonionized concentrations at
equilibrium and taking concentration of water as unity. Changes in the pKa’s due to temperature, ionic strength, and solvent isotope effects have
been considered. c pD. d µ ) 1.0 (KCl).
Table 4. Pseudo-First-Order Rate Constantsa for Formation (kf) and Hydrolysis (kr) of p-Nitroformanilide at 79 ( 1 °C and the Equilibrium
Constantsb (K′eq) at Various pHs and Buffer Concentrations in Aqueous Formate Buffer
pH
[formate]totalc (M)
kr (s-1
)
kf (s-1
)
K′eq (M-1
)
2.80 ( 0.03
3.18 ( 0.03
3.20 ( 0.03
3.20 ( 0.03
3.60 ( 0.03
1.00
0.10
0.50
1.00
1.00
(2.33 ( 0.05) × 10-4
(9.86 ( 0.02) × 10-5
(1.12 ( 0.03) × 10-4
(1.28 ( 0.03) × 10-4
(7.22 ( 0.07) × 10-5
(5.34 ( 0.04) × 10-5
(2.16 ( 0.20) × 10-6
(1.17 ( 0.05) × 10-5
(2.40 ( 0.06) × 10-5
(9.33 ( 0.60) × 10-6
0.26
0.29
0.27
0.26
0.25
a Calculated from the pseudo-first-order rate constants (kobs) for establishment of equilibrium, using kobs ) kr + kf and K′ ) kf/kr. Errors are
deviation from the mean for duplicate values and standard deviations for triplicate values. b Calculated using the nonionized of concentrations at
equilibrium and taking concentration of water as unity. Changes in the pKa’s due to temperature have been considered. For formic acid, the change
in pKa due to ionic strength variation has also been considered. However, the effect of ionic strength on the pKa of p-nitroanilinium ion was not
considered since at these pHs, a change of (0.35 in pKa unit will have less than 2% effect on the K′eq. c µ ) 1.0 (KCl).
Table 5. Pseudo-First-Order Rate Constantsa for Formation (kf)
and Hydrolysis (kr) of p-Methoxyformanilide at 79 ( 1 °C and the
Equilibrium Constantsb (K′eq) at pHs 2.80 and 3.20 in Aqueous
Formate Buffer
pHc
kr (s-1
)
kf (s-1
)
K′eq (M-1
)
2.82 ( 0.03 (4.08 ( 0.05) × 10-5 (4.01 ( 0.04) × 10-5
41.0
38.9
3.18 ( 0.03 (2.13 ( 0.03) × 10-5 (4.11 ( 0.06) × 10-5
a Calculated from the pseudo-first-order rate constants (kobs) for
establishment of equilibrium, using kobs ) kr + kf and K′ ) kf/kr. Errors
are deviation from the mean for duplicate values and standard deviations
for triplicate values. b Calculated using the nonionized of concentrations
at equilibrium and taking concentration of water as unity and accounting
for changes in the pKa’s due to temperature and ionic strength.
c [buffer]total ) 1.0, µ ) 1.0 (KCl).
Figure 2. Pseudo-first-order rate constants of formation (kf, 2) and
hydrolysis (kr, 9) of formanilide vs [formate]total at 98 ( 2 °C (pH )
3.60 and µ ) 1.0 (KCl)). The lines were obtained from the fits to eqs
5 and 6.
the K′eq for formanilide is 1.1 ( 0.1 but that the DKIE on K′ )
kf/kr is 1.5-1.6 ([formate]total ) 0.5-1.0 M).
Given in Tables 4 and 5 are the kf, kr, and K′eq values for the
equilibration of p-nitroformanilide and p-methoxyformanilide
at 79 °C at various pHs. The respective K′eq values of 0.26 and
40 M-1, coupled with the value of 20.1 M-1 for formanilide at
79 °C, indicate a strong dependence on the basicity of the aniline
substituent.
can be fit to eqs 5 and 6,
kr ) k0r + k1r[formate]total
(5)
(6)
kf ) [formate]total(k1f + k2f[formate]total
)
B. Effect of [formate] on kf and kr. Shown in Figure 2 are
typical plots of the pseudo-first-order rate constants for forma-
tion (kf) and hydrolysis (kr) of formanilide as a function of
[formate]total at 98 °C, pH ) 3.60, and µ ) 1.0 (KCl). Similar
plots (not shown) can be made for the processes at 79 °C. The
data indicate that the hydrolysis follows a linear dependency
on the [formate], while the formation follows an exponential
behavior. The kinetic/concentration data for the two processes
where k0r, the intercept for eq 5, is the spontaneous pseudo-
first-order rate constant for hydrolysis at pH 3.60, k1r is the
second-order rate constant for formate-catalyzed hydrolysis, and
k1f and k2f are the second- and third-order rate constants for
formate-dependent hydrolysis and formation of formanilide.
Fitting of the appropriate hydrolysis and formation data at 98
and 79 °C (from Tables 1 and 3) to eqs 5 and 6 gives the values

4602 J. Am. Chem. Soc., Vol. 121, No. 19, 1999
Aman and Brown
Table 6. Pseudo-First-Order Rate Constants for the Acid Catalyzed Hydrolysis, k0r (s-1),a of Formanilide at a Given pH or pD, Second-Order
Rate Constants, k1r (M-1 s-1),a of Formate (buffer) Catalysis on Hydrolysis, and Second- and Third-Order Rate Constants, k1f (M-1 s-1)a and k2f
(M-2 s-1),a of Formate (buffer) Catalysis on Formation of Formanilide from Aniline and Formic Acid at Various Temperatures and pHs or pDs
k0r × 105
k1r × 106
K1f × 105
k2f × 105
temp (°C)
pH or pD
(s-1
)
(M-1 s-1
)
(M-1 s-1
)
(M-2 s-1
)
79 ( 1
79 ( 1
79 ( 1
79 ( 1
79 ( 1
79 ( 1
79 ( 1
98 ( 2
2.80
3.20
3.60
4.00
3.60c,d
2.00d
1.96c
3.60
7.02 ( 0.21
3.24 ( 0.01
1.35 ( 0.01
0.54 ( 0.01
1.38 ( 0.08
43.8 ( 0.01
53.2 ( 0.01
3.72 ( 0.56
8.50 ( 3.50b
8.45 ( 0.01
8.01 ( 0.14
5.19 ( 0.13
8.31 ( 1.10
7.84 ( 0.09
6.24 ( 0.05
4.45 ( 0.14
2.02 ( 0.08
3.04 ( 0.10
1.45 ( 0.11
1.95 ( 0.06
1.47 ( 0.15
0.93 ( 0.08
0.91 ( 0.10
95.0 ( 18.0
13.7 ( 0.10
19.9 ( 0.20
a The errors calculated from the standard deviation of the fit of kr vs [formate]total and kf vs [formate]total to eqs 5 and 6, respectively, at the given
pH or pD. b At this pH the acid-catalyzed hydrolysis, k0r, is much higher than the buffer catalyzed hydrolysis, k1r (less than 11% at highest buffer
concentration). c pD. d HCl or DCl was used as buffer; no formate added, µ ) 1.0 (KCl).
Table 7. Pseudo-First-Order Rate Constants of Formation (kf) and Hydrolysis (kr) of Formanilide and the Equilibrium Constants (K′eq) in the
Presence of Phosphate at 79 ( 1 °C in Aqueous Formate Buffer and the Second-Order Rate Constant for Phosphate Catalysis on Hydrolysis
b
(krphos)b and Formation (kfphos
)
pHa
[phosphate]total (M)
kr × 105 (s-1
)
kf × 105 (s-1
)
K′eq (M-1
)
krphos × 105 (M-1 s-1
6.14 ( 0.05
)
kfphos × 105 (M-1 s-1
13.4 ( 0.10
)
3.21
3.21
0.10
0.30
(4.66 ( 0.02)
(5.90 ( 0.10)
(9.50 ( 0.03)
(12.2 ( 0.05)
18.5
18.6
3.58
3.60
0.10
0.50
(2.47 ( 0.03)
(3.52 ( 0.05)
(6.71 ( 0.02)
(9.44 ( 0.01)
19.1
18.8
2.70 ( 0.09
6.90 ( 0.09
a [buffer]total ) 1.00 M, µ ) 1.0 (KCl). b The errors are calculated from the standard deviation of the fit of kr vs [phosphate]total and kf vs
[phosphate]total to a linear equation, kr ) kr0 + krphos[phosphate]total or kf ) kf0 + kfphos[phosphate]total, where kr0 and kf0 are the pseudo-first-order rate
constants of hydrolysis and formation without added phosphate under the experimental conditions at the given pH. The data at zero phosphate
concentration were obtained from Table 4.
Table 8. Pseudo-First-Order Rate Constantsa of Formation (kf) and Hydrolysis (kr) of Formanilide and the Equilibrium Constantsb (K′eq) in
Aqueous Ethanolic, Formate Buffer at 60.0 ( 0.3 °C
% (v/v)
ethanol
d
pHc
kr (s-1
)
kf (s-1
)
K′eq (M-1
)
3.59 ( 0.03
4.92 ( 0.05
3.60 ( 0.05
3.60 ( 0.05
20
80
80
0
(5.32 ( 0.55) × 10-6
(2.82 ( 0.03) × 10-6
(7.01 ( 0.02) × 10-6
(5.86 ( 0.03) × 10-6
(3.30 ( 0.02) × 10-5
(1.85 ( 0.02) × 10-5
(7.14 ( 0.02) × 10-5
(1.65 ( 0.02) × 10-5
23.9
14.4e
9.25e
32.7
a Calculated from the pseudo-first-order rate constants (kobs) for establishment of equilibrium, using kobs ) kr + kf and K′ ) kf/kr. Errors are
deviations from the mean for duplicate values and standard deviations for triplicate values. b Calculated using the nonionized of concentrations at
equilibrium and taking concentration of water as unity, accounting for the changes in the pKa’s due to the percent ethanol present in the solvent and
variation of temperature. 20% v/v ethanol water corresponds to 16.4 wt %, and 80% v/v corresponds to 75.8 wt %. c [buffer]total ) 1.0 M. d Given
the possibility that ethyl formate is produced during the reaction, that species could be involved in the formation of formanilide, and that the water
concentration can no longer be taken as unity, the K′eq should be written as K′eq) ([amide][H2O])/(([acid] + [ethyl formate])[amine]). e µ uncorrected.
for these rate constants compiled in Table 6. Two things are
apparent from the data in Table 6. First, the k0r constants increase
linearly with increasing [H3O+], indicating that the hydrolysis
in the absence of buffer is specific acid catalyzed. Second, from
the D2O data given in the table, the (k0rH+/k0rD+) ratio at pH
(pD) 3.6 is 1.0 ( 0.1, while that at pH 2.0 is 0.9. The solvent
DKIE on the buffer catalysis for hydrolysis at pH (pD) 3.6,
Given in Table 8 are the pseudo-first-order rate constants for
formation and hydrolysis of formanilide in aqueous ethanolic
solutions of [formate]total ) 1.0 M, T ) 60 °C, and ionic strength
not held constant. In keeping with the trends of temperature on
K′eq noted in part A above, in the absence of alcohol, the
corrected equilibrium constant increases at lower temperature
although the rate of attainment of equilibrium decreases. Added
ethanol has the effect of decreasing both kr and kf, but by
different amounts, the net effect being to reduce K′eq as the
ethanol content increases.
(k1rH O/k1rD O), is 0.96 ( 0.13, while the DKIE for formation
2
2
are (k1fH O/k1fD O) ) 1.46 ( 0.10 and (k2fH O/k2fD O) ) 1.61 (
0.31.
2
2
2
2
C. Effect of Additives on the kf, kr, and K′eq Constants.
Given in Table 7 are the kf, kr, and K′eq constants for the
equilibration of formanilide at 79 °C at pHs 3.20 and 3.60 in a
1.0 M formate buffer in the presence of K2PO4. At these pH
values, the phosphate exists as the monoanionic form and the
ionic strength was held at 1.0 with added KCl. Although
equilibrium is attained faster in the presence of phosphate, the
K′ and K′eq values are unchanged relative to those obtained in
the absence of phosphate. Plots (not shown) of the kf and kr
values against [phosphate] give straight lines, the slopes of which
give the phosphate catalysis on the formation (kfphos) and
hydrolysis (krphos) processes: these values are also listed in Table
7.
Discussion
A summary of the conditional equilibrium constants (K′,
defined as in eq 3) and corrected equilibrium constants (K′eq,
defined as in eq 4) for the anilides under all conditions
investigated is presented in Table 9. The K′ constant gives an
indication of the expected [anilide]/[aniline]total at a given set
of conditions when there is an excess of carboxylic acid present.
It is known that the formation of the anilides depends on the
concentrations of the nonionized acid and amine2a,3,11,13,14 as
depicted in Scheme 2. Therefore factors that alter the pKa’s of
the amines and acids play an important role in controlling the
extent of amide formation under a given set of conditions and

Equilibrium Formation of Anilides
J. Am. Chem. Soc., Vol. 121, No. 19, 1999 4603
Table 9. Conditional Equilibrium Constants (K′) and Equilibrium
Constants (K′eq)a of Formanilide, Acetanilide, p-Nitroformanilide,
and p-Methoxyformanilide under Various Conditions
Scheme 3
temp [RCOOH]total
K′eq
compound
formanilide
formanilide
formanilide
formanilide
formanilide
acetanilide
acetanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
p-nitroformanilide
p-nitroformanilide
p-nitroformanilide
p-nitroformanilide
p-nitroformanilide
p-methoxyformanilide 2.82
p-methoxyformanilide 3.18
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
formanilide
pH
(°C)
(M)
K′
(M-1
)
4.17
3.60
3.18
3.60
3.60
1.95
3.75
2.80
2.79
2.81
3.20
3.20
3.20
3.60
3.61
3.59
4.01
3.20
3.20
98
98
98
98
98
98
98
79
79
79
79
79
79
79
79
79
79
79
79
0.10
0.10
0.10
0.50
1.00
1.00b
1.00b
0.10
0.50
0.90
0.10
0.50
1.00
0.10
0.50
1.00
0.10
0.50
1.00
0.50
0.75
1.00
1.00
0.10
0.50
1.00
1.00
1.00
1.00
1.00d
1.00e
1.00d
1.00f
1.00
1.00g
1.00h
1.00h
0.19 12.3
0.27 13.1
0.20 12.9
1.23 11.9
2.64 12.7
0.06
1.34
3.60
3.20
0.11 20.6
0.57 22.7
1.07 20.7
0.21 19.1
1.00 18.2
2.05 18.6
0.29 20.4
1.49 20.9
2.77 19.5
0.30 20.9
1.58 21.7
2.78 19.0
0.93 22.3
1.37 21.9
1.80 21.6
240-fold. (4) For formanilide at a given pH and [HCOOH]total
,
reduction in the temperature from 100 to 60 °C produces almost
no change in K′ but increases K′eq by a factor of about 3. (5)
Added phosphate increases the rate of attainment of equilibrium
without affecting the K′ or K′eq. (6) Added ethanol increases
the rate of attainment of equilibrium and the K′ but reduces
K′eq. In the following we will deal with these observations in
detail.
3.58c 79
3.61c 79
3.60c 79
2.82
3.20
3.20
3.21
3.60
79
79
79
79
79
79
79
79
79
79
79
60
60
60
60
0.23
0.02
0.10
0.19
0.13
0.26
0.29
0.27
0.26
0.25
A. Mechanism of Specific Acid Catalyzed and General
Catalyzed Hydrolysis and Formation. The hydrolysis of
amides is specific acid catalyzed and adheres to the mechanism
given in Scheme 3,1,20,21 where preequilibrium protonation of
the amide, followed by attack of H2O assisted by one20 or more21
water molecules, generates a neutral tetrahedral intermediate
(TO) and an encounter complex with H3O+. Rapid protonation
of the N to yield TN+ followed by water-assisted C-N cleavage
generates the acid, amine, and H3O+. Consistent with such a
mechanism is the fact that the solvent DKIE for H3O+-catalyzed
hydrolysis is ∼1.0 and little 18OdC exchange is detected in
labeled amides recovered from solution after partial hydrolysis,20
indicating that the rate-limiting step is largely formation of TO
from the protonated amide. The data of Table 6 for formanilide
indicate that the hydrolysis at zero added formate is linearly
dependent on [H3O+] and that the solvent DKIE for that process
is, as expected,20 1.0 ( 0.1.
It is required that the reaction of acid-catalyzed amide
formation follows the microscopic reverse of the hydrolysis,
proceeding through the same intermediates with the same rate-
limiting step(s). Thus the reaction must proceed through a
reversible H3O+-catalyzed attack of the amine on RCOOH to
yield TN+, which then deprotonates to give a TO, which
undergoes rate-limiting general acid catalyzed (by H3O+) C-OH
cleavage to yield the protonated amide.
0.98 41.0
1.88 38.9
2.04 18.5
2.04 18.5
2.72 19.1
2.70 19.0
2.82 32.7
6.21 23.9i
3.21
3.21
3.58
3.58
3.60
3.59
3.60
4.92
10.2
6.54
14.4i
9.3i
a Obtained using K′eq ) [formanilide]/[aniline][formic acid] and
taking activity of water as unity. The concentrations of nonionized acid
and nonprotonated amine were calculated using the corrected pKa under
the given conditions. b Acetate. c pD. d In presence of 0.10 M KH2PO4.
e In presence of 0.30 M KH2PO4. f In presence of 0.50 M KH2PO4.g In
20% aqueous ethanol. h In 80% aqueous ethanol. i Given the possibility
that ethyl formate is produced during the reaction, that species could
be involved in the formation of formanilide, and that the water
concentration can no longer be taken as unity, the K′eq should be written
as K′eq) ([amide][H2O])/(([acid] + [ethyl formate])[amine]).
Scheme 2
As previously mentioned in the Results, the observed rate
constant for the attainment of equilibrium, kobs, is the sum of
the pseudo-first-order rate constants kr and kf. Figure 2 illustrates
that both the latter constants are dependent on [RCOOH]total
,
but in different ways. Fitting of the kr and kf vs [formate]total
data in Tables 1 and 3 to eqs 5 and 6 gives the values for the
formate-dependent rate constants presented in Table 6. The k0r
and k1f terms relate to the H3O+-catalyzed hydrolysis of
formanilide (discussed above) and its microscopic reverse, the
this is reflected in the corrected equilibrium constant, K′eq.
Several points of note are apparent from inspection of the data.
(1) For a given amide, K′ increases as the [formate]total increases,
(20) (a) Bennet, A. J.; Slebocka-Tilk, H.; Brown, R. S.; Guthrie, J. P.;
Jodhan, A. J. Am. Chem. Soc. 1990, 112, 8497. (b) Bennet, A. J.; Slebocka-
Tilk, H.; Brown, R. S. J. Am. Chem. Soc. 1992, 114, 3088. (c) Brown, R.
S.; Bennet, A. J.; Slebocka-Tilk, H. Acc. Chem. Res. 1992, 25, 481.
(21) (a) Cox, R. A.; Yates, K. Can. J. Chem. 1981, 59, 2853 and
references therein. (b) Moodie, R. B.; Wale, P. D.; Whaite, T. J. J. Chem.
Soc. 1963, 4273.
but K′eq remains constant irrespective of the pH and [formate]total
.
(2) In the set of formanilides, the p-substituent alters both K′
and K′eq. (3) Alteration of the acyl portion from formanilide to
acetanilide decreases the equilibrium constant by a factor of
3.7 but decreases the rate of attainment of equilibrium by about

4604 J. Am. Chem. Soc., Vol. 121, No. 19, 1999
Aman and Brown
H3O+-catalyzed formation of formanilide. Similarly, the k1r and
k2f terms relate to a formate-catalyzed hydrolysis of formanilide
and its microscopic reverse, the formate-catalyzed aminolysis
of formic acid. At least three possible kinetically equivalent
mechanisms can be presented for the formate-catalyzed hy-
drolysis as given in 1-3. Mechanism 1 involves general acid
solvent DKIE for (k1rH O/k1rD O). Mechanism 3 is predicted to
have a more inverse value than mechanism 2 which results from
the pre-equilibrium protonation DKIE superimposed on a factor
2
2
5
of /3 to account for an increased [formate anion] in H2O.
B. Effect of pH on the Conditional Equilibrium Constant,
K′. As shown in Table 9, at constant [formate]total, the conditional
equilibrium constant maximizes at about pH 3.60 and then
decreases at higher and lower pH. The conditional equilibrium
constant includes the [formate]total so eq 3 becomes
K′ ) kf/kr ) K′′[formate]total ) [anilide]/[aniline]total
where K′′ ) [anilide]/([HCOOH + HCOO-])([H3N+-Ar +
H2N-Ar]). Because the formation reaction depends on the
concentrations of RCOOH and amine, the optimum pH for
synthesis of a given amide from its constituents occurs at values
between the pKa’s of RCOOH and H3N+-Ar, expressed as pHopt
) (pKa + pKaammonium)/2.2b,3 Using a pKa value of 3.63 for
acid
HCOOH and 3.87 (100 °C) or 3.78 (80 °C) for anilinium ion
(see Appendix), the calculated optimum pHs for formation of
formanilide are calculated to be 3.83 and 3.71 at 80 and 100
°C, close to the experimentally observed optimum value of 3.60
given in Table 9.
catalyzed attack of H2O on formanilide or, in the reverse
direction, general base plus H3O+-promoted decomposition of
TO. Mechanism 2 involves formate anion acting as a general
base to deliver H2O to a protonated amide and, in the reverse
direction, formic acid acting as a general acid to assist in the
decomposition of TO. Finally, in mechanism 3, formate acts as
a nucleophile on the protonated amide to give a transient
anhydride intermediate.
There is an apparent solvent deuterium equilibrium isotope
effect (EIE) on the formanilide K′H/K′D ) 1.56 at pH (pD) 3.6
and T ) 80 °C (Table 9). The data of Tables 3 and 6 clearly
show that the source of the DKIE is in the kf term since the kr
term has a unit kinetic isotope effect. Consideration of the effects
of D2O on the pKa’s of formic acid and anilinium24 indicates
that at pH (pD) 3.6 there is 50% less of the required neutral
amine in D2O than in H2O (14% vs 29%) but 7/5 more HCOOD.
Thus the forward reaction, kf[HCOOL][L2N-Ar] should show
The solvent DKIE of (k1rH O/k1rD O) ) 0.96 ( 0.13 for the
formate-promoted hydrolysis of formanilide at pH (pD) 3.60
and 79 °C permits us to favor mechanism 2 after consideration
of the effect of D2O on the concentrations of protonated amide,
formic acid, and formate anion. Since it is generally observed
that oxygen and nitrogen acids are stronger by a factor of ∼3
in H2O relative to D2O22 (since D3O+ is a stronger acid in D2O
than is H3O+ in H2O), one expects that the amide would be
more protonated by a factor of ∼3 in D2O. Also, since the pKa
for formic acid under our conditions is 3.63 in water and 4.09
in D2O,23 for a given [formate]total, there will be more HCOOD
at pD 3.60 in D2O than HCOOH at pH 3.60 in H2O (70% vs
50%).24 Mechanism 1 is disfavored since it should exhibit a
large primary DKIE of kH/kD ) 2-5 attributable to at least two
2
2
5
2
a solvent EIE of /7 × /1 ) 1.4 due to simple concentration
effects alone. It is notable that the solvent EIE on K′eq, which
depends on the true concentrations of the neutral acid and amine,
is unity.
C. Effect of Structure on K′ and K′eq. Fersht and Requena11
showed that, for a number of aliphatic amines, the corrected
equilibrium constant for amide formation was related to the pKa
of the corresponding ammonium ion by eq 7. The experimental
log K′eq ) 0.50 + 0.51pKa(ammonium)
(7)
5
protons in flight reduced by /7 to account for the increase of
[HCOOL] in D2O. By contrast, mechanism 2 is similar to what
is proposed in Scheme 3 for the specific acid-catalyzed
hydrolysis of amides, but with formate replacing H2O as the
base in assisting the delivery of water to protonated amide. In
this case, the DKIE should be ∼1 due to the competing effects
of an inverse DKIE associated with preequilibrium protonation
of the amide and a normal DKIE associated with the general
base delivery of the attacking water,20 modified by 5/3 to account
for the increased [formate anion] in H2O. This same mechanism
was found to be operative for the acetate-promoted hydrolysis
of distorted anilide 4.25
K′eq from that study for formanilide is 8 times smaller than
would be predicted using eq 7 probably due to the fact that
aliphatic and aromatic amines are described by slightly different
Brønsted relationships. In the present study of formanilides at
80 °C, the relationship between K′eq and aniline basicity is
log K′eq ) -0.95 + 0.57pKa(anilinium)
(8)
It is also seen (Tables 1 and 2) that attainment of equilibrium
is 240 times faster for formanilide than for acetamide under
comparable conditions of temperature, pH, and reagent con-
centrations. The higher reactivity of HC(dO)X can be attributed
to steric factors since nucleophilic attack on the formate carbonyl
(24) Given that pKa ) pH + log[A-H]/[A], it can be calculated that for
formic acid, pKa ) 3.6 in water, at pH 3.6 the ratio of [HCOOL]/[HCOO-]
is 1.0; in D2O, where the pKa of formic acid is 4.0, the ratio is 2.51.
Similarly, for anilinium, pKa ) 4.02 in water at 79 °C, the ratio of [L3N+-
Ar]/[L2N-Ar] is 2.51 at pH 3.6, but in D2O, where the pKa ) 4.4, the
ratio is 6.3.
Mechanisms 2 and 3 are kinetically indistinguishable, but
there is a preference for 2 based on the near unit value of the
(22) Gold, V. AdV. Phys. Org. Chem. 1969, 7, 259.
(23) Bell, R. P.; Kuhn, A. T. Trans. Faraday Soc. 1963, 59, 1789.
(25) Wang, Q.-P.; Bennet, A. J.; Brown, R. S.; Santarsiero, B. D. J. Am.
Chem. Soc. 1991, 113, 5757.

Equilibrium Formation of Anilides
J. Am. Chem. Soc., Vol. 121, No. 19, 1999 4605
Scheme 4
intermediates. In some cases it is known that phosphate can act
nucleophilically toward RC(dO)X to give acyl phosphate
intermediates. In a recent paper, we provided evidence that
phosphate can act as a nucleophilic catalyst toward ester and
thiolester hydrolysis.31 This mechanism generally occurs when
the conjugate acid of the departing group, X, has a low pKa
and can leave readily as an anion which is not the case for
phosphate catalysis of amide equilibration where X would be
OH or NHR. In an earlier study we suggested14 that the catalytic
mechanism of phosphate-assisted formation and hydrolysis of
amides proceeds via either sequential general base/general acid
process or a bifunctional concerted general base/general acid
processes (Scheme 4).
To study the effect of the solvent polarity on the K′ and K′eq,
the formation and hydrolysis of formanilide was carried out in
both 20 and 80% (v/v) ethanol/water mixtures at 60 °C and pH
3.60 (Table 8). Addition of cosolvents complicates the defini-
tions of pH and pKa and also modifies the water concentration
in the equilibrium process. Given the possibility that ethyl
formate is produced during the reaction and that species could
be involved in the formation of formanilide (vide infra) and
that the water concentration can no longer be taken as unity,
the K′eq should be written as K′eq ) ([amide][H2O])/(([acid] +
[ethyl formate])[amine]). However, the values for K′eq given in
Tables 8 and 9 are computed in the same way as for the aqueous
conditions to facilitate comparison. Under comparable conditions
of pH and [formate]total, both the rate constant for attainment
of equilibrium (kobs) and the conditional equilibrium constant
(K′) increased about 3.5 times in changing from 0 to 80%
ethanol (see Table 9). The pKa of formic acid increases with
the addition of ethanol in aqueous media,32 and under the
experimental conditions, the pKa of formic acid changes to
∼5.33 in 80% ethanol from 3.63 in aqueous media, T ) 25 °C.
On the other hand the pKa of anilinium ion decreases to ∼3.78
in 80% ethanol from 4.30 in aqueous media.33 The changes in
pKa’s due to the addition of ethanol increase the concentration
of nonionized forms of both aniline and formic acid at a given
pH (here 3.6), and so the equilibrium should lie more toward
anilide formation. However, it should be noted that, if the
observed change in K′ were only due to the increases in effective
concentration of reactive species ([HCOOH] and [H2NPh]), the
change would be higher than 3.5-fold in 80% ethanol. The other
competing factor which counterbalances the overall effect on
K′ is the drop in basicity of the amine.
Figure 3. Plot of log of formation equilibrium constant (K′eq) against
1/T for equilibrium formation of formanilide, µ ) 1.0 (KCl): 9, this
work; 0, at 25 °C (ref 11).
is less hindered than that on the CH3C(dO)X.26 However, the
respective corrected equilibrium constants vary only by a factor
of 3.7, indicating that the formation and hydrolysis processes
are nearly equally sensitive to the steric effects. Morawetz and
Otaki12 came to the same conclusion after they observed that
the amide equilibrium formation constants for a series of higher
fatty acids under aqueous basic conditions were nearly the same
but that CH3NH2 reacts 140 times faster with formate than with
acetate.
D. Effect of Temperature on K′ and K′eq. The corrected
equilibrium constant (K′eq) for formanilide is inversely propor-
tional to temperature, increasing 2.6 times from 100 to 60 °C
(Table 9). A plot of log K′eq vs 1/T (Figure 3) gives a straight
line with positive slope from which can be computed a ∆Heq
of 6.60 ( 0.25 kcal/mol. This value is very close to the ∆H of
ionization for anilinium (7.38 kcal/mol).27 The pKa of anilinium
ion decreases from 3.78 to 4.30 from 100 to 60 °C, whereas
the change in pKa of formic acid over the same range is
negligible because its ∆H of ionization is -0.04 kcal/mol27 (see
Appendix). The anticipated K′eq from eq 8 for an aniline having
a pKa of 4.30 would be 31.6 M-1, essentially the same as what
is found experimentally (32.7 M-1, Table 9). Thus the over-
whelming result of the temperature effect is to alter the K′eq
due to changes in the amine basicity.
In practical terms, there is a near independence of the
conditional equilibrium constant, K′, on temperature, the values
at pH 3.6 and [formate]total being 2.82, 2.77, and 2.64 at 60, 79,
and 100 °C. This is due to an almost complete counterbalancing
of the two competing effects of (a) increasing the amount of
neutral amine as the temperature increases due to the drop in
pKa, which should serve to increase K′, and (b) reducing the
nucleophilicity of the amine which should serve to decrease
the K′ as in eq 8.
E. Effects of Phosphate and Ethanol Additives. The data
in Table 7 show that phosphate catalyzes the formation and
hydrolysis of formanilide, but as expected, the equilibrium
position is not changed by the addition of KH2PO4. The
Similar observations were made in studies of enzymatic
formation of amide bonds,2,3 and it is reported that, in the
enzyme-catalyzed peptide formation, the addition of organic
accelerating effect is not great since, at pH 3.6, [formate]total
)
1.0 M, and µ ) 1.0, the addition of 0.5 M phosphate only
increases the rate of attainment of equilibrium by 50%. It is
known that phosphate ions are able to mediate acyl trans-
fer,28,29,30 in this case to amine or water, through proton addition
and removal during the formation and breakdown of tetrahedral
(28) (a) Schepartz, A.; Breslow, R. J. Am. Chem. Soc. 1987, 109, 1814.
(b) Cunningham, B.; Schmir, G. L. J. Am. Chem. Soc. 1967, 89, 917.
(29) (a) Lee, Y.; Schmir, G. L. J. Am. Chem. Soc. 1979, 101, 3026. (b)
Cunninghum, B.; Schmir, G. L. J. Am. Chem. Soc. 1966, 88, 551.
(30) (a) Kirby, A. J.; Mujahid, T. G.; Camilleri, P. J. Chem. Soc., Perkin
Trans. 2 1979, 1610. (b) Camilleri, P.; Ellul, R.; Kirby, A.; Mujahid, T.
G.J. Chem. Soc., Perkin Trans. 2 1979, 1617.
(31) Gill, M. S.; Neverov, A.; Brown, R. S. J. Org. Chem. 1997, 62,
7351.
(32) Grunwald, E.; Berkowitz, B. J. Am. Chem. Soc. 1951, 73, 4939.
(33) Gutbezahl, B.; Grunwald, E. J. Am. Chem. Soc. 1953, 75, 1953.
(26) The Taft steric parameters (defined as log(k/kCH ) ) Es, determined
3
from the acid-catalyzed hydrolysis of esters) for H and CH3, are 1.24 and
0, respectively: Shorter, J. In AdVances in Linear Free Energy Relationships;
Chapman, N. B., Shorter, J., Eds.; Plenum Press: New York, 1972.
(27) Isaacs, N. S. In Physical Organic Chemistry; Longmans Group UK
Ltd.: New York, 1987; pp 210-213.

4606 J. Am. Chem. Soc., Vol. 121, No. 19, 1999
Aman and Brown
Table 10. Computed pKa’s of Anilines and Acids at Various
Scheme 5
Temperatures and at µ ) 1.0
pKa in 20% (v/v)
pKa in 80% (v/v)
temp (°C)
pKa in water
ethanol
ethanol
Aniline
4.42g
25
60
80
4.89 ( 0.05a
4.30b
3.78g,k
3.19h,k
3.83h
4.02b
100
3.78b
Formic Acid
3.97i
25
60
80
3.63 ( 0.03a
3.63b
5.33i,k
5.33j,k
solvents shifted the equilibrium more toward amide bond
formation. For example,3b in the chymotrypsin-catalyzed forma-
tion of the dipeptide benzyloxycarbonyl-L-tryptophanylglycin-
amide (from benzyloxycarbonyl-L-tryptophan (5) and glycin-
amide (6)), the addition of 85% 1,4-butanediol increased the
3.97j
3.63b
100
3.64b
Acetic Acid
25
100
4.60c
4.62b
p-Nitroaniline
25
80
1.00d
0.65d
p-Methoxyaniline
25
80
5.34e
4.35e,f
a Measured at µ ) 1.0 (KCl). b Calculated using ∆Hionization and
∆Sionization values from ref 27. c Available data35 indicates that the pKa
of acetic acid drops from µ ) 0 to µ ) 0.2, but above µ ) 0.2, the pKa
is independent of ionic strength. The value used was that given at µ )
0.2. d Reference 37. e Reference 38 at µ ) 0.10 M. f Extrapolated value
using data from ref 36 (from 20 to 40 °C). g Calculated using data from
ref 33. 20% v/v ethanol water corresponds to 16.4 wt % and 80% v/v
corresponds to 75.8 wt %. pKa values interpolated from data given
between 0 and 20 wt %, and 65 and 80 wt % assuming a linear
interpolation. h Assuming effect of temperature same as for aqueous
media. i Calculated using data from ref 33. 20% v/v ethanol water
corresponds to 16.4 wt % and 80% v/v corresponds to 75.8 wt %. pKa
values interpolated from data given between 0 and 20.3 wt %, and
extent of amide formation by a factor of 84. The pKa of the
carboxylic acid group of 5 shifted from 3.6 in aqueous solution
to 5.3 in 85% 1,4-butanediol, without a significant change in
the pKa of the amine group of 6 (from 8.2 to 8.1). In that study
the enhancement in dipeptide formation in the above solvent
system was attributed mostly to the change in pKa of the
carboxylic acid group of benzyloxycarbonyl-L-tryptophan. Pre-
sumably such effect could be operative in the nonenzymatic
coupling of these two species as well.
Although the effects of the addition of ethanol can be
explained in terms of the change in pKa’s, another possibility
exists wherein ethyl formate is formed as a reactive intermediate
and amide formation will occur through the aminolysis of ethyl
formate (Scheme 5). Kistiakowsky studied34 the equilibrium
formation and hydrolysis of ethyl formate in acidic aqueous
ethanol and showed that the acid-catalyzed rate of attainment
of equilibrium ((k2 + k-2) in Scheme 5) varies from 2.3 × 10-3
to 3.3 × 10-3 M-1 s-1 when the solvent composition changes
from 21 to 80% ethanol at 25 °C. The equilibrium positions
lay 17, 61, and 72% toward formation of ethyl formate in 21,
73, and 84% ethanol, respectively. The author also reported an
8-10% increase in the rate of attainment of equilibrium when
the temperature was increased to 30 °C from 25 °C, with no
change in the equilibrium position. From these data, it is clear
that ethyl formate will be formed under our reaction conditions
so the possibility of formanilide formation via aminolysis of
ethyl formate cannot be ruled out at this point.
j
65.1 and 79.9 wt % assuming a linear interpolation. Assuming no effect
of temperature change. k µ not corrected.
the attainment of equilibrium for the various formanilides varies
from 1.6 to 8.6 h, and so for practical purposes, equilibrium
can be established within 9-42 h (5t1/2). The rate of attainment
of equilibrium is subject to a large steric effect in the acyl portion
such that acetanilide equilibrium is established 240 times slower
than is formanilide equilibrium under comparable conditions
of temperature and pH. Nevertheless, the corrected equilibrium
constant, K′eq, is only marginally affected by the steric factor,
the value for acetanilide being 3.7 times smaller than forma-
nilide.
Reducing the temperature slows the rate of establishing
equilibrium but has virtually no effect on K′, although K′eq
increases. This is due largely to the effect of increasing the
basicity of the aniline at lower temperature which makes it more
nucleophilic toward the acid although the RCOOH pKa is
essentially invariant with temperature. Added phosphate (0.5
M) increases the rate of establishing equilibrium for formanilide
by 1.6 times at pH 3.6 and 80 °C but does not influence the
equilibrium position. Both the rate of establishing equilibrium
and the K′ increase as the ethanol content increases in the
medium. The effect results primarily from increasing the
concentrations of the neutral forms of RCOOH and amine. From
a practical point of view, an 80% ethanol solution at pH 3.6
containing 1.0 M [formate]total at 60 °C enhances the yield of
formanilide to 91% from 73% in water, with equilibrium being
attained in 12 h relative to 42 h.
Conclusions
In this study we have shown that it is possible to form
formanilide, acetanilide, p-nitroformanilide, and p-methoxyfor-
manilide in reasonable amounts and in reasonable times from
their amine and carboxylic acid constituents in aqueous acidic
solution. The conditional equilibrium constant, K′, gives an
indication of the [anilide]/[aniline]total ratio expected once
equilibrium is established. The equilibrium position for forma-
nilides shifts more to formation with increasing basicity of the
aniline, and at 1.0 M [formate]total and pH 3.60, the respective
K′ values for formanilide, p-methoxyformanilide, and p-nitro-
formanilide are 2.05, 1.88, and 0.19 at 80 °C. The half-time for
All the experiments in this study were conducted using excess
formate and concentrations of aniline or anilide which ensured
homogeneous conditions for purposes of the kinetic work.
(34) Kistiakowsky, Z. Physik. Chem. 1898, 27, 250.

Equilibrium Formation of Anilides
J. Am. Chem. Soc., Vol. 121, No. 19, 1999 4607
Synthetically, for amide formation in water, one can use much
higher concentrations of acid and aniline since the forming
anilide will quickly separate from the water layer and drive the
equilibrium further to the side of amide formation.
Assuming that ∆S is insensitive to ionic strength, ∆H at µ )
1.0 (KCl) becomes 7.77 kcal/mol. Therefore, the pKa of
anilinium ion at 100 °C becomes 3.75. An average pKa of 3.78
was used for further calculations. Similar calculations were done
at 80 °C (pKa ) 4.02) and at 60 °C (pKa ) 4.30).
2. Formic Acid. The pKa of formic acid, 3.63 ( 0.03, was
obtained by potentiometric titration at 24 ( 1 °C and µ ) 1.0
(KCl). It is reported27 that ∆Hionization ) -0.04 kcal/mol and
∆Sionization ) -17 cal/(mol K) for formic acid at µ ) 0. From
calculations similar to those above, the pKa at 100 °C is 3.63.
Therefore, although the change in ionic strength (from 0 to 1.0
M) alters the pKa of formic acid by 0.12 unit, there is no effect
due to the change in temperature.35
Acknowledgment. The authors gratefully acknowledge the
financial assistance of the Natural and Engineering Sciences
research Council of Canada and Queen’s University.
Supporting Information Available: Tables of pseudo-first-
order rate constants for the establishment of equilibrium and
K′ values for formanilide, acetanilide, p-methoxyformanilide,
and p-nitroformanilide under various conditions (12 tables)
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
3. p-Methoxyanilinium Ion. Biggs has shown that the pKa
of p-methoxyanilinium ion decreases with increasing temper-
ature36 and has obtained pKa’s of p-methoxyanilinium ion at
five different temperatures ranging from 20 to 40 °C and µ )
0.10 M (KCl). The pKa vs temperature plot follows a linear
equation, pKa ) a + bΤ, where T is temperature in °C, a )
0.018, and b ) 5.79. Using these data pKa of p-methoxya-
nilinium was calculated to be 4.35 at 80 °C.
Appendix: Calculation of pKa’s Listed in Table 10
1. Aniline. Isaacs has listed27 the enthalpies and entropies
for dissociation of various acids and ammonium ions: for
anilinium ion ∆Hionization ) 7.38 kcal/mol and ∆Sionization ) 3.7
cal/(mol K) at µ ) 0. By titration in this study, the pKa of
anilinium ion was determined to be 4.89 ( 0.05 at 24 ( 1 °C
and µ ) 1.0 (KCl). Using the expression -∆G ) RT ln Ka,
∆G of anilinium dissociation is calculated to be 6.67 kcal/mol
(Ka ) 1.29 × 10-5) at µ ) 1.0 and T ) 24 ( 1 °C.
Since ∆G ) ∆H - T∆S, and assuming the ∆H is unaffected
by ionic strength, ∆S can be calculated as 2.39 cal/(mol K) at
µ ) 1.0 (KCl). Thus the ∆G at 100 °C and µ ) 1.0 becomes
6.49 kcal/mol. Hence, the pKa of aniline under these conditions
is 3.80.
JA990104D
(35) Perrin, D. D.; Dempsey, B.; Serjeant, E. P. In pKa Prediction for
Organic Acids and Bases; Chapman and Hall: London, 1981; pp 6-9.
(36) Biggs, A. I. J. Chem. Soc. 1961, 2572.
(37) Johnson, C. D.; Katritzky, A. R.; Shapiro, S. A. J. Am. Chem. Soc.
1969, 91, 6654.
(38) Perrin, D. D. In Dissociation Constants of Organic Bases in Aqueous
Solution; Butterworth: London, 1965.