715
Koh et al.
dispersion, but it also assists leaving-group departure (17).
This assistance is particularly important in the second-order path,
k1, when the (aprotic) solvent cannot assist by hydrogen bonding
(17).
benzylamine nucleophiles, Aldrich GR, were used without fur-
ther purification. Preparation of deuterated benzylamines was
as described previously (5). The analysis (NMR and GC mass
spectroscopy) of the deuterated benzylamines showed more
than 99% deuterium content, so that no corrections to kinetic
isotope effects for incomplete deuteration were made. Ethyl
aryl carbonates were prepared by reacting phenols with ethyl
chloroformates. The substrates synthesized were confirmed by
IR, NMR, and mass spectral analyses as follows.
For the catalyzed path, the kH/kD values are further in-
creased to 2.0–2.5, reflecting correctly the partial bond cleav-
age of the N—H(D) bond in the rate-determining step,
kd[RNH2] in Scheme 1. Among several conceivable TS struc-
tures for the catalyzed path, the cyclic hydrogen bonded struc-
ture, 6, is considered to be the most likely candidate. The larger
−
kH/kD and smaller ρz , ρX, and ρXZ values for 6 than for 5 can
CH3CH2OC(O)OC6H4–p-CN: mp 40–42°C. νmax: 3100 (C-
H), 3000 (C-H, aromatic), 2250 (C;N), 1760 (CTO).
δH(CDCl3): 7.2–7.7 (C6H4, 4H, m), 4.3 (CH2, 2H, q), 1.3 (CH3,
3H, t). m/z = 119 (M+).
be explained as follows. Two hydrogen bonds are involved so
that kH/kD should increase, but positive charge on N and nega-
tive charge on O should be partially offset by partial deproto-
nation from N and partial protonation on O. The decrease in
−
the two ρ values, ρX and ρz , should result in a substantial
CH3CH2OC(O)OC6H4–p-COCH3: mp 44–45°C. νmax: 3100
(C-H), 3000 (C-H, aromatic), 1750 (CTO), 1680 (COCH3).
δH(CDCl3): 7.2–8.0 (C6H4, 4H, m), 4.3 (CH2, 2H, q), 2.6
(COCH3, 3H, s), 1.3 (CH3, 3H, t). m/z = 208 (M+).
decrease in the intensity of interaction leading to a large de-
crease in the ρXZ values (18). The trends of change in the
magnitude of kH/kD in Tables 3 and 4 are also in line with our
proposed TS structures, 5 and 6. In structure 5, the greater the
leaving ability (δσZ > 0) of the phenoxide, the greater will be
the extent of deprotonation and consequently the larger is the
kH/kD value (Table 3). However, in 6, the stronger the electron
acceptor in the phenoxide (δσZ > 0), the less will be the nega-
tive charge on O due to the greater delocalization, and hence
the weaker is the hydrogen bond between O and the catalyst
amine nitrogen. Thus the kH/kD value will decrease as the elec-
tron-withdrawing power of Z increases (Table 4). In both
structures, an electron acceptor X (δσX > 0) should increase
the acidity of a substituted benzylammonium cation, i.e., de-
protonation becomes more facile, so that kH/kD increases.
The proposed mechanism is also supported by the activa-
tion parameters (Table 5) determined from rate data at three
temperatures, 25.0, 35.0, and 45.0°C. The relatively low posi-
tive ∆H‡ and large negative ∆S‡ values are in accord with the
stepwise mechanism proposed (11, 19). For the catalyzed path,
the ∆H‡ values are lower than those for the uncatalyzed path
and the ∆S‡ values are large and negative. This is consistent
with a simple proton transfer process involved in TS 6, where
there is a two-proton bridge that is highly structured but very
little energy will be required for the deprotonation.
CH3CH2OC(O)OC6H4–m-NO2: mp 50–52°C. νmax: 3100 (C-
H), 3000 (C-H, aromatic), 1750 (CTO), 1680 (COCH3).
δH(CDCl3): 7.3–8.1 (C6H4, 4H, m), 4.3 (CH2, 2H, q), 1.2 (CH3,
3H, t). m/z = 208(M+).
CH3CH2OC(O)OC6H3–(p-Cl, m-NO2): liquid. νmax: 3100
(C-H), 3000 (C-H, aromatic), 1760 (CTO). δH(CDCl3):
7.2–7.7 (C6H3, 3H, m), 4.2–4.4 (CH2, 2H, q), 1.3 (CH3, 3H, t).
m/z = 245 (M+).
CH3CH2OC(O)OC6H4–p-NO2: mp 60–61°C. νmax: 3100 (C-
H), 3000 (C-H, aromatic), 1750 (CTO). δH(CDCl3): 7.3–8.3
(C6H4, 4H, m), 4.2 (CH2, 2H, q), 1.2 (CH3, 3H, t). m/z= 211 (M+).
Kinetic procedures
Rates were measured conductimetrically in acetonitrile. The
conductivity bridge used in this work was a homemade com-
puter-automatic A/D converter conductivity bridge. Pseudo-
first-order rate constants, kobs, were determined by the
Guggenheim (20) method with a large excess of benzylamine.
The k1 and k2 values were reproducible to ± 5%.
In summary, the aminolysis of ethyl aryl carbonates in ace-
tonitrile proceeds by a stepwise mechanism with rate-limiting
expulsion of the phenoxide leaving group from the zwitter-
ionic tetrahedral intermediate, T±. Two reaction pathways,
catalyzed and uncatalyzed paths, compete in most of the reac-
tions; base catalysis is not required for the reactions involving
strong nucleophiles (X = p-CH3O, p-CH3) with a strong nu-
cleofuge (Z = p-NO2). The proposed mechanism is based on
Product analysis
Ethyl p-nitrobenzyl carbonate was reacted with excess benzyl-
amine with stirring for more than 15 half-lives at 25.0°C in
acetonitrile, and the products were isolated by evaporating the
solvent under reduced pressure. The product mixture was
treated with column chromatography (silica gel, 20% ethyl
acetate – n-hexane). Analysis of the product gave the follow-
ing results.
−
(i) the large magnitude of ρX and especially of ρz , (ii) the
relatively large normal kinetic isotope effects (kH/kD > 1.0)
involving a deuterated benzylamine nucleophile, (iii) the
greater kH/kD values for the catalyzed path, and (iv) the low ∆H‡
and ∆S‡ values for both pathways, but with lower values for
the base-catalyzed path than for the uncatalyzed reaction.
CH3CH2OC(O)NHCH2C6H5: Liquid. νmax: 3300 (N-H), 3000
(C-H, aromatic), 1700 (CTO). δH(CDCl3): 7.2–7.5 (C6H5,
5H, m), 5.2 (NH, 1H, s), 4.2 (CH2, 2H, q), 4.4 (CH2, 2H, s),
1.2 CH3, 3H, t). m/z = 227 (M+).
Acknowledgements
Experimental
We thank the Korea Science and Engineering Foundation,
Inha University, and Chonju National University of Education
for support of this work.
Materials
Merck GR acetonitrile was used after three distillations. The
© 1998 NRC Canada