Kinetics and mechanism of aminolysis of phenyl acetates in aqueous solutions of poly(ethylenimine)

Article abstract of DOI:10.1021/jo9514681

Second-order rate constants (k_n) for the aminolysis of some phenyl acetates with poly(ethylenimine) (PEI) were obtained in a pH range 4.36-11.20 at 25 °C in 1 M KC1. Linear Bronsted-type plots (log k_n vs pK_N of PEI) were found for less reactive esters 2-nitrophenyl acetate, 4-acetoxy-3-chlorobenzoic acid, and 4-acetoxybenzenesulfonate with slopes of 0.92, 0.99, and 0.82, respectively. Curved plots were obtained for 3-acetoxy-2,6-dinitrobenzoic acid and 4-acetoxy-3-nitrobenzene-sulfonate, which are consistent with a stepwise reaction. The most likely mechanism involves the existence of a tetrahedral intermediate (T^±) and a change in the rate-determining step from its breakdown to its formation when the basicity of the polyamine increases. A semiempirical equation was used to calculate the values of limiting slopes of the plots (0.9 and 0.1 for both esters) and pK_N at the center of the curvature of the plots (p-K_N° = 7.94 and 9.02, respectively). The values of pK_N° are lower than those estimated for the aminolysis of the same esters with simple monomeric amines (pK_n° > 11) because of a better leaving ability of the aryl oxide ion from the tetrahedral intermediate when amino groups of PEI instead of simple amines are involved. Estimation of the pK's of the reactive intermediates and of the microscopic rate constants for the proton transfer from T_± to PEI or from PEIH⁺ to T^± indicates that either base or acid catalysis is unimportant in the aminolysis of these esters by PEI.

Full text of DOI:10.1021/jo9514681

1682
J . Org. Chem. 1996, 61, 1682-1688
Kin etics a n d Mech a n ism of Am in olysis of P h en yl Aceta tes in
Aqu eou s Solu tion s of P oly(eth ylen im in e)
Antonio Arcelli* and Carlo Concilio
Dipartimento di Chimica “G. Ciamician”, via Selmi 2, 40126 Bologna, Italy
Received August 8, 1995X
Second-order rate constants (kn) for the aminolysis of some phenyl acetates with poly(ethylenimine)
(PEI) were obtained in a pH range 4.36-11.20 at 25 °C in 1 M KCl. Linear Bronsted-type plots
(log kn vs pKN of PEI) were found for less reactive esters 2-nitrophenyl acetate, 4-acetoxy-3-
chlorobenzoic acid, and 4-acetoxybenzenesulfonate with slopes of 0.92, 0.99, and 0.82, respectively.
Curved plots were obtained for 3-acetoxy-2,6-dinitrobenzoic acid and 4-acetoxy-3-nitrobenzene-
sulfonate, which are consistent with a stepwise reaction. The most likely mechanism involves the
existence of a tetrahedral intermediate (T() and a change in the rate-determining step from its
breakdown to its formation when the basicity of the polyamine increases. A semiempirical equation
was used to calculate the values of limiting slopes of the plots (0.9 and 0.1 for both esters) and pKN
at the center of the curvature of the plots (pKN° ) 7.94 and 9.02, respectively). The values of pKN°
are lower than those estimated for the aminolysis of the same esters with simple monomeric amines
(pKn° > 11) because of a better leaving ability of the aryl oxide ion from the tetrahedral intermediate
when amino groups of PEI instead of simple amines are involved. Estimation of the pK’s of the
reactive intermediates and of the microscopic rate constants for the proton transfer from T( to PEI
or from PEIH+ to T( indicates that either base or acid catalysis is unimportant in the aminolysis
of these esters by PEI.
In tr od u ction
the basicity of PEI changes with the ionization degree
(i.e., the pH) because of the strong interactions between
vicinal charged and uncharged amino groups along the
polymeric chain.7a,b This peculiarity leads to a large
interval of basicity of the nucleophilic groups without a
change in the structure as generally occurs with different
amines. In a continuation of our studies on the aminoly-
sis of phenyl acetates by PEI,5a we investigated the
dependence of the rate of aminolysis in 1 M KCl on the
basicity of PEI and on the leaving group ability, with the
aim of identifying structure-reactivity relationships
useful to the elucidation of the reaction mechanism. The
final purpose is to extend this approach in order to
analyze the behavior of the said substrates with PEI in
the absence of added electrolytes, when the polyamine
causes a 105-fold rate enhancement, but association and/
or inhibition phenomena result in a complex kinetic
behavior.9 One of the esters examined was also reacted
with a set of monomeric amines in order to compare the
behavior of the two kinds of nucleophiles.
The mechanism of acetyl-transfer reactions involving
small molecules has been subject to extensive investiga-
tion owing to the importance of these reactions in
chemistry as well as in biochemistry.1,2 Structure-
reactivity correlations such as Hammett or Bronsted-type
relationships help to clarify the rate-determining steps
and to determine whether the reaction proceeds stepwise
or by a concerted pathway.3 Indeed, only a few papers
have been published which deal with these subjects in
systems involving esters of low molecular weight and
polyelectrolytes which have reactive nucleophilic groups
in the chain.4,5a-c Substrate-polyelectrolyte interactions
create inhomogeneity in the solution; the local concentra-
tion of the reagents, the basicity of the nucleophilic
groups, or even the solvent properties in the polyelectro-
lyte microenvironment cannot be determined quantita-
tively, so it is difficult to investigate the reaction mech-
anism in detail.6 However, when the ionic strength is
largely increased by addition of strong electrolytes such
as KCl, the substrate-polyelectrolyte interactions are
suppressed and accurate selective rate constants can be
measured.5a
Exp er im en ta l Section
Ma t er ia ls. 3-Ch lor op h en yl acetate, 2-nitrophenyl ace-
tate, 3-acetoxy-2,6-dinitrobenzoic acid, 4-acetoxy-3-nitroben-
zenesulfonate (natrium salt), 4-nitrophenyl acetate; 4-acetoxy-
benzoic acid, 4-acetoxy-3-nitrobenzoic acid, 4-acetoxybenzene-
sulfonate (natrium salt), and 4-acetoxy-3-chlorobenzoic acid
were synthesized and purified according to ref 5a.
Poly(ethylenimine) (PEI) was “Polymin P” 47.6% by weight
from B.D.H. A monomer molecular weight of 59 was deter-
mined by titration with HCl.5b Amine hydrochlorides were
recrystallized before use. All other reagents were of analytical
grade. Water was deionized and then redistilled from KMnO4.
Buffer solutions of amines were prepared from amine
hydrochlorides with the pH adjusted with 0.1-1 M NaOH or
A rather well known polyelectrolyte is poly(ethylen-
imine) (PEI) (Chart 1) which has a branched, very
compact structure and only
a limited expansion
possibility.7a,8 It contains 25% primary, 50% secondary,
and 25% tertiary amino groups. Unlike simple amines,
X Abstract published in Advance ACS Abstracts, February 15, 1996.
(1) J encks, W. P. Catalysis in Chemistry and Enzimology; McGraw
Hill: New York, 1969; p 463.
(2) Page, M. I.; Williams, A. Enzyme Mechanism; Royal Society of
Chemistry: London, 1987; p 159.
(3) (a) Williams, A. Chem. Soc. Rev. 1994, 93; (b) Adv. Phys. Org.
Chem. 1992, 27, 1.
(4) Lege, C. S.; Deyrup, J . A. Macromolecules 1981, 14, 1629, 1634.
(5) (a) Arcelli, A.; Concilio, C. J . Chem. Res., Synop. 1992, 8; (b) J .
Chem. Soc., Perkin Trans. 2 1983, 1327; (c) Gazz. Chim. Ital. 1984,
114, 271; (d) J . Chem. Soc., Perkin Trans. 2 1989, 887.
(6) Sugimara, M.; Okubo, T.; Ise, N. Macromolecules, 1981, 14, 124.
(7) (a) Bloys van Treslong, C. J . Recl. Trav. Chim. Pays-Bas 1978,
97, 13. (b) Bloys van Treslong, C. J .; Staverman, A. J . Recl. Trav.
Chim. Pays-Bas 1974, 93, 171.
(8) Kobayashi, S.; Hiroishi, K.; Tokunoh, M.; Saegusa, T. Macromol-
ecules 1987, 20, 1496.
(9) Arcelli, A.; Concilio, C. Manuscript in preparation.
0022-3263/96/1961-1682$12.00/0 © 1996 American Chemical Society

Aminolysis of Phenyl Acetates
J . Org. Chem., Vol. 61, No. 5, 1996 1683
Ch a r t 1
in 0.05 M KH2PO4/Na2HPO4 or 0.05 M NaHCO3-NaOH buffer
carrier. Ionic strength was maintained at 1.0 M with added
KCl. PEI buffers were prepared from a stock solution (0.5
monomer mol L-1) at 1.0 M KCl final concentration, with the
pH adjusted with 0.1-5 M HCl. Solutions were prepared and
kept under nitrogen atmosphere and used within 1 day.
P oten tiom etr ic Titr a tion s. pH measurements and po-
tentiometric titrations were performed with a Knick 802 pH-
meter and an Ingold HA 405 combined glass electrode stand-
ardized according to ref 10. The apparent pK of the conjugate
acid of the amino groups of PEI (pKN) at each pH was
determined at 25 °C in 1.0 M KCl in a nitrogen atmosphere
by the equation
water-dioxane was injected by a Hamilton syringe into a
cuvette containing 3 mL of buffer solution thermally equili-
brated at 25 ( 0.1 °C. The pseudo-first-order rate constants
(kobs) were calculated by least-squares analysis of plots of
ln(O.D.∞ - O.D.0)/(O.D.∞ - O.D.t) vs time and followed for 3-4
half-lives. Very slow kinetics were followed by the initial rate
method; the optical density at infinite time was determined
independently on a solution of phenol in the same buffer. Fast
kinetics were followed by a Sigma WZS-II (Biochem) dual-
wavelength spectrophotometer equipped with a thermostated
stopped-flow unit and digital data storage, interfaced with an
Apple IIe PC. The solutions of ester and of nucleophile were
mixed in the observation chamber in a 1:10 ratio. Ester
concentration after mixing was between 2 × 10-4 and 1 × 10-5
M. The change of pH at the end of all the kinetics never
exceeded 0.05 units. The pseudo-first-order kobs was deter-
mined by the method of Guggenheim.11 Other calculations and
plots of the curves were performed by FigP program by Biosoft.
Experimental details and values of rate constants are reported
in Tables 1 and 2.
pKN ) pH - log [(1 - R)/R]
where R ) [NH+]/[Ntot] is the ionization degree calculated by
the equation
7b
R ) ([H+]add - [H+]f + [OH-]f)/[Ntot
]
Here, [H+]add is the proton concentration as resulting from
added HCl, [H+]f and [OH-]f are the concentrations of free
protons and hydroxyl ions, as determined from titrations of
strong acid and base, respectively, at the same ionic strength
with respect to added salt, [Ntot] is the total PEI concentration,
and pKN values are reported in Table 1.
Resu lts
The experimental pseudo-first-order rate constants
(kobs) for the aminolysis by PEI of anionic and neutral
phenyl acetates were determined at 25 °C in 1.0 M KCl
at a fixed pH (4.36 - 11.20) in an excess of PEI (0.09-
0.4 monomer mol L-1). A linear dependence of rate
constants on the concentration of the polyamine was
observed which can be expressed by eq 1
Id en tifica tion of th e Rea ction P r od u cts. The identifi-
cation and the quantitative determination of the phenolic
moiety at the end of reaction was performed by comparison of
UV spectra with authentic samples. Aryl acetates acetylate
the primary and secondary amino groups of PEI,5c while they
could give free acetic acid when reacted with tertiary amino
groups. To verify this possibility, a 0.02-0.04 M solution of
each ester was reacted with 5 mL of a 0.4 monomer mol L-1
PEI buffer solution at pH 10.7 at 60 °C. At the end of the
reaction the solutions were acidified to pH 2 with concd HCl.
Acetic acid was analyzed by a Carlo Erba HRGC 5300 mega
series gas chromatograph using a capillary column (Megaacid
15 m × 0.32 mm). Comparative experiments with added acetic
acid to PEI buffers gave a quantitative recovery of the acid,
assuring the effectiveness of the procedure. For 3-acetoxy-
2,6-dinitrobenzoic acid and 4-acetoxy-3-nitrobenzenesulfonate,
the presence of acetic acid was also tested after aminolysis at
pH 4.0 at 25 °C. The presence of free acetic acid at the end of
the reactions (2-5%) also suggested that tertiary amino groups
of PEI were involved in the reaction.
kobs ) k0 + (kn + k′OH[OH-]) [PEI] (1 - R) (1)
where k0 represents the spontaneous hydrolysis, kn is the
nucleophilic second-order rate constant, k′OH[OH-] is the
rate of the hydroxide ion-catalyzed aminolysis, [PEI] is
the concentration of the polyamine in monomer mol L-1
,
and R is its ionization degree at each pH. k0 was always
negligible even at basic pH as is shown also by the low
values of the rates of the alkaline hydrolysis (kOH in
-
Table 1). k′OH[OH-], which cannot be calculated, was
considered unimportant because of the relatively low
values of pH used. The linearity of the function (1) in
the large concentration range of PEI excludes second-
order terms in amine due to intermolecular general base
or general acid catalysis. With 3-acetoxy-2,6-dinitroben-
zoic acid at pH > 8.75, a downward curvature of the plot
was observed, suggesting the presence of a substrate-
Kin etic Meth od s. Reactions were followed on a Perkin-
Elmer Lambda 6 UV/vis spectrophotometer equipped with a
thermostated cell holder, measuring at suitable wavelengths
the release of phenol or phenolate ion. Thirty µL of a 0.01-
0.03 M stock solution of the ester in dioxane or in 50% v/v
(10) Bates, R. G. J . Res. Nat. Bur. Stand. A 1962, 66, 179.
(11) Guggenheim, E. A. Philos. Mag. 1926, 2, 538.

1684 J . Org. Chem., Vol. 61, No. 5, 1996
Arcelli and Concilio
Ta ble 1. Exp er im en ta l Con d ition s a n d Kin etic P a r a m eter s for th e Am in olysis of P h en yl Aceta tes by
P oly(eth ylen eim in e) in 1 M KCl a t 25 °C
R1R2R3C6H2OCOCH3
e
concn range/
no. of
kOH /
R1
R2
R3
λa/nm
252
pH pKN monomerc/mol L-1 runs
kobsd/s-1
kn/s-1 M-1
s-1 M-1
b
-
4-SO3
H
H
H
H
7.20 7.62
8.40 8.53
10.00 9.42
10.70 9.54
0.1-0.4
8
(1.34-4.64) × 10-4 (3.45 ( 0.05) × 10
4.54 ( 0.35
(1.86 ( 0.06) × 10-2 f
252
256
0.1-0.4
12
14
10
8
0.01-0.0367
0.0139-0.0474
0.196-0.323
0.112 ( 0.0009
0.119 ( 0.02
0.139 ( 0.013
0.091-0.364
1.30-2.20
0.1-0.4
265-350 11.20g 9.88
2-Cl
4-CO2H
250
250
4.36 5.54
6.00 6.68
(3.64-9.57) × 10-6 (7.87 ( 0.18) × 10-5,h
(1.13-2.39) × 10-4 (2.92 ( 0.03) × 10-3
1.6 ( 0.1f
4.89 ( 0.30
25.2 ( 1.5
0.1-0.4
9
280-350 8.40
280-350 10.00
280-350 11.20g
345
345
400
400
0.1-0.4
15
16
8
8
8
0.048-0.181
0.631-1.26
0.566 ( 0.023
0.749 ( 0.073
0.863-1.72
0.1-0.4
2-NO2 4-SO3
4.36
(0.771-2.61) × 10-4 (4.87 ( 0.34) × 10-3
(0.364-1.21) × 10-4 (2.72 ( 0.2) × 10-3
(3.22-8.51) × 10-4 (1.1 ( 0.1) × 10-2
-
4.00 5.22
5.00 5.92
6.00 6.68
0.1-0.4
0.1-0.4
0.1-0.4
12 (2.34-6.62) × 10-3 (7.0 ( 1) × 10-2
400-650 7.92 8.17
0.091-0.364
12
0.0534-0.157
1.05 ( 0.13
2.43 ( 0.12f
5.05 ( 0.53
7.72 ( 0.67
7.81 ( 0.47
10.0 ( 0.6
8.40
400-650 9.21 9.05
400-650 10.00
0.091-0.364
0.091-0.364
0.091-0.364
0.804-2.29
0.1-0.4
13
15
16
14
8
0.408-1.24
0.691-2.5
0.917-2.88
8.23-22.8
400-650 10.70 9.54
400-650 11.20g 9.88
4-CO2H
H
H
H
250
250
280
350
410
4.36
6.00
10.00
4.36
6.00
8.40
(0.327-1.19) × 10-6 (7.17 ( 0.19) × 10-6 h
0.1-0.4
8
(0.397-1.45) × 10-5 (1.7 ( 0.28) × 10-4
0.1-0.4
12 (0.399-1.44) × 10-2 (4.36 ( 0.04) × 10-2
2-NO2 4-CO2H
0.1-0.4
8
8
(0.443-1.44) × 10-4 (1.75 ( 0.5) × 10-3
(0.988-3.31) × 10-3 (3.71 ( 0.3) × 10-2
0.758 ( 0.03f
0.1-0.4
2-NO2 4-NO2 3-CO2H 380
4.36 5.54
5.40 6.22
6.00
0.1-0.4
8
8
(0.785-2.4) × 10-3 (4.26 ( 0.069) × 10-2 109 ( 8
380
377
0.1-0.4
(0.484-1.64) × 10-2 0.216 ( 0.02
0.1-0.4
14
14
16
16
15
0.015-0.0524
0.048-0.11
0.105-0.405
0.268-0.876
0.523-1.29
0.599 ( 0.07
1.05 ( 0.08
3.5 ( 0.06
5.55 ( 0.18
6.42 ( 0.46
6.14 ( 1f
377-570 6.55 7.11
377-570 7.25 7.62
377-570 8.01 8.23
377-570 8.33 8.48
377-570 8.40
377-570 8.75 8.78
377-570 10.00 9.42
0.095-0.364
0.095-0.364
0.091-0.364
0.091-0.367
0.0545-0.227
0.0727-0.294
0.1-0.4
0.091-0.364
0.1-0.4
0.1-0.4
0.1-0.4
18
22
8
8
8
0.607-1.34
2.27-8.59
29.8 ( 0.5i
4-NO2
2-NO2
H
H
H
H
318
4.36
(0.295-1.2) × 10-5 (1.96 ( 0.19) × 10-4
420-550 10.00
0.095-0.33
1.1 ( 0.009
350
420
420
5.14 6.04
(1.56-6.56) × 10-5 (1.05 ( 0.07) × 10-3
(1.40-4.80) × 10-4 (5.17 ( 0.4) × 10-3
(1.13-4.30) × 10-3 (3.35 ( 0.08) × 10-2
0.216 ( 0.02f
10.5 ( 0.73
6.13 6.79
7.23 7.62
8.40
9
8
420-550 9.44 9.18
420-550 9.93 9.38
420-550 11.20g 9.88
0.091-0.364
0.1-0.3
1.30-2.17
0.1-0.4
12
14
12
8
0.0567-0.255
0.118-0.313
4.03-6.63
1.01 ( 0.05
1.26 ( 0.07
2.98 ( 0.67
3-Cl
H
H
273
4.36 5.54
(1.71-4.83) × 10-7 (8.34 ( 0.53) × 10-6
a
Analytical wavelength for phenol or phenolate release. The second value reported was used for dual wavelength spectroscopy.
Calculated from pKN ) pH - log[(1 - a)/a] were a is the ionization degree (see text). The accuracy was within (0.05 pK units. c Total
b
polyamine concentration. Calculated from eq 1. Standard errors are reported. e From kobs ) kOH[OH-] at pH ) 11.30 in 0.05 M NaHCO3/
d
NaOH buffer. f From ref 5a. In 2 M KCl. Values corrected for the contribution of the protonated form by eq 2. Uncertain value due
g
h
i
to a probable substrate-polyamine association.
Ta ble 2. Exp er im en ta l Con d ition s a n d Ra te Con sta n ts for th e Am in olysis of 4-Acetoxy-3-n itr oben zen esu lfon a te by
P r im a r y Am in es a t 25 °C a n d 1.0 M Ion ic Str en gth a
b
amine
aniline
trifluoroethylamine
ethylenediamine
pKN
pH
102[N]totc/M
no. of runs
103kobs/s-1
102 knd/s-1 M-1
4.85
5.84
7.42
6.45e
7.00e
7.50
6.95
3.9-19.1
3.8-14.7
4.8-24.0
7.94-31.7
6
6
8
8
0.558-2.31
1.14 ( 0.05
2.82 ( 0.04
1.15-4.22
35-191
51.1-105
113 ( 2
glycylglycine
glycine
8.25
9.76
204 ( 12f
9.70
9.22
11.42
10.51
1.8-7.2
3.6-9.0
3.6-14.5
3.6-14.5
10
8
12
12
184-736
203-459
(2.07 ( 0.21) × 103 g
(1.25 ( 0.12) × 104 g
methylamine
10.66
(5.92-20.7) × 103
(1.61-7.17) × 103
Kinetics were followed at λ 400 nm. Standard errors of estimate are reported. From refs 12a and 13. c Total amine concentration.
a
b
d
g
In 0.05 M KH2PO4/Na2HPO4 buffer carrier. e Calculated from kobs ) ko + kn [N]f. f From lit. 5a. Calculated from the plots of kobs ) ko
-
-
-
-
-
+ (kn + k′OH [OH ])[N]f vs [N]f at two pHs and then from the intercept of the plot (kn + k′OH [OH ]) vs [OH ].
polyelectrolyte association. Protogenic esters react in the
anionic form in all the pH ranges except for 4-acetoxy-
3-nitrobenzoic acid (pKa 3.15), 4-acetoxy-3-chlorobenzoic
acid (pKa 3.52), and 4-acetoxybenzoic acid (pKa 3.89) for
which at pH 4.36 the contribution of the protonated form
must also be considered. The rate constants knCOOH and
-
knCOO , for the aminolysis of the acid and anionic forms,
were obtained by eq 2

Aminolysis of Phenyl Acetates
J . Org. Chem., Vol. 61, No. 5, 1996 1685
Ka
kobs - k0 ) knCOO
+
-
(
)
[
aH + Ka
aH
+
knCOOH
[PEI](1 - R) (2)
(
)
]
aH+ + Ka
where Ka is the dissociation constant of the carboxylic
group calculated according to ref 12b. Due to the pH
-
dependence of pKN of PEI, knCOO could not be calculated
the same way as simple amines.14 It was obtained from
-
the slope of eq 2 and from the ratio knCOOH/knCOO 9.5
-
calculated by the Hammett relationship log(knCOOH/knCOO
)
14
-
) F(σp-COOH - σp-COO ) taking F ) 2 for simple amines
and σp-COOH ) 0.44 and σp-COO ) -0.05.12b The values
of the second-order rate constants kn for the aminolysis
by PEI of the neutral and anionic phenyl acetates are
reported in Table 1. The aminolysis of 4-acetoxy-3-
nitrobenzenesulfonate was also performed with a set of
primary amines. Under our experimental conditions, the
spontaneous hydrolysis was negligible, and only the first-
order term in amine was highlighted according to the eq
kobs ) (kn + k′OH[OH-])[N]f, where [N]f is the free base
amine concentration (results are listed in Table 2).
-
F igu r e 1. Bronsted-type plots for the aminolysis by poly-
(ethylenimine) in 1.0 M KCl at 25 °C of 4-acetoxy-2,6-
dinitrobenzoic acid (*); 4-acetoxy-3-nitrobenzensulfonate ([);
2-nitrophenyl acetate (b); 4-acetoxy-3-chlorobenzoic acid (2);
and 4-acetoxybenzenesulfonate (9) (cf. Table 1). Aminolysis
of 4-acetoxy-3-nitrobenzenesulfonate by primary amines (O)
at ionic strength 1.0 M with added KCl (cf. Table 2). Points
are experimental, and curves were calculated according to eq
4 or by linear least-squares method excluding points in
parentheses.
Discu ssion
It is generally accepted that the aminolysis of a phenyl
acetate occurs through a zwitterionic tetrahedral inter-
mediate (T() produced by the primary or secondary amine
attack on the carbonyl carbon:
inflection of the plot. In the aminolysis of phenyl
acetates, nitrogen nucleophiles and leaving aryl oxide
ions have the same nucleofugality from the T( intermedi-
ate (k-1 ) k2) when the difference between the pK of the
nucleophile and that of the conjugate acid of the leaving
group (pK1g) is 4-5 pK units.17a
O–
O
O
k1
k2
OAr
HN+
R
fast
HN+
R
HN + C
C + ArO–
CH3
C
OAr
k–1
R
CH3
CH3
T+
In the aminolysis of 4-acetoxy-3-nitrobenzenesulfonate
with simple amines in the range of pKN 4.85-10.66 we
found a linear Bronsted-type plot with slope 0.70 (Figure
1). The value of the slope is still comparable with that
reported for the aminolysis of most phenyl acetates with
monomeric amines (ânuc 0.8-0.9 ( 0.113,18,19). Since the
pK of the leaving group of the ester is 6.08,5d an inflection
of the plot for an amine of pKN 4-5 pK units higher was
expected, beyond the range of basicity examined by us.
Linear Bronsted-type plots were also found for the
aminolysis by PEI of 2-nitrophenyl acetate (pK1g 7.235d),
4-acetoxybenzenesulfonate (pK1g 9.065d), and 4-acetoxy-
3-chlorobenzoic acid (pK1g 7.895d) with slopes of 0.92, 0.99,
and 0.82 (Figure 1). The small negative deviation from
the linearity of the values of kn at pKN 9.54-9.88 (points
not computed in the plots) could indicate an incipient
inflection of the plots which should be centered at higher
values of pKN unattainable with PEI. The values of kn
and of pKN were not statistically corrected because the
number of equivalent protons which can be transferred
from PEI and that of the sites on the polyamine which
can accept protons are unknown.
O
(3)
N
C + ArOH
CH3
R
From steady state treatment the rate constant for such
a reaction scheme gives kn ) k2k1(k-1 + k2). The rate
constant reduces to kn ) k2k1/k-1 when k-1 . k2; i.e., when
the rate-determining step is the decomposition of the T(
intermediate (less basic nucleophiles and esters with poor
leaving groups), and to kn ) k1 when k-1 , k2, i.e., when
the rate-determining step is the formation of the T(
intermediate (more basic nucleophiles and esters with
good leaving groups).15a,b,16a,b,17a Nonlinear Bronsted-type
plots (log kn vs pKN of the nucleophile) are observed when
there is a change from one to the other rate-determining
step on increasing pKN. In this case, eq 4 was proposed
log(kn/k°n) ) â2(pKN - pKN°) +
log 2- log[1 + 10(â -â )(pK
°)] (4)
- pKN
N
2
1
by Castro16b for structurally homogeneous nucleophiles
where â2 and â1 are the slopes of the Bronsted plot when
k2 or k1 are rate determining and log k°n and pKN° are
the values of log kn and of pKn at the center of the
(15) (a) Castro, E. A.; Ureta, C. J . Org. Chem. 1990, 55, 1676; (b) J .
Chem. Soc., Perkin Trans. 2 1991, 63; (c) J . Chem. Res., Synop. 1987,
358.
(16) (a) Castro, E. A.; Ibanez, F.; Salas, M.; Santos, J . G.; Sepulveda,
P. J . Org. Chem. 1977, 42, 459. (b) Castro, E. A.; Gil, F. J . J . Am.
Chem. Soc. 1993, 99, 7611. (c) Castro, E. A.; Ibanez, F.; Santos, J . S.;
Ureta, C. J . Org. Chem. 1992, 57, 7024.
(12) Perrin, D. D.; Dempsey, B.; Serjeant, E. P. pKa Prediction For
Organic Acids and Bases; Chapman and Hall: London, 1981; (a) p 22;
(b) p 104; (c) p 27.
(13) J encks, W. P.; Gilchrist, M. J . Am. Chem. Soc. 1968, 90, 2622.
(14) St. Pierre, T.; J encks, W. P. J . Am. Chem. Soc. 1968, 90, 3817.
(17) (a) Gresser, M. J .; J encks, W. P. J . Am. Chem. Soc. 1977, 99,
6963; (b) 6970.
(18) Fersht, A. R.; J encks, W. P. J . Am. Chem. Soc. 1970, 92, 5442.
(19) Satterthwait, A. C.; J encks, W. P. J . Am. Chem. Soc. 1974, 96,
7018.

1686 J . Org. Chem., Vol. 61, No. 5, 1996
Arcelli and Concilio
In the aminolysis of 3-acetoxy-2,6-dinitrobenzoic acid
and 4-acetoxy-3-nitrobenzenesulfonate by PEI-curved
Bronsted-type plots were obtained (Figure 1). The values
of the limiting slopes of the curves fitting the experimen-
tal points according to eq 4 and the values of the
parameters at the inflection points were â1 ) 0.1, â2 )
0.90, pKN° ) 7.94, and log kN° ) 0.618 for the former ester
and â1 ) 0.1, â2 ) 0.9 pKN° ) 9.02, and log k°n ) 0.687
for the latter. Since the pK1g of the two esters are 6.085a
and 4.40,5d respectively, the change of the rate-determin-
ing step occurs when the difference pKN° - pK1g is 3-3.5,
about one pK unit lower than the value found for simple
amines. A tentative explanation of this difference can
result from the knowledge of the rate constants for the
expulsion of the nucleophile and of the aryl oxide ion from
the T( intermediate according to eqs 5 and 6, obtained
as detailed in the Appendix.
log k-1 ) 14.4 + 0.5pK1g - 0.8pKN
log k2 ) 10.69-0.1pK1g
(5)
(6)
F igu r e 2. Bronsted-type plots for the aminolysis of some
phenyl acetates of different leaving group acidity (pK1g from
ref 5a,d) by poly(ethylenimine) in 1 M KCl at 25 °C at various
PEI basicity. Points are experimental, and lines were calcu-
lated according to eq 4 for each value of pK1g. Uncertain values
are in parentheses. Esters and respective symbols are the
same as in Figure 1. Other esters are 4-acetoxy-3-nitrobenzoic
acid (4), 4-nitrophenyl acetate (~); 3-chlorophenyl acetate (]);
4-acetoxybenzoic acid (O); 3-acetoxybenzoic acid (1) (ref 5a);
and 4-acetoxyphenylacetic acid (half circle) (ref 5a).
The value of pKN at the center of the Bronsted-type
plots in Figure 1, calculated by equating eq 5 and 6,
results in 9.19 for 4-acetoxy-3-nitrobenzenesulfonate
(pK1g 6.08), in fair agreement with the value of 9.02
obtained from eq 4. Values of pKN° > 10 were obtained
for esters giving linear plots.
In the comparison of the values of k-1 and k2 calculated
for PEI with those obtained for alicyclic amines by eqs 6
and 8 in ref 16a, the ratio k2(PEI)/k2(amines is 9 for 3-acetoxy-
2,6-dinitrobenzoic acid and 41 for 4-acetoxy-3-nitroben-
zenesulfonate, while the ratio k-1(PEI)/k-1(amines) (calculated
in the pKN range 7-9) is about 0.5 for both esters. These
findings support a better nucleofugality of the aryl oxide
from the T( intermediate rather than a worse nucleo-
fugality of PEI with respect to an isobasic amine.
Electrostatic interactions and/or hydrogen bonds involv-
ing PEI and charged groups of the esters may perhaps
facilitate the aryl oxide over the amine expulsion. The
lower polarity in the microenvironment20 of the polyion
chain cannot be responsible for the lower value of pKN°
found with PEI: an increase of just 0.2-0.3 units was
reported for pKN° in the aminolysis of some esters when
the polarity of the solvent decreases.15c,17b In addition,
neither a pH-induced change of substrate-polyelectrolyte
interactions nor the change of the conformation of the
polyelectrolyte can be involved, taking into account that
linear plots were observed with the other structurally
similar phenyl esters under the same experimental
conditions (Figure 1).
values of the slopes were comparable with those found
for the aminolysis of a set of phenyl acetates with
glycylglycine5a (slope -0.88 which becomes -0.28 for the
most reactive ester) or with monomeric amines (slope -1
( 0.1 changing to -0.2 ( 0.2).18,19 The lower sensitivity
of log kn to the change of pK1g (slope -0.73), observed
with PEI at high basicity (pKN ) 9.42), could suggest a
lower C-O bond breaking in the transition state in
agreement with the Hammond postulate.
The curvature of the Bronsted-type plots does not prove
the presence of any intermediate in the reaction path,16a,17a
but the existence of different inflection points for different
esters agrees with the presence of a stepwise mechanism
as reported for simple amines,21 where the leaving group
expulsion or the proton transfer in the tetrahedral
intermediate decomposition is believed to be rate limit-
ing.
Our results do not offer evidence for the contribution
of an intermolecular base or acid catalysis in the range
of used concentration of PEI. Such types of catalysis are
also excluded by the comparison of the microscopic rate
constants estimated for the two processes (k3 and k5) with
the rate of decomposition of the T( intermediate (k2)
according to the paths shown in Scheme 1. To this
purpose we first consider the aminolysis of 4-acetoxy-2,3-
dinitrobenzoic acid. For the deprotonation step of T( to
T- by a base we calculate pKN(T(), the acidity of the
aminium ion in T(. According to Castro,16a the introduc-
tion of the CH3CH(O-)- group on the nitrogen atom of
an amine increases the pKN of the parent aminium ion
by 2.2 pK units. We then assume the same increment
of pKN when going from PEI to 1. Since FI of pKN-
substituted aminium ions is -8.4,22 from σI 0 for H and
σI 0.52 for -OC6H2-2,4-NO2-3-COO- (ref 23), a dif-
ference of -8.4 × 0.52 ) -4.4 units is expected for
In agreement with the above results is the effect of the
change of the leaving group of the esters on the rate of
aminolysis by PEI at constant pH. Figure 2 shows the
dependence of log kn on the pK of the conjugate acid of
the leaving aryl oxide ion (pK1g). Solid curves are the
least-squares fitting of experimental data according to
an equation corresponding to (4), where pKN and pKN°
are substituted by pK1g and pK1g°
. The values of the
optimized parameters pK1g°, log kn°, and â1g2 were 4.98,
1.27, and -0.73 at pKN 9.42, 5.95, 0.28, and -0.76 at pKN
8.53, 5.06, -0.51, and -0.80 at pKN 6.80, and 4.46, -1.37,
and -0.85 at pKN 5.52. Slopes â1g1 around -0.24 are
estimated for the most reactive ester at all pKN. The
(20) Everaerts, A.; Samyn, C.; Smets, G. Macromol. Chem. 1984,
185, 1881.
(21) Song, B. D.; J encks, W. P. J . Am. Chem. Soc. 1989, 111, 8479.
(22) Fox, J . P.; J encks, W. P. J . Am. Chem. Soc. 1974, 96, 1436.

Aminolysis of Phenyl Acetates
Sch em e 1
J . Org. Chem., Vol. 61, No. 5, 1996 1687
less basic secondary amino groups (intrinsic pKN 8.57a).
We estimate29 pKOH(T+) 6.1 at pH 8.75 (for 2) and
pKOH(T+) 4.9 at pH 4.36 (for 3). At both pH’s the proton
transfer from PEIH+ to T( is unfavored even in the best
conditions for the proton transfer. At pH 4.36, we
assume for k-5 the value 109 s-1 M-1 (refs 15a and 26),
obtaining log k5 ) log k-5 - pKOH(T+) ) 9 - 4.9 ) 4.1
and k5[PEI]R ) 4.7 × 103 s-1, lower than k2. Also for the
proton transfer from H3O+ to T(, assuming for k′5 the
value 109 s-1 M-1 (refs 16a and 27), we obtain k′5[H3O+]
) 4.4 × 104 s-1, negligible in comparison with k2. Since
the paths described do not appear to influence the
spontaneous decomposition of T( to products, we did not
evaluate k4 and k6, the rate constants of the subsequent
steps of the catalyzed paths.
the pKN of the aminium ion on going from 1 to T(, and
one obtains pKN(T() ) pKN(PEI) -4.4 + 2.2 ) pKN(PEI)
- 2.2. At the lowest and at the highest pH investigated
for this ester (4.36 and 8.75), pKN(PEI) is 5.52 and 8.78,
respectively, and for pKN(T() the values 3.3 and 6.6 are
obtained. It follows that the proton transfer from T( to
PEI is thermodynamically favored at both pH’s, and
according to Eigen,26 the value 2 × 109 s-1 M-1 can be
assumed for the rate constant k3.27 Consequently, under
the best conditions for the proton transfer (pH 8.75),
k3[PEI] (1 - R) ) (2 × 109)(0.23)(0.48) ) 2.2 × 108 s-1
,
lower than k2 ) 1.8 × 1010 s-1 calculated according to eq
6 for a pK1g 4.40. The base catalysis by PEI on a
tetrahedral intermediate can then be ignored. The
proton transfer from T( to water is not favored even at
pH 4.36 where the pKN(T() has the lowest value. Ac-
cording to ref 26 one calculates log k′3 ) log k′-3
-
pKN(T() + pKH
) 9.3 - 3.3 - 1.75 ) 4.25, i.e., k′3 )
3O+
1.8 × 104 s-1 M-1, and finally k′3[H2O] ) 1 × 106 s-1, a
value still negligible in comparison with k2.
To evaluate the contribution of the acid catalysis, i.e.,
the protonation of T( to give T+, we first estimate
pKOH(T+), the acidity of the -OH group in T+ (Scheme
1). We consider only polar effects since the electrostatic
potential in the environment of the polycation, which
mainly contributes to decrease the pKa of ionizable groups
of covalently incorporated moieties, is considerably re-
duced even at low concentrations of added salt.28 Struc-
tures 2 and 3 are chosen to simulate those of the
intermediate T( at basic or acid pH, when the attacking
amino group is primary or secondary because at low
ionization degree of PEI the more basic primary amino
groups (intrinsic pKN 9.57a) react preferentially over the
For the aminolysis of 4-acetoxy-3-nitrobenzenesulfonate
the rate constants k2, k3, and k5 are similarly obtained.
In the best conditions for the proton transfer from T(
intermediate to PEI (pH ) 11.20, [PEI] ) 2.3 monomer
mol L-1, a ) 0.05, pKN(PEI) ) 9.88, pKN(T() ) 8.1),
assuming k3 ) 2 × 109 s-1 M-1, it gives k3[PEI](1 - a) )
4.35 × 109 lower than k2 ) 1.2 × 1010 s-1. For the proton
transfer from PEIH+ to T( to obtain T+ (pH ) 4.36, [PEI]
) 0.4 monomer mol L-1, pKN(PEI) ) 5.52, a ) 0.9,
pKOH(T+) ) 5.2) one obtains k5[PEI]a ) 2.2 × 103 s-1
,
again lower than k2. Also, the rate constants for the
proton transfer at pH 4.36 from T( to water (k′3[H2O] )
3.9 × 105 s-1) and from H3O+ to T( (k′5[H3O+] ) 4.4 ×
104 s-1), are much lower than k2. In the aminolysis at
pH 11.20, the highest pH used with this ester, the proton
-
(23) The value of σI of the group -OC6H2-2,4-NO2-3-CO2 was
obtained as in ref 24. From the pKOH 4.40 of the phenolic group one
calculates σAr* 1.86 for -C6H2-2,4-NO2-3-CO2-, and from the linear
relationship of σArO* vs σAr* (from ref 12b), σ* ) 3.27 was obtained
which led to σI ) 3.27/6.23 ) 0.52, σI ) σ*/6.23 in ref 25.
(24) Gurthrie, J . P.; Pike, D. C. Can. J . Chem. 1987, 65, 1951.
(25) Charton, M. J . J . Org. Chem. 1964, 29, 1222.
(29) Starting from pKOH 7.7 of CH3NH2+C(CH3)(OH)(OPh) calculated
by Castro,16a taking σI ) 0.39 for PhO- and σI ) 0.52 for -OC6H2-2,4-
(NO2)2-3-COO-, and considering that the sensitivity of alcohols to σI
(26) Eigen, M. Angew. Chem., Int. Ed. Engl. 1964, 3, 1.
(27) The rate constant should be even lower than this value owing
to the shielding effect of the high number of positive charges on the
polycation. A decrease from 1010 to 2 × 109 s-1 M-1 of the rate constant
is reported for the proton transfer from the T( intermediate, originating
from 4-nitrophenyl dithioacetate and piperazine or piperazine cation,
to the corresponding aminium ion.16c
is expressed by FI
)
-8.4,22 one obtains for CH3NH2+C-
(CH3)(OH)(ODNPhCO2-) pKOH ) 7.7 - 8.4 (0.52 - 0.39) ) 6.60. From
σ* -N(CH3)2
) 0.32, considering the interposition of the two
methylenes,12c it results that σI -CH2CH2N(CH3)2 ) 0.32 × 0.42/6.23
) 0.01. Since σI of CH3 is -0.0525 and FI ) -8.4, we obtained for 2 a
pKOH(T+) ) 6.6 - 8.4 (0.01 + 0.05) ) 6.1, Similarly, for 3, from a σ*
+
(28) Morishima, Y.; Kobayashi, T.; Nozakura, S.-i. Macromolecules,
1988, 21, 101.
2.24 of -CH2NH3+, we calculated σI of -CH2CH2NH3 ) 2.24 × 0.4/
6.23 ) 0.14 and pKOH(T() ) 6.60 - 8.4 (0.01) + 0.05 + 0.14) ) 4.9.

1688 J . Org. Chem., Vol. 61, No. 5, 1996
Arcelli and Concilio
transfer from T( to OH- is also considered: since k3 )
1010 s-1 M-1 (ref 26), k3[OH-] ) 1.6 × 107 s-1 is calculated,
also lower than k2.
Similar calculations for the less reactive 4-acetoxy-
benzenesulfonate (pK1g 9.06) give for the decomposition
of the tetrahedral intermediate at pH 11.20, a ) 0.05,
[PEI] ) 2.2 monomer mol L-1, the rate constant k2 ) 6
× 109 s-1, pKN(T() ) 8.6 and then k3[PEI](1 - a) ) 4.2 ×
109 s-1 not far from k2, suggesting that the base-catalyzed
path could be partially rate determining at least at these
extreme concentrations of PEI.
We conclude that the intermolecular acid or base
catalyses in the aminolysis of phenyl acetates by PEI are
less important than the uncatalyzed decomposition of the
T( intermediate to products. An undetectable intra-
molecular acid or base catalysis can also be considered
improbable because in the aminolysis of 4-acetoxy-3-
nitrobenzenesulfonate the poly(ethylenimine) was from
two to five times less reactive than isobasic monofunc-
tional amines.
the nucleophilic nitrogen in the transition state 4 (on the
path from reagents to T() were obtained by adding to the
charges that the atoms have in the reagents, respectively,
the values â0 -0.2 that is the slope of the left hand side
of the plots in Figure 2 and âN 0.1, the slope of the right
hand side of the two curved plots in Figure 1. In both
cases kn ) k1. Charges of -0.1 for the oxygen atom and
0.9 for the nitrogen atom in the transition state 5, on
the path from T( to products, were obtained by adding
to the charges in the reagents the mean values â0 -0.8
found for the right hand side of the plots in Figure 2 and
âN +0.9 found for the left hand side of the plots in Figure
1, where kn ) k1k2/k-.1 Value â0 -0.1 to be used in eq 6
was obtained from the difference of charges on the
leaving oxygen in the transition state 5 and in T(.
Assuming that the basicity of PEI does not affect the rate
of expulsion of the aryl oxide ion, as was reported for
simple amines,16a,17a, âN is 0. Since the nucleofugality of
the aryl oxide ion from T( formed in the reaction of
phenyl acetates with PEI is unknown, the intercept 10.69
in eq 6 was obtained using for â0 the value -0.1 obtained
above and for k2 the value 3 × 1010 s-1 reported19 for the
aminolysis of 1-acetoxy-4-methoxypyridinium (pK1g 2.1)
by methylamine. Similarly, â0 0.5 and âN -0.8 to
calculate k-1 in eq 5 were obtained by subtracting from
the charges of oxygen and of nitrogen atoms in the
transition state 4 the charges that the atoms have in T(.
Finally, the value 14.4 for the intercept of eq 5 was
obtained by equating k-1 and k2 using the values pK1g
4.40 and pKN° 7.9 found for 3-acetoxy-2,6-dinitrobenzoic
acid and the values of â0 and âN calculated above.
Ap p en d ix
The parameters in eqs 5 and 6 were obtained according
to eq 12 in ref 17a for aminolysis of phenyl acetates with
simple amines were â0 was the sensitivity of the rate
constants of each reaction step in Scheme 1 to the acidity
of the conjugate acid of the leaving aryl oxide ion and âN
the sensitivity of the rate constants to the pKN of the
nucleophile. We assumed in the reagents a charge of 0.7
for the oxygen atom of the leaving aryl oxide ion and 0
for the nitrogen atom of the nucleophile and charges 0
and 0.9 for the two atoms in the T( intermediate. The
effective charges 0.5 on the leaving oxygen and 0.1 on
J O9514681