Effect of Cation-Complexing Agents on the Ba(II)-Assisted Basic Ethanolysis of Phenyl Acetate: From Cation Deactivation to Cation Activation

Article abstract of DOI:10.1021/jo9619995

A thorough investigation of the effect of adding cation complexing agents on the barium-assisted basic ethanolysis of phenyl acetate shows that any catalytic activity disappears upon sequestration of the metal ion by cryptand 222, but addition of the crown ethers 18C6, 15C5, and 12C4 yields ternary complexes of 1:1:1 (crown ether)-(metal ion)-ethoxide composition, in which instead of the expected cation deactivation a definite cation activation takes place. Clear-cut evidence was obtained that the quaternary complex [EtOBa(15C5)₂]⁺, albeit substantially less reactive than the uncomplexed barium-bound ethoxide, is still more reactive than free ethoxide.

Full text of DOI:10.1021/jo9619995

J . Org. Chem. 1997, 62, 3089-3092
3089
Effect of Ca tion -Com p lexin g Agen ts on th e Ba (II)-Assisted Ba sic
Eth a n olysis of P h en yl Aceta te: F r om Ca tion Dea ctiva tion to
Ca tion Activa tion
Roberta Cacciapaglia,* Luigi Mandolini,* and Valeria Van Axel Castelli
Centro CNR di Studio sui Meccanismi di Reazione and Dipartimento di Chimica,
Universita` La Sapienza, 00185 Roma, Italy
Received October 25, 1996X
A thorough investigation of the effect of adding cation complexing agents on the barium-assisted
basic ethanolysis of phenyl acetate shows that any catalytic activity disappears upon sequestration
of the metal ion by cryptand 222, but addition of the crown ethers 18C6, 15C5, and 12C4 yields
ternary complexes of 1:1:1 (crown ether)-(metal ion)-ethoxide composition, in which instead of
the expected cation deactivation a definite cation activation takes place. Clear-cut evidence was
obtained that the quaternary complex [EtOBa(15C5)2]+, albeit substantially less reactive than the
uncomplexed barium-bound ethoxide, is still more reactive than free ethoxide.
In tr od u ction
We have now discovered a reaction system, namely,
the reaction of barium ethoxide with phenyl acetate in
ethanol, in which addition of the metal ion complexing
agents 18-crown-6 (18C6), 15-crown-5 (15C5), and 12-
crown-4 (12C4) does not result in the expected inhibition
of cation participation. The results of this investigation
are reported here and compared with the contrasting
behavior observed with the 222 cryptate of barium ion.
It has been widely recognized that anion properties
(nucleophilicity and basicity) are deeply influenced by
metal counterions1 and, consequently, that addition of
cation complexing agents can significantly alter rates and
equilibria of reactions involving anionic species.2
The concept of anion activation upon addition of crown
ethers and cryptands is well established.3,4 It strictly
applies to cases where free anions are more reactive than
ion pairs and higher aggregates. Less common, yet well
documented, are examples of cation-assisted reactions,
where the cation-paired anion is more reactive than the
free ion. Here cation capture by host ligands is believed
to cause rate decrease through inhibition of cation
participation.3,5 A widely accepted mechanistic criterion
is based on the idea that a rate decrease upon addition
of cation complexing agents provides compelling evidence
for a mechanism in which the metal cation plays a role
as electrophilic catalyst.5,6 Clear-cut examples of reac-
tions where the metal counterion plays an unequivocal
role as Lewis acid catalyst are provided by our studies
of the influence of alkali and alkaline-earth metal ions
on rates of basic methanolysis and ethanolysis of esters.7
Our studies have revealed that in these reactions metal-
bound alkoxide species are more reactive than free
alkoxides, most likely on account of transition state
stabilization via chelate structures having the form of a
four-membered contact ion pair I or of the kinetically
equivalent six-membered solvent shared ion pairs II and
III.
Resu lts a n d Discu ssion
The Me4NOEt-catalyzed ethanolysis of phenyl acetate
(eq 1) is strictly first order in ester substrate and first
order in Me4NOEt. The previously determined8 second-
CH3CO2C6H5 + EtOH Me NOEt8
4
CH3CO2Et + C6H5OH (1)
order rate constant ko ) 1.44 ( 0.03 M-1 s-1 at 25.0 °C
was reproduced with good precision by control experi-
ments in the present work. Rate measurements carried
out in the presence of excess 1.00 and 5.06 mM Me4NOEt
gave rate constant values of 1.41 and 1.43 M-1 s-1
,
respectively. Solutions for rate measurements in the
presence of barium ion were prepared by mixing equimo-
lar amounts of Ba(SCN)2 and Me4NOEt immediately
before use.9,10 These solutions showed enhanced rates
of ethanolysis of phenyl acetate relative to solutions
containing Me4NOEt alone. As shown in previous stud-
(4) Vo¨gtle, F.; Weber, E. Host Guest Complex Chemistry: Synthesis,
Structures, Applications; Springer-Verlag: Berlin, 1985.
(5) Lefour, J .-M.; Loupy, A. Tetrahedron 1978, 34, 2597.
(6) (a) Suh, J .; Mun, B. S. J . Org. Chem. 1989, 54, 2009. (b) Suh, J .;
Heo, J . S. J . Org. Chem. 1990, 55, 5531.
(7) Cacciapaglia, R.; Mandolini, L. Chem. Soc. Rev. 1993, 22, 221.
(8) Cacciapaglia, R.; Mandolini, L.; Reinhoudt, D. N.; Verboom, W.
J . Phys. Org. Chem. 1992, 5, 663.
(9) The thiocyanate salt was given preference over the bromide as
a source of barium because mixtures of the former with Me4NOEt
showed a reduced tendency to yield a crystalline material on standing
(see ref 10).
* To whom correspondence should be addressed. Phone: +39 6
4957808. Fax: +39 6 490421. E-mail: d42500@netmgr.ced.rm.cnr.it.
X Abstract published in Advance ACS Abstracts, April 1, 1997.
(1) Gordon, J . E. The Organic Chemistry of Electrolyte Solutions;
Wiley: New York, 1975; Chapter 3.
(2) Lehn, J .-M. Pure Appl. Chem. 1980, 52, 2303.
(3) Gokel, G. W. Crown Ethers and Cryptands; The Royal Society
of Chemistry: Cambridge, 1991; Chapter 5.
(10) Kraft, D.; Cacciapaglia, R.; Bo¨hmer, V.; Abu El-Fadl, A.;
Harkema, S.; Mandolini, L.; Reinhoudt, D. N.; Verboom, W.; Vogt, W.
J . Org. Chem. 1992, 57, 826.
S0022-3263(96)01999-8 CCC: $14.00 © 1997 American Chemical Society

3090 J . Org. Chem., Vol. 62, No. 10, 1997
Cacciapaglia et al.
Ta ble 2. Effect of Ad d in g Cr yp ta n d 222 a n d 18C6, 15C5,
a n d 12C4 Eth er s to th e Rea ction of P h en yl Aceta te w ith
a 5.0 m M Equ im ola r Mixtu r e of Me4NOEt a n d Ba (SCN)2
in Eth a n ol a t 25.0 °Ca
[ligand]
(mM)
k
[ligand]
(mM)
k
[ligand]
(mM)
k
(M-1 s-1
)
(M-1 s-1
)
(M-1 s-1
)
222
0.89
1.57
1.77
1.92
48.5
44.7
40.7
39.8
2.90
3.55
3.84
4.97
29.4
20.4
17.8
5.88
11.2
16.8
1.55
1.45
1.43
1.60
18C6
1.23
2.49
3.78
49.8
53.0
66.9
4.98
6.14
10.2
81.1
85.1
87.4
14.9
20.5
85.0
85.2
F igu r e 1. Kinetic data for the reaction of phenyl acetate with
equimolar mixtures of Me4NOEt and Ba(SCN)2 in ethanol at
25.0 °C (data from Table 1, entries 1-4). From the slope of
15C5
12C4
2.47
4.93
7.50
49.5
42.1
26.1
9.87
15.0
20.0
9.65
5.03
4.88
30.7
39.9
49.9
4.63
4.25
4.04
the regression line kEtOBa ) 48 M-1 s-1
.
Ta ble 1. P seu d o-F ir st-Or d er Ra te Con sta n ts for th e
Rea ction of P h en yl Aceta te w ith Mixtu r es of Me4NOEt
a n d Ba (SCN)2 in Eth a n ol a t 25.0 °C
2.54
4.97
7.51
59.0
62.4
60.3
12.2
20.3
29.4
50.1
34.8
24.1
49.7
13.3
entry
[Me4NOEt] (mM)
[Ba(SCN)2] (mM)
kobs (s-1
)
a
In the absence of additives kEtOBa ) 48 M-1 s-1
.
1
2
3
4
5
6
7
2.48
4.98
7.60
9.86
5.07
5.05
10.1
2.48
4.98
7.60
9.86
15.1
2.51
5.02
0.121
0.239
0.359
0.461
0.251
0.192
0.385
ies,8,10 a barium-bound ethoxide species is quantitatively
formed upon mixing (eq 2). This species, whose concen-
Me4NOEt + Ba(SCN)2 u
EtOBaSCN + Me4NSCN (2)
tration is stoichiometrically determined by that of the
added reactants, behaves kinetically as a single species.
This is clearly confirmed by the strict linearity of the plot
of kobs vs reactant concentration (Table 1, entries 1-4,
and Figure 1), from which a second-order rate constant
kEtOBa ) 48 M-1 s-1 is obtained. This value is somewhat
smaller than the value of 65 M-1 s-1 obtained for the
ethanolysis of phenyl acetate catalyzed by a barium-
bound ethoxide species prepared using BaBr2 as the
source of barium, but under otherwise identical condi-
tions.8 It seems likely therefore, that the metal-ethoxide
complex is significantly paired to the anion, either
bromide or thiocyanate, contained in the added salt. Also
consistent with the formulation of eq 2 is the finding that
the reaction is insensitive to the amount of Ba(SCN)2
exceeding the molar concentration of ethoxide (Table 1,
entry 5 vs 2). On the other hand, cleavage of phenyl
acetate in solutions in which the ethoxide/barium ratio
is 2:1 (Table 1, entries 6 and 7) takes place with first-
order specific rates, which are significantly higher than
the values calculated for equimolar mixtures of EtO-
BaSCN and Me4NOEt. It is apparent that 1 equiv of
barium ion activates 2 equiv of ethoxide, which strongly
suggests that extensive formation of the ion triplet
(EtO)2Ba takes place (eq 3). If one assumes that ion
F igu r e 2. Influence of added cryptand 222 (data from Table
2). The dashed line is based on eq 4 and the piece-wise full
line is based on eqs 9-11. The horizontal dotted line represents
the reactivity of free ethoxide.
in the 2:1 complex. Comparison of this value with the
values of 48 and 65 M-1 s-1 obtained for the species
EtOBaSCN and EtOBaBr, respectively, shows that the
reactivity of an EtOBaX species is influenced by the
nature of X, but not to a dramatic extent.
In flu en ce of Ca tion -Com p lexin g Agen ts. Addition
of a series of cation complexing agents including cryptand
222 and the crown ethers 18C6, 15C5, and 12C411,12
caused remarkable and largely unpredictable effects on
rates of ethanolysis of phenyl acetate, as shown by the
data listed in Table 2 and plotted in Figures 2-5. In all
these experiments initial concentrations were as fol-
lows: [PhOAc] ) 0.17 mM; [Me4NOEt] ) [Ba(SCN)2] )
5.0 mM.
The only ligand showing the expected cation deactiva-
tion is cryptand 222 (Figure 2). Here, the reactivity drops
upon addition of the ligand until the stoichiometric
amount is added and eventually reaches a plateau value
2Me4NOEt + Ba(SCN)2 u (EtO)2Ba + 2Me4NSCN
(11) No binding data to barium ion in EtOH are available for these
ligands in the extensive compilation published by Izatt and co-workers
(ref 12), where the following log K values in MeOH at 25 °C are
reported: 2.2.2, 12.9; 18C6, 7.3; 15C5, 4.1 (1) and 2.6 (2); 12C4, 2.6 (1)
and 2.4 (2). (1) and (2) refer to 1:1 and 2:1 ligand-metal complexes,
respectively. It is assumed that the known data in MeOH can be taken
as useful approximations for the unknown data in EtOH.
(12) Izatt, R. M.; Pawlak, K.; Bradshaw, J . S.; Breuning, R. L. Chem.
Rev. 1991, 91, 1721.
(3)
triplet formation is quantitative, the kobs values listed in
Table 1, entries 6 and 7, translate into a statistically
corrected second-order rate constant k(EtO) Ba ) 38 M-1
2
s-1 for reaction of phenyl acetate with the ethoxide ion

Ba(II)-Assisted Basic Ethanolysis of Phenyl Acetate
J . Org. Chem., Vol. 62, No. 10, 1997 3091
magnitude as that of the tetramethylammonium cation
is a possibility that, albeit unlikely, cannot be ruled out.
It is worth noting that the reactivity drop upon ligand
addition (Figure 2) does not follow the dashed line of eq
4 expected for a kinetic titration taking place according
k ) kEtOBa - (kEtOBa - ko)x
if 0 e x e 1 (4)
to the simple process described by eq 5. In eq 4 x is
defined as the mole ratio of added ligand to barium. In
the early part of the titration, the rate appears to be
hardly sensitive to the addition of the cryptand. This is
due to the fact that the ethoxide liberated during the first
half-titration (eq 5) reacts with (EtOBa)+ to give the ion
triplet (EtO)2Ba (eq 6), whose statistically corrected
reactivity is only slightly lower than that of (EtOBa)+.
F igu r e 3. Influence of added 18C6 (data from Table 2). The
dashed line is based on eq 12, and the piece-wise dotted line
is based on eqs 9 and 15.
(EtOBa)+ + 222 u EtO- + (222‚Ba)2+
EtO- + (EtOBa)+ u (EtO)2Ba
(5)
(6)
2(EtOBa)+ + 222 u (EtO)2Ba + (222‚Ba)2+ (7)
(EtO)2Ba + 222 u (222‚Ba)2+ + 2EtO-
(8)
The overall process is given in eq 7. Counterions of
charged species are omitted henceforth for simplicity. In
the second half-titration there is a steep reactivity drop
because destruction of (EtO)2Ba takes place according to
eq 8. On the basis of eqs 7 and 8, eqs 9-11 are easily
derived. Figure 2 shows that data points fit the idealized
rate profile given by the piecewise full line based on eqs
9-11 reasonably well.
F igu r e 4. Influence of added 15C5 (data from Table 2). The
horizontal dotted line represents the reactivity of free ethoxide.
k ) kEtOBa - (kEtOBa - k(EtO) Ba)2x
if 0 e x e 0.5
(9)
2
k ) 2k(EtO) Ba - ko - (k(EtO) Ba - ko)2x if 0.5 e x e 1
2
2
(10)
k ) ko
if 1 e x (11)
In the case of 18C6 (Figure 3) addition of the ligand
causes a rate increase, until a plateau value (kEtOBa18C6
) 86 M-1 s-1) is reached when 18C6 is equimolar with
respect to EtOBaSCN. This implies that the crown-
complexed barium ion is still able to bind ethoxide to give
the ternary complex (EtOBa‚18C6)+, which is signifi-
cantly more reactive than the uncomplexed ion pair
(EtOBa)+. The rate increase in the region of x from 0 to
1 does not exhibit the linear dependence (eq 12) expected
on the basis of a complete conversion of (EtOBa)+ into
(EtOBa‚18C6)+ (eq 13), shown as a dashed line in Figure
3.
F igu r e 5. Influence of added 12C4 (data from Table 2). The
horizontal dotted line represents the reactivity of free ethoxide.
(ko ) 1.44 M-1 s-1) which is the same as that measured
in the presence of Me4NOEt alone. This value can be
taken as a measure of the intrinsic reactivity of free
ethoxide ion. It appears therefore that the barium
cryptate behaves as a kinetically inert species,13 which
indicates that “through the window” interactions of the
cryptated metal ion with the ethoxide reactant and with
the anionic transition state are negligibly small. It is
clear, however, that a situation in which the differential
interaction of the cryptated barium ion with reactant and
transition state happens to be exactly of the same
k ) kEtOBa + (kEtOBa18C6 - kEtOBa)x if 0 e x e 1 (12)
(EtOBa)+ + 18C6 u (EtOBa‚18C6)+
(13)
In the first half-titration k appears to be rather
insensitive to the concentration of added ligand, but in
the second half-titration a steep rate increase is observed.
Since it seems very likely that binding of 18C6 to barium
ion is complete under the reaction conditions,11 the
nonlinear shape of the titration plot provides unequivocal
indication that the composition of the system in the
titration experiment is not one in which the fraction of
the crown-complexed metal-bound ethoxide species in-
creases in proportion to the amount of added crown ether.
(13) Interestingly, the 222 cryptate of potassium ion has been
reported to act as a catalyst in the reaction of ethoxide ion with
p-nitrophenyl methanesulfonate in ethanol: Pregel, M. J .; Buncel, E.
J . Am. Chem. Soc. 1993, 115, 10.

3092 J . Org. Chem., Vol. 62, No. 10, 1997
Cacciapaglia et al.
As a reasonable explanation of the observed results, we
suggest that a certain amount of the (EtOBa‚18C6)+
species is transformed into (EtO)2Ba as shown by eq 14,
of significantly lower stability than those formed by 15C5.
Upon addition of 12C4 the rate increases until an x value
of 1 is reached. This clearly indicates the formation of a
more reactive ternary complex (EtOBa·12C4)+. Further
addition of 12C4 causes the gradual conversion of the
ternary complex into the less reactive quaternary com-
plex [EtOBa(12C4)2]+. No plateau is reached in the high
12C4 concentration region, on account of its being a much
weaker binder than 15C5.
(EtOBa‚18C6)+ + (EtOBa)+ a
(Ba‚18C6)2+ + (EtO)2Ba (14)
where two metal species compete for the common EtO-
ligand. That the equilibrium of eq 14 is not completely
shifted to the right is clearly shown by the fact that the
titration profile does not follow the piecewise dotted line
of eq 9 in the first half-titration and eq 15 in the second
half-titration (Figure 3). The actual shape of the titration
Con clu sion s
The results reported in this work show that the barium
ion-catalyzed ethanolysis of phenyl acetate exhibits the
expected cation deactivation upon complexation when the
barium is either sequestered by cryptand 222 or sand-
wiched between two crown ether ligands. However, rate
enhancements are obtained upon formation of 1:1 com-
plexes with 18C6, 12C4, and possibly, also with 15C5.14
This does not necessarily imply that a crown-complexed
barium ion binds to the transition state resembling the
tetrahedral intermediate more strongly than uncom-
plexed barium. It simply means that the differential
energy between transition state and reactant is (slightly)
in favor of the reaction carried out in the presence of 1
mol equiv of crown ether. Our data do not allow a
dissection of catalysis into reactant state and transition-
state effects, but since it seems very likely that ion
pairing is weakened by cation binding to a crown ether,
the enhanced catalysis upon addition of 1 mol equiv of
crown ether is essentially an initial state effect. In other
words, cation binding to a crown ether destabilizes the
barium ethoxide pair more strongly than the barium-
(transition state) pair, with the net result that the
activation free energy is decreased. The finding that a
barium ion can be incorporated into a crown ether
complex of defined geometry and still be catalytically
active will be useful in the design and construction of
supramolecular transacylation catalysts in which the
catalytic site is provided by an alkaline-earth metal ion
bound to a crown ether moiety.
k ) 2k(EtO) Ba - kEtOBa18C6 + (kEtOBa18C6 - k(EtO) Ba)2x
2
2
if 0.5 e x e 1 (15)
plot suggests a more balanced situation. In other words,
addition of 18C6 up to 0.5 molar equiv transforms a
reactive (EtOBa)+ species into a mixture of comparable
amounts of more reactive (EtOBa‚18C6)+ and less reac-
tive (EtO)2Ba, with the net result that modest changes
of the overall reactivity are observed. But in the second
half-titration a gradual transformation of the least reac-
tive (EtO)2Ba into the most reactive (EtOBa‚18C6)+ takes
place (eq 16), and consequently, a steep reactivity in-
18C6 + (Ba‚18C6)2+ + (EtO)2Ba a
2(EtOBa‚18C6)+ (16)
crease is observed. The shape of the titration plot
indicates that transformation of the initial (EtOBa)+
species into the ternary complex (EtOBa‚18C6)+ is virtu-
ally complete upon addition of 1 molar equiv of 18C6 (x
) 1).
A markedly different behavior is experienced by 15C5
(Figure 4). The rate profile shows an apparent insensi-
tivity to addition of small amounts of ligand, but further
addition causes a marked reactivity drop. The plateau
value of 4.1 M-1 s-1 reached when x > 2 is substantially
higher, however, than the second-order rate constant of
1.44 M-1 s-1 for reaction of free ethoxide. Interpretation
of these observations involves the formation of the
ternary complex (EtOBa·15C5)+, which is presumably
more reactive than (EtOBa)+ but involved in a ligand-
exchange process analogous to that described in eq 14.
The two factors appear to be closely balanced, with the
net result of a substantial insensitivity to the addition
of the crown ether in the initial portion of the profile.
The steep reactivity drop caused by further addition of
the ligand is due to the formation of the quaternary
complex [EtOBa(15C5)2]+, which is much less reactive
than (EtOBa‚15C5)+ but still more reactive than free
ethoxide. This interpretation is supported by the well-
documented tendency of 15C5, not experienced by 18C6,
to give stable 2:1 complexes with metal ions.11 A striking
feature emerging from the data is that the metal ion is
still capable of assisting the attack of ethoxide ion on the
ester carbonyl, in spite of its being sandwiched between
two crown ether molecules. It is remarkable that the
transition state of the latter reaction is a complex formed
by no less than five species, namely, substrate ester,
ethoxide nucleophile, metal ion, and two ligand mol-
ecules.
Exp er im en ta l Section
Ma ter ia ls. Phenyl acetate was distilled under vacuum.
Reagent-grade commercial samples of Ba(SCN)2‚3H2O, cryptand
222, and crown ethers were used as received. Other materials,
apparatus, and techniques were as reported previously.10
Kin etics. Reaction progress was monitored by following
absorbance changes due to the release of phenoxide ion at λ )
300 nm. All reactions were carried out under pseudo-first-
order conditions with the base concentration ca. 30 times
greater than substrate concentration. Absorbance readings
spanning 8-15 half-lives were fitted to the integrated first-
order eq (17) by means of a nonlinear least-squares procedure
(At - A0) ) (A∞ - A0)[1 - exp(-kobst)]
(17)
in which A∞ and kobs were treated as adjustable parameters.
Nonlinear least-squares calculations were carried out with the
program SigmaPlot for Windows, 1.02 (J andel Scientific). The
error in kobs is estimated to be not greater than (3%.
Ack n ow led gm en t. Financial contributions from
MURST and from CNR, Progetto Strategico Tecnologie
Chimiche Innovative are greatly acknowledged.
J O9619995
(14) We have deliberately avoided the use of the expression “ligand-
accelerated catalysis” (LAC), because the concept of LAC, as defined
and thoroughly discussed by Sharpless and co-workers (ref 15) in the
context of asymmetric reactions catalyzed by transition metals, is not
strictly applicable to our system.
(15) Berrisford, D. J .; Bolm, C.; Sharpless, K. B. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1059.
The reactivity profile for the 12C4 case (Figure 5)
shares many features with the profile reported in Figure
4 for the 15C5 case. This is not surprising, as 12C4 is
also known to form 2:1 complexes with metal ions,11 albeit