Kinetics of the pH-independent hydrolyses of 4-nitrophenyl chloroformate and 4-nitrophenyl heptafluorobutyrate in water-acetonitrile mixtures: Consequences of solvent composition and ester hydrophobicity

Article abstract of DOI:10.1002/poc.1081

The pH-independent hydrolyses of 4-nitrophenyl chloroformate, NPCF and 4-nitrophenyl heptafluorobutyrate, NPFB in aqueous acetonitrile were studied spectrophotometrically from 15 to 45 °C. The binary solvent composition covers water concentrations from 2.

Full text of DOI:10.1002/poc.1081

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2006; 19: 793–802
Published online 23 November 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.1081
Kinetics of the pH-independent hydrolyses of
4-nitrophenyl chloroformate and 4-nitrophenyl
heptafluorobutyrate in water-acetonitrile
mixtures: consequences of solvent composition
and ester hydrophobicity
*
´
Omar A. El Seoud and Fabio Siviero
´
Instituto de Quımica, Universidade de Sa˜o Paulo, C.P. 26077, 05513-970, Sa˜o Paulo, S.P., Brazil
Received 17 January 2006; revised 27 February 2006; accepted 10 March 2006
ABSTRACT: The pH-independent hydrolyses of 4-nitrophenyl chloroformate, NPCF and 4-nitrophenyl heptafluor-
obutyrate, NPFB in aqueous acetonitrile were studied spectrophotometrically from 15 to 45 8C. The binary solvent
composition covers water concentrations from 2.349 to 53.207 and from 2.745 to 53.333 mol Lꢀ1 for NPCF and
NPFB, respectively. For both esters, the dependence of log (kobs), the observed rate constant, on log [water] is
sigmoidal. The approximate kinetic orders with respect to water were found to be 2 and 3 for NPCF and NPFB,
¼
respectively. DG gradually decreases as a function of increasing [water], due to a complex, quasi-mirror image
¼
¼
compensation of DH and DS ; both parameters increase. The structures of the transition states were probed by a
proton inventory study, carried out in the presence of L2O mole fractions (L ¼ H or D) of 0.190, 0.540, 0.890 and
0.180, 0.529, 0.890, for NPCF and NPFB, respectively. Plots of observed rate constants versus the atom fraction of
deuterium in the solvent curve downward. Cyclic transition state models were fitted to the kinetic data; these models
contain the ester and two water molecules (NPCF) or three water molecules (NPFB). Thus, the sigmoidal dependences
of log (kobs) on log [water] are not due to a change in the number of water molecules in the transition states as a function
of increasing [water]. The binary solvent mixture is micro-heterogeneous; there exists two ‘‘micro-domains,’’ one
consists predominantly of coordinated water molecules, the other consists mostly of acetonitrile hydrogen-bonded to
water molecules. NPCF is 232 times more soluble in water than NPFB. That is, the former ester is dissolved in the
outer, more polar periphery of these micro-domains whereas the more hydrophobic NPFB is dissolved in their inner,
less polar interiors. This conclusion is corroborated by comparing the dependence on log [water] of log [kobs], and of
ET, the empirical solvent polarity parameter, as measured by solvatochromic probes of increasing hydrophobicity.
Copyright # 2006 John Wiley & Sons, Ltd.
KEYWORDS: 4-nitrophenyl chloroformate; hydrolysis of 4-nitrophenyl heptafluorobutyrate; hydrolysis of proton
inventory; solvent effect; solvatochromism
INTRODUCTION
binary mixture affect the solvation of the species of
interest, for example, reactants and activated complexes,
hence reaction rates and equilibria. In this regard, studies
of solvatochromism (effects of the medium on the spectra,
absorption or emission, of solvatochromic probes) and
thermo-solvatochromism (effects of temperature on
solvatochromism) have contributed a great deal to our
understanding of the above-mentioned interactions.5,6
The reason is that solvatochromic probes may be
employed as models for the species of interest, for
example, polar reagents and activated complexes. These
studies have indicated that the hydrophobic/hydrophilic
character and pKa of both solvatochromic probe and
components of the solvent play a central role in
determining the composition of the solvation micro-
sphere of the probe. This composition is not related in a
Binary mixtures of water and organic solvents are
extensively employed in organic syntheses and in studies
of reaction mechanisms of, for example, acyl transfer
reactions. The reasons are partially practical and are often
connected with solubility constraints. On the other hand,
there is an increased interest in understanding the effects
of reaction medium, of both pure solvents and binary
mixtures on rate and equilibrium constants of chemical
and biochemical reactions.1–4 The physico-chemical
properties of both substrate and components of the
´
*Correspondence to: O. A. El Seoud, Instituto de Quımica,
Universidade de Sa˜o Paulo, C.P. 26077, 05513-970, Sa˜o Paulo, S.P.,
Brazil.
E-mail: elseoud@iq.usp.br
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

794
O. A. EL SEOUD AND F. SIVIERO
simple way to bulk solvent composition because most
mixtures of water with organic solvents are micro-
heterogeneous.6
or in DCl/D2O, hence the deuterium contents of these
media, were determined by 1H-NMR (Varian Innova-300
NMR spectrometer) by using 1,4-dioxane as an internal
reference.8
As a part of our interest in studying medium effects on
reactivity we report here on the pH-independent
hydrolyses of two reactive esters, 4-nitrophenyl chlor-
oformate (NPCF) and 4-nitrophenyl heptafluorobutyrate
(NPFB) in binary mixtures of water (W) and acetonitrile
(MeCN) over almost the entire range of water concen-
tration, from 2.349 to 53.333 mol Lꢀ1. Effects of medium
composition on the reactions were deduced from kinetic
orders with respect to water; kinetic solvent isotope
Kinetic runs (at least triplicate) were carried out with
the aid of Beckman DU-70, or Applied Photophysics SX-
18 MV stopped-flow Uv–Vis spectrometers; both are
interfaced to PC, and equipped with thermostated cell
holders, whose temperatures were controlled within
ꢁ0.05 8C. Preliminary experiments in W-MeCN mixtures
([W] ¼ 53.333 mol Lꢀ1) have shown that the observed
rate constant, kobs, was independent of ester concentration
in the range (2–8 ꢂ 10ꢀ5 mol Lꢀ1). In subsequent runs the
final [ester] was ꢃ5 ꢂ 10ꢀ5 mol Lꢀ1. The reaction
progress was followed by monitoring the liberation of
4-nitrophenol at 320 nm, as a function of time. Values of
kobs were calculated from log (absorbance) versus time
plots; these were rigorously linear over more than five
half-lives. The agreement between calculated and
experimental ‘‘infinity’’ absorbances was routinely
checked. The relative standard deviation in kobs, that is,
((standard deviation/kobs) 100), was ꢄ0.2, that between
kobs of triplicate runs was ꢄ1.0%.
effects, KSICs, that is, ðkobs
Þ
H O MeCN=ðkobsÞD O MeCN
;
ꢀ
ꢀ
2
the dependence of the activa2tion parameters on log
[water], and by applying the proton inventory technique
to probe the structures of the transition states. Information
on the relationship between ester hydrophobicity and the
polarity of its reaction site was secured by comparing the
dependence on log [water] of log (kobs) and of ET, the
empirical solvent polarity parameter; the latter was
measured by two solvatochromic indicators of different
hydrophobicity.
EXPERIMENTAL
Materials
Spectrophotometric determination of the
empirical solvent polarity parameter, ET
The reagents were purchased from Acros or Merck. The
solvatochromic probes 2,6-dichloro-4-(2,4,6-triphenyl
pyridinium-1-yl) phenolate (WB); 1-methylquinolinium-
8-olate, QB (structures depicted in Results and Discussion)
were available from previous studies.6 MeCN was further
purified by distillation from P4O10, then from anhydrous
K2CO3.7 D2O was distilled under nitrogen. All-glass
double distilled water was employed throughout.
This was carried out as described in detail elsewhere,6 by
using the Beckman DU-70 spectrometer. The final
solvatochromic
probe
concentration
was
2–
5 ꢂ 10ꢀ4 mol dmꢀ3. Values of lmax (calculated from
the first derivative of the absorption spectrum) were
determined at 25 8C as a function of solvent composition.
ET was calculated from:5,6
NPCF was purified by sublimation under reduced
pressure. NPFB was synthesized by refluxing a mixture of
10 mmol distilled heptaflourobutyryl chloride and
12 mmol of sodium 4-nitrophenoxide in 50 mL of
anhydrous acetonitrile for 6 h. The solvent was
evaporated, the residue was suspended in 25 mL dry
chloroform, the suspension filtered, and the ester purified
by flash column chromatography, using chloroform as
eluent. Yellow oil; yield 78%. Calculated for
C10H4F7NO4, %: C, 35.84; H, 1.20, N, 4.18. Analyzed:
ETprobeðkcal=molÞ ¼ 28 591:5=lmaxðnmÞ:
RESULTS AND DISCUSSION
In W-MeCN mixtures, at [W] ¼ 11.076, 33.440 and
53.207 mol Lꢀ1, values of kobs for both esters were found
to independent of [HCl] in the range 0.001–0.1 mol Lꢀ1
This shows that both reactions are water-catalyzed
hydrolyses. Simple kinetic pathways for the reactions
studied are depicted in Scheme 1.
Hydrolysis of each ester gave one equivalent of 4-
nitrophenol and the spectra of both reactions (lvs. time)
showed sharp isosbestic points at 285 nm. Additionally,
rigorous pseudo-first order kinetics were always
observed; identical kobs were obtained in the presence,
or absence of 5 ꢂ 10ꢀ5 mol Lꢀ1 4-nitrophenol, added to
the solution before the start of the kinetic run. That is, the
intermediates of the first step of both reactions of Scheme
1 do not accumulate; this (first) step is rate limiting, in
.
C, 35.63; H, 1.26; N, 4.25. IR (Bruker Victor-22 FTIR,
0.025 mol Lꢀ1 solution in CD3CN) 1802 cmꢀ1 (nC—
);
—
O
1521 and 1342 cmꢀ1 (asymmetric and symmetric yNO2 ).
Kinetic measurements
All binary solvent mixtures were prepared at 25 8C by
volume and by weight. In order to suppress any base-
catalyzed reaction, the mixtures contained 0.01 mol Lꢀ1
HCl or DCl. The (residual) protium concentration in D2O,
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

HYDROLYSES OF ACTIVATED ESTERS IN WATER-ACETONITRILE MIXTURES
795
O
C
OH
k1
k2
C
Cl
O
NO2 + H2O
Cl
HCl + CO
NO2
NO2
2 + HO
O
NO2
(1)
(2)
k-1
OH
O
OH
k1
C
k2
C
O
NO2 + H2O
C3F7
C3F7CO2H +HO
NO2
O
C3F7
k-1
OH
Scheme 1. Pathways for the pH-independent hydrolyses of NPCF and NPFB
agreement with the fact that the 4-nitrophenoxide ion is a
better leaving group than the hydroxide ion. Accordingly,
the reaction rate is given by:
respectively; the corresponding activation parameters are
listed in Tables 3 and 4. Figures 1 and 2 show plots of log
kobs versus log [W] for NPCF and NPFB, respectively; the
corresponding dependence of the activation parameters on
log [W] is shown in Figs. 3 and 4, respectively. All these
results show the following relevant points:
x
Rate ¼ k1½esterꢅ½Wꢅ ¼ kobs½esterꢅ; since½Wꢅ
ꢆ ½esterꢅ
(3)
i. In all solvent mixtures, NPFB is much more reactive
than NPCF, in agreement with the stronger (-I) effect
of the n-C3F7- group relative to the Cl atom;9
ii. As shown by Eqn (4), (x) may be calculated from plot
of log (kobs) versus log [W]. For NPCF these plots are
x
kobs ¼ k1½Wꢅ or log kobs ¼ log k1 þ x log½Wꢅ (4)
Where (x) is the kinetic order with respect to water.
Tables 1 and 2 show the dependence of kobs on solvent
composition at different temperatures, for NPCF and NPFB,
Table 1. Hydrolysis of 4-nitrophenyl chloroformate in water (W)-acetonitrile (MeCN) mixtures: observed rate constants (kobs
,
a
secꢀ1) as a function of temperature and medium composition, listed as water molar concentration, [W], and mole fraction, xW
kobs ꢂ 102,
15 8C
kobs ꢂ 102,
25 8C
kobs ꢂ 102,
35 8C
kobs ꢂ 102,
[W], mol Lꢀ1
xW
45 8C
2.350
4.447
7.489
0.115
0.202
0.313
0.418
0.504
0.614
0.717
0.807
0.903
0.948
0.986
0.019
0.071
0.165
0.261
0.342
0.487
0.733
1.163
1.983
2.637
3.201
0.033
0.126
0.319
0.543
0.736
1.078
1.646
2.490
4.638
6.314
7.247
0.054
0.221
0.579
1.024
1.421
2.242
3.485
5.450
10.098
13.280
16.520
0.086
0.356
0.966
1.733
2.537
4.043
6.712
10.910
20.096
27.520
33.358
11.076
14.427
20.006
26.430
33.440
42.848
48.409
53.207
a The rate constants reported are the mean of at least triplicate runs. The relative standard deviation between kobs of a triplicate, that is, ((standard deviation/
kobs)100) was ꢄ1.0%.
Table 2. Hydrolysis of 4-nitrophenyl heptafluorobutyrate in W-MeCN mixtures: Observed rate constants (kobs, secꢀ1) as a
function of temperature and medium compositiona
kobs ꢂ 102,
kobs ꢂ 102,
kobs ꢂ 102,
kobs ꢂ 102,
[W], mol Lꢀ1
xW
secꢀ1, 15 8C
secꢀ1, 25 8C
secꢀ1, 35 8C
secꢀ1, 45 8C
2.745
6.230
9.496
13.294
17.163
25.914
33.223
43.091
48.394
53.333
0.130
0.266
0.371
0.472
0.557
0.710
0.806
0.904
0.946
0.986
0.011
0.080
0.174
0.295
0.417
0.847
1.894
7.531
16.40
30.00
0.013
0.107
0.250
0.435
0.644
1.447
3.077
11.745
24.50
36.450
0.016
0.137
0.338
0.623
0.977
2.268
4.832
17.810
35.330
63.610
0.018
0.162
0.442
0.841
1.404
3.335
7.351
25.290
49.570
94.320
a The rate constants reported are the mean of at least triplicate runs. The relative standard deviation between kobs of a triplicate, that is, ((standard deviation/
kobs)100) was ꢄ1.0%.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

796
O. A. EL SEOUD AND F. SIVIERO
Table 3. Activation parameters for the hydrolysis of 4-nitrophenyl chloroformate in W-MeCN mixtures as a function of medium
compositiona
¼
¼
¼
¼
[W], mol Lꢀ1
DH , kcal molꢀ1
DS , cal Kꢀ1 molꢀ1
ꢀTDS , kcal molꢀ1, 25 8C
DG , kcal molꢀ1
2.350
4.447
7.489
8.5
9.3
ꢀ46.0
ꢀ40.7
ꢀ35.9
ꢀ32.3
ꢀ29.6
ꢀ26.2
ꢀ23.5
ꢀ20.6
ꢀ19.4
ꢀ18.5
ꢀ17.7
13.7
12.1
10.7
9.6
8.8
7.8
7.0
6.7
5.8
5.5
22.2
21.4
20.9
20.6
20.4
20.1
19.9
19.5
19.3
19.1
19.0
10.2
10.9
11.6
12.3
12.9
13.3
13.5
13.6
13.7
11.076
14.427
20.006
26.430
33.440
42.848
48.409
53.207
5.3
a The errors are ꢁ 0.1 kcal molꢀ1 (DH and DG ) and 0.5 cal Kꢀ1 molꢀ1 (DS ).
¼
¼
¼
Table 4. Activation parameters for the hydrolysis of 4-nitrophenyl heptafluorobutyrate in W-MeCN mixtures as a function of
medium compositiona
¼
¼
¼
¼
[W], mol Lꢀ1
DH , kcal molꢀ1
DS , cal Kꢀ1 molꢀ1
ꢀTDS , kcal molꢀ1, 25 8C
DG , kcal molꢀ1
2.745
6.230
9.496
13.294
17.163
25.914
33.223
43.091
48.394
53.333
2.4
3.7
5.1
5.8
6.8
7.7
7.6
6.8
6.1
6.3
ꢀ63.6
ꢀ55.1
ꢀ48.9
ꢀ45.3
ꢀ41.2
ꢀ36.5
ꢀ35.3
ꢀ35.4
ꢀ36.2
ꢀ34.6
19.0
16.4
14.6
13.5
12.3
10.9
10.5
10.6
10.8
10.3
21.4
20.2
19.6
19.3
19.1
18.6
18.2
17.4
16.9
16.6
a The errors are ꢁ0.1 kcal molꢀ1 (DH , and DG ) and 0.5 cal Kꢀ1 molꢀ1 (DS ).
¼
¼
¼
not strictly linear; a third order polynomial may be
conveniently fitted to the data (r2 > 0.999). If linear
regression is employed instead, (x) was found to
increase from 1.5 to 1.8 as T is increased from 15
to 45 8C. A fourth order polynomial is required to fit
the data of NPFB (r2 > 0.999); for the same T range,
(x) obtained by linear regression of the data increases
from 2.5 to 2.7. These linear regressions, although
only approximate, indicate that the transition state,
TS, for the hydrolysis of NPFB contains more water
molecules than its NPCF counterpart. Note that (x) for
the hydrolysis (in W-MeCN) and methanolysis (in
1.5
45ºC
35ºC
45ºC
35ºC
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
-4.0
1.0
25ºC
15ºC
0.5
0.0
25ºC
15ºC
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
log [Water]
Log [Water]
Figure 1. Plots of log (kobs) versus log [water] for the
hydrolysis of 4-nitrophenyl chloroformate in W-MeCN at
different temperatures
Figure 2. Plots of log (kobs) versus log [water] for the
hydrolysis of 4-nitrophenyl heptafluorobutyrate in W-MeCN
at different temperatures
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

HYDROLYSES OF ACTIVATED ESTERS IN WATER-ACETONITRILE MIXTURES
797
¼
¼
24
22
20
18
16
14
12
10
8
trations DG is dominated by variation in the DH
term, Tables 3 and 4. The relative importance of these
contributions is apparent from Figs. 3 and 4. For
≠
≠
∆
G
H
¼
NPCF, DG decreases monotonically as a function
of increasing log [W], due to an increase of DH ,
¼
accompanied by quasi mirror-image increase of the
¼
∆
corresponding DS . This enthalpy–entropy compen-
sation has been observed for several water-catalyzed
reactions; among other factors, they reflect variations
in the interactions of the components of the binary
solvent mixture with the species of interest, reagent
≠
-T ∆S
6
state, RS, and/or TS.2,11 For NPFB, DG also
¼
4
decreases monotonically as a function of increasing
log W], due to partial enthalpy–entropy compen-
sation. The energy difference between the DH
and TDS curves in the water poor-region is, however,
large so that the two curves do not cross (as in case of
NPCF). This results in the above-mentioned entropy
dominance of the hydrolysis of NPFB;
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
Log [Water]
¼
¼
Figure 3. Dependence of the activation parameters for the
hydrolysis of 4-nitrophenyl chloroformate on log [W]. The
TDS term was calculated at 25 8C
¼
iv. In principle, the above-mentioned non-linear depen-
dence of log (kobs) on log [W], Figs. 1 and 2, may be
taken to indicate that the number of water molecules
in the TS depends on [W]. Examples of such depen-
dence are known, for example, (x) for the spontaneous
hydration of chloral varies from ca. 4 to ca. 2 when
MeOH–MeCN) of (structurally similar) 4-nitrophe-
nyl trifluoroacetate are 3.2 and 2.9, respectively. The
plots that generated these (x) were linear because of
the very limited protic solvent concentration range
employed, 1–5 mol Lꢀ1 10
;
iii. The above-mentioned difference in (x) also agrees
[W] is increased from 0.25 to 46.3 mol L ;
ꢀ1 12a the TS
¼
with the large differences between TDS for both
for the hydration of 1,3-dichloracetone contains one,
two, or three molecules of water, depending on
whether the reaction is carried out in the presence
of surfactant (inverse) aggregates in n-hexane, or in
aqueous dioxane, respectively.12b–d Finally, the num-
ber of water molecules in the TS of the imidazole-
catalyzed hydrolysis of N-4-nitrophenyltrifluoroace-
tamide in W-MeCN was reported to increase from
one to four as a function of increasing [W];12e
esters. Over the whole solvent composition range, the
hydrolysis of NPFB is associated with much more
negative entropy of activation than for NPCF, pre-
sumably because of the larger number of water
¼
molecules involved in its TS. In fact, DG for NPFB
is dominated by the TDS term, over the entire water
¼
concentration range. This contrasts with the hydroly-
sis of NPCF, where the above-mentioned entropy
dominance is observed only for the water-poor
regions, [W] ꢄ 5 mol Lꢀ1; at higher water concen-
v. A handle on the structures of the TSs may be obtained
by submitting the kinetic data to analysis by the
proton inventory technique, PI. The use of PI to probe
the structures of TSs is documented,3,4,8,13 so that
only a brief account will be given here. The observed
KSIE is employed to estimate the number of hydro-
gens (hence water molecules in the present reactions)
that participate in the TS. This number and other
chemical information, for example, activation
parameters, enable one to suggest a structure for
the TS. The observed rate constant, kn, in the presence
of H2O–D2O mixtures is related to the observed rate
constant in pure H2O, ko, by the equation:
22
20
≠
∆
G
18
16
14
12
10
8
≠
≠
-T
∆
S
H
∆
6
TS
Q
¼
4
ð1 ꢀ n þ n’i Þ
i
2
kn=ko ¼
(5)
RS
Q
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
ð1 ꢀ n þ n’jÞ
j
Log [Water]
Where (n) is the atom fraction of deuterium in the solvent.
All TS hydrogens (i) and all RS hydrogens (j) that
contribute to the isotope effect possess a term in the
Figure 4. Dependence of the activation parameters for the
hydrolysis of 4-nitrophenyl heptafluorobutyrate on log [W].
The TDS term was calculated at 25 8C
¼
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

798
O. A. EL SEOUD AND F. SIVIERO
Table 5. Dependence of the observed rate constants (kn) for the hydrolysis of 4-nitrophenyl chloroformate on the atom fraction
of deuterium (n) in mixtures of MeCN–H2O–D2O at different L2O mole fractions, x
at 25 8Ca
L2O
xL O ¼ 0:190
xL O ¼ 0:540
xL O ¼ 0:890
2
2
2
Expl.
Calc.
Expl.
Calc.
Expl.
Calc.
kn ꢂ 102
kn ꢂ 102
kn ꢂ 102
kn ꢂ 102
kn ꢂ 102
kn ꢂ 102
(n)
(secꢀ1
)
(secꢀ1
)
(n)
(secꢀ1
)
(secꢀ1
)
xD
(secꢀ1
)
(secꢀ1
)
0.0b
0.116
0.101
0.087
0.075
0.062
0.116
0.101
0.087
0.074
0.062
0.0b
0.886
0.767
0.665
0.561
0.480
0.886
0.771
0.666
0.567
0.473
0.0b
4.571
4.050
3.550
3.134
2.731
4.571
4.049
3.568
3.131
2.723
0.238
0.487
0.724
0.969
0.241
0.478
0.720
0.969
0.250
0.495
0.732
0.969
KSIE
wa
1.89
0.726 ꢁ 0.001
1.89
0.722 ꢁ 0.003
1.71
0.764 ꢁ 0.001
a Expl. ¼ Experimental kn; Calc. ¼ Calculated kn by using Eq. (8); kn/ko ¼ (1 ꢀ n þ nwa)2, where wa is the fractionation factor of protons Ha of TS2 of Fig. 5;
KSIE ¼ kinetic solvent isotope effect, that is, ðkobs
Þ
=ðkobs
Þ
MeCN; L ¼ H or D.
H2OꢀMeCN
D2Oꢀ
b
Calculated rate constant for the reaction in pure H2O.
numerator and denominator, respectively, of right hand
side of Eqn (5). Each hydrogen site is associated with a
fractionation factor, defined by Eqn (6):
Tables 5 and 6 show the dependence of kn on (n) at three
different mole fractions of L2O, xL O, where L¼ H or D;
2
these cover the three regions of the plots of Figs. 1 and 2.
The procedure usually employed is to draw TS structures
that are chemically sound, and to test how these structures
agree with experimental KSIE. Our objective is to determine
½D=Hꢅsite
’
¼
(6)
site
½D=Hꢅsolvent
the dependence, if any, of the structure of the TS on xL O
.
2
That is, wsite expresses the preference of deuterium for
the site relative to that of bulk solvent. Analysis of the
present reaction is simplified by the following: under
reaction conditions, MeCN does not undergo H/D
exchange with D2O (conclusion based on integration of
the 1H-NMR peak of MeCN as a function of time); NPCF
and NPFB carry no exchangeable hydrogens; w of the
other reactant, water, is unity, by definition. Con-
sequently, the denominator of Eqn (5) is unity, and we
need consider only the hydrogens of the TS.
Based on the above-mentioned values of (x), any TS
containing one water molecule may be ruled out.
Therefore, we consider transition states that contain
two, or three water molecules, as depicted in Fig. 5 (for
sake of simplicity, we draw (Ha) symmetrically between
the oxygen atoms).
In TS1, the proton bridge, Ha, contributes a primary
isotope effect. Hb which is converted from an OH of water
to an OH of the tetrahedral intermediate does not
contribute to the isotope effect, i.e., wb ¼ 1. Hc of the
Table 6. Dependence of the observed rate constants (kn) for the hydrolysis of 4-nitrophenyl heptafluorobutyrate on the atom
fraction of deuterium (n) in mixtures of MeCN–H2O–D2O at different L2O mole fractions, x
at 25 8Ca
L2O
xL O ¼ 0:180
xL O ¼ 0:529
xL O ¼ 0:890
2
2
2
Expl.
kn ꢂ 10,
Calc.
kn ꢂ 10
Expl.
kn ꢂ 10
Calc.
kn ꢂ 10
Expl.
kn ꢂ 10
Calc.
kn ꢂ 10
(n)
(secꢀ1
)
(secꢀ1
)
(n)
(secꢀ1
)
(secꢀ1
)
(n)
(secꢀ1
)
(secꢀ1
)
0.0b
0.036
0.036
0.028
0.022
0.016
0.012
0.0b
0.572
0.446
0.342
0.251
0.179
0.572
0.445
0.341
0.252
0.179
0.0b
10.090
7.965
6.155
4.758
3.582
10.090
7.986
6.178
4.760
3.554
0.238
0.487
0.724
0.969
KSIE
wa
0.029
0.022
0.016
0.012
3.22
0.241
0.478
0.720
0.969
3.34
0.247
0.497
0.731
0.969
2.92
0.679 ꢁ 0.001
0.668 ꢁ0.001
0.697 ꢁ 0.002
a Expl. ¼ Experimental kn; Calc. ¼ Calculated kn by using Eq. (9); kn/ko ¼ (1 ꢀ n þ nwa)3, where wa is the fractionation factor of protons Ha of TS3 or TS4 of
Fig. 5; KSIE ¼ kinetic solvent isotope effect, that is, ðkobs
Þ
=ðkobs
Þ
MeCN; L ¼ H or D.
H2OꢀMeCN
D2Oꢀ
b Calculated rate constant for the reaction in pure H2O.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

HYDROLYSES OF ACTIVATED ESTERS IN WATER-ACETONITRILE MIXTURES
799
Ha
similar, ꢃ0.81; this is unreasonable for a primary and a
O
C
δ−
O
Hb
secondary KSIE. For xL O ¼ 0:190 and 0:890; wa ¼
O
C
2
0.989 and 0.928 and wc ¼ 0.720 and 0.797, respectively.
Again, wa is much larger than expected for this proton (ca.
0.5), whereas wc is less than expected, at least for
Ha
O
Cl
Hb
NO2
Cl
O
O
O
Ha
Hb
δ+
xL O ¼ 0:190. Provided that the second water molecule
Hc
2
O
acts as a ‘‘general base,’’ that is, its rate constant lies on
the line of the Brønsted equation plot, the fractionation
factor for Hc of TS1 may be calculated from: wc ¼ 0.69b,8
and should be ca. 0.8. The latter calculation is based on
b ¼ 0.59, calculated for general base-catalyzed hydrolysis
of 4-nitrophenyl trifluoroacetate in 0.56 mol Lꢀ1 water in
acetonitrile.14 The fact that the fractionation factors
Hc
NO2
TS1
TS2
Hb
Hb
O
Ha
Ha
O
O
Ha
O
C3F7
HO
C3F7
Ha
O
C
Ha
C
calculated by Eqn (7) depend on xL O mean that the
degree of proton transfer in the TS depends (somewhat
erratically) on xL O, there is no chemical ground for this
2
O
O
Ha
O
O
Hb
Hb
O2N
Hb
2
dependence, that is, TS1 may be safely rejected.
The cyclic TS2 also involves the ester plus two water
molecules, protons Ha are transferred in a six-membered
ring. The corresponding dependence of kn/ko on (n) is
given by:
NO2
TS3
TS4
Figure 5. Suggested transition state structures for the
hydrolysis of 4-nitrophenyl chloroformate, TS1 and TS2,
and of 4-nitrophenyl heptafluorobutyrate, TS3 and TS4
2
kn=ko ¼ ð1 ꢀ n þ n’aÞ
(8)
This equation reproduces the data satisfactorily, as can
be seen from the calculated kn of Table 5 and from Fig. 6.
Values of wa are practically independent of xL O, in the
‘‘general base’’ water molecule contributes a secondary
isotope effect. By the same line of reasoning, only Ha of
the other TSs contributes to the KSIE. According to TS1,
the dependence of kn/ko on (n) is given by:
2
range 0.190–0.540, and is only slightly higher (by 5%) in
the water-rich region. Consequently, the structure of the
TS appears to be the same, over the whole water
concentration range. Equation (9) (NPCF plus three water
molecules) also reproduces kn/ko, with wa ¼ 0.82 ꢁ 0.01,
although with larger errors in w, and larger differences
between experimental and calculated kn (rate constants
not listed). The value of (x), and the fact that
kn
2
¼ ð1 ꢀ n þ n’aÞð1 ꢀ n þ n’cÞ
(7)
ko
Although this equation reproduces the kinetic data
reasonably, the errors in the fractionation factors
calculated are much larger (between 5 and 10 times)
than wa and wc. Additionally, w calculated are different
¼
¼
(DS NPCF) ꢆ (DS NPFB) over the whole [W] range,
argues against TSs that contain the same number of
water molecules, namely three, for both esters. That is,
from what is expected. For xL O ¼ 0:540, wa and wc are
2
1.0
0.9
0.8
0.7
0.6
0.5
0.4
1.0
0.9
0.8
0.7
0.6
0.3
0.5
0.0 0.2 0.4 0.6 0.8 1.0
0.0 0.2 0.4 0.6 0.8 1.0
Atom Fraction of Deuterium (n)
Atom Fraction of Deuterium (n)
Figure 6. Typical results for the application of the proton inventory technique to the hydrolysis at xL O (L ¼ H or D) of 0.540 and
0.529, for 4-nitrophenyl chloroformate and 4-nitrophenyl heptafluorobutyrate, respectively. The p2oints are experimental and
the solid lines were calculated by Eqs. (8) and (9), respectively
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

800
O. A. EL SEOUD AND F. SIVIERO
assigning TS3 or TS4 to the hydrolysis of NPCF is not
warranted.
sets in, and there exists two ‘‘micro-domains,’’ one
highly structured consisting predominantly of coordi-
nated water molecules, and a relatively disordered one
containing mostly acetonitrile, hydrogen-bonded to
water. Addition of more MeCN leads to a change in the
relative concentrations of the two micro-domains but
their compositions do not vary appreciably. At xW ꢄ 0.3
the water clusters have become few and far apart that
new interactions set in. W-MeCN interactions become
important, this results in the formation of complexes,
for example, (MeCN)m-W where m ¼ 1–4. At still lower
xW values, the structure of acetonitrile dominates, and
species such as (MeCN)m are abundant. It should be
born in mind, however, that the onset of formation of the
different regions is not sharp, and is dependent on the
method used to study the solvent system.16–18 Using
electron impact mass spectroscopy, it has been shown
that hydrophobic substrates, partition in favor of
hydrophobic micro-domain of the binary mixture.18
Consequently, if NPCF and NPFB are present in the
same nano-droplet, then the former ester is expected to
be solubilized at its (more polar) periphery, whereas the
latter one is expected to be solubilized in its (less polar)
central part. A similar difference in solubilization sites
has been shown to occur in aqueous micelles. Thus
NPCF is solubilized at the (polar) micellar interface,
whereas a relatively hydrophobic ester, 4-nitrophenyl
2,2-dichloropropionate, is solubilized in the (less polar)
micellar interior.19 Another approach to corroborate the
presence of the two esters in micro-environments of
different polarities is to compare the response to solvent
composition of two independent processes: hydrolysis
of the esters and solvatochromism of two probes, 2,6-
dichloro-4-(2,4,6-triphenyl pyridinium-1-yl) phenolate
(WB); 1-methylquinolinium-8-olate, QB, see structures
in Fig. 7.
kn
3
¼ ð1 ꢀ n þ n’aÞ
(9)
ko
The same line of reasoning may be employed in order
to consider TS2 unlikely for the hydrolysis of NPFB.
Additionally, data treatment according to Eqn (8) resulted
in larger errors in both (wa) and kn (rate constants not
listed), compared with those calculated by Eqn (9), see
Fig. 6. Application of the latter equation to the kn/ko data
resulted in fractionation factors that are independent of
x
L O, similar to what has been obtained for NPCF. The
2
results may be also conveniently reproduced according to
TS4, an alternative to TS3. In this model, NPFB is rapidly
hydrated by a single water molecule, followed by a rate-
limiting decomposition of the hydrate involving two
additional water molecules. A point in favor of TS4 is that
the equilibrium constants for the hydration of trifluor-
omethyl derivatives of carbonyl compounds are >1, and
these reactions are exothermic.12 The exothermic pre-
equilibrium may explain the relatively small values of
¼
DH , Table 4, since the latter quantity is the sum of the
enthalpies of (exothermic) pre-equilibrium and
(endothermic) rate limiting step. It is worth noting that
the average wa for NPFB, 0.682 ꢁ 0.014, is similar to
those calculated for the pH-independent hydrolyses in
aqueous acetonitrile of 4-nitrophenyl trifluoroacetate
(0.697 ꢁ 0.005), and 4-methylphenyl trichloroacetate
(0.707 ꢁ 0.003) for which cyclic TSs, similar to TS3
and TS4 have been suggested.3i,10
vi. Finally, we address the relevance of ester hydropho-
bicity and the micro-heterogeneity of the binary mix-
ture to the kinetic data, in particular to log (kobs) versus
log [W] plots. A measure of the hydrophobic character
of the ester is its partition coefficient between n-
octanol and water, log P, defined as:15
Log P of these two probes were found to be 1.79 and
ꢀ0.63, for WB and QB, respectively,6c that is, the latter
probe is 263 times more soluble in water than the former
one. Figure 8 shows plots of the dependence of log (kobs
)
½substanceꢅnꢀoctanol
log P ¼
(10)
and ET(probe) on log [W], at 25 8C. For convenience,
we employed reduced log (kobs) and reduced ET(probe),
so that results of different substrates (esters and probes)
½substanceꢅwater
Although the partition coefficient experiment could
not be carried out because of the susceptibility of these
esters toward hydrolysis, values of log P may be
conveniently calculated from group contribution;15 they
were found to be 1.658 and 4.023 for NPCF and NPFB,
respectively. That is, the former ester is 232 times more
soluble in water than the latter one. Before discussing
the relevance of this large difference in hydrophobicity,
it is instructive to briefly discuss the microscopic
structure of W-MeCN mixtures as a function of their
composition. When the organic solvent is added to
water it replaces the uncoordinated water molecules.
The limit of xW beyond which MeCN cannot be
accommodated within the cavities of water is
xW ꢃ 0.85. Below this xW, solvent micro-heterogeneity
+
N
+
N
O
-
CH3
Cl
Cl
O -
QB
WB
Figure 7. Structures of solvatochromic probes employed
J. Phys. Org. Chem. 2006; 19: 793–802
Copyright # 2006 John Wiley & Sons, Ltd.

HYDROLYSES OF ACTIVATED ESTERS IN WATER-ACETONITRILE MIXTURES
801
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
NPCF
QB
NPFB
WB
0.0 0.3 0.6 0.9 1.2 1.5 1.8
Log [Water]
0.0 0.3 0.6 0.9 1.2 1.5 1.8
Log [Water]
Figure 8. Plots of the dependence of log (kobs) and of ET(probe) on log [W], at 25 8C. Reduced log (kobs) and reduced ET(probe)
are employed, so that results of different species (esters, probes) may be directly compared. Values of ET in pure water and pure
MeCN are, respectively, 64.62 and 52.97 kcal molꢀ1 (QB) and 70.24 and 54.72 kcal molꢀ1 (WB).
may be directly compared. The plots show a remarkable
resemblance, the more hydrophobic substrates (NPFB
and WB) deviate more from the ideal (i.e., linear)
behavior than the less hydrophobic ones (NPCF and
QB).
CONCLUSIONS
(i) The kinetic order with respect to one of the
components of a binary solvent mixture should be
based on rate constants obtained over a sufficiently wide
range of solvent composition. Care should be exercised
when kinetic orders with respect to water are interpreted
in terms of the number of water molecules in the TS.
The reason is that slopes of plots of log (k) versus log
[water] are most probably affected by the micro-
heterogeneity of the medium, and the concomitant
preferential solvation of RS and/or TS by one
component of the mixture; (ii) the complex (sigmoidal)
dependence of log (kobs) on log [W] is not due to a
change in the structure of the corresponding TS as a
function of increasing [W]; (iii) the TS of both esters
contain different numbers of water molecules, 2 and 3
for NPCF and NPFB, respectively; (iv) solvatochromic
probes may be fruitfully employed in order to mimic the
response of chemical reactions to variation of solvent
composition. Thus solvation of WB and QB and
hydrolysis of both esters seem to be sensitive to the
same solute-solvent interactions, namely, hydrogen
bonding and dipolar interactions, these are modulated
by the water-acetonitrile interactions.
The non-linear dependence of log (kobs) on log [W]
probably reflects the effects of preferential solvation of
RS and/or TS; a similar rational has been employed to
explain the non-linear dependence of ET on xW in several
binary mixtures.5,6 Because Gibbs free energies of
transfer of several non-electrolytes from water to
mixtures of water and aprotic solvents are relatively
insensitive to solvent composition,20 it is plausible that
kinetic solvent effects on the RS are not dominant and we
concentrate on the TS. Acetonitrile is known to solvate
positive centers stronger than negative ones. On the other
hand, the water molecule is capable of solvating both
types of centers effectively. Thus the dipoles that are
generated in TS2 to TS4 are more efficiently solvated by
water, so that the activation parameters should be
sensitive to the state of water in the solvent mixture.
The initial addition of acetonitrile to water results in a
¼
¼
decrease of kobs and increase in DH and DS since
uncoordinated water molecules would be considerably
more polar, that is, of higher kinetic reactivity than their
coordinated counterparts.2 After this region, the effect of
preferential solvation on kobs and the activation
parameters becomes apparent. For example, most of
Acknowledgements
¼
¼
the variation in DH and DS occurs below xW ꢃ 0.7. The
differences between the enthalpies and entropies of
activation, that is, between those calculated for the largest
We thank FAPESP (State of Sa˜o Paulo Research Founda-
tion) for financial support; CNPq (National Council for
Scientific and Technological Research) for a research
productivity fellowship to O. A. El Seoud, and a BIPIC
fellowship to F. Siviero; Dr. Paulo A. R. Pires and Mrs.
Clarissa T. Martin for their help during the preparation of
this manuscript.
¼
and smallest water concentrations are: (DDH ) ¼ 5.25,
3.85 kcal molꢀ1; T(DDS ) ¼ 8.45 and 8.66 kcal molꢀ1
,
¼
¼
for NPCF and NPFB, respectively. DDH may reflect the
activities of water in the different solubilization sites in
MeCN-W nano-clusters.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802

802
O. A. EL SEOUD AND F. SIVIERO
10. Venkatasubban KS, Bush M, Ross E, Schultz M, Graza O. J. Org.
Chem. 1998; 63: 6115.
REFERENCES
11. (a) Blokzijl W, Engberts JBN, Jager J, Blandamer MJ. J. Phys.
Chem. 1987; 91: 6022; (b) Blokzijl W, Blandamer MJ, Engberts
JBN. J. Org. Chem. 1991; 56: 1832.
12. (a) Sørensen OS. Acta Chem. Scand. 1976; A30: 673; (b) El Seoud
AO, Vieira RC, Farah JPS. J. Org. Chem. 1981; 46: 1231; (c) Bell
RP, Critchlow JE, Proc. R. Soc. London. Ser A. 1971, 325; (d)
Guthrie JP, J. Am. Chem. Soc. 2000; 122: 5529; (e) Henederson JW,
Haake P. J. Org. Chem. 1977; 42: 3989.
1. (a) Husky WP, Hogg JL. J. Org. Chem. 1981; 46: 53; (b) Husky
WP, Warren CT, Hogg JL. J. Org. Chem. 1981; 46: 59;
(c) Gopalakrishnan G, Hogg JL. J. Org. Chem. 1989; 54: 768;
(d) Fernnadez MA, de Rossi RH. J. Org. Chem. 1999; 64: 6000.
2. (a) Engberts JBFN. Pure Appl. Chem. 1982; 54: 1797 and refer-
ences cited therein; (b) Galema SA, Blandamer MJ, Engberts
JBFN. J. Org. Chem. 1989; 54: 1227.
3. (a) Zucco C, Lima CF, Rezende MC, Vianna JF, Nome F. J. Org.
Chem. 1987; 52: 5356; (b) El Seoud OA, El Seoud MI, Farah JPS.
J. Org. Chem. 1997; 62: 5928; (c) Mancini PME, Terenzani A,
Adam C, Vottero LR. J. Phys. Org. Chem. 1999; 12: 430;
(d) Humeres E, Nunes RJ, Machado VG, Gasques MDG, Machado
C. J. Org. Chem. 2001; 66: 1163; (e) Litwinienko G, Ingold KU. J.
Org. Chem. 2003; 68: 3433; (f) Harrisson S, Mackenzie SR,
Haddleton DM. Macromol. 2003; 36: 5072; (g) Tian J, Tan J,
Schown KB, Schowen RL. Can. J. Chem. 1999; 77: 781; (h)
Longhinotti E, Domingos JB, da Silva PLF, Szpoganicz B, Nome
F. J. Phys. Org. Chem. 2005; 18: 167; (i) Frasson CML, Branda˜o
TAS, Zucco C, Nome F. J. Phys. Org. Chem. 2006; 19: 143.
4. Anslyn EV, Dougherty DA. Modern Physical Organic Chemsitry.
University Science Books: Sausalito, 2004, p. 437.
5. (a) Reichardt C. Solvents and Solvent Effects in Organic Chemistry
(3rd edn). VCH: Weinheim, 2003 (i) p. 147, (ii) p. 389 and references
cited therein; (b) Reichardt C. Pure Appl. Chem. 2004; 76: 1903.
6. (a) Tada EB, Novaki LP, El Seoud OA. J. Phys. Org. Chem. 2000;
13: 679; (b) Antonius MS, Tada EB, El Seoud OA. Ibid. 2002; 15:
403; (c) Tada EB, Silva PL, El Seoud OA. Ibid. 2003; 16: 691; (d)
Tada EB, Silva PL, El Seoud OA. Phys. Chem. Chem. Phys. 2003;
5: 5378; (e) Tada EB, Silva PL, Tavares C, El Seoud OA. J. Phys.
Org. Chem. 2005; 18: 398.
13. (a) Albery J. In Proton Transfer reactions, Caldin E, Gold V (eds).
Wiley: New York, 1975, p. 263; (b) Theodorou LG, Lymper-
opoulos K, Bieth JG, Papamichael EM. Biochemistry 2001; 40:
3996; (c) Enyedy EJ, Kovach IM. J. Am. Chem. Soc. 2004; 126:
6017; (d) Zhang D, Kovach IM. J. Am. Chem. Soc. 2005; 127:
3760.
14. Neuvonen H. J. Chem. Soc., Perkin Trans. 1987; 2: 1723.
15. Leo AJ, Hansch C. Perspect. Drug Discov. 1999; 17: 1.
16. (a) Gorbunov BZ, Naberukhin YuI. J. Mol. Struct. 1972; 14:
113; (b) Gorbunov BZ, Naberukhin YuI. J. Struct. Chem. 1975;
16: 755; (c) Easteal AJ. Aust. J. Chem. 1979; 32: 1379;
(d) Balakrishnan S, Easteal AJ. Aust. J. Chem. 1981; 34: 943;
(e) Easteal AJ, Woolf LA. J. Chem. Thermodynamics 1982;
14: 755.
17. Marcus Y. Monatsh. Chem. 2001, 132: 1387 and references cited
therein.
18. (a) Wakisaka A, Shimizu Y, Nishi N, Tokumaru K, Sakuragi H. J.
Chem. Soc., Faraday Trans. 1992; 88: 1129; (b) Wakisaka A,
Takahashi S, Nishi N. J. Chem. Soc. Faraday Trans 1995; 91: 4063;
(c) Shin DN, Wijnen JW, Engberts JBFN, Wakisaka A. J. Phys.
Chem. B. 2001; 105: 6759.
19. El Seoud OA, Ruasse MF, Possidonio S. J. Phys. Org. Chem. 2001;
14: 525.
20. (a) Abraham MH. Prog. Phys. Org. Chem. 1974; 11: 2; (b) Yates K,
McClelland RA. Ibid. 1974; 11: 324; (c) Buncel E, Wilson H. Acc.
Chem. Res. 1979; 12: 42; (d) Cox BG. J. Chem. Soc. Perkin Trans.
1973; 2L 607.
7. Armarego WLF, Chai CLL. Purification of Laboratory Chemicals
(5th edn). Elsevier: New York, 2003.
8. Schown KB. In Transition States for Biochemical Processes, Gandour
RD, Schown RL (eds). Plenum Press: New York, 1978, p. 225.
9. Hansch C, Leo A, Taft RW. Chem. Rev. 1991; 91: 165.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 793–802