Water catalyzed hydrolysis of p-nitrotrifluoroacetanilide and trifluoroacetanilide. Carbonyl 18O-exchange does not accompany the water reaction

Article abstract of DOI:10.1021/ja971463g

The water catalyzed hydrolyses of p-nitrotrifluoroacetanilide (3) and trifluoroacetanilide (8) were studied at various pH's and temperatures. The activation parameters for the water reactions of 3 and 8 are: ΔH = 14.4 ± 0.6 and 11.7 ± 0.3 kcal/mol and ΔS = -36.1 ± 1.8 and -52.3 ± 0.7 calmol·K, respectively. These are consistent with reactions that involve considerable restriction of degrees of freedom of the solvent/substrate in the transition state. A proton inventory analysis of the rate constants for hydrolysis of 3 in media of different mole fraction D₂O indicates the process involves two or more protons in flight or undergoing loosening of their bonding in the transition state. ¹⁸O-Labeled amides were subjected to the hydrolytic conditions for various times up to 3 half-times of hydrolysis and recovered. Mass analysis showed that the 180 content in the recovered amide did not change during the course of the reaction. All the data support a process where the rate-limiting step for the water reaction involves a concerted or nearly concerted formation of a diol which undergoes subsequent C-N cleavage in preference to OH expulsion.

Full text of DOI:10.1021/ja971463g

J. Am. Chem. Soc. 1997, 119, 10969-10975
10969
Water Catalyzed Hydrolysis of p-Nitrotrifluoroacetanilide and
1
8
Trifluoroacetanilide. Carbonyl O-Exchange Does Not
Accompany the Water Reaction
†
‡
H. Slebocka-Tilk, Christopher G. Rescorla, Salma Shirin, Andrew J. Bennet, and
R. S. Brown*
Contribution from the Department of Chemistry, Queen’s UniVersity,
Kingston, Ontario, Canada, K7L 3N6
X
ReceiVed May 6, 1997
Abstract: The water catalyzed hydrolyses of p-nitrotrifluoroacetanilide (3) and trifluoroacetanilide (8) were studied
q
at various pH’s and temperatures. The activation parameters for the water reactions of 3 and 8 are: ∆H ) 14.4 (
q
0
.6 and 11.7 ( 0.3 kcal/mol and ∆S ) -36.1 ( 1.8 and -52.3 ( 0.7 cal/mol‚K, respectively. These are consistent
with reactions that involve considerable restriction of degrees of freedom of the solvent/substrate in the transition
state. A proton inventory analysis of the rate constants for hydrolysis of 3 in media of different mole fraction D2O
indicates the process involves two or more protons in flight or undergoing loosening of their bonding in the transition
18
state. O-Labeled amides were subjected to the hydrolytic conditions for various times up to 3 half-times of hydrolysis
1
8
and recovered. Mass analysis showed that the O content in the recovered amide did not change during the course
of the reaction. All the data support a process where the rate-limiting step for the water reaction involves a concerted
or nearly concerted formation of a diol which undergoes subsequent C-N cleavage in preference to OH expulsion.
8
Introduction
co-workers have shown that the simplest amide, formamide,
exhibits a plateau around neutrality in the pH/rate profile for
Due to the importance of the amide bond in biological systems
it is not surprising that considerable effort has been expended
to understand the mechanisms for its hydrolysis in both acid
and base, eq 1. Experimental evidence pertaining to possible
mechanisms for the water promoted hydrolysis of amides is
much less available. This is due to the fact that amide hydrolysis
1,2
H O+
R NC(dO)R′ 7
OH-
7
3
-
+
R NH + R′CO
R NH2 +
2
2
2
2
R′CO H (1)
2
in water in the absence of acid or base is an exceedingly slow
process, and there are very few amides where the water reaction
or its kinetic equivalent, OH attack on the protonated amide,
hydrolysis at 80 °C which corresponds to rate constant for water
-
8
-1
-
hydrolysis (ko) of 8.4 × 10 s , (t1/2 ) 95 days). Quite
possibly the water reaction for amides is a general one in the
neutral pH region, but it is not usually considered due to the
difficulty in detecting it.
There is considerable interest in determining the water
hydrolysis rate constants for certain peptides so that these values
has been identified. For the most part, the identified amides
capable of hydrolyzing without acid or base catalysis activated
3
3b
4
5
6
7
in some way, such as 1, 2, 3, 4, 5, and 6. Without
structural activation, water promoted hydrolysis is very slow
but observable in some selected cases. For example, Hine and
could be used as comparisons for their enzymatic hydrolyses
*
Author to whom correspondence should be sent.
Undergraduate researcher, Queen’s University, 1996-1997.
Present address: Department of Chemistry, Simon Fraser University,
9-11
under comparable conditions.
New, highly sensitive assay
†
14
9
‡
techniques using C radioisotopic detection or post-hydrolytic
reaction to form highly fluorescent tracers have been developed
10
Burnaby, British Columbia, Canada, V5A 1S6.
X
Abstract published in AdVance ACS Abstracts, October 15, 1997.
to follow reactions to a few percent completion at 25 °C. In a
(1) For leading works summarizing the mechanisms of acid and base-
promoted amide hydrolysis see: (a) Jencks, W. P. Catalysis in Chemistry
and Enzymology: McGraw- Hill Inc.: New York, 1969; pp 523-527. (b)
Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 3rd. ed.; Harper and Row Inc.: New York, 1987; pp 714-717.
(4) (a) Komiyama, M.; Bender, M. L. Bioorg. Chem. 1978, 7, 133. (b)
Komiyama, M.; Bender, M. L. Bioorg. Chem. 1979, 8, 141. (c) Segretain,
A. M.; Beugelmans-Verrier, M.; Laloidiard, M. Bull. Soc. Chim. Fr. 1972,
3367.
(5) (a) Somayaji, V.; Brown, R. S. J. Org. Chem. 1986, 51, 2676. (b)
Blackburn, G. M.; Skaife, C. J.; Kay, I. T. J. Chem. Res. Synop. 1980, 294.
(6) (a) Cipiciani, A.; Linda, P.; Savelli, G. J. Heterocycl. Chem. 1978,
15, 1541. (b) Cipiciani, A.; Savelli, G.; Bunton, C. A. J. Heterocycl. Chem.
1984, 21, 975.
(
c) O’Connor, C. J. Q. ReV. Chem. Soc. 1970, 24, 553. (d) Bender, M. L.
Chem. ReV. 1960, 60, 53. (e) Isaacs, N. Physical Organic Chemistry, 2nd
ed.; Longman Scientific and Technical and J. Wiley and Sons: New York,
1
995; pp 529-531. (f) March, J. AdVanced Organic Chemistry 4th. ed.;
Wiley-Interscience: New York, 1992; pp 383-386.
2) (a) Brown, R. S.; Bennet, A. J.; Slebocka-Tilk, H. Acc. Chem. Res.
992, 25, 481. (b) Bennet, A. J.; Brown, R. S. Physical Organic Chemistry
(
(7) Patterson, J. F.; Huskey, W. P.; Hogg, J. L. J. Org. Chem. 1980, 45,
4675.
1
of Acyl Transfer Reactions In ComprehensiVe Biological Catalysis; Sinnott,
M. L., Ed.; Academic Press Inc.: London, 1997; Vol. 2, in press.
(8) Hine, J.; King, R. S.-M.; Midden, R.; Sinha, A. J. Org. Chem. 1981,
46, 3186.
(
3) (a) Jencks, W. P.; Carriuolo, J. J. Biol. Chem. 1959, 1272. (b) Fife,
(9) Kahne, D.; Still, W. C. J. Am. Chem. Soc. 1988, 110, 7529.
(10) Bryant, R. A. R.; Hansen, D. E. J. Am. Chem. Soc. 1996, 118, 5498.
(11) Radzicka, A.; Wolfenden, R. J. Am. Chem. Soc. 1996, 118, 6105.
T. H.; Natarajan, R.; Werner, M. H. J. Org. Chem. 1987, 52, 740. (c) Fife,
T. H. Acc. Chem. Res. 1993, 26, 325.
S0002-7863(97)01463-7 CCC: $14.00 © 1997 American Chemical Society

10970 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Slebocka-Tilk et al.
recent study, Radzicka and Wolfenden11 have determined the
activation parameters for water hydrolysis of some simple
dipeptides by monitoring the hydrolyses between pH 4.2 and
18
O-Labeled trifluoroacetic acid was made by exchange of trifluo-
roacetic acid (0.9 g, 0.008 mol) with 18O-water (0.4 g, 0.022 mol).
The mixture was heated at reflux overnight with a little dry HCl as a
catalyst.
7.8 at several temperatures between 120 and 200 °C. Depending
18O-Labeled p-nitrotrifluoroacetanilide was made exactly as above
on the structures of the peptides in the latter three studies, the
from p-nitroaniline and 18O-labeled trifluoroacetyltriflate (TFAT). The
9
t1/2 for the hydrolyses at 25 °C varied from 7 years to 600
synthesis of TFAT was done according to a procedure of Taylor et
years.11
17
al. A larger amount of P
O
2 5
(13.0 g, 0.046 mol) was added to account
18
Despite the interest in determining the rate of the water
reactions, there is no mechanistic information about how this
process occurs except in the case of the activated amides
for extra water introduced with O-labeled trifluoroacetic acid. The
crude amide was purified by column chromatography using silica gel
and an eluant of 20% (by volume) ethyl acetate in chloroform. Final
crystallization from acetone-hexane yielded 0.43 g of pure material
with about 12% O-enrichment in the carbonyl group (by mass
spectrometry). O-labeled trifluoroacetanilide was made twice from
-6.3-7
1
Of particular importance would be the detection of
18
intermediates produced by the addition of water and knowledge
of their partitioning between product formation and reversal to
form starting material. So far as we know, in only one case,
that of N-trifluoroacetylpyrrole (5), has it been unambiguously
18
1
8
O-Labeled TFAT and aniline in the manner described above, yield
4
5%, with about 16% and 44% 18O incorporation.
DCl (Aldrich, 37 wt % solution in D
2
O, 99.5% D), D
18O ( O, 95-98%, Cambridge Isotope Laboratories) were used
without further purification.
2
O (99.9%),
1
2
determined that a hydrated intermediate can be formed,
18
and H
2
13
although from simple mechanistic considerations and by
analogy with aldehydes and ketones,14 a gem-diol intermediate
All solutions used for kinetics studies were made using degassed
(CO free, O free), deionized water (Osmonics-Aries water purifying
(To, 7) is likely to be formed on the reaction pathway for water
2
2
promoted amide hydrolysis. Should partitioning of the diol
between reversal and product formation occur, it could be readily
detected through the use of 18OdC exchange studies such as
those we have used previously to determine the reversal of
intermediate formation in acid and base promoted hydrolysis
system).
Dimethoxyethane (HPLC grade, Fisher) was dried over Na (48 h,
reflux) under Ar atmosphere. Fresh, dry DME was used to make stock
solutions of the amides.
(b) Kinetics. (i) H
2
O. Kinetic data were obtained by observing
2,15a-f
15gh
the rate of change (increase for 3, decrease for 8) in absorbance of
of amides
and esters.
Herein we report detailed studies
-
4
0
.5-1.0 × 10 M aqueous solution of amides at the wavelength of
of the water hydrolyses of two trifluormethyl activated acetani-
maximum change (400 nm, 3; 250 nm, 8) with a Cary 210 UV-vis
spectrophotometer interfaced as previously described.15 At temperatures
below 80 °C the reactions were initiated by injecting 30 µL of a stock
4
18
lides, 3 and 8, as well as OdC studies that show that there
is no exchange in the carbonyl oxygen of the amides recovered
from the medium during the course of hydrolysis.
-2
solution of amide in DME (1 × 10 M) to a 1.0 cm quartz cuvette
containing 3.0 mL of acid solution of appropriate concentration (µ )
0.2 (KCl), 3; µ ) 1.0 (KCl), 8) which had been previously equilibrated
in the instrument cell holder for 15 min. Each determination was
repeated in triplicate, and the rate constants are derived from the fits
of the abs. vs time profiles to a standard exponential model: the values
reported in the tables are the averages with errors being the standard
deviation of the mean.
Experimental Section
1H NMR and 13C NMR spectra were obtained on a Bruker AC-200,
infrared spectra on a BOMEM MB-120 spectrometer, and mass spectra
on a Fisons Quattro Triple Quadrupole Mass Spectrometer coupled to
a Fisons GC 8000 gas chromatograph.
(
a) Materials. p-Nitrotrifluoroacetanilide (3) and trifluoroacetanilide
Data for slower reactions of 8 and for 3 at temperatures above 80
°C were obtained by measuring absorbance of individual 5 mL aliquots
of sample in the appropriate medium which were thermostated in sealed
ampoules (flushed with Argon, then flame sealed) in a boiler designed
for conducting isothermal experiments at elevated temperatures.
Ampoules were removed at various times and cooled, and the
absorbance of the contents measured at 25 °C. Rate constants were
obtained by fitting the absorbance vs time profile to a standard
exponential model using nonlinear least squares methods. The reactions
were followed to at least two half-times and in all cases exhibited clean
first-order kinetics.
(
8) were prepared by the dropwise addition (at 0 °C) of trifluoroacetic
anhydride to an equimolar amount of the appropriate aniline dissolved
in pyridine. In the case of p-nitrotrifluoroacetanilide (3) crystallization
from acetone-hexane (40:60%, by volume) gave yellowish crystals,
4
,17
1
yield 34%, mp 149-150 °C (lit. mp 147 °C).
δ ) 7.84 (m, 2H,), 8.23 (m, 2H,), 9.53 (bs, 1H, NH); 13C NMR
CD CN): δ ) 156.3 (q, JF,CdO ) 38 Hz), 145.7, 142.9, 125.7, 121.7,
16.6 (q, JF,C ) 285 Hz); IR (Nujol) 1744.6 cm
Trifluoroacetanilide (8) was crystallized also from acetone-hexane
60:40%, by volume) to give pinkish crystals, yield 66%, mp 89-90
3
H NMR (CDCl ):
(
1
3
-
1
.
(
°
(
1
1
6
1
2
(ii) D O. DCl solutions were made by adding aliquots of concen-
C (lit. mp 86.5 °C) . H NMR (CDCl
3
): δ 7.23-7.72 (m, 5H), 8.23
): δ ) 155.2 (q, JF,CdO ) 37 Hz), 135.0,
29.2, 126.4, 120.0, 115.7(q, JF,C ) 287 Hz).
1
3
trated DCl (37 wt %) to D O (µ ) 0.2 (KCl)). The exact acid
s, NH); C NMR (CDCl
3
2
concentrations were determined by titration against standardized base
using phenolphthalein as an indicator. Hydrolysis kinetics in D
O-H O mixtures were performed as above. Solutions of different
mole fraction D O were made up by mixing the appropriate amounts
of H O and D O and adding to 100 mL of this 1.0 mL of 1.00 N HCl
to bring the [L
2
O and
(12) Cipiciani, A.; Linda, P.; Savelli, G. J. Chem. Soc., Chem. Commun.
D
2
2
1
977, 857.
2
(
13) Guthrie, J. P. J. Am. Chem. Soc., 1974, 96, 3608.
2
2
(14) (a) Bell, R. P. AdV. Phys. Chem. 1966, 4, 1. (b) Patai, S. The
+
-2
3
O ] to 10 M.
c) O-Exchange. A typical exchange experiment was conducted
Chemistry of the Carbonyl Group; Wiley Interscience: London, 1970; Vol.
18
(
II, p 1.
(
15) (a) Yao-Liu, B.; Brown, R. S. J. Org. Chem. 1993, 58, 732. (b)
Bennet, A. J.; Slebocka-Tilk, H.; Brown, R. S. J. Am. Chem. Soc. 1992,
14, 3088. (c) Brown, R. S.; Bennet, A. J.; Slebocka-Tilk, H.; Jodhan, A.
as follows. To 192 mL of the acid solution of desired concentration
(µ ) 0.2 (KCl) for 3 and µ ) 1.0 (KCl) for 8) in a 200 mL volumetric
flask was added 8 mL of 0.0125 M solution of the appropriate amide
1
J. Am. Chem. Soc. 1992, 114, 3092. (d) Slebocka-Tilk, H.; Bennet, A. J.;
Hogg, H. J.; Brown, R. S. J. Am. Chem. Soc. 1991, 113, 1288. (e) Bennet,
A. J.; Slebocka-Tilk, H.; Brown, R. S.; Guthrie, J. P.; Jodhan, A. J. Am.
Chem. Soc. 1990, 112, 8497. (f) Slebocka-Tilk, H.; Bennet, A. J.; Keillor,
J. W.; Brown, R. S.; Guthrie, J. P.; Jodhan, A. J. Am. Chem. Soc. 1990,
12, 8507. (g) Kellogg, B. A.; Tse, J. E.; Brown, R. S. J. Am. Chem. Soc.
995, 117, 1731. (h) Kellogg, B. A.; Brown, R. S.; MacDonald, R. S. J.
2 2
in DME. The aqueous solutions were made CO and O free by
bubbling argon through them for 15 min. The contents of the
volumetric flask were transferred in 10 mL portions to several ampoules
(16-20) which were then flame-sealed and placed in the boiler. The
ampoules were kept at a given temperature for 10 min for thermal
equilibration after which four ampoules were removed at times
corresponding to 1/2, 1, 2, and 3 half-times of hydrolysis. At each
time of withdrawal, the contents of the four ampoules were combined
1
1
Org. Chem. 1994, 59, 4652.
16) (a) Rutherford, K. G.; Ing, S. Y-S.; Thibert, R. J. Can. J. Chem.
965, 43, 541. (b) Saxby, M. J. Org. Mass Spectrom. 1969, 2, 835.
17) Taylor, S. L.; Forbus, Jr., T. R.; Martin, J. C. Organic Syntheses;
Wiley: New York, 1990; Collect. Vol. VIII, p 506.
(
1
(
and extracted with 3 × 10 mL of freshly distilled CH
2 2
Cl . The
combined extracts were washed with saturated NaCl and water. The

p-Nitrophenyl Trifluoroacetanilide and Trifluoroacetanilide
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 10971
Table 2. Rate Constants for Acid (kH3O+) and Water (kH2O
)
Hydrolysis of Trifluoroacetanilide (8) at Different Temperatures (µ
)
1.0 (KCl))
H O+ × 10 a(M-1 s-1
4
H O × 106 a(s-1
1.01 ( 0.14
1.67 ( 0.13
3.92 ( 0.09
temp (°C)
k
)
k
)
3
2
7
8
2.0
2.4
1.69 ( 0.04
2.77 ( 0.12
8.36 ( 0.13
1
00.0
a
E+rrors computed from standard deviations on the plots of kobs vs
3
[H O ], data of Table 1.
Table 3. Pseudo-First-Order Rate Constants of the Hydrolysis of
p-Nitrotrifluoroacetanilide (3) at Different pH Values, T ) 70.2 (
0
.2 °C (µ ) 0.2 (KCl))
5
obs × 10 (s-1) pH
5
obs × 10 (s-1)
5
pH
k
obs × 10 (s-1) pH
k
k
a
b
-
0.28
.0
.0
.30
64.1 ( 0.1b
0.70
0.70
1.00
1.52
11.6 ( 0.1d
1.73 6.05 ( 0.05
1.97 5.6 ( 0.2
2.25 5.6 ( 0.2
0
0
0
34.0 ( 0.1
12.1 ( 0.1
c
d
36.2 ( 0.1
8.3 ( 0.1
c
3.0e
21.9 ( 0.1
6.5 ( 0.2
5.2 ( 0.3
12.1 ( 0.1c
Figure 1. Plot of the log kobs vs pH for the hydrolysis of trifluoroac-
etanilide (8) at three different temperatures. Data of Table 1: 9, T )
0.70
a
Determined at 300 nm, by observing disappearance of 3; µ ) 1.91.
7
2 °C; 2, T ) 82.4 °C; b, T ) 100 °C.
b
Determined at 400 nm by observing appearance of p-nitroaniline; µ
d
)
1.0. c Determined at 300 nm; µ ) 1.0 (KCl). Determined at 400
Table 1. Pseudo-First-Order Rate Constants of the Hydrolysis of
Trifluoroacetanilide (8) at Different pH Values and Temperatures (µ
e
nm; µ ) 0.2 (KCl). Value quoted in the table extrapolated to [buffer]
)
0 from formate buffer: three different concentrations of formate
)
1.0 (KCl))
-5 -1
buffer were used: 0.05 M, kobs ) 5.80 ( 0.01 × 10
s ; 0.10 M, kobs
; 0.20 M, kobs ) 8.05 ( 0.01 × 10
obs × 10 (s-1)
6
-5 -1
-5
s
-1.
k
) 6.92 ( 0.01 × 10
s
pH
t ) 72.0 °C
143 ( 3
t ) 82.4 °C
t ) 100.0 °C
799 ( 93
Table 4. Pseudo-First-Order Rate Constants for the Water
Catalyzed Hydrolysis of p-Nitrotrifluoroacetanilide (3) at Different
pH Values and Temperatures (µ ) 0.2 (KCl))
0
.00
.004
.301
.305
.70
.00
.12
.31
.52
.70
.82
.00
.31
.70
.80
363 ( 12
315 ( 35
0
0
0
0
1
1
1
1
1
1
2
2
2
2
78.3 ( 0.5
449 ( 20
obs × 105 (s-1) pH temp (°C)
obs × 105 (s-1)
1.52 ( 0.02
3.23 ( 0.06
5.6 ( 0.2
7.70 ( 0.09
1.38 ( 0.01
3.12 ( 0.03
5.6 ( 0.2
pH temp (°C)
k
k
212 ( 30
73.1 ( 2.1
35.1 ( 2.5
1
1
1
1
1
1
1
.64
.64
.64
.80
.80
.80
.80
46.3
50.1
64.7
46.3
50.1
64.7
82.4
1.16 ( 0.02 1.97
1.77 ( 0.01 1.97
3.93 ( 0.21 1.97
1.14 ( 0.04 1.97
1.78 ( 0.01 2.25
4.00 ( 0.10 2.25
50.4
61.2
70.2
75.3
51.2
61.1
70.4
32.3 ( 0.1
16.3 ( 0.2
12.4 ( 0.4
9.40 ( 0.22
5.92 ( 0.02
4.33 ( 0.08
3.53 ( 0.09
2.89 ( 0.34
107 ( 3
16.9 ( 3.0
7.2 ( 2.6
20.7 ( 1.2
15.9 ( 2.7
2.25
4.85 ( 0.25
3.12 ( 0.28
2.9 ( 1.6
12.4 ( 0.5
7.79 ( 0.21
5.43 ( 1.04
Table 5. Activation Parameters for the Hydrolysis of
p-Nitrotrifluoroacetanilide (3) and Trifluoroacetanilide (8)
1.73 ( 0.15
q
q
q
amide
∆H (kcal/mol)
∆S (cal/K mol)
∆G 100 C (kcal/mol)
1
.6 ( 1.7
4.93 ( 0.22
3
.10
1.81 ( 0.07
3
8
8
14.4 ( 0.6
11.7 ( 0.3
14.0 ( 1.3
-36.1 ( 1.8
-52.3 ( 0.7
-35.4 ( 3.7
27.9 ( 1.1
31.2 ( 0.6
27.2 ( 2.4
a
4
organic layer was dried (MgSO ) and stripped of solvent (rotary
a
H
3
O+ catalyzed reaction.
evaporator) to yield the residual amide which was subjected to analysis
with a Fisons quattro triple quadrupole mass spectrometer coupled to
a Fisons GC 8000 gas chromatograph (DB-5HS column, 30 m, i.d.
region from 1.0-4.5 in that pH profile resulting from the water
catalyzed reaction was of special importance to us. Therefore
the rate constants at different temperatures, solvent deuterium
0
1
.25 mm, 40 °C, 4 psi with helium carrier gas). For mass analysis,
1
6
18
5-20 scans of the O and O peaks were done for each sample, and
1
8
18
the O-content was evaluated as % O ) IM+2/(IM+ + IM+2) where I
)
%
18
kinetic isotope effect (SKIE), and O-exchange studies were
peak intensity and M and M + 2 refer to the molecular ions. The
performed in this region to learn more about the mechanism.
We controlled the pH with known [HCl] to avoid using buffers
such as formate which were used in the previous study.4 The
rate constants for hydrolysis of 3 in H2O (pH ) -0.28 to 3.0)
18
O vs time data are given in Table 7.
a,b
Results
1
8
(a) Hydrolysis. Given in Table 1 are the pseudo-first-order
at 70.2 °C are presented in Table 3, while the rate constants
rate constants, kobs, for the hydrolysis of trifluoroacetanilide (8)
at various acid concentrations (µ ) 1.0 (KCl): in Figure 1 are
graphical representations of the kobs vs pH data at three different
temperatures (72.0, 82.4, and 100.0 °C). These data can be
in H O at four different pH’s (1.64, 1.80, 1.97 and 2.25) in the
2
temperature range 46.3-82.4 °C are given in Table 4. Given
q
q
q
in Table 5 are the activation parameters ∆H , ∆S , and ∆G
(calculated using the standard Eyring equation) for the acid and
water catalyzed hydrolysis of trifluoroacetanilide and for the
water catalyzed reaction of p-nitrotrifluoroacetanilide (using the
data of Tables 2 and 4, respectively).
+
satisfactorily fit by linear regressions of the type: kobs ) kH O
3
+
[
[
(
H3O ] + kH O. Linear least squares fitting of the kobs versus
2
+
H3O ] data in the region of maximum curvature of the plots
pH 1-3) yields the rate constants, kH O+ and kH O, given in
(b) SKIE and Proton Inventory Studies. The solvent
kinetic isotope effects (SKIE) for the hydrolysis of 3 were
measured in the pH independent region (pH ) 2.0) at 70.2 °C.
The rate constants in DCl/D2O are given in Table 6. The
3
2
Table 2.
The pH-rate constant profile of the hydrolysis of p-nitrotri-
fluoroacetanilide (3) was examined in the pH range 0.0-11.4
by Komiyama and Bender.4
a,b
The reported pH-independent
(kobs)H O/(kobs)D O ratio is 3.3 ( 0.2 at [LCl] ) 10 M.
-2
2
2

10972 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Slebocka-Tilk et al.
Table 6. Pseudo-First-Order Rate Constants for the Water
in Table 5 it can be seen that the enthalpies lie in the 11.5-15
Catalyzed Hydrolysis of p-Nitrotrifluoroacetanilide (3) at Different
kcal/mol range, while the entropies of activation for both amides
+
-2
Mole Fractions of D
2 3
O, [L O ] ) 10 M, T ) 70.2 ( 0.3 °C, µ )
are large and negative: at 100.0 °C, the entropies contribute
0
.2 (KCl)
q
4
8% (3) and 62% (8) to the free energy of activation, ∆G .
mole fraction (n)
k
obs × 10 (s-1) mole fraction (n)
5
k
obs × 10 (s-1)
5
Water reactions have been also observed for the hydrolyses
0.0
0.2
0.4
0.6
5.59 ( 0.18
4.3 ( 0.1
3.53 ( 0.08
2.7 ( 0.2
0.8
1.0
1.0
2.3 ( 0.1
of activated esters such as alkyl and aryl haloacetates and
1.71 ( 0.05
1.59 ( 0.02
19a
19b
19b,c
substituted phenyl acetates, as well as for methyl, ethyl,
a
isopropyl, and tert-butyl trifluoroacetates.19b It is notable that
19b
for these the activation enthalpies are fairly low (7-14 kcal/
mol) while the entropies are large and negative (-34 to -50
cal/mol‚K), the exception being tert-butyl trifluoroacetate which
a
At pL ) 2.20.
Table 7. 18O Content in Amide Recovered at Various Times from
the Hydrolysis Medium during the Hydrolysis of p-Nitrotrifluoro-
acetanilide (3), µ ) 0.2 (KCl), and Trifluoroacetanilide (8), µ ) 1.0
1
9b
hydrolyses by an SN1 mechanism.
The former activation
parameters would be the expected ones for a hydrolysis process
involving a transition state that restricts the degrees of freedom
(KCl).
19a,b
tempa
% 18
O
of more than one water molecule.
The activation parameters
amide
temp (°C)
pH
can be rationalized by processes involving ordered transition
states involving several waters such as in 9-11. Whether these
processes are noncyclic (as in 9), concerted (10), or cyclic
leading to diol tetrahedral intermediates (11 to 12) is not known.
Analogous structures and transition states could rationalize the
activation parameters for the two amides in this study.
3
100.0
1.80
0
1
11.3 ( 0.2
11.1 ( 0.3
11.5 ( 0.3
11.1 ( 0.3
11.2 ( 0.4
11.3 ( 0.3
16.1 ( 0.2
16.1 ( 0.2
15.8 ( 0.1
15.5 ( 0.1
15.9 ( 0.1
16.0 ( 0.2
43.3 ( 0.3
43.8 ( 0.3
43.8 ( 0.4
44.0 ( 0.3
43.7 ( 0.3
43.9 ( 0.2
/2t1/2
t
1/2
1/
t
t
1
2
3
0
t
2 1/2
1/2
1/2
8b
70.0
0.0
t
1/2
1
2
.0
.0
0
t
1/2
0
t
1/2
1
00.0
1.0
0
t
2
0
1/2
t
1/2
3.0
2) In earlier studies Komiyama and Bender4a,b had deter-
(
t
2
1/2
mined that the pH independent region for hydrolysis of 3 extends
from 1.0 to 4.5 with a reported SKIE of 3.7 in this domain at
T ) 70 °C. Given that our initial hydrolysis experiments with
t
1/2
a
t
1/2 hydrolysis computed from the data in Tables 1 and 3.
3
indicated that the reaction rate constant for the water reaction
A proton inventory study was performed with 3 in the pH
4a
-5 -1
was roughly half that reported, (5.6 ( 0.2 × 10
s
vs 1.0
independent region (pH ) 1.97). In Table 6 are presented the
rate constants of water catalyzed reaction at various mole
fractions (n) of D2O in H2O at 70.2 °C: a graphical illustration
of these (not shown) is bowed down.
-4 -1
(
0.05 × 10 s ), we decided to repeat the hydrolysis
experiments with 3 at pH 1-3 using HCl so that we could avoid
20
the use of formate buffers. It was necessary to repeat all the
hydrolysis measurements since the 2-fold disparity between our
c) O-Exchange. Whether 18O-exchange accompanies the
1
8
(
4a,b
data and the previous ones
could have significant bearing
hydrolyses of either amide was determined by mass analysis of
amide recovered from the hydrolytic media at various times up
to 3t1/2 hydrolysis. The hydrolysis solutions contained 4% DME
for reasons of solubility. The experimental data for O-
exchange for 3 and 8 (the latter in both 1 M acid and at pH 1,
on the interpretation of the SKIE. The SKIE that we determined
for the hydrolysis of 3 at [LCl] ) 10 M, T ) 70.2 °C is 3.3
-2
(
0.2. These data are consistent with the SKIE of 3.7 at pH )
1
8
4a,b
4
.0, T )70 °C that was originally reported, the only difference
being the absolute values of the rate constants.
3) The fact that there is no observable 18O-exchange for the
2
, and 3) are given in Table 7. In all cases, the results show
(
18
that the percentage of O remains, within experimental error,
unchanged up to 3 half-times of hydrolysis.
water reaction of either amide comes as a surprise and rules
out some of the available mechanisms while placing constraints
on others. Most easily ruled out is a two- or more-step process
such as in eq 2 where a symmetric diol intermediate (hydrate)
is reversibly produced from the addition of water with rate
Discussion
There are three essential pieces of information that need to
be considered in order to establish the mechanistic possibilities
for the water reaction for anilides 3 and 8. These are (1) the
activation parameters for the reaction for both amides; (2) the
solvent SKIE and proton inventory conducted for 3 (this work
and ref 4b); and (3) the lack of any observable 18O-exchange
in either amide recovered from the hydrolysis medium. We
will deal with these data in turn.
(18) The data in Table 3 indicate that below pH 1 there is an acid
promoted reaction. At higher acid concentrations, the abs. vs time profile
at 400 nm shows some upward drift indicative of some decomposition of
the p-nitroaniline product. At 300 nm, the disappearance of 3 exhibits clean
first-order kinetics. The dependence of the rate on [HCl] seems slightly
less than first-order, and the cause of this is under investigation. We thank
a referee for alerting us to inconsistencies in our original data reported for
the acid domain.
(
19) (a) Kirby, A. J. In ComprehensiVe Chemical Kinetics, Bamford, C.
H., Tipper, C. F. H., Eds.; Elsevier: Amsterdam, 1972; Vol. 10, pp 57-
07. (b) Barnes, D. J.; Cole, M.; Lobo, S.; Winter, J. G.; Scott, J. M. W.
(1) In the case of the hydrolysis of trifluoroacetanilide (8),
2
there is no plateau in the pH vs kobs profiles shown in Figure 1,
but fitting of the pH vs kobs data in Table 1 to the linear
Can. J. Chem. 1972, 50, 2175. (c) Hegazi, M.; Mata-Segreda, J. F.;
Schowen, R. L. J. Org. Chem. 1980, 45, 307 and references therein.
(20) The possible source for the discrepancy between our results and
those of Komiyama and Bender4 could arise from the use of formate
buffers. Experimentally this requires that the kobs values be several different
+
+
regression, kobs ) kH O [ H3O ] + kH O, allows one to determine
3
2
a,b
the acid and water hydrolytic constants at the three temperatures
(
8
s
Table 2). By comparison, at 72 °C the water reaction for amide
[
buffer], the reported value being found by extrapolation to [formate] ) 0.
-
6
is some 60-fold slower than that for 3 at 70 °C: 1.0 × 10
In our study we have avoided the use of buffers by controlling the pH with
HCl.
-
1 -5 -1
vs 5.6 × 10 s . From the activation parameters given

p-Nitrophenyl Trifluoroacetanilide and Trifluoroacetanilide
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 10973
Scheme 1
limiting breakdown to product. This leaves us with a few
alternative possibilities that need to be considered in the light
of the proton inventory data accumulated here and in the
previous study.4
a,b
(
4) As has been well established, the proton inventory
2
1
technique provides information about the number of protons
undergoing change in their bonding in the transition state relative
to the ground state. Equation 3 expresses the relationship
between the rate constant observed in mixtures of L2O (L ) H,
D) with known isotopic composition and the fractionation factors
Mechanistic Possibilities. (a) Diol with Equivalent Oxy-
gens. Should the diol actually be the intermediate, its formation
18
must be rate limiting, otherwise it is expected that O-exchange
would be observed. From the proton inventory and activation
parameter data, the process requires two or more protons in flight
and restriction of at least two waters in the transition state.
Viable possibilities are given in 13 and 14 which are cyclic
and open versions of a two-or-more-water process. An analo-
gous cyclic TS involving three H2O molecules is suggested for
the hydration of 1,3-dichloroacetone: this process shows a kH/
(
φ) for the exchangeable protons undergoing change in their
bonding; n is the atom fraction of D in the medium, and i and
j represent the sites undergoing exchange in the transition and
reactant states, respectively.
TS
RS
2
5
kn ) ko (1 - n + nΦi)/ (1 - n + nΦj)
(3)
kD ) 2.7. Transition states for hydration involving two or
three waters could be feasible geometrically. An analogy might
be the proton exchange between the oxygens of benzoic acid
in methanol, for which it has been shown that there are two
solvent molecules involved in an eight-membered ring.26 Fitting
all the proton inventory data of Table 6 to the simplified form
∏i
∏J
The φ’s for the various hydrogens relate to the tightness of
their bonding and are less than unity if the overall bonding is
weak as is the case for H’s in flight in the transition state or for
H’s involved in hydrogen bonds where the overall bonding is
loose.22 The former case relates to primary kinetic isotope
effects, while the latter relates to secondary kie’s which can
contribute significantly to the overall effect. The data for the
hydrolysis of 3 determined at [LCl] ) 10 M (Table 6), when
plotted (not shown), indicate a significant downward bowing
which is consistent with a transition state having two or more
TS
of eq 3, kD/kH ) Π(1 - n-nφ ), where the fractionation factors
-
2
2
0
protons in flight or undergoing significant bond loosening.
We can analyze the expected SKIE on kobs for various
possibilities for the hydrolysis of 3 in a fashion analogous to
what we have presented before for the acid and base promoted
hydrolyses of amides.15 The φ values for the exchangeable H’s
in the ground state (H2O) are unity which considerably simplifies
for the protons in flight in TS 13 are equal gives best fit values
TS
of φ ) 0.50 if there are two protons in flight from two waters,
TS
and φ ) 0.67 if there are three waters involved, the fit for
the latter case being slightly better than for two waters. Fitting
TS
the overall SKIE based on eq 3 into kD/kH ) Πφ . We assume
of the limited data to a model for TS 14 leading to the diol/
that the transition state for water addition is rate limiting with
the position of the TS being roughly 0.5-0.7 along the reaction
coordinate. The primary DKIE for a proton in flight between
O and O or O and N is based on a TS fractionation factor of
+
-
H3O /HO (H2O)2 encounter complex 15 is not possible given
the fact that there are at least two unequal protons in flight,
three more undergoing changes in bonding in the developing
hydronium and hydroxide ions, and at least two waters of
0
.4-0.5, or slightly higher (0.6-0.7) if the proton transfers are
- 27
solvation for the developing OH . Large SKIEs for kH/kD
-
nonlinear as required for a cyclic transition state. For OH ,
the φ is 1.22 with each of its three solvating water H-bonds
having φ ) 0.7;23 a developing hydroxide or one involved in
proton abstraction then has two waters of solvation rather than
between 6.9 and 8.5 can be computed for this process on the
basis of the assumptions made above which probably rules out
this mechanism. Furthermore, from an intuitive perspective,
TS 14 seems an unlikely way of creating the diol because of
-
the three in HO (H2O)3. Finally, φ for each of the hydrogens
-
+
the simultaneous generation of OH and H3O in 15, although
+
24
in H3O is 0.69, while those for the OH’s in a diol and any
anionic tetrahedral intermediate are 1.0. In what follows we
will use bold H’s to signify the hydrogens in the TS with φ
values different from unity.
interestingly, in the microscopic reverse direction, the require-
-
+
ment for collecting both OH and H3O for the reversal of 15
18
provides a convenient explanation for the lack of O-exchange.
18
In reality, the lack of O-exchange simply means that
protonation of the leaving group and C-N cleavage is faster
than oxygen equilibration and expulsion from the tetrahedral
intermediate. Due to the strong electron withdrawing properties
(
21) (a) Schowen, R. L. In Isotope Effects on Enzyme-Catalyzed
Reactions; Cleland, W. W., O’Leary, M. H., Northrop, D. B., Eds.;
University Park Press: Baltimore, MD, 1977. (b) Alvarez, F. J.; Schowen,
R. L. In Isotopes in Organic Chemistry; Buncel, E., Lee, C. C., Eds.;
Elsevier: Amsterdam, 1987; Vol. 7, pp 1-60. (c) Kresge, A. J.; More-
O’Ferrall, R. A.; Powell, M. F. In Isotopes in Organic Chemistry; Buncel,
E., Lee, C. C., Eds.; Elsevier: Amsterdam, 1987; Vol. 7, pp 177-274. (d)
Alvarez, F. J.; Schowen, R. L. In Isotopes in Organic Chemistry Buncel,
E., Lee, C. C., Eds.; Elsevier: Amsterdam, 1987; Vol. 7, pp 1-60.
(25) Bell. R. P.; Millington, J. P.; Pink, J. M. Proc. R. Soc. London Ser.
A 1968, 303, 1.
(26) (a) Grunwald, E. Prog. Phys. Org. Chem. 1966, 3, 317. (b)
Grunwald, E.; Jumper, C. F.; Meiboom, S. J. Am. Chem. Soc. 1963, 85,
522.
(27) Note that for any of the models for the various TS, the limited data
available means that the derived fractionation factors from fitting are
seriously correlated.
(
(
(
22) Kresge, A. J. J. Am. Chem. Soc. 1973, 95, 3065.
23) Gold, V.; Grist, S. J. Chem. Soc., Perkin Trans. 2 1972, 89.
24) Bone, R.; Wolfenden, R. J. Am. Chem. Soc. 1985, 107, 4772.

10974 J. Am. Chem. Soc., Vol. 119, No. 45, 1997
Slebocka-Tilk et al.
of the CF3 group, the starting materials are destabilized28 which
both encourages nucleophilic addition and discourages regenera-
tion of the amide. Thus, the rate constant for expulsion of OH
from the diol may be abnormally slow. For related cases where
we know diols are formed readily, such as in the hydration of
anionic oxygen be protonated either by external H3O or by
+
+
the nascent H3O in the encounter complex 18. The former
pathway is thermodynamically favorable and should occur at
the diffusion limit, which, at pH 1, sets the rate constant at
9
10 -1 32
+
roughly 10 -10 s . Proton transfer from the nascent H3O
5
to form 76 or in the acid-promoted hydrolysis of N-
would require rotation within the complex and disruption/
reformation of hydrogen bonds which could occur in competi-
tion with diffusional separation of the encounter complex (i.e.,
15b
18
toluoylpyrrole (16, Scheme 1), reversal with concomitant O-
exchange1 occurs. Acid promoted amide hydrolysis in general
5b
+
10
11 -1
gives a T0/H3O encounter complex which, if the amine is basic,
10 -10 s or possibly slower since this process involves
3
2
undergoes fast proton transfer to N prior to diffusional separa-
overcoming electrostatic attraction). Because of the rapidity
of these processes, the only way that the two oxygens could be
prevented from becoming protonically equilibrated under the
reaction conditions would be if the lifetime of the intermediate
tion.1
5,29
This accounts for the fact that acid-catalyzed hydroly-
sis of amides, including 8 (Table 7), is not usually accompanied
by significant 18O-exchange unless protonation of the amine is
inhibited sterically or by electronic factors that reduce the ability
to install a proton at the N. Such is the case with 16 which
-11
with respect to C-N cleavage was less than 10 s. This would
require that the H O in the encounter complex become
+
3
15
exchanges about 50-fold faster than it hydrolyzes. In the
associated with the developing, but weakly basic, nitrogen lone
pair to assist in the aniline’s departure. In an effort to explain
30
present case, the amine leaving groups are more basic than is
3
1
4b
pyrrole. Apparently, however it is created, once the diol of 3
or 8 enters the hydrolysis manifold, C-N cleavage with general
their proton inventory data, Komiyama and Bender have
implied simultaneous proton transfer from the attacking water
to the departing leaving group in the hydrolysis of 3 via
transition state 19. In the limit where this occurs so quickly as
to prevent equilibration of the oxygens, the energy barrier for
the breakdown process must be sufficiently low that water attack
and leaving group departure are essentially concerted. Con-
certed processes have been suggested for the water promoted
hydrolyses of acyl imidazoles (1,2) and acyl triazole (6) via
a cyclic transition state as in 20, but these amides have remote
basic nitrogens with fully formed lone pairs to which H-bonds
can be directed. The scenario of a concerted reaction with no
intermediate formed during the hydrolysis of the anilides used
in this study would require that there be a preexisting hydrogen
bond network between the attacking water(s) and the anilino
lone pair, which, given its extensive delocalization into the
CF3CdO and aromatic ring, is unfavorable. We conclude that
a truly concerted displacement of the anilide by water is unlikely
and that the reaction, even if proceeding via the immature
+
-4
or specific acid catalysis by H3O (0.1-10 M) occurs more
easily than OH expulsion.
(b) Tetrahedral Intermediate with Inequivalent Oxygens.
A possibility exists that attack of one water is assisted by a
second as in 17 to give the encounter complex 18. The
mechanism is similar to that proposed and termed “immature
1
9c
3b,c
7
hydronium ion” by Hegazi, Mata-Segreda, and Schowen for
the water hydrolysis of ethyl trifluoroacetate for which a SKIE
of 3.6 is observed. Transition states 17 and 14 are similar except
for the degree of transfer of the proton from the three waters
solvating the developing oxyanion. For TS 17, the data can be
1
9c
hydronium ion mechanism
must lead to the diol which
irreversibly breaks down to product.
fit by a model having a single proton in flight (fixed value of
TS
φ
) 0.4-0.5) and five others changing their degree of bonding
TS
24
(
computed values of φ ) 0.95-0.91). If the three waters
solvating the oxyanion are omitted, as they were in the
immature hydronium ion” mechanism, and a φ of 0.5 is
set for Ha, a computed φ of 0.8 for Hb is obtained; these values
19c
TS
“
TS
compare favorably with those computed for the analogous H’s
Ha
Hb
-
in the hydrolysis of ethyl trifluoroacetate (φ ) 0.47, φ
)
(c) OH Attack on a Protonated Amide. As a final topic
9c
0
.78).1 Note also that for any fits for the latter cases, the φ
we consider the possibility that the water reaction really results
from its kinetic equivalent, HO attack on the protonated amide
as in 21. One can rule this out, at least in the case of 3, by
considering that at [H3O ] ) 10 M, the onset of the water
-
values of Ha and Hb are heavily correlated. Even the case of
a very late transition state resembling 18, with its fully formed
Ha
Hb
+
-2
hydronium ion (φ ) φ ) 0.69) and an unspecified number
of weakly associated waters all having φ ) 1.0, provides a
satisfactory fit to the proton inventory data.
-
-11
33a
reaction, the [HO ], is ∼1.59 × 10
M at 70 °C and the
-
5
amount of protonated amide is calculated to be <10 M if a
3
3b
A convenient, but not necessarily correct, rationalization for
pKa of less than -3 is appropriate for the latter.
Since kobs
) 5.6 × 10 s ) kOH-[HO ][A-H ], a value for kOH- of >
3.5 × 10 would be necessary to explain the observed
rate, this value being about 35- to 100-fold greater than the
1
8
-5 -1
1 -1 -1
-
+
the lack of O-exchange is a virtue of this mechanism since
the carbonyl oxygen need never become fully equivalent to the
one in the attacking water. To do so would require that the
1
M
s
3
2
diffusion limit. An analogous calculation for 8 cannot rule
out this possibility for the water reaction, but the strong structural
(28) Gambaryan, N. P.; Rokhlin, E. M.; Zeifman, Yu. V.; Ching-Yun,
C.; Knunyants, I. L. Angew. Chem., Int. Ed. Engl. 1966, 5, 947.
(
(
29) Satterthwait, A. C.; Jencks, W. P. J. Am. Chem. Soc. 1974, 96, 7031.
30) pKa of anilinium and p-nitroanilinium are 4.63 and 1.0 (25 °C):
(32) Eigen, M. Angew. Chem., Int. Ed. Engl. 1964, 3, 1.
CRC Handbook of Chemistry and Physics, 50th ed.; Weast, R. C., Ed.;
The Chemical Rubber Co.: Cleveland, OH, 1969-1970; p D115.
(33) (a) At 70 °C the Kw of water is computed to be 1.59 × 10-11.
Harned, H. S.; Owen, B. B. The Physical Chemistry of Electrolytic Solutions,
3rd ed., ACS Monograph Series, No. 137, Reinhold: New York, 1958; p
645. (b) pKa for amides is usually between 0 and -2; see: Guthrie, J. P J.
Am. Chem. Soc. 1974, 96, 3608 and references therein.
(
31) pKa of the conjugate acid of pyrrole is -3.8, but this is for
protonation on the ring; if the protonation were on N, the pKa would be
lower: Chiang, Y.; Whipple, E. B. J. Am. Chem. Soc. 1963, 85, 2763.

p-Nitrophenyl Trifluoroacetanilide and Trifluoroacetanilide
J. Am. Chem. Soc., Vol. 119, No. 45, 1997 10975
analogies between 3 and 8 suggest the two should hydrolyze
by the same mechanisms.
to the diol prior to diffusional separation. The most interesting
1
8
aspect of this work is the lack of observed O-exchange in the
CdO group of the amide recovered from solution at incomplete
hydrolysis. This indicates that even though the amine leaving
groups are weakly basic, their departure from the diol interme-
diate, which undoubtedly requires protonation prior to, or
concerted with C-N cleavage, occurs in preference to regenera-
tion of the amide starting material. It remains to be seen how
well the information obtained for these amides relates to the
water reactions of amides in general.
Conclusion
The current study provides the most complete set of data
available that have bearing on the ill-defined process of water
promoted hydrolysis of an amide. Admittedly the amides in
question are trifluoroacetanilides which are both highly disposed
toward the formation of hydrated forms and resistant to
expulsion of the aniline and OH groups. The available evidence
for these systems indicates that the most probable mechanism
is a two-step one where the rate limiting step involves two or
more waters attacking the amide via an ordered, possibly cyclic,
transition state. It is most likely that the first intermediate having
any lifetime is the diol (hydrate), and this could be formed in
one concerted step or in two steps from a anionic tetrahedral
Acknowledgment. The authors gratefully acknowledge the
financial support of the Natural Sciences and Engineering
Research Council of Canada (NSERC), the University of
Alberta, and Queen’s University. In addition they acknowledge
the helpful suggestions of the referees.
+
intermediate/H3O encounter complex which quickly collapses
JA971463G