Kinetics and mechanisms of the reactions of benzoyl derivatives of nucleophiles: Dependence of the solvation requirement of the reaction on the structures of the nucleophile and the acyl group

Article abstract of DOI:10.1002/poc.864

The reactions of two series of benzoyl derivatives of nucleophiles were investigated and the results were compared with those for N-benzoylimidazole. The general structure is ArCO-Nu, where Ar = X-C₆H₄ and/ or X₂C₆H₃; X = 4-Me, H, 4-C1, 4-CN, 4-NO ₂ and X₂ = 3,5-dinitro; Nu = iodosobenzoate, ArCO-Iba, and phosphate dianion, ArCO-Phos. Catalytic rate constants, kinetic solvent isotope effects, kinetic substituent effects (Hammett equation) and the dependence of HΔ^≠ and ΔS^≠ on σ_x were determined. For ArCO-Iba, the hydrolysis occurs via two pathways, uncatalyzed (κ H₂O) and specific base-catalyzed (κ _OH) water attack on the iodine atom of the iodosobenzoate ring. This conclusion is based on theoretical calculations of the partial charges on ArCO-Iba, and the small ρ values calculated at 25°C, -0.22 and 0.92 for κ H₂O and k_oH, respectively. The data for the reaction of ArCO-Phos are consistent with a dissociative transition state, leading to elimination of the metaphosphate monoanion (PO₃ ). The dependence of the mechanistic pathway on the nucleophile is discussed. Two results are relevant to the reactions of ArCO-Iba: (i) moderate to large substituent effects on the activation entropies suggest that solvation of the leaving benzoate anion and desolvation of the entering nucleophile contribute to the OH -mediated reaction; (ii) the negative and positive signs of ΔS^≠ indicate large differences in solvation of the transition states of the κH₂O and κ_OH pathways. For the spontaneous decomposition of ArCO-Phos, sizeable substituent effects on both ΔH^≠ and ΔS^≠ were observed. This shows the contribution of solvation of the leaving benzoate and substituent-induced shift of the structure of the transition state. Copyright

Full text of DOI:10.1002/poc.864

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2005; 18: 173–182
Published online 5 October 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.864
Kinetics and mechanisms of the reactions of benzoyl
derivatives of nucleophiles: dependence of the solvation
requirement of the reaction on the structures of the
nucleophile and the acyl groupy
Omar A. El Seoud,1* Monica Ferreira,2 Wagner A. Rodrigues1 and Marie-Franc¸oise Ruasse3
1Instituto de Quı´mica, Universidade de Sa˜o Paulo, C.P. 26077, 05513-970 Sa˜ o Paulo, S.P., Brazil
2Fuctional Chemicals Division, Clariant SA, Sa˜ o Paulo, Brazil
3Institut de Topologie et de Dynamiques des Syste`mes, Universite´ de Paris VII–CNRS, 1 Rue Guy de la Brosse, F-75005 Paris, France
Received 30 January 2004; revised 19 April 2004; accepted 11 May 2004
ABSTRACT: The reactions of two series of benzoyl derivatives of nucleophiles were investigated and the results were
compared with those for N-benzoylimidazole. The general structure is ArCO-Nu, where Ar ¼ X-C6H4- and/or X2-
C6H3-; X ¼ 4-Me, H, 4-Cl, 4-CN, 4-NO2 and X2 ¼ 3,5-dinitro; Nu ¼ iodosobenzoate, ArCO-Iba, and phosphate
dianion, ArCO-Phos. Catalytic rate constants, kinetic solvent isotope effects, kinetic substituent effects (Hammett
¼
¼
equation) and the dependence of ÁH and ÁS on ꢀX were determined. For ArCO-Iba, the hydrolysis occurs via
two pathways, uncatalyzed (kH O) and specific base-catalyzed (kOH) water attack on the iodine atom of the
iodosobenzoate ring. This conclusion is based on theoretical calculations of the partial charges on ArCO-Iba, and
2
the small ꢁ values calculated at 25 ꢀC, ꢁ0.22 and 0.92 for kH O and kOH, respectively. The data for the reaction of
2
ArCO-Phos are consistent with a dissociative transition state, leading to elimination of the metaphosphate monoanion
(POꢁ3 ). The dependence of the mechanistic pathway on the nucleophile is discussed. Two results are relevant to the
reactions of ArCO-Iba: (i) moderate to large substituent effects on the activation entropies suggest that solvation of
the leaving benzoate anion and desolvation of the entering nucleophile contribute to the OHꢁ-mediated reaction; (ii)
¼
the negative and positive signs of ÁS indicate large differences in solvation of the transition states of the kH O and
2
¼
kOH pathways. For the spontaneous decomposition of ArCO-Phos, sizeable substituent effects on both ÁH and
ÁS were observed. This shows the contribution of solvation of the leaving benzoate and substituent-induced shift of
¼
the structure of the transition state. Copyright # 2004 John Wiley & Sons, Ltd.
KEYWORDS: benzoyl phosphates; decomposition; benzoyl iodosobenzoates; hydrolysis; acyl derivatives of nucleophiles;
reaction mechanisms
INTRODUCTION
formation and subsequent decomposition have been the
subject of intense research, because of its relevance to
mechanism of chemical and enzymatic catalysis.4–9
Much attention has been devoted to the mechanism of
formation of RCO-Nu and/or ArCO-Nu (concerted or
stepwise; R ¼ alkyl moiety) than to its subsequent de-
composition,9b,d although the latter step is important,
since it determines the catalyst turnover. Mechanistic
studies of the decomposition of R- and/or ArCO-Nu
have been carried out mainly on enzymes, or enzyme
models, in order to understand acyl transfers between
nucleophiles. It is not surprising, therefore, that the
investigated intermediates were mainly R- and/or
ArCO-Imz and R- and/or ArCO-Phos (Nu ¼ imidazole
and phosphate, respectively) since these two nucleophiles
play a major role in biochemical reactions.5–10 These
studies have indicated that acyl derivatives of nucleo-
philes may decompose by a variety of mechanisms,
depending on the structures of the acyl moiety, the
nucleophile and reaction conditions. Figure 1 depicts
Studies of the mechanistic details of acyl-transfer reac-
tions have contributed a great deal to our understanding
of several problems of central importance in chemistry
and biochemistry, including effects of structure on reac-
tivity and the mechanism of action of hydrolytic en-
zymes.1–3 In addition to uncatalyzed water addition, kH O
these reactions may be subject to specific, e.g. kOH
general, kGB, and nucleophilic catalysis. In ester hydro-
lysis, the last pathway leads to the formation of an
intermediate, an acyl derivative of the nucleophile, whose
,
,
2
´
*Correspondence to: O. A. El Seoud, Instituto de Quımica, Universi-
dade de Sa˜o Paulo, C.P. 26077, 05513-970 Sa˜o Paulo, S.P., Brazil.
E-mail: elseoud@iq.usp.br
ySelected article presented at the Seventh Latin American Conference
on Physical Organic Chemistry (CLAFQO-7), 21–26 September 2003,
´
Florianopolis, Brazil.
Contract/grant sponsor: Sa˜o Paulo State Research Foundation, FA-
PESP.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

174
O. A. EL SEOUD ET AL.
information on the kinetics and mechanism of the hydro-
lysis of ArCO-Iba.
The general reaction scheme is given in Fig. 2.
Catalytic rate constants, activation parameters, kinetic
solvent isotope effects (KSIE) and Hammett ꢁ values
have been determined for the formation of ArCO-Nu and
for the hydrolysis of ArCO-Imz8 (formally, these ꢁ values
are equivalent to Brønsted ꢂleaving group, since ꢀ is
obtained from the effect of X on the pKa of benzoic
acid, the leaving group in our reactions; nevertheless, we
discuss substituent effects in terms of ꢁ, rather than
Figure 1. Mechanistic pathways for the decomposition of a
typical acyl derivative of nucleophile, ArCO-Nu, in basic
media. For simplicity, we do not show the negative sign of
the hydroxide ion in kOH
ꢂ
leaving group, since the latter is not very informative in
the absence of the corresponding ꢂentering nucleophile). We
now address the kinetics and mechanism of reactions of
the two other intermediates, namely ArCO-Iba and
ArCO-Phos, where Ar ¼ X-C6H4- and/or X2-C6H3-;
X ¼ 4-Me, H, 4-Cl, 4-CN, 4-NO2 and X2 ¼ 3,5-dinitro.
We anticipated that ArCO-Phos dianion may decompose
spontaneously, as shown previously for the dianions of
acetyl phosphate and phosphate monoesters.5b,10 No
prediction was made regarding the hydrolysis of ArCO-
Iba, as the reaction is open to different pathways shown
in Fig. 1.
the main pathways for the decomposition reactions of
ArCO-Nu in basic media, where kspon refers to sponta-
neous decomposition.
Hydrolysis of ArCO-Imz is well documented; the
reaction is subjected to specific and general acid–base
catalysis.7,8a The reactions of dianionic ArCO-Phos and
of phosphate monoester dianions appear to proceed by a
dissociative mechanism, via a metaphosphate-like transi-
tion state.9,10 It is not possible, a priori, to predict the
favored catalytic mechanism for nucleophiles other than
Phos, Imz or amines, although this knowledge adds to our
understanding of catalysis in chemistry and enzymology.
We have studied the hydrolysis of phenyl benzoates,
substituted in the acyl (by X) and in the leaving phenol
(by Y).8 The three buffers employed were Imz, the o-
iodosobenzoate anion, Iba, and monohydrogenphosphate
dianion, Phos. Our interest in these buffers is due to the
following reasons: (i) although their pKa values are
similar, namely 7.53, 7.05 and 7.20 for Iba, Imz and
Phos, respectively, their catalytic efficiencies (in ester
hydrolysis) are very different, IBA > Imz ꢂ Phos;8d (ii)
Iba is a powerful nucleophile, with potential use in large-
scale decontamination of toxic chemical substances and
wastes, including phosphorelated pesticides and chemical
warfare agents;11–13 (iii) to our knowledge, there is no
We show that the hydrolysis of ArCO-Iba proceeds by
the kH O and kOH pathways, but do not show measurable
2
kGB. The behavior of ArCO-Phos agrees with our predic-
tion (spontaneous decomposition). A comparison of the
above-mentioned quantities, i.e. ꢁ, KSIE and activation
parameters, for the three intermediates indicates that (i) ꢁ
does not depend much on the mechanism of catalysis, (ii)
substituent effects on the activation parameters are more
characteristic of each of the catalytic pathways and (iii)
solvation of the appropriate species (entering nucleophile
and leaving group) plays an important role, independent
of the catalysis mechanism.
EXPERIMENTAL
The solvents and other reagents were purchased from
Aldrich and were purified by standard procedures.14 The
Figure 2. Complete reaction scheme for the hydrolysis of benzoate esters, catalyzed by phosphate, imidazole and o-iodoso-
benzoate buffers
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

REACTIONS OF BENZOYL DERIVATIVES OF NUCLEOPHILES
175
purity of all solid reagents was checked by measuring
their melting-points. Iba was available from a previous
study.8c
ArCO-Phos. These were synthesized in situ by reacting
the appropriate acyl chloride with (n-C4H9)4Nþ H2PO4ꢁ
in dry acetonitrile. The organic phosphate salt was pre-
pared by reacting 2 ml of 85% orthophosphoric acid with
19.5 ml of a 40% aqueous solution of (n-C4H9)4Nþ OHꢁ,
followed by adjustment of the pH to 4.3 (with H3PO4)
and evaporation of water. The hygroscopic product was
dried over P4O10 under reduced pressure. Calculated for
C16H38NPO4: C, 56.66; H, 11.20; N, 4.12. Found: C,
56.38; H, 11.08; N, 4.12%.
Synthesis of ArCO-Iba and ArCO-Phos
Acyl chlorides. These were prepared by refluxing the
appropriate benzoic acid with excess purified thionyl
chloride (molar ratio 1:4, bath temperature ca 130 ꢀC)
for 6 h, followed by removal of SOCl2 and distillation of
the acyl chloride. PCl5 was employed for the preparation
of 3,5-dichlorobenzoyl chloride, as given elsewhere,15a
and was purified by crystallization from a mixture of
CCl4 and light petroleum (b.p. 30–60 ꢀC). The b.p.s and
m.p.s of the acyl chlorides were in agreement with the
values published elsewhere.15
The following example shows the completeness of the
reaction between 4-nitrobenzoyl chloride and (n-
C4H9)4Nþ H2PO4ꢁ. A solution containing 0.03 mol of
the acyl chloride and 0.0315 mol of the organic phosphate
salt was mixed and left to react at room temperature. The
IR spectrum was periodically scanned in order to deter-
mine the reaction progress. It showed the complete dis-
appearance of the acyl chloride ꢃC— (ca 1760 cmꢁ1),
O
—
ArCO-Iba. To a solution of the acyl chloride (2 mmol) in
CH2Cl2 (25 ml) were added 25 ml of an aqueous solution
containing Iba (2 mmol), NaOH (2 mmol) and 0.02 g of
tetra(n-butyl)ammonium hydrogensulfate. The two-
phase mixture was vigorously stirred at room tempera-
ture, until a solid compound separates at the CH2Cl2/
water interface. The reaction time was found to depend
on the nature of X, ranging from 1 min for X ¼ NO2 to
3 min for X ¼ CH3. The solid formed was quickly fil-
tered, washed with cold CH2Cl2 and dried over P4O10
under reduced pressure. The products gave satisfactory
elemental analyses (Perkin-Elmer Model 2400 CHN,
Elemental Analysis Laboratory, University of Sa˜o Paulo)
and the expected IR bands (KBr pellets, Perkin-Elmer
1750 FTIR), as shown in Table 1.
with concomitant appearance of the mixed anhydride
ꢃC— (ca 1739 cmꢁ1).16
O
—
Kinetic measurements
Reaction kinetics were studied with Zeiss PM6KS,
Beckman DU-70 UV–visible or Applied Photophysics
stopped-flow spectrophotometers; the last is provided
with syringes of unequal volumes. These instruments
are fitted with thermostated cell holders whose tempera-
ture was kept constant to within ꢃ 0.05 ꢀC. The condi-
tions employed are shown in Table 2. For comparison, we
have also included the conditions previously employed in
the hydrolysis of ArCO-Imz.8b
Table 1. Elemental analysis and IR frequencies of ArCO-Iba
Calculated (%)
H
Found (%)
X
M.p. ( ꢀC)
C
N
C
H
N
IR frequencies (cmꢁ1
)
4-CH3 198–199
47.41
45.68
41.77
45.83
40.47
2.90
2.46
2.00
2.05
1.95
47.05
45.42
41.55
45.64
40.54
2.85
2.36
1.92
2.00 3.16
1.86 3.42
33101 (ꢃC–H); 1685 (ꢃC—O); 1631 (ꢃC—
O
—
3109 (ꢃC–H); 1679 (ꢃC—O); 1642 (ꢃC— )
O
— —
)
O
—
3100 (ꢃC–H); 1689 (ꢃC—O); 1633 (ꢃC—
—
H
192–193
199–200
4-CN 205–206
4-NO2 206–207
)
—
4-Cl
3.58
3.39
3084 (ꢃC–H); 2229 (ꢃCꢄN) 1683 (ꢃC—O); 1657 (ꢃC—
)
O
—
—
3073 (ꢃC–H); 1717 (ꢃC—O); 1633 (ꢃC—O); 1528
—
—
(ꢃNO
,
assym); 1329 (ꢃNO
,
)
sym
2
2
Table 2. Experimental conditions employed in the kinetic studies
Experimental variable
ArCO-Ibaz
ArCO-Imza
ArCO-Phos
Solvent
13% (v/v) CH3CN in water
N-methylmorpholine (NMeM); Imz; 0.07; KCl; 7.0–7.8
10% (v/v) CH3CN in water Water
Buffer; ꢄ (mol lꢁ1);
electrolyte to adjust ꢄ; pH range 0.2; KCl; 7.6–8.8
Potassium phosphate;
0.245–1.0; buffer and KCl, 7.2
Reagent disappearance at: 234
253, CH3; 245, H; 252, Cl; H; 262, Cl, CN and NO2;
X and ꢅ employed (nm)
Reagent disappearance at: 250, Reagent disappearance at:
CH3; H; Cl and CN; at 260,
NO2
Product formation at 303, NO2
255, CN; 260, and NO2
at 222, 3,5-dinitro
Starting [RCO-Nu] (mol lꢁ1
)
(4–6) ꢅ 10ꢁ5
(1–3) ꢅ 10ꢁ5
(1–4) ꢅ 10ꢁ5
a
Data taken from Ref. 8b.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

176
O. A. EL SEOUD ET AL.
All kinetic runs were carried out in triplicate, under
regression of a large number of experimental points
(ꢇ 80), their literature counterparts were based on ‘five
to ten points’.5b In summary, in situ preparation of ArCO-
Phos has been successfully achieved; the kinetic data
obtained, kobs, activation parameters and ꢁ, are reliable.
pseudo-first-order conditions. Slow reactions were in-
itiated by injecting a solution of ArCO-Nu in anhydrous
acetonitrile into the appropriate buffer solution, followed
by homogenization with a hand-held micro-stirrer
(Hellma, type 338.004). We verified that observed rate
constants, kobs, were independent of [ArCO-Nu] in the
range (1–6) ꢅ 10ꢁ5 mol lꢁ1. Plots of log(A ꢁ At) and/or
Hydrolysis of ArCO-Iba
1
log(At ꢁ A ) versus time were rigorously linear over
1
more than five half-lives and their slopes gave kobs. For
the same run, the relative standard deviation in kobs, i.e.
(standard deviation/kobs) ꢅ 100, was ꢆ 0.2%; for tripli-
cate runs, the difference between kobs was ꢆ 1%. The
spectra of representative kinetic runs, recorded after
reaction completion, were found to be identical with
those of authentic mixtures of the expected products.
Consider the hydrolysis of ArCO-Nu in the presence of
Nu–NuH buffer. Catalysis by the latter, if it occurs,
cannot be nucleophilic since the catalyst and the leaving
group are the same. Detection of a GB catalytic pathway
involves hydrolysis of ArCO-Iba in the presence of Iba–
IbaH buffer. We were unable, however, to carry out this
experiment because the buffer absorbs strongly in the
same UV–visible region of ArCO-Iba (note that
[buffer] ꢂ [substrate]). Therefore, we investigated the
reaction in N-methylmorpholine (NMeM) buffer, whose
pKa, 7.38, is close to that of Iba. We found that kobs
(0.065 ꢃ 0.002 sꢁ1) did not change as a function of
increasing buffer concentration (X ¼ CH3, T ¼ 25 ꢀC,
RESULTS AND DISCUSSION
The discussion is organized in the following order: first,
we comment on the substituents employed and the reason
for using in situ synthesis for ArCO-Phos; second, we
discuss mechanistic details of the reactions of ArCO-Iba
and ArCO-Phos; finally, we compare the data of these
two intermediates with those of ArCO-Imz, with the
objective of acquiring a better understanding of the
factors controlling their reactions (Fig. 1).
[NMeM] ¼ 0.02–0.22 mol lꢁ1
, pH 7.8). This result
indicates the absence of GB catalysis, a conclusion
corroborated by the following experiment: hydrolysis of
ArCO-Iba (X ¼ CN) was studied at a constant concentra-
tion of NMeM (0.05 mol lꢁ1, pH 7.8), as a function of
increasing the concentration of Iba (employed as a
reagent). The rate constants calculated were 102kobs
(sꢁ1) ¼ 6.9 ꢃ 0.2, 7.0 ꢃ 0.2, 7.5 ꢃ 0.4, 7.3 ꢃ 0.3, 7.4 ꢃ
0.3 for the reaction in buffer solutions in the absence and
in the presence of (6, 9, 12 and 15) ꢅ 10ꢁ5 mol lꢁ1 Iba,
respectively. The uncertainties in these rate constants are
relatively large, most certainly because of the high initial
absorbance (of Iba plus ArCO-Iba), coupled with the small
variation of absorbance during the reaction (disappearance
of the intermediate and liberation of Iba). The rate con-
stants calculated show, however, no significant dependence
on the concentration of Iba. In summary, hydrolysis of
ArCO-Iba is not subject to measurable GB catalysis (by
NMeM or Iba), and may proceed by solvolytic and specific
base-catalyzed pathways.
Choice of the substituents and in situ synthesis
of ArCO-Phos
The acyl group of ArCO-Iba was either benzoyl or 4-
substituted benzoyl. Because the reaction of ArCO-Phos
is slow, the lowest reaction temperature employed was
35 ꢀC, and we also included the 3,5-dinitrobenzoyl group.
We attempted the synthesis of 4-nitrobenzoyl phosphate
by reaction of the appropriate acyl chloride with lithium
phosphate in dry acetonitrile and also by phase-transfer
catalysis, as given for ArCO-Iba. Although the targeted
mixed anhydride was obtained by the latter procedure, we
did not pursue this approach because of the very low yield
obtained, ca 5% for X ¼ NO2. This is probably because of
the unfavorable phase transfer of the phosphate dianion.17
The above-mentioned IR data (see Experimental) show
that the in situ procedure led to complete formation of the
acyl phosphoric mixed anhydride. Additionally, we com-
pared the rate constants for the hydrolysis of three mixed
anhydrides with literature values (where the mixed anhy-
drides were also synthesized in situ), at 39 ꢀC.5b The
following results were obtained for 105kobs (sꢁ1): 1.0,
46.7, 2505b and 0.52, 41.5 and 243.3 (present study) for 4-
chloro-, 4-nitro- and 3,5-dinitrobenzoyl phosphate, re-
spectively. We attribute the differences between the two
sets of data to the availability of better instrumentation at
our disposal. Whereas we calculated kobs by non-linear
Rate constants of the OHꢁ-catalyzed hydrolysis of
ArCO-Iba were determined in the pH range 7.6–8.8 in
the presence of 0.05 mol lꢁ1 NMeM buffer in the tem-
perature range 18–45 ꢀC. Table 3 shows the catalytic rate
constants, kOH (obtained from the slopes of the plots of
kobs versus [OHꢁ]) along with the corresponding activa-
tion parameters and Hammett ꢁ values. An experiment in
D2O (X ¼ NO2; pD ¼ 8.0–9.2) indicated that kOD
/
kOH ¼ 0.92.
In principle, it should be possible to determine the rate
constant of the water reaction, kH O, from the intercepts
2
of the plots of kobs versus [OHꢁ]. Depending on the
relative magnitudes of the slope and intercept, the rate
constants calculated by this procedure may be subject to
relatively large uncertainties. We found, however, that the
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

REACTIONS OF BENZOYL DERIVATIVES OF NUCLEOPHILES
177
Table 3. Catalytic rate constants and activation parameters
for the OHꢁ-catalyzed hydrolysis of ArCO-Iba, kOH, at different
temperatures and the corresponding Hammett ꢁ valuesa
ꢁ3
10
k
(l molꢁ1 sꢁ1 b
)
OH
Temperature
( ꢀC)
CH3
H
Cl
CN
NO2
ꢁc
Figure 3. Possible sites of attack on ArCO-Iba
18
25
35
45
3.06
4.76 5.95 19.02 22.01 0.92
Regarding mechanistic details of this reaction, we
address the following questions.
6.22 10.51 13.56 42.20 46.04 0.92
17.67 27.14 36.43 119.28 133.31 0.95
40.49 68.05 88.21 279.54 325.16 0.95
Activation
Detection of an intermediate during hydrolysis.
Experiments were carried out with ArCO-Iba (X ¼ NO2)
in 50% aqueous acetonitrile, at ‘nominal’ pH ¼ 7.8
(0.05 mol lꢁ1 NMeM). The reason for changing the solvent
composition was to slow the reaction, so that it can be
conveniently monitored with a UV–visible spectrophot-
ometer. In one experiment, the spectra of the reaction were
recorded as a function of time. A sharp isosbestic point was
observed at 288 nm. The hydrolysis rate constant was also
determined by following the disappearance of the reactant at
260 nm and/or the appearance of the products at 303 nm.
The rate constants calculated were in close agreement,
0.0113 and 0.0110 sꢁ1, respectively. Both experiments in-
dicate either a concerted mechanism or a putative one
without accumulation of a detectable intermediate.
parametersd
ÁH
¼
17.2
16.5
12.3
17.4 17.7
18.2 19.7
12.0 11.8
17.8
22.2
11.2
17.9
22.8
11.1
(kcal molꢁ1
ÁS
)
¼
(cal Kꢁ1 molꢁ1
ÁG
)
¼
(kcal molꢁ1
)
a
See Experimental for the uncertainty in these rate constants.
The following hydroxide ion concentrations were employed (107 [OHꢁ],
b
X): 5.81–37.7, CH3; 4.93–53.6, H; 5.48–58.9, Cl; 3.75–40.6, CN; 4.95–
50.1, NO2.
c
The uncertainty in ꢁ is ꢆ 0.06.
d
The uncertainties in the activation parameters are ꢃ 0.1 kcal molꢁ1
(ÁH and ÁG ) and ꢃ 0.5 cal Kꢁ1 molꢁ1 (ÁS ). In this and subsequent
¼
¼
¼
¼
tables, ÁS is that calculated at 25 ꢀC.
Table 4. Second-order rate constants and activation para-
meters for the uncatalyzed water hydrolysis of ArCO-Iba,
k
H O, at different temperatures and the corresponding
Site of reagent (water and/or OHꢁ) attack on
ArCO-Iba. As shown in Fig. 3, there are, in principle,
2
Hammett ꢁ valuesa
104kH O (l molꢁ1 sꢁ1
)
—
—
three possible sites of attack on ArCO-Iba: (a) the C
—
O
2
group of the heterocyclic ring; (b) at the C O group of
—
Temperature
( ꢀC)
CH3
H
Cl
CN NO2
ꢁc
the acyl group; (c) at the iodine atom of the heterocyclic
ring. Possibility (a) can easily be ruled out, since the
hydrolysis reaction should be negligibly dependent on the
nature of X, at variance with the results of Table 3. Ab
initio calculations on model compounds for Iba have
shown that the partial positive charge on the iodine
atom is larger than the corresponding one on the carbon
18
25
35
45
7.35 6.52 5.52 4.33
13.7 13.1 10.8 9.08
33.3 32.1 27.5 21.9
81.8 77.3 67.5 58.0
4.03 ꢁ0.26
8.42 ꢁ0.23
19.8 ꢁ0.24
54.3 ꢁ0.19
Activation
parametersb
ÁH
atom of the C O group of the heterocyclic ring.11b Our
—
¼
—
15.8 16.2 16.5 16.9
16.9
(kcal molꢁ1
ÁS
)
PM3 semi-empirical calculations (Table 5) on Iba and on
ArCO-Iba show (at least qualitatively) that the partial
positive charge on the iodine atom is larger than the
¼
ꢁ18.5 ꢁ17.4 ꢁ16.8 ꢁ15.8 ꢁ16.0
)
(cal Kꢁ1 molꢁ1
ÁG
¼
21.3 21.4 21.5 21.6
21.7
—
—
(kcal molꢁ1
)
corresponding ones on the carbon atoms of the C
O
group of the heterocyclic ring and/or the acyl moiety.
Finally, results of the following experiments show that
the attack of nucleophiles on derivatives of Iba occurs at
the iodine atom, as shown in Fig. 4.11a,c
a
See Experimental for the uncertainty in these rate constants. In this and
subsequent tables, Ar ¼ X-C6H4- and/or X2-C6H3-; X ¼ 4-Me, H, 4-Cl, 4-
CN, 4-NO2 and X2 ¼ 3,5-dinitro.
b
The uncertainties in the activation parameters are ꢃ 0.1 kcal molꢁ1
(ÁH and ÁG ) and ꢃ 0.5 cal Kꢁ1 molꢁ1 (ÁS ).
¼
¼
¼
c
The uncertainty in ꢁ is ꢆ 0.02.
Structure of the transition state (TS) of the
hydrolysis reaction. We suggest structures TS1 and
TS2 (Fig. 5) for the transition states of uncatalyzed, and
OHꢁ-mediated hydrolysis of ArCO-Iba, respectively. For
simplicity, we only show the species that is directly
attacking the iodine atom. Other solvent molecules are
most certainly involved. For example, proton inventory
studies of several water-catalyzed hydrolyses have shown
that the attack of the ‘nucleophilic’ water molecule is
hydrolysis reaction is independent of pH in the range 0.9–
4.3, and determined kH O directly, by carrying out the
2
reaction at pH 2.80 (adjusted with HCl, ꢄ ¼ 0.2, KCl).
Table 4 shows the rate constants calculated, along with
the corresponding activation parameters and Hammett
ꢁ. Experiments in D2O, at ꢄ ¼ 0.2 and pD ¼ 3.21
( ¼ pH þ 0.41),18 showed small, positive KSIE of 1.2
and 1.3 for X ¼ CH3 and NO2, respectively.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

178
O. A. EL SEOUD ET AL.
Table 5. Partial charges on selected atoms of Iba, acetyl-Iba and ArCO-Iba, calculated by the PM3 semi-empirical method
Atom
Compound
I
O3
O1
O18
—
C6
C16
—
Iba
ArCO-Iba
Me
H
Cl
0.8248
ꢁ0.4574
ꢁ0.8718
0.6117
0.8457
0.8464
0.8474
0.8492
0.8510
0.8531
ꢁ0.4690
ꢁ0.4678
ꢁ0.4663
ꢁ0.4620
ꢁ0.4585
ꢁ0.3969
ꢁ0.6816
ꢁ0.6812
ꢁ0.6808
ꢁ0.6800
ꢁ0.6799
ꢁ0.6816
ꢁ0.5870
ꢁ0.5852
ꢁ0.5830
ꢁ0.5750
ꢁ0.5721
ꢁ0.5804
0.6118
0.6116
0.6114
0.6103
0.6094
0.6078
0.6796
0.6781
0.6752
0.6701
0.6672
0.6275
CN
NO2
CH3CO-Ibaa
a
The atom numbering for CH3CO-Iba is the same as that for ArCO-Iba.
following reasoning. (i) Since the TSs are more polar than
the precursor organic substrates, ꢁ essentially reflects the
effects of X (stabilization/destabilization) on the charges
that are being developed on the reaction center. For the
watꢁer reaction ꢁ is close to zero, whereas ꢁ ¼ 0.92 for the
OH -mediated hydrolysis (at 25 ꢀC). Both results reflect
the attenuated transmission of the electronic effects of X
to the iodine atom in TS1 and TS2. One measure of this
attenuation is the slope of plot of partial charge on the
atom of concern (Table 5) versus ꢀX, 0.0051 and
ꢁ0.0128, for the iodine atom and the acyl carbon atom,
respectively. Therefore, the electronic effects of X are
transmitted to the iodine atom less efficiently than to the
carbon atom of the acyl group, which decreases jꢁj.
Additionally, the smaller degree of charge development
in TS1 further decreases jꢁj. (ii) The TSs suggested also
agree with the entropies of activation and the KSIE. The
Figure 4. Mechanism of attack of nucleophiles on deriva-
tives of Iba
¼
water reaction is associated with a negative ÁS and a
relatively small but positive KSIE (i.e., kH O > kD O).
2
2
¼
Negative ÁS are expected for bimolecular reactions
involving ArCO-Iba and one or more water molecules,
but the magnitude of KSIE (ꢆ1.3) is smaller than that
expected for
a
typical water-catalyzed reaction
(>2.0).7d,e,18 The small KSIE may be the consequence
of the reactant-like structure of TS1, with little bond
formation and bond breaking. Alternatively, the reaction
coordinate might be dominated by the movement of
heavy atoms, especially the oxygen, with subordinate
proton transfer, characterized by a low amplitude of
motion at the TS. This mode of GB catalysis (e.g. by a
second water molecule) may be associated with a small
KSIE.20
Figure 5. Suggested structures for the transition states of
attack of water (TS1) and OHꢁ-mediated reaction of ArCO-
Iba (TS2), spontaneous decomposition of ArCO-Phos (TS3),
and GB-catalyzed hydrolysis of ArCO-Imz, TS4
The OHꢁ-mediated reaction is associated with positive
¼
ÁS , inverse KSIE (kOH < kOD), and ꢁ that is lower than
the range reported for acyl-transfer reactions of substi-
tuted benzoates (ꢁ ꢇ 1.2). TS2 shows more bond forma-
tion and breaking than TS1. Therefore, the sign of ꢁ is
expected, since an electron-withdrawing group should
favor both the formation of the OHꢁ–iodine bond and the
usually catalyzed by one or more ‘GB’ water molecu-
les.7d,e,18 Additionally, the hydroxide ion is known to be
strongly solvated by several water molecules.19 Our
suggestion is based on the data in Tables 3–5 and the
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

REACTIONS OF BENZOYL DERIVATIVES OF NUCLEOPHILES
179
Table 6. Rate constants and activation parameters for the
spontaneous breakdown of ArCO-Phos, kspon, at different
temperatures and the corresponding Hammett ꢁ valuesa
breaking of the iodine–carboxylate group bond. The
¼
positive ÁS is interesting because negative values are
expected for bimolecular reactions, provided that the
reaction is controlled by its intrinsic energy barrier, i.e.
that due to bond breaking and formation. However, for
reactions involving a neutral molecule and a strongly
solvated species, in particular the alkoxide and hydroxide
ions, the energetics may be partially or completely
dominated by the desolvation barrier, i.e. that required
to desolvate the base, with concomitant introduction of
the neutral reagent into its solvation microsphere.21
Therefore, the positive entropy of activation seems to
be due to the increase in the degree of freedom of the
water molecules, liberated by the (partial) desolvation of
the hydroxide ion. In this regard, it is relevant that several
bimolecular reactions are accompanied with positive
entropies of activation.22 Additionally, in OHꢁ-mediated
reactions, the H/D fractionation factor of the attacking
OHꢁ usually increases on going from the reactant
state, RS, to the TS.18,20 This increase reflects the
diminished role of the solvent shell on covalent bond
formation between oxygen and iodine, in agreement
with a positive entropy of activation. Finally, there is
reason to expect an inverse KSIE since ODꢁ is a stronger
base than OHꢁ.
105kspon (sꢁ1
Cl
CN NO2 Dinitro ꢁc
)
Temperature
( ꢀC)
H
35
45
50
55
1.2 2.8 15.2 24.2 144.6 1.48
6.9 14.8 68.2 91.7 509.0 1.32
16.2 33.1 139.7 173.2 927.7 1.24
37.0 72.0 279.9 320.7 1659.9 1.16
Activation parametersb
¼
ÁH
33.8 32.0 28.7 25.4
28.7 24.5 17.0 7.2
25.3 24.7 23.6 23.2
23.9
6.1
(kcal molꢁ1
ÁS
)
¼
(cal Kꢁ1 molꢁ1
ÁG
)
¼
22.1
(kcal molꢁ1
)
a
See Experimental for the uncertainty in these rate constants.
The uncertainties in the activation parameters calculated at 25 ꢀC are
b
ꢃ 0.1 kcal molꢁ1 (ÁH and ÁG ) and ꢃ 0.5 cal Kꢁ1 molꢁ1 (ÁS ).
¼
¼
¼
c
The uncertainty in ꢁ is ꢆ 0.08.
be calculated from kobs (not kobs/[water]), a point of
conflict in the literature.5b,23 Table 6 shows the rate
constants for the spontaneous decomposition of ArCO-
Phos, along with the corresponding Hammett ꢁ values
and the activation parameters.
Breakdown of ArCO-Phos
Reactions of acyl phosphates have been studied in some
detail, including the hydrolysis of acetyl phosphate, and
of some benzoyl phosphates at a single temperature,
39 ꢀC, in the pH range 6.9–7.5.5b In the present study,
the rate constants were determined more accurately and
the activation parameters were calculated and are dis-
cussed in terms of reaction mechanism. The pKa values of
4-methoxybenzoyl-, 4-nitrobenzoyl- and 3,5-dinitroben-
zoyl phosphate were estimated as 4.8, 4.3 and 4.0,
respectively.5b Therefore, at the pH employed in the
present study, 7.2, all our kinetic data refer to the
reaction of ArCO-Phos dianion. The following results
show that the rate of reaction of 3,5-dinitrobenzoyl
phosphate is insensitive to the variation of buffer con-
centration (ꢄ ¼ 0.78 mol lꢁ1; pH ¼ 7.20; T ¼ 39 ꢀC;):
[HPO42ꢁ] ¼ 0.1, 0.2, 0.3 mol lꢁ1; kobs ¼ 0.0260 ꢃ 0.001;
0.0256 ꢃ 0.001; 0.0250 ꢃ 0.001 sꢁ1, respectively. Under
the same reaction conditions, using 0.1 mol lꢁ1 phosphate
buffer, the reaction was found to be independent of pH in
the range 6–8; there was only a 5.4% rate constant
increase when ꢄ was increased (KCl) from 0.245 to
1.0 mol lꢁ1; and a 2.5% decrease in kobs when the solvent
was changed from H2O to D2O. These data agree with
those observed for the hydrolysis of acetyl phosphate
dianion, which indicated that the reaction is not subject to
specific-base and/or GB catalysis, and that water does not
participate in the slow step of hydrolysis.5b The latter
conclusion means that the activation parameters should
In agreement with these data, and with previous results,
we propose TS3 (Fig. 5) for the spontaneous decomposi-
tion of ArCO-Phos. This TS structure agrees with the
¼
positive ÁS because the degrees of freedom increase on
going from RS to TS. Note that several solvation/deso-
¼
lvation interactions contribute to ÁS , and may cancel
out partially, including desolvation of the reactant, and
solvation of the produced benzoate and metaphosphate
anions (the latter rapidly hydrolyzes to phosphate).
Since there is no nucleophilic addition of water to
ArCO-Phos, the practically unity of KSIE is not unex-
pected. The large, positive ꢁ indicates a significant
negative charge development in the benzoate moiety.
The noticeable sensitivity of ꢁ to the temperature, not
observed for the other ArCO-Nu, is interesting and will
be considered later. Finally TS4 is that suggested for the
GB-catalyzed hydrolysis of ArCO-Imz.
Comparison of the reactions of RCO-Iba,
RCO-Imz and RCO-Phos
We now compare the data obtained under experimentally
similar conditions (Table 7) for the reactions of the three
species, ArCO-Imz, ArCO-Iba and ArCO-Phos, which
occur via the different mechanisms depicted in Fig. 1. In
what follows, we are always concerned with the changes
that occur between the reactant state, RS, and the TS. For
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

180
O. A. EL SEOUD ET AL.
Table 7. Kinetic criteria for the mechanisms of reactions of ArCO-Nua
ArCO-Imz
ArCO-Ibad
ArCO-Phosd
kspon
b
c
c
Parameter
KSIE
kGB
kH O
kOH
kH O
kOH
2
2
2.94
2.43
1.25
2.54
2.36
1.41
1.13
1.03
1.47
1.2
1.3
ꢁ0.22
—
0.92
0.92
1.03
0.96
1.65
1.2e
ꢁ
¼
ÁH (kcal molꢁ1
)
10.4
7.9b
10.4
12.1
7.0
12.0
15.8
17.1
17.2
18.1
35.5
25.4
26.6f
33.0
7.2
¼
ÁS (cal Kꢁ1 molꢁ1) ꢁ36.9
ꢁ44.6
ꢁ32.7
ꢁ43.4
ꢁ20.3
ꢁ18.7
ꢁ15.2
16.5
22.6
ꢁ40.0
3.7f
¼
ÁG (kcal molꢁ1
)
21.4
19.8
23.7
21.8
19.9
18.1
21.3
21.6
12.3
11.1
25.8
23.2
25.4f
ꢁ9.9
ꢁ24.6
¼
ÁH versus ꢀg
ꢁ2.6
ꢁ2.8
1.8
12.6
ꢁ0.4
ꢁ4.8
1.2
3.1
0.7
6.4
¼
ÁS versus ꢀg
a
In each column, the upper and lower figures refer to X ¼ 4-CH3 and 4-NO2, respectively. All data are at 25 ꢀC.
b
c
Data taken from Ref. 8b.
The data for kH O and kOH were taken from Ref. 7b and 7c. Where needed, the Hammett equation was employed in order to calculate the required rate
2
constant.
d
Data from the present work.
e
Data for ArCO-Phos at 39 ꢀC, from Ref. 5b.
Data for acetyl phosphate, from Ref. 5b.
These are the slopes of plots of the appropriate activation parameters of ArCO-Nu versus Hammett ꢀ of the substituents.
f
g
brevity, we employ the terms benzoate and Iba anions to
indicate the corresponding incipient anions in the TS.
Additionally, ArCO-Phos refers to the acyl phosphate
dianion. The substituent effects on ÁH and ÁS are
listed in the last two rows of Table 7 and correspond
to the slopes of the (reasonably linear) dependence of
these parameters on ꢀX. For comparison, we also include
hydroxide ion (for kOH pathway) and the leaving
benzoate and Iba anions play an important role in
the reactivity of the ArCO-Nu investigated. As the
¼
¼
¼
ÁS versus ꢀX plots clearly indicate, changes in the
solvation of the benzoate anion is substituent depen-
dent since the developing negative charge is competi-
tively stabilized by the solvent and by the substituent
X. Accordingly, electron-withdrawing substituents are
expected to stabilize the benzoate anion and decrease
its solvation. We now examine how this interpretation
applies to the data for the above-discussed reaction
pathways.
published data for the following reaction paths: kH O and
kOH for ArCO-Imz and kspon for acetyl phosphate dianion.
The following points are relevant:
2
1. The relationship between ꢁ and the catalytic pathway
is not straightforward. Therefore, the usefulness of
this parameter in understanding the pathway-deter-
mining factors of these reactions is limited. However,
all ꢁ clearly indicate a moderate to large charge
development in the benzoate moiety of the TSs.
4. First we consider the OHꢁ-mediated reactions of
ArCO-Imz and ArCO-Iba. In these cases, there is no
change in the overall charge but a large charge
reorganization, leading to changes in solvation. Since
the charge on the OHꢁ decreases, its solvation de-
¼
¼
2. The dependence of KSIE and of ÁS on the mechan-
creases and the corresponding contribution to ÁS is
ism is clear. That is, water attack, when catalyzed by a
GB (either another water molecule and/or the buffer)
is associated with a large KSIE and highly negative
entropy of activation. In contrast, the decomposition
of ArCO-Phos is associated with a negligible KSIE
expected to be positive. On the other hand, the charges
on the forming benzoate and/or Iba anions increase,
leading to an increase in solvation and a correspond-
¼
ing decrease in ÁS . The results obtained (magnitude
of ÁS and its dependence on ꢀX) indicate that the
¼
¼
and positive ÁS since its slow step does not involve
water. The less than expected KSIE for kH O of ArCO-
balance between the two opposite effects is different
for the above-mentioned substrates. For ArCO-Imz,
2
¼
Iba has been discussed.
3. More interestingly, the activation parameters, ÁH
ÁS is highly negative and exhibits a small depen-
¼
dence on ꢀX. This indicates little desolvation of the
entering OHꢁ associated with a small solvation of
the leaving benzoate. In contrast, the same pathway
¼
and ÁS , are substituent dependent, as shown in the
last two rows of Table 7. Except for one case, kGB of
¼
¼
ArCO-Imz, the jslopej of the ÁS versus ꢀX plot is
for ArCO-Iba is associated with positive ÁS and
high dependence on the substituent. Both results in-
dicate extensive OHꢁ desolvation, which more than
¼
much larger the corresponding value for ÁH . There-
fore, changes in solvation of the relevant species, the
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

REACTIONS OF BENZOYL DERIVATIVES OF NUCLEOPHILES
181
¼
counterbalances the effects on ÁS of the solvation
of the benzoate and Iba anions. Note that attack of
nucleophiles on the Iba moiety is unhindered since the
can be drawn for the kH O path of the same two ArCO-Nu.
2
In particular, the sites of nucleophilic attack on ArCO-Iba
and ArCO-Imz are different, which leads to very different
ꢁ values. Regarding the spontaneous decomposition of
ArCO-Phos we have shown, for the first time, that the
highly positive activation entropy decreases noticeably
when X is electron withdrawing. We interpreted our
result not only in terms of reactant desolvation (a dianion)
but also of the Hammond postulate, i.e. electron-with-
drawing X induce earlier TSs. Therefore, the previous
assumption on the large solvation requirements of the
phosphate reactions, mainly based on the data for a single
compound (e.g. an acyl phosphate or a phosphate ester),
should be revised since our results indicate little solvation
when X is strongly electron withdrawing. In conclusion,
our data and, in particular, those on the substituent
dependence of the activation parameters shed light on
the importance of solvation in reactions of biochemical
interest.
I—O bond protrudes almost linearly (the O—I—O
24
ꢀ
˚
angle is 165 ), ca 2 A from the heterocyclic ring.
5. Some uncatalyzed water reactions involve large
¼
charge development, therefore ÁS is expected to
be negative and substituent dependent. This is clearly
the case for the kH O pathway of ArCO-Imz. For the
2
¼
ArCO-Iba, ÁS is negative, although less dependent
on X, relative to N-benzoylimidazole. This essentially
reflects the smaller degree of charge development on
the benzoate anion, as indicated by the negligible jꢁj
of this reaction.
6. The unimolecular reaction of ArCO-Phos behaves
differently from the bimolecular reactions, e.g. all
¼
ÁS values are positive and decrease on going from
X ¼ 4-methyl to 4-nitro. This is contrary to what is
expected, based on the effects of X on solvation of the
benzoate anion (see above). Additionally, the asso-
¼
ciated substituent effect on ÁH is not as negligible
as those on the above-discussed bimolecular reactions.
As recently argued10e for monophosphate ester dia-
nions, e.g. 4-XC6H4O-PO32ꢁ, the interaction of an
electron-withdrawing X with the bridging oxygen
leads to slight stretching of the bridging O—P bond.
In the TS, this bond is nearly broken, i.e. the interac-
tion energy with X is expected to be much stronger
than in the RS. Therefore, relative to an electron-
releasing X, there is more charge dispersion in the RS
and an extensive one in the TS when X is electron
withdrawing, in agreement with the very large depen-
Acknowledgments
We thank the Sa˜o Paulo State Research Foundation,
FAPESP, for financial support, and the National Research
Council, CNPq, for research fellowships to W. A.
Rodrigues and M. Ferreira and a research productivity
fellowship to O. A. El Seoud. This work was carried
out within a CAPES/COFECUB bilateral cooperation
project.
¼
dence of ÁS on ꢀX. The positive entropy is a
consequence of these differences in solvation and
the dissociative nature of the TS.
REFERENCES
1. Bender ML, Bergeron RJ, Komiyama M. The Bioorganic Chem-
istry of Enzyme Catalysis. Plenum Press: New York, 1984; 45.
2. Jencks WP. Catalysis in Chemistry and Enzymology. Dover: New
York, 1987; 42, 463.
3. Isaacs NS. Physical Organic Chemistry. Longman: Harlow, 1987;
375.
CONCLUSIONS
4. Carry FA, Richard JS. Advanced Organic Chemistry, Part A (4th
edn). Kluwer: New York, 2000; 449.
The mechanism of reactions of ArCO-Nu is strongly
dependent on the nucleophile and the nature of ArCO.
For example, the OHꢁ-mediated reactions of ArCO-Iba
and ArCO-Imz exhibit very different features. At 25 ꢀC, ꢁ
for ArCO-Iba (0.92) is markedly smaller than that for
ArCO-Imz (1.47), in agreement with different sites of
nucleophilic attack: iodine atom and benzoyl carbon
atom, respectively. More surprisingly, the calculated
5. (a) Jencks WP, Carriuolo J. J. Am. Chem. Soc. 1961; 83: 1743–
1750; (b) Di Sabato G, Jencks WP. J. Am. Chem. Soc. 1961; 83:
4400–4405; (c) Kirby AJ, Jencks WP. J. Am. Chem. Soc. 1965; 87:
3209–3216, 3217–3224; (d) O’Connor CJ, Wallace RG. Aust. J.
Chem. 1984; 37: 2559–2569; (e) Herschlag D, Jencks WP. Aust. J.
Chem. 1989; 111: 7579–7587; (f) Herschlag D, Jencks WP. Aust.
J. Chem. 1990; 112: 1942–1950.
6. (a) Hubbard CD, Kirsch JF. Biochemistry 1972; 11: 2483–2493;
(b) Castro EA, Steinfort GB. J. Chem. Soc., Perkin Trans. 2 1983;
453–457.
¼
ÁS are moderately positive and significantly substitu-
7. (a) Klinmann JP, Thornton ER. J. Am. Chem. Soc. 1968; 90:
4390–4394; (b) Choi M-U, Thornton ER. J. Am. Chem. Soc. 1974;
96: 1428–1436; (c) Palaitis W, Thornton EER. J. Am. Chem. Soc.
1975; 97: 1193–1196; (d) Gopalakrishnan G, Hogg Jl. J. Org.
Chem. 1983; 48: 2038–2043; (e) Barbaro J, Gopalakrishnan G,
Hogg Jl. J. Org. Chem. 1989; 54: 4438–4443.
8. (a) Menegheli P, Farah JPS, El Seoud OA. Ber. Bunsenges. Phys.
Chem. 1991; 95: 1610–1615; (b) El Seoud OA, Menegheli P, Pires
PAR, Kiyan NZ. J. Phys. Org. Chem. 1994; 7: 431–437; (c) El
Seoud OA, Martins MF. J. Phys. Org. Chem. 1995; 8: 637–646;
(d) El Seoud OA, Ruasse M-F, Rodrigues WA. J. Chem. Soc.,
Perkin Trans. 2 2002; 1053–1058.
ent dependent in the reaction of ArCO-Iba and highly
negative and much less dependent on ꢅ for ArCO-Imz.
These entropy data are interpreted in terms of a balance
between OH desolvation and solvation of the forming
benzoate and/or Iba in the TS. Whereas these two
opposite effects are large for ArCO-Iba, they do not
play an important role in case of ArCO-Imz, probably
because of the very different requirements of solvation of
(neutral) Imz and (anionic) Iba. Analogous conclusions
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182

182
O. A. EL SEOUD ET AL.
9. (a) Benkovic SJ, Schray KJ. In Transition States of Biochemical
15. (a) Saunders BC, Stacey GJ, Wilding IGE. Biochem. J. 1942; 36:
368–375; (b) Ashley JN, Barber H, Oanea D, Johnson D. J. Chem.
Soc. 1942; 103–116; (c) Weast RC (ed). Handbook of Chemistry
and Physics (73rd edn). CRC Press: Boca Raton, FL, 1993.
16. Camici G, Manao G, Cappugi G, Ramponi G. Experientia 1976;
32: 535–536.
17. Demhlow EV, Demhlow SS. Phase Transfer Catalysis (2nd edn).
Verlag Chemie: Weinheim, 1983; 14.
18. Showen K. In Transition States of Biochemical Processes,
Gandour RD, Schowen RL (eds). Plenum Press: New York,
1978; 225.
19. (a) Marcus Y. Ion Solvation. Wiley: New York, 1985; 87; (b)
Krestov GA. Thermodynamics of Solvation. Ellis Horwood:
Chichester, 1991; 97.
20. (a) Alvarez F, Schowen RL. In Isotopes in Organic Chemistry,
Buncel E, Lee CC (eds). Elsevier: New York, 1984; 1–60; (b)
Kovach IM, Bennet AJ, Bibbs JA, Zhao Q. J. Am. Chem. Soc.
1993; 115: 5138–5144.
21. Dewar MJS, Storch DM. J. Chem. Soc., Perkin Trans. 2 1989;
877–885.
22. (a) Queen A. Can. J. Chem. 1967; 45: 1619–1629; (b) Felton SM,
Bruice TC. J. Am. Chem. Soc. 1969; 91: 6721; (c) Buncel E,
Raoult A. Can. J. Chem. 1972; 50: 1907–1911; (d) Bhatt MV, Rao
KS, Rao V. J. Org. Chem. 1977; 42: 2697–2702.
Processes, Gandour RD, Schowen RL (eds). Plenum Press: New
York, 1978; 493; (b) Williams A. Acc. Chem. Res. 1989; 22: 387–
392; (c) Thatcher GRJ, Kluger R. Adv. Phys. Org. Chem. 1989;
25: 99–265; (d) Castro EA. Chem. Rev. 1999; 99: 3505–3524.
10. (a) Hollfelder F, Herschlag D. Biochemistry 1995; 34: 12255–
12264; (b) Admiraal SJ, Herschlag D. J. Am. Chem. Soc. 1999;
121: 5837–5845; (c) Admiraal SJ, Herschlag D. J. Am. Chem. Soc.
2000; 122: 2146–2148; (d) O’Brien PJ, Herschlag D. Biochem-
istry 2002; 41: 3207–3225; (e) Cheng H, Nikolic-Hughes I, Wang
JH, Deng H, O’Brien PJ, Wu L, Zhang Z-Y, Herschlag D,
Callender R. J. Am. Chem. Soc. 2002; 124: 11295–11306.
11. (a) Mackay RA, Longo FR, Knier BL, Durst HD. J. Phys. Chem.
1987; 91: 861–864; (b) Moss RA, Scrimin P, Rosen RT. Tetra-
hedron Lett. 1987; 28: 251–254; (c) Moss RA, Wilk B, Krogh-
Jespersen K, Blair JT, Westbrook JD. J. Am. Chem. Soc. 1989;
111: 250–258; (d) Yang Y-C, Baker JA, Ward R. Chem. Rev. 1992;
92: 1729–1743; (e) Moss RA, Zhang H. J. Am. Chem. Soc. 1994;
116: 4471–4472.
12. Berg FJ, Moss RA, Yang YC, Zhang HM. Langmuir 1995; 11:
411–413.
13. (a) Moss RA, Morales-Rojas H, Zhang HM, Park BD. Langmuir
1999; 15: 2738–2744; (b) Moss RA, Morales-Rojas H. Langmuir
2000; 16: 6485–6491.
14. Perrin DD, Armarego WLF, Perrin DR. Purification of Labora-
tory Chemicals (2nd edn). Wiley–Interscience: New York, 1988.
23. Kurz JL, Gutsche CD. J. Am. Chem. Soc. 1960; 82: 2175–2181.
24. Banks DF. Chem. Rev. 1966; 66: 243–266.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 173–182