Hydroxide as general base in the saponification of ethyl acetate

Article abstract of DOI:10.1021/ja011931t

The second-order rate constant for the saponification of ethyl acetate at 30.0°C in H₂O/D₂O mixtures of deuterium atom fraction n (a proton inventory experiment) obeys the relation k₂(n) = 0.122 s^-1 M^-1 (1 - n + 1.2n) (1 - n + 0.48n)/(1 - n + 1.4n) (1 - n + 0.68n)³. This result is interpreted as a process where formation of the tetrahedral intermediate is the rate-determining step and the transition-state complex is formed via nucleophilic interaction of a water molecule with general-base assistance from hydroxide ion, opposite to the direct nucleophilic collision commonly accepted. This mechanistic picture agrees with previous heavy-atom kinetic isotope effect data of Marlier on the alkaline hydrolysis of methyl formate (Marlier, J. F. J. Am. Chem. Soc. 1993, 115, 5953).

Full text of DOI:10.1021/ja011931t

Published on Web 02/14/2002
Hydroxide as General Base in the Saponification of Ethyl
Acetate
Julio F. Mata-Segreda*
Contribution from the School of Chemistry, UniVersity of Costa Rica, 2060 Costa Rica
Received August 9, 2001
2 2
Abstract: The second-order rate constant for the saponification of ethyl acetate at 30.0 °C in H O/D O
-
1
mixtures of deuterium atom fraction n (a proton inventory experiment) obeys the relation k
M
2
(n) ) 0.122 s
-
1
3
(1 - n + 1.2n) (1 - n + 0.48n)/(1 - n + 1.4n) (1 - n + 0.68n) . This result is interpreted as a process
where formation of the tetrahedral intermediate is the rate-determining step and the transition-state complex
is formed via nucleophilic interaction of a water molecule with general-base assistance from hydroxide ion,
opposite to the direct nucleophilic collision commonly accepted. This mechanistic picture agrees with previous
heavy-atom kinetic isotope effect data of Marlier on the alkaline hydrolysis of methyl formate (Marlier, J. F.
J. Am. Chem. Soc. 1993, 115, 5953).
The alkaline hydrolysis of carboxylic esters in solution is
commonly accepted to take place by a two-step mechanism,
where formation of the tetrahedral intermediate is strongly rate-
acids undergo alkaline hydrolysis with relatively low enthalpies
of activation, e.g., 40 kJ/mol for ethyl formate, 43 kJ/mol for
ethyl acetate, 44 kJ/mol for ethyl isobutyrate, and 52 kJ/mol
1
1
determining and occurs through direct nucleophilic collisions
for ethyl pivalate, too low a value to account for extensive
between hydroxide ions and ester molecules (eq 1, the BAC2
mechanism).
desolvation.
Kinetic data on the saponification of ethyl acetate in mixtures
of water and organic solvents show that one water molecule is
involved in the formation of the transition-state (TS) complex.
These cases include acetone-H2O mixtures in the range from
4
pure water to a water concentration of 22 M, dioxane-H2O
mixtures over the water concentration range from 24 to 53 M,4
and DMSO-H2O mixtures with water concentrations from 5
5
to 25 M. This idea of general-base catalysis also fits well with
the intuitive picture of a moving negatively charged “OH”
atomic configuration, randomly diffusing in aqueous media via
a Grotthuss mechanism, until it becomes attached to the
2
Marlier challenged this commonly held view in 1993 on the
basis of heavy-atom kinetic isotope effect experiments. The
author measured carbonyl carbon and oxygen as well as
nucleophilic oxygen kinetic isotope effects for the alkaline
hydrolysis of methyl formate in water solvent at 25 °C. Carbonyl
oxygen k16/k18 ) 0.999, carbonyl carbon k12/k13 ) 1.038, and
oxygen nucleophile k16/k18 ) 1.023 were interpreted as consis-
tent with a stepwise mechanism with rate-limiting formation
of the tetrahedral intermediate. The isotope effect on the oxygen
nucleophile was interpreted as a molecular event where the
attacking nucleophile in aqueous alkali is water, with general
base assistance from hydroxide.
6
aqueous solvation sphere of the ester. Once in place, the TS
complex for ester alkaline hydrolysis may be formed through
the intermediacy of a bridging water molecule. Thus, consid-
erations other than Marlier’s kinetic isotope effects also suggest
the direct nucleophilic collision as a less likely mechanism for
the saponification of carboxylic esters.
7
A proton inVentory experiment for the saponification of a
common carboxylic ester will clearly discriminate the direct
nucleophilic collision from the general-base-catalyzed mecha-
nism, since this kind of probe allows the determination of the
minimal structure of the TS complex. A minimal structure
includes all the reacting molecules and ions involved in the
formation of the TS complex, without details about the config-
uration of loosely bound solvent molecules.
Hydroxide ion is strongly solvated in aqueous environment,
as can be inferred from its high enthalpy of hydration of -423.4
kJ/mol. A direct ester/OH nucleophilic collision must in-
3
-
volve a great amount of desolvation energy. Esters of carboxylic
*
To whom correspondence should be addressed. E-mail: jmata@
cariari.ucr.ac.cr.
(4) Tommila, E.; Koivisto, A.; Lyyra, J. P.; Antell, K.; Heimo, S. Ann. Acad.
Sci. Fennicae A II 1952, No. 47.
(
(
(
1) Kirby, A. J. In ComprehensiVe Chemical Kinetics; Bamford, C. H., Tipper,
C. F. H., Eds.; Elsevier: Amsterdam, 1972; Vol. 10.
(5) Tommila, E.; Murto, Acta Chem. Scand. 1963, 17, 1947.
(6) Laidler, K. J. Chemical Kinetics, 3rd ed.; Harper & Row: New York, 1987;
p 221.
2) (a) Marlier, J. F. J. Am. Chem. Soc. 1993, 115, 5953. (b) Marlier, J. F.
Acc. Chem. Res. 2001, 34, 283.
3) Friedman, H. L.; Krishnan, C. V. In Water. A ComprehensiVe Treatise;
Franks, F., Ed.; Plenum: New York, 1973; Vol. 3, p 56.
(7) Schowen, K. B. J. In Transition States of Biochemical Processes; Gandour,
R. D., Schowen, R. L., Eds.; Plenum: New York, 1978.
10.1021/ja011931t CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC.
9
VOL. 124, NO. 10, 2002 2259

A R T I C L E S
Mata-Segreda
Table 1. Second-Order Specific Rates for the Saponification of
Ethyl Acetate at 30.0 °C in Mixtures of H2O/D2O of Deuterium
Atom Fraction n
species (R). Each φ value gives the ratio of the preference for
deuterium over protium relative to the similar preference in bulk
solvent molecules.
1
n
k
2
(n)/s- M-1
For the system under study, one has
0.000
0.190
0.237
0.438
0.609
0.998
0.122 ( 0.007
0.128 ( 0.007
0.13 ( 0.01
0.138 ( 0.003
0.144 ( 0.002
0.160 ( 0.001
TS
k (n) ) k (n ) 0) Π (1 - n +
2
2
n φ )/reactant-state contribution (3)
TS
The structural model of Gold and Grist for aqueous hydroxide
is the most appropriate for the interpretation of proton inventory
data because it takes into account the strong specific solute-
solvent interactions inherent to the structure of aqueous hy-
droxide. The ion is considered as being tightly solvated by three
water molecules, as shown in the following structure:
This paper presents the proton inventory for the saponification
of ethyl acetate at 30.0 °C. The results agree with the mechanism
proposed by Marlier.
9
Experimental Section
Materials. Ethyl acetate (BDH) was distilled under normal pressure.
Deuterium oxide (99.8%) was from Aldrich.
Preparation of Solutions. The mixtures were prepared gravimetri-
cally in 100 or 50-mL volumetric flasks. A predetermined amount of
aqueous 0.1 M NaOH was used in their preparation to yield final NaOL
(L ) H, D) concentrations of about 10 mM, except the case of 99.8%
D
2
O, where metallic sodium was used to make NaOD.
Kinetic Runs and Data Treatment. The reaction temperature was
3
0.0 °C. The reactions were started by addition of enough ethyl acetate
The Gold-Grist model gives the values φa ) 1.2-1.5, φb )
.65-0.70, and φc ) 1. The mean values 1.4, 0.68, and 1.0
will be used, respectively. Thus
to the NaOL/L O solution, so that equimolar initial quantities of both
reagents were present. Sampling was done every five minutes. Titration
of the remaining amount of NaOL with 0.0100 M HCl/neutral red in
2
0
2
000-µL aliquot portions was done by means of a Gilmore microburet.
Each run was followed to at least 75% of completion. The rate
determinations in the L O mixtures were carried out in duplicate, the
O experiment was done in triplicate, and the H O experiment was
run five times.
reactant-state contribution, RSC(n) )
3
2
(1 - n + 1.4n) (1 - n + 0.68n) (4)
D
2
2
and
Second-order rate constants were obtained as the slopes of 1/[NaOL]
vs reaction time plots by linear least-squares fitting.
TS
transition-state contribution, TSC(n) ) Π (1 - n + φ n)
TS
Results and Discussion
(5)
Specific-Rate Values. The saponification of ethyl acetate with
NaOH in water solvent obeys the rate law
Inserting eqs 4 and 5 into eq 3 and solving for TSC(n) gives:
3
TSC(n) ) k (n)(1 - n + 1.4n)(1 - n + 0.68n) /k (n ) 0)
-
d[NaOH]/dt ) k [NaOH][AcOEt]
2
2
2
(6)
Table 1 gives the second-order rate constants in different H2O/
D2O mixtures of deuterium atom fraction n. The inverse overall
kinetic solvent isotope effect kH O/kD O ) 0.76 ( 0.01 agrees
Once the contributions from all reactant species are stated
from known experimental or calculated fractionation factors,7
the next step in the interpretation of the data is carrying out a
polynomial analysis of TSC(n). This analysis determines the
number of hydrogenic sites that, in the transition state, make
the greatest contributions to the overall kinetic isotope effect.
The mathematical description of the TSC(n) data as a function
of n should lead to the minimal structure of the TS complex.
Figure 1 shows the rate data as a curve increasing with n.
The qualitative meaning of the positive slope is that TSC(n) >
RSC(n) and the curvature suggests a multiproton mechanism.
Polynomial fitting was performed on the TSC(n) data. The
values of total standard deviation are 0.0138 for linear fit,
0.00269 for quadratic, and 0.00308 for the cubic function.
Intuition calls for quadratic fit, since no reasonable mechanism
can be envisioned for a process involving three protons
undergoing bonding changes in the TS, while it is easy to
visualize a two-proton process.
2
2
8
quite well with the reported value of 0.75 for this reaction.
The rate data have standard errors ranging from 0.3% up to 8%
average 3% standard error). Thus, there is no need to analyze
(
this set of data at statistical certainties greater than 95% (vide
infra).
Proton Inventory. The purpose of a proton inventory
experiment is to determine how many exchangeable hydrogenic
sites contribute to the observed overall solvent kinetic hydrogen
isotope effect. In this kind of study, one relates the observed
7
k2(n) and n data pairs through the Gross-Butler equation
TS
R
k (n) ) k (n ) 0) Π (1 - n + n φ )/Π (1 - n + n φ )
2
2
TS
R
(2)
where the φ values are the isotopic fractionation factors for all
exchangeable hydrogenic sites in the TS complex and reactant
(
9) (a) Gold, V.; Grist, S. J. Chem. Soc., Perkin Trans. 2 1972, 89. (b) Bell,
R. P. The Proton in Chemistry, 2nd ed.; Chapman & Hall: London, 1973;
p 248.
(
8) Cited in: Laughton, P. M.; Robertson, R. E. In Solute-SolVent Interactions;
Coetzee, J. F., Ritchie, C. D., Eds.; Dekker: New York, 1969; p 490.
2260 J. AM. CHEM. SOC.
9
VOL. 124, NO. 10, 2002

Saponification of Ethyl Acetate
A R T I C L E S
Figure 1. Proton inventory of the saponification of ethyl acetate at 30.0
Figure 2. TSC (n) data fitted to (1 - n + 1.2n)(1 - n + 0.48n).
°
C. Data points are fitted to eq 8.
available for the saponification of methyl formate in the gas
phase suggest that it proceeds with little or no activation
energy. Quantum mechanical calculations of Ornstein et al.
have yielded results that indicate the breakdown of the
tetrahedral intermediate as rate determining in the gas-phase
alkaline hydrolysis of methyl, ethyl, and isopropyl acetates. This
theoretical result agrees with those of Riveros and contrasts with
The analysis of variance allows us to make the final choice.
Two analyses of variance were carried out. Since the precision
of the k2(n) values was fair but far from optimal, the analyses
were run separately with the six TSC(n) values (from the
nominal six rate values) and with the 18 data pairs calculated
from the six k2(n)’s plus the 12 values corresponding to k2(n)
11a
12
(
one standard error.
The first calculation rendered the quadratic term to be
experience in solution.
11b
Fink and Hadad
studied the gas-phase reaction of the
statistically significant at the 99% certainty level. This result
was doubtlessly due to a fortunate cancellation of errors, but
speaks in favor of the validity of the quadratic term in TSC(n).
The second calculation took into account the actual quality of
the data. In this case, for the quadratic term, Fisher F ) 4.17,
p ) 0.06.10
simple esters HCO2Me, HCO2Et, CH3CO2Me, and CH3CO2Et
-
with different nucleophiles. The authors found the hard NH2
-
to react faster than the soft N≡CCH2 anion via the BAC2
mechanism, followed by a proton transfer within the ion-
molecule complex, to yield the corresponding alcohols and
-
-
RCONH and RC(O )dCH2CN, respectively. However, the
The best quadratic fit to the data resulted as
extent of BAC2 reaction among the possible reaction channels
(
proton-transfer, SN2, E2, and BAC2) was larger for the bulkier
2
TSC(n) ) 1 - 0.32n - 0.093n
(7)
nucleophile. Carbonyl groups are considered hard centers,
1
3
responsive, to the basicity of the nucleophile. The observed
kinetic trends suggest that formation of the nucleophile-
carbonyl bond is not rate determining in the gas phase, a result
Thus, the relationship between k2(n) and n for the saponifica-
tion of ethyl acetate at 30.0 °C is given by
-
1
-1
2
11a
in agreement with the early proposal of Riveros.
k2(n) ) (0.122 s M )(1 - 0.32n - 0.093n )/(1 - n +
Solvent plays a fundamental role in the saponification of
carboxylic esters, much beyond the na ¨ı ve consideration as only
a material medium, that has an effect on both the frequency of
molecular collisions and long-range chemical interactions with
reactants and TS complexes. This assertion is based on experi-
mental results that point to the formation of the tetrahedral in-
3
1
.4n)(1 - n + 0.68n) (8)
Mechanism for the Saponification of Carboxylic Esters.
The basic hydrolysis of esters has been extensively studied in
the liquid phase as well as experimentally and theoretically
in the gas phase.
Riveros and Takashima observed that the methyl esters of
formic, pivalic, and benzoic acids react with OH- in the gas
phase via the BAC2 hydrolytic pathway and SN2 at the methyl
group. The BAC2/SN2 product ratios were found to be 2.7, 9.0,
and 12, respectively.1 This result indicates that, unlike in the
solution process, there is negligible steric effect for the BAC2
process in the gas phase. This observation led Takashima and
Riveros to state that the salient dynamic feature of the BAC2
mechanism in the gas phase was “the ease with which the
tetrahedral intermediate evolves into the products”. The data
1
11
12
1
termediate as rate-determining step in solution and the existence
of steric hindrance to its formation. An appropriate mechanistic
picture for this system must take into account the solvent
molecules that should be included in the minimal TS structure.
1a
(
10) Spiridonov, V. P.; Lopatkin, A. A. Tratamiento Matem a´ tico de Datos F ´ı sico-
Qu ´ı micos, Spanish ed., Mir: Moscow, 1973; pp 42-43.
(11) (a) Takashima, K.; Riveros, J. M. J. Am. Chem. Soc. 1978, 100, 6128.(b)
Frink, B. T.; Hadad, C. M. J. Chem. Soc., Perkin Trans. 2 1999, 2397.
(12) Zhan, C.-G.; Landry, D. W.; Ornstein, R. L. J. Am. Chem. Soc. 2000, 122,
1522.
(13) Fleming, I. Frontier Orbitals and Organic Reactions; Wiley: Chichester,
U.K., 1990; p 47.
J. AM. CHEM. SOC.
9
VOL. 124, NO. 10, 2002 2261

A R T I C L E S
Mata-Segreda
The TSC(n) data were fitted to eq 9 with the computer
program package ENZFITTER.14 The best fractionation factors
for the quadratic polynomial were
0.5. The proton attached to the nucleophile oxygen atom
resembles those in gem-diols and hemiketals, for which a φ
1
5
16
value of 1.2 was reported, though Wolfenden showed this
value to have been overestimated by about 10%. Therefore,
its magnitude can be taken as close to 1.0. The third proton in
the “HO” configuration must have a fractionation factor be-
tween 1.4 (as in hydroxide) and the value 1.0, as in a fully
formed water molecule. Formyl hydrogen kinetic isotope effect
for the alkaline hydrolysis of methyl formate allowed Kirsch et
TSC(n) ) (1 - n + φ n)(1 - n + φ n)
(9)
1
2
φ ) (1.20 ( 0.01)φ ) 0.482 ( 0.006
1
2
Figure 2 shows the result. These numerical values can be
translated into a TS structure.
A TS complex resulting from the direct nucleophilic collision
could be depicted as follows:
1
7
al. to estimate that the TS is located at about 40% along the
reaction coordinate toward the tetrahedral intermediate. This
means the hydroxide character has been reduced to about 60%
0
.6
in the TS. Thus, φ ) 1.4 ) 1.2 in fine agreement with
experiment.
The results of this work support the predictions derived from
the mechanism of Marlier for the alkaline hydrolysis of
carboxylic esters.
Such a structure requires a linear TSC(n) function because
of the involvement of only one hydrogen, a situation in
disagreement with experiment.
The mechanism of Marlier implies the following minimal TS
structure:
Acknowledgment. I am grateful to Professor John F. Marlier
for his interest in this work.
JA011931T
(
(
(
14) Leatherbarrow, R. J. ENZFITTER. A Non-Linear Regression Data Analysis
Program; Elsevier-Biosoft: Cambridge, U.K., 1987.
15) Mata-Segreda, J. F.; Wint, S.; Schowen, R. L. J. Am. Chem. Soc. 1974,
96, 5608.
16) Bone, R.; Wolfenden, R., J. Am. Chem. Soc. 1985, 107, 4772. See also:
a) Gruen, L. C.; McTigue, P. T., J. Chem. Soc. 1963, 5217. (b) Lienhard,
(
G. E.; Jencks, W. P. J. Am. Chem. Soc. 1966, 88, 3892. (c) Kurz, J. L. J.
Am. Chem. Soc. 1967, 89, 3524. (d) Burkey, T. J.; Fahey, R. C. J. Am.
Chem. Soc. 1983, 105, 868. The φ value for gem-diols can be taken as
The values of φ can be predicted from independent mecha-
nistic and structural studies. The type of in-flight bridging
protons is known to have a fractionation factor of the order of
1.08 ( 0.01.
(
17) Bilkadi, Z.; de Lorimier, R.; Kirsch, J. F. J. Am. Chem. Soc. 1975, 97,
4317.
2262 J. AM. CHEM. SOC.
9
VOL. 124, NO. 10, 2002