Mechanism of Alkaline Hydrolysis of Some HO-π-COOAr Acyl Derivatives

Article abstract of DOI:10.1021/jo990151o

To gain knowledge on the role played by the nature of the bridge interposed between hydroxyl and carbonyl groups in esters of the title type, in principle able to hydrolyze through dissociative pathways via the conjugate base of the substrate (E1cB mechanism), we have studied the alkaline hydrolyses of 2,4-dinitrophenyl esters in which the π-system is a biphenyl, azobenzene, benzylide-neaniline, or stilbene skeleton. Kinetic data, such as reactivity comparisons and Arrhenius parameters, show that these substrates react through the usual, associative, B_Ac2 mechanism. This outcome is discussed and interpreted from both structural and energetic standpoints. The data suggest that a value of 0.0 is the most appropriate assignment of the σ_p value for the benzylidenamino substituent (C₆H₅CH=N-).

Full text of DOI:10.1021/jo990151o

5422
J . Org. Chem. 1999, 64, 5422-5426
Mech a n ism of Alk a lin e Hyd r olysis of Som e HO-π-COOAr Acyl
Der iva tives
Giorgio Cevasco* and Sergio Thea*
C.N.R., Centro di Studio per la Chimica dei Composti Cicloalifatici ed Aromatici,† Dipartimento di
Chimica e Chimica Industriale, Universita` di Genova, Via Dodecaneso 31, 16146 Genova, Italy
Received J anuary 27, 1999
To gain knowledge on the role played by the nature of the bridge interposed between hydroxyl and
carbonyl groups in esters of the title type, in principle able to hydrolyze through dissociative
pathways via the conjugate base of the substrate (E1cB mechanism), we have studied the alkaline
hydrolyses of 2,4-dinitrophenyl esters in which the π-system is a biphenyl, azobenzene, benzylide-
neaniline, or stilbene skeleton. Kinetic data, such as reactivity comparisons and Arrhenius
parameters, show that these substrates react through the usual, associative, BAc2 mechanism. This
outcome is discussed and interpreted from both structural and energetic standpoints. The data
suggest that a value of 0.0 is the most appropriate assignment of the σp value for the
benzylidenamino substituent (C6H5CHdN-).
We have been interested in the dissociative (E1cB)
of 4-hydroxybenzoates or two π-systems linked by a single
bond in the case of hydroxycinnamates.
mechanism of acyl transfer reactions for over a decade.
We have previously found1 that the hydrolyses of aryl
4-hydroxybenzoates in moderately to strongly alkaline
aqueous solution follow the usual associative route (BAc2)
when the esters possess leaving groups having pKa higher
than about 6.5, whereas esters having leaving groups
with lower pKa hydrolyze through the E1cB mechanism
with the participation of the unprecedented p-oxo ketene
intermediate (1). We have subsequently demonstrated
that the hydrolyses of aryl 4-hydroxycinnamates2 and
2-hydroxycinnamates3 in dioxane-water mixtures be-
have almost in the same way with the participation, in
their dissociative paths, of the “extended” oxo ketene
intermediates 2 and 3, respectively.
It is generally accepted4 that an adequate internal
nucleophilicity of the substrate (which represents the
ability of its ionized form to expel the leaving group), high
nucleofugality of the leaving group, and relatively high
stability of the putative intermediate are the driving
forces for the dissociative pathway.
Our studies1-3 indicate that interposition of an extra
vinylene group between the internal nucleophile and the
reaction center favors the dissociative mechanism. Com-
parisons between the hydrolysis rates of 2′,4′-dinitrophe-
nyl esters of 4-hydroxybenzoic and 2- and 4-hydroxycin-
namic acid indicate that, taking into account the different
acidities of the esters (i.e., the acidity of the phenolic
groups in the esters, which is related to nucleophilicity),
the observed rate enhancement could be due to an
increased stability of the “extended” intermediates that,
in turn, can be ascribed to a more extended delocalization
of π electrons. Analogous results come from our recent
studies5 on dissociative sulfonyl group transfer reactions
where “extended” sulfoquinone intermediates are in-
volved.
Moreover, we have recently suggested6 the occurrence
of the E1cB mechanism in the hydrolysis of an ester, 2,4-
dinitrophenyl 4′-hydroxyphenylpropiolate, possessing a
C-C triple bond as part of the π-conjugated system.
To improve understanding of the effects of specific
structural modifications of the π-backbone on the hydro-
lytic pathways, we have directed our attention toward
other esters possessing different molecular architecture
such as two aromatic π-systems linked by a single bond
or by Z)Y π-conjugated systems with Z and Y being N
or CH.
In this study, we report the results of a kinetic
investigation on the alkaline hydrolyses of some HO-π-
COOAr esters where π is either the biphenyl moiety in
compound 4a or the isoelectronic bridge in azobenzene
(5a ), benzylideneaniline (6a ), and stilbene (7a ) deriva-
Although the presence of an ionizable R-hydrogen had
been previously considered a prerequisite for the occur-
rence of E1cB pathways in ester hydrolysis,4 we have
been able to demonstrate that the dissociative mecha-
nism carries the reaction flux also when the hydroxyl
group is π-conjugated with the reaction center in esters
of the type HO-π-COOAr: the intervening backbone
being a simple π-system, the aromatic ring in the case
† Associated with the Istituto Nazionale di Coordinamento per la
Chimica dei Sistemi Biologici.
(1) (a) Cevasco, G.; Guanti, G.; Hopkins A. R.; Thea, S.; Williams,
A. J . Org. Chem. 1985, 50, 479. (b) Thea, S.; Cevasco, G.; Guanti, G.;
Kashefi-Naini, N.; Williams, A. Ibid. 1985, 50, 1867.
(2) Cevasco, G.; Thea, S. J . Org. Chem. 1994, 59, 6274.
(3) Cevasco, G.; Thea, S. J . Org. Chem. 1995, 60, 70.
(4) Williams, A.; Douglas, K. T. Chem. Rev. 1975, 75, 627-649.
(5) Cevasco, G.; Thea, S. J . Org. Chem. 1998, 63, 2125.
(6) Cevasco, G.; Pardini, R.; Thea, S. Eur. J . Org. Chem. 1998,
5, 665.
10.1021/jo990151o CCC: $18.00 © 1999 American Chemical Society
Published on Web 06/24/1999

Alkaline Hydrolysis of Acyl Derivatives
J . Org. Chem., Vol. 64, No. 15, 1999 5423
tives. Since the E1cB mechanism requires π-system
F igu r e 1. pH-rate profiles for the hydrolysis of 2,4-dinitro-
phenyl esters 5a (triangles) and 6a (circles) in 40% dioxane
buffers at 25 °C and ionic strength 0.1 M (KCl). Lines are
calculated from eq 1.
Resu lts a n d Discu ssion
The hydrolyses of 2,4-dinitrophenyl derivatives 4a -
7a and, for the sake of comparison, those of the corre-
sponding esters devoid of acidic hydrogen 4b-7b were
carried out under pseudo-first-order conditions in 40%
dioxane/water (v/v) at 25 °C and ionic strength held
constant (0.1 M) with added potassium chloride. The
progress of the reactions was monitored by following the
variation of absorbance due to the disappearance of the
substrate or liberation of products. The products of the
reactions of esters 4-7 in alkaline solution were identi-
fied as 2,4-dinitrophenoxide ion and the appropriate acid.
This was achieved by comparison of the UV-vis spectra
after completion of the reactions with authentic samples
of these products under the same conditions. Esters 6a ,b
are Schiff bases, and therefore, they may undergo hy-
drolysis of the azomethine group; however, control ex-
periments clearly showed that, under the conditions
employed in this work, this is a secondary reaction that
takes place only after completion of the hydrolysis of
carboxylic ester group. However, for this reason, the pH-
rate profile (see below) for hydrolysis of ester 6a was
stopped at pH ca. 9.5.
planarity to allow the conjugative interaction between
the hydroxyl group and the carbonyl carbon atom and
(E)-azobenzene and (E)-stilbene derivatives are thought
to be essentially planar in the solid state and in solu-
tion,7,8 it is in principle possible that the hydrolyses of
(E)-5a and (E)-7a follow the dissociative pathway with
the participation of unsaturated intermediates such as
8. On the contrary, benzylideneanilines are generally not
planar in those phases,7-9 and since the aniline ring is
severely twisted out of the C-NdC-C plane, 6a should
hydrolyze through the usual associative BAc2 mechanism.
As far as 4a is concerned, it is well-known that biphenyl
itself is planar in the solid state, whereas in the gas phase
and in solution the phenyl rings are twisted.10 These
observations are in agreement with the outcome of
theoretical molecular mechanics calculations11 that rota-
tion around the bond between the phenyl rings occurs
readily. In this case, it is reasonable to suppose that for
the alkaline hydrolysis of 4a a dissociative mechanism
through the intermediate 9 is a possible alternative to
the usual associative route.
The rates of hydrolysis of the esters 4a -7a were found
to obey eq 1 and are illustrated as pH-rate profiles in
Figures 1 and 2.
kobs ) (ka + kb[OH-])/(1 + aH/Ka)
(?)
Finally, it must be emphasized that analogues of
compounds 5-7 are currently target molecules not only
for conformational8 and for Z/ E isomerization12 studies
but also for the nonlinear optical (NLO) properties of
push-pull derivatives.7,13
In eq 1, aH is the proton activity and Ka is the ionization
constant of the hydroxyl group of the ester, which is
responsible for the curvature in the pH-rate profiles. The
Ka values were separately determined, and the kinetic
constants can be calculated, knowing the Ka value, from
primary kinetic data, by iterative nonlinear curve-fitting
performed with the Fig.P program.14 The values of the
kinetic parameters ka, the pseudo-first-order rate con-
stant in the plateau region of the pH-rate profile
(actually the plateau region is only an inflection in the
case of the stilbene derivative 7a and is very narrow in
the other cases), and kb, the second-order term related
to the bimolecular attack of hydroxide ion on the ionized
ester, are reported in Table 1 together with parameters
and conditions relevant to the hydrolyses of these esters.
(7) For example: van Walree, C. A.; Franssen, O.; Marsman, A. W.;
Flipse, M. C.; J enneskens, L. W. J . Chem. Soc., Perkin Trans. 2 1997,
799 and the following article in the issue.
(8) For example: Zamir, S.; Bernstein, J .; Ioffe, A.; Brunvoll, J .;
Kolonits, M.; Hargittai, I. J . Chem. Soc., Perkin Trans. 2 1994, 895
and references therein.
(9) El-Bayoumi, M. A.; El-Aasser, M.; Abdel-Halim, F. J . Am. Chem.
Soc. 1971, 93, 586.
(10) J affe`, H. H.; Orchin, M. Theory and Applications of Ultraviolet
Spectroscopy; Wiley: New York, 1962; pp 402-403.
(11) PCModel, Serena Software, Bloomington, IN, 1993.
(12) For example: (a) Cimiraglia, R.; Asano, T.; Hofmann, H.-G.
Gazz. Chim. Ital. 1996, 126, 679. (b) Sanchez, A. M.; de Rossi, R. H. J .
Org. Chem. 1995, 60, 2974.
(13) For example: Albert, I. D. L.; Marks, T. J .; Ratner, M. A. J .
Am. Chem. Soc. 1998, 120, 11174 and references therein.
(14) Program Fig.P from Biosoft, Cambridge, U.K., 1991.

5424 J . Org. Chem., Vol. 64, No. 15, 1999
Cevasco and Thea
Ta ble 1. Ion iza tion a n d Ra te Con sta n ts for Ester s 4a -7a in 40% Dioxa n e a t 25 °C a n d µ ) 0.1 (p Kw ) 15.00)
a
c
substrate
pKa
λ,b nm
ka, s-1
kb, M-1 s-1
kapp
Nd
pHe
4a
5a
6a
7a
10.33
8.81
360
480
400
400
(2.76 ( 0.44) × 10-4
(8.07 ( 0.87) × 10-5
(5.89 ( 1.33) × 10-5
(2.05( 0.12) × 10-4
1.90 ( 0.30
8.42 ( 0.62
3.34 ( 0.37
3.01 ( 0.14
13
125
41
10
10
8
8.4-14.0
8.3-14.0
9.5-14.0
8.7-14.0
9.16
10.46
7
11
The error on these values is less than 0.01 pK unit. Wavelength employed for pKa determination. c See text. Number of data points,
a
b
d
not including duplicates. e pH range employed. See text for the wavelength employed for kinetics.
customary, to the calculated apparent second-order rate
constant for the hydrolysis of esters 4a -7a (kapp ) kaKa/
Kw). As shown by the data in Table 2, such values are
equal or only slightly higher than the kOH values deter-
mined for the esters 4b-7b.
On the contrary, in the case of 2,4-dinitrophenyl 4′-
hydroxybenzoate, the ratio between kapp and kOH for the
corresponding methoxy derivative is large (ca. 280),6 thus
suggesting that the two esters hydrolyze through a
different mechanism. Therefore, the present results seem
to indicate that all the substrates investigated in this
work (4-7) hydrolyze through the same mechanism, i.e.,
the associative one; in particular, there is no reason to
invoke the participation of the E1cB mechanism in the
hydrolysis of esters 4a , 5a , and 7a .
To further substantiate this conclusion, we have also
calculated the activation parameters for the hydrolysis
reaction of the 2,4-dinitrophenyl ester of 4-(4′-hydrox-
yphenyl)azobenzoic acid (5a ). Arrhenius parameters have
been used16 to distinguish E1cB from BAc2 mechanisms
since the bimolecular process should show a considerably
more negative entropy of activation than E1cB process,
in which the rate-determining step is unimolecular. The
activation parameters for the hydrolysis of 5a have been
determined in the range 17-35 °C and are as follows:
∆Sq ) -25.7 ( 1.7 eu, ∆Hq ) 15.3 ( 0.5 kcal/mol. The
largely negative value of ∆Sq for the hydrolysis of this
ester is well within the range expected for an associative
process,17 thus excluding the incursion of a dissociative
pathway in this reaction.
F igu r e 2. pH-rate profiles for the hydrolysis of 2,4-dinitro-
phenyl esters 7a (triangles) and 4a (circles) in 40% dioxane
buffers at 25 °C and ionic strength 0.1 M (KCl). Lines are
calculated from eq 1 (one of them is dashed for the sake of
clarity).
Ta ble 2. Secon d -Or d er Ra te Con sta n ts for th e
Hyd r olysis of Ester s 4b-7b in 40% Dioxa n e a t 25 °C
a n d µ ) 0.1
b
substrate
λ,a nm
kOH, M-1 s-1
kapp/ kOH
4b
5b
6b
7b
400
360
400
410
12.06 ( 0.04
38.07 ( 1.80
8.95 ( 0.03
6.91 ( 0.25
1.1
3.3
4.6
1.0
a
b
Wavelength employed for the kinetic runs. Values of kapp are
taken from Table 1.
In Table 1 the spectrophotometrically determined pKa
values are reported as well.
Therefore, although possible in principle, the E1cB
mechanism does not carry the reaction flux in the
hydrolyses of the esters studied in this work.
Since BAc2 and E1cB mechanisms in acyl group trans-
fer reactions of ionizable substrates obey the same rate
equation (i.e., eq 1), they cannot be differentiated kineti-
cally, and therefore, other tools must be employed in
order to elucidate the reaction mechanism. However, the
limited width of plateau region in the pH-rate profiles
would not suggest that the dissociative mechanism is
occurring.
One of these criteria is offered by reactivity compari-
sons with substrate models lacking the ionizable center
and that, therefore, cannot react through the E1cB
mechanism. In the present case, esters 4b-7b represent
suitable models. Their alkaline hydrolyses obey eq 2,
where Kw is the ionic product of water in the employed
medium (pKw )15.00 at 25 °C)15 and kOH is the second-
order rate constant related to the BAc2 attack of hydroxide
ion on the substrate. The relevant parameters are
recorded in Table 2.
As far as compounds 4a and 6a are involved, this result
could be ascribed to a diminished conjugative interaction
between the hydroxyl and the ester groups because of
deviation from planarity of the π-system (vide supra). On
the other hand, comparison between the hydrolyses of
ester 7a (whose π-system is planar or nearly so) and the
parent 2,4-dinitrophenyl 4′-hydroxycinnamate (which
follows the E1cB hydrolytic mechanism2) seems to indi-
cate that loss of aromaticity involved in reaching the rate-
determining transition state of the E1cB route is large
enough to make this pathway unfavorable, with respect
to the associative one, when two (rather than one)
p-phenylene units are part of the π-bridge. Of course, the
same factor can be invoked for ester 5a , and esters 4a
and 6a as well, as their π-systems include two aromatic
rings.
Since the present results give no specific indication on
the effect of substitution of sp2 nitrogen atoms for sp2
kobs ) kOHKw/aH
(2)
(16) Vlasak, P.; Mindl, J . J . Chem. Soc., Perkin Trans. 2 1997, 1401-
1403. Safraoui, A.; Calmon, M.; Calmon, J .-P. J . Chem. Soc., Perkin
Trans. 2 1991, 1349-1352. Broxton, T. J . Aust. J . Chem. 1985, 38,
77-83.
(17) Isaacs, N. S. Physical Organic Chemistry; Longman S&T:
Harlow, 1987; Chapter 10.6.
To carry out the aforementioned reactivity comparisons
between the hydrolyses of these esters we resort, as is
(15) Arned, H. S.; Fallon, L. J . Am. Chem. Soc. 1931, 61, 2374-
2378.

Alkaline Hydrolysis of Acyl Derivatives
J . Org. Chem., Vol. 64, No. 15, 1999 5425
analytical reagent grade. Water was double distilled and
preboiled to free it from dissolved carbon dioxide. Dioxane was
purged of peroxides by passage of the analytical-grade product
through an activated alumina column under nitrogen; the
absence of peroxides was checked by the KI test. The 1H NMR
spectra were recorded with a Varian Gemini 200 spectrometer
(200 MHz) with TMS as internal standard and acetone-d6 or
DMSO-d6 as solvent.
Syn th esis. The esters were prepared by condensation of
the appropriate acid (commercial, unless otherwise stated)
with 2,4-dinitrophenol by means of dicyclohexylcarbodiimide
in ethyl acetate with pyridine as catalyst. After usual workup
of the reaction, the resulting ester was recrystallized to
constant mp. The structures of the synthesized acids and esters
1
were confirmed by H NMR spectroscopy. The esters were as
follows; mp is given together with analytical data.
2,4-Din itr op h en yl 4-(4′-h yd r oxyp h en yl)ben zoa te (4a ):
mp 186-7 °C (from toluene). Anal. Calcd for C19H12N2O7: C,
60.0; H, 3.2; N, 7.4. Found: C, 60.4; H, 3.3; N, 7.2. 2,4-
Din itr op h en yl 4-p h en ylben zoa te (4b): mp 138-9 °C (from
toluene). Anal. Calcd for C19H12N2O6: C, 62.6; H, 3.3; N, 7.7.
Found: C, 63.2; H, 3.5; N, 7.7. 2,4-Din itr op h en yl 4-(4′-
h yd r oxyp h en yl)a zoben zoa te (5a ): mp 166-7 °C (from
methanol). Anal. Calcd for C19H12N4O7: C, 55.9; H, 2.9; N, 13.7.
Found: C, 55.6; H, 2.9; N, 13.5. The starting 4-(4′-hydrox-
yphenyl)azobenzoic acid (in the stable E form)12b,21 was pre-
pared as described in the literature.22 2,4-Din itr op h en yl
4-(4′-m eth oxyp h en yl)a zoben zoa te (5b): mp 183-4 °C (from
toluene). Anal. Calcd for C20H14N4O7: C, 56.9; H, 3.3; N, 13.3.
Found: C, 57.3; H, 3.4; N, 13.0. The starting acid was obtained
by methylation with dimethyl sulfate of 4-(4′-hydroxyphenyl)-
azobenzoic acid (see above). 2,4-Din itr op h en yl N-(4′-h y-
d r oxyb en zylid en e)-4-a m in ob en zoa t e (6a ): this ester,
repeatedly crystallized from toluene, does not have a clear
mp, most likely owing to mesomorphism. Anal. Calcd for
F igu r e 3. Hammett-σ relationship for the alkaline hydrolysis
of 2,4-dinitrophenyl para-substituted benzoates. Kinetic data
from Table 2 and ref 6; the σ values are taken from ref 20
(however, see text).
carbon atoms in the conjugated backbone on the dissocia-
tive mechanism, further work is required to shed light
on this topic.
However, our data allow the correct assignment of the
σp value for the benzylidenamino group (C6H5CHdN-)
to be made; indeed, two strikingly different values have
been reported in the literature for this group, namely
0.018 and -0.55,19 although they were both based on
ionization of substituted benzoic acids. The last value,
although the authors themselves suspected it could be
affected by hydrolysis of the azomethine group during
pK determination, seems to be currently given credit (it
has been recently reported in an authoritative review20).
Since it is reasonable to assume that in compounds
such as 5-7 the electronic effect of the whole group
XC6H4ZdY is primarily due to the nature of Z and Y (the
effect of X should be quite attenuated by the two
interposed groups), the polar effect of MeOC6H4ZdY
should not be too much different from that of C6H5Z)Y
(other things being equal). Now, from the data shown in
Table 2, and previous data of ours on the hydrolysis of
2,4-dinitrophenyl 4′-methoxybenzoate under the same
conditions (kOH ) 2.25 ( 0.03 M-1 s-1),6 it is possible to
ascertain that the second-order rate constant for the
hydrolysis of 2,4-dinitrophenyl N-(4′-methoxybenzylidene)-
p-aminobenzoate (6b) fits well a Hammett relationship,
as shown in Figure 3, provided that the value σp ) 0.0
for this substrate is employed. It is worth noting that
the derived Hammett sensitivity (F ) +2.1) is in good
agreement with the value previously determined for the
alkaline hydrolysis of 2,4-dinitrophenyl benzoates in
water (+1.97).1a Therefore, in light of the above assump-
tion, we suggest that 0.0 is the most reliable value of σp
for the benzylideneamino substituent. This outcome
fulfills the expectation that the electron-withdrawing
effect of the whole group should increase as the number
of nitrogen atoms in the bridge increases.
C
20H13N3O7: C, 59.0; H, 3.2; N, 10.3. Found: C, 58.6; H, 3.4;
N, 10.0. Starting N-(4′-hydroxybenzylidene)-4-aminobenzoic
acid was prepared, according to standard procedure, by re-
fluxing overnight an ethanolic solution of p-hydroxybenzalde-
hyde and p-aminobenzoic acid. The cooled solution offered,
after removal of the solvent, a crystalline, yellow solid. The
1H NMR spectrum (DMSO-d6) of this product, melting with
decomposition at 247-8 °C (lit.23 mp 229-230 °C dec),
confirmed the expected structure. 2,4-Din itr op h en yl N-(4′-
m eth oxyben zylid en e)-4-a m in oben zoa te (6b): mp 170-1
°C (from toluene). Anal. Calcd for C21H15N3O7: C, 59.9; H, 3.6;
N, 10.0. Found: C, 59.9; H, 3.5; N, 10.0. Starting N-(4′-
methoxybenzylidene)-4-aminobenzoic acid was prepared by
refluxing overnight an ethanolic solution of p-methoxybenzal-
dehyde and p-aminobenzoic acid. A pinkish crystalline solid
melting at 188-9 °C was obtained upon cooling. 2,4-Din itr o-
p h en yl 4′-h yd r oxy-tr a n s-stilben e-4-ca r boxyla te (7a ): mp
185-6 °C (from toluene). Anal. Calcd for C21H14N2O7: C, 62.1;
H, 3.5; N, 6.9. Found: C, 62.3; H, 3.6; N, 6.8. The 1H NMR
spectrum (acetone-d6) of this ester showed, together with the
typical, expected doublets due to the rings protons, the
following signals: δ 7.46 (d, 1, J ) 16.5 Hz), 7.20 (d, 1, J )
16.4 Hz). These vinylic coupling constants undoubtedly indi-
cate a trans stereochemistry.
4′-Hydroxystilbene-4-carboxylic acid was prepared as fol-
lows. Ethyl R-bromo-p-toluate (obtained by esterification of the
commercial acid) was treated with triphenylphosphine in
anhydrous acetonitrile, under nitrogen at rt for 2 days,
affording, in good yield, the corresponding benzyltriphenylphos-
phonium bromide. This salt was reacted with 4-acetoxyben-
zaldehyde following a recently reported24 modified Wittig
reaction employing 18-crown-6 and solid potassium hydroxide
in dichloromethane. This reaction furnished, in good yield,
Exp er im en ta l Section
Gen er a l Meth od s. Starting reagents and solvents were
purified and/or distilled before use. Buffer materials were of
(21) Zollinger, H. Azo and Diazo Chemistry; Interscience Publish-
ers: New York, 1961; Chapter 4.
(18) Ryan, J . J .; Humffray, A. A. J . Chem. Soc. B 1966, 842.
(19) Exner, O.; Lakomy, J . Collect. Czech. Chem. Commun. 1970,
35, 1371.
(22) Cohen, P. P.; McGilvery, R. W. J . Biol. Chem. 1946, 166, 261.
(23) Senier, A.; Forster, R. B. J . Chem. Soc. 1914, 105, 2469.
(24) Bellucci, G.; Chiappe, C.; Lo Moro, G. Tetrahedron Lett. 1996,
37, 4225.
(20) Hansch, C.; Leo, A.; Taft, R. V. Chem. Rev. 1991, 91, 165.

5426 J . Org. Chem., Vol. 64, No. 15, 1999
Cevasco and Thea
1
nearly a 1:1 mixture of the E/Z isomers as judged by H NMR
spectroscopy (fortunately, in acetone-d6 the two isomers have
well-resolved and easily interpretable spectra). Our attempts
to obtain complete E stereoselectivity by means of benzyl-
diphenylchlorophosphonium bromide24 failed. The mixture of
isomers was subsequently hydrolyzed in aqueous sodium
hydroxide by refluxing until complete dissolution occurred. The
cooled solution was acidified, affording a crystalline precipitate
that was identified as 4′-hydroxystilbene-4-carboxylic acid in
ly: the choice of the appropriate wavelength was dictated by
the pH of the buffers employed in the particular kinetic run
since the ionization of the hydroxyl group of both substrates
and liberated acids in alkaline solutions causes large shift in
the UV-vis spectra. The buffered solution (2.5 mL) was
equilibrated to the required temperature ((0.1 °C) in a 1-cm
path-length quartz cell placed in the thermostated cell holder
of a Kontron Uvikon 941 spectrophotometer. The reaction was
initiated by adding 10 µL of a stock solution of the substrate
ca. 0.01 M in dioxane to the buffer, and automated acquisition
of 50-200 data points for each kinetic run was performed.
Reactions were carried out with potassium hydroxide at
different concentrations and with phosphate, borate, carbon-
ate, ammonia, and Tris buffers. In all cases, at least three
different buffer concentrations, at constant pH, were em-
ployed: when buffer effects were observed the rate constants
at zero buffer concentration were obtained by extrapolation.
The ionic strength was kept at 0.1 M with KCl. The pH of the
buffered solutions were measured before and after each kinetic
run using a Radiometer PHM62 meter equipped with a Ross
combined electrode, calibrated with standard buffers. All pH
values quoted for the dioxane-water solutions are relative
values measured directly, no further corrections being applied.
The pseudo-first-order rate constants (kobs) were obtained by
NLLSQ fitting of absorbance vs time data, and the values
reported are the averages of at least duplicate runs. Reactions
were normally followed over about 7 half-lives.
1
a ca. 1:1 mixture of the E/Z isomers as estimated by H NMR
spectroscopy (again the spectra of the two isomers are well
resolved in DMSO-d6). After condensation with 2,4-dinitro-
phenol in the presence of dicyclohexylcarbodiimide, the usual
workup of the reaction afforded again a mixture of the E/Z
isomers of the desired ester as shown by the 1H NMR
spectrum. Finally, the cis isomer was easily and completely
converted into the trans isomer simply by treating a solution
of the mixture of isomers in dichloromethane with a catalytic
amount of iodine in daylight for a few hours.25 2,4-Din itr o-
p h en yl 4′-m eth oxy-tr a n s-stilben e-4-ca r boxyla te (7b): mp
201-2 °C (from toluene). Anal. Calcd for C22H16N2O7: C, 62.9;
H, 3.8; N, 6.7. Found: C, 62.9; H, 4.0; N, 6.9. The 1H NMR
spectrum (DMSO-d6) of this ester showed, together with the
typical, expected doublets due to the rings protons, the
following signals: δ 7.51 (d, 1, J ) 16.6 Hz), 7.27 (d, 1, J )
16.4 Hz), 3.82 (s, 3). These vinylic coupling constants again
indicate a trans stereochemistry. 4′-Methoxystilbene-4-car-
boxylic acid and the required E ester were prepared following
the procedure described above for the corresponding hydroxylic
acid employing 4-methoxybenzaldehyde instead of 4-acetoxy-
benzaldehyde.
Ion iza tion Con sta n ts. The determinations of pKa values
were carried out by spectrophotometric titration of the ioniz-
able substrates employing at least seven buffered solutions
for each determination and extrapolating the absorption to
zero time.
Meth od s
Kin etics. The hydrolyses of the esters 4-7 in 40% v/v
dioxane-water solvent were followed spectrophotometrical-
(25) Yang, J .-H.; Shi, L.-L.; Xiao, W.-J .; Wen, X.-Q.; Huang, Y.-Z.
Heteroatom. Chem. 1990, 1, 75.
J O990151O