Kinetics and mechanism of the anilinolysis of aryl 4-nitrophenyl thionocarbonates in aqueous ethanol

Article abstract of DOI:10.1002/poc.1814

The reactions of bis(4-nitrophenyl), 3-chlorophenyl 4-nitrophenyl, and 3-methoxyphenyl 4-nitrophenyl thionocarbonates (1, 2, and 3, respectively) with a series of anilines are subjected to a kinetic investigation in 44wt.% ethanol-water, at 25.0°C and an

Full text of DOI:10.1002/poc.1814

Research Article
Received: 13 April 2010,
Revised: 27 August 2010,
Accepted: 23 September 2010,
Published online in Wiley Online Library: 23 November 2010
(wileyonlinelibrary.com) DOI 10.1002/poc.1814
Kinetics and mechanism of the anilinolysis of
aryl 4-nitrophenyl thionocarbonates in
aqueous ethanol
a
a
a
*
´
Enrique A. Castro , Rayen Acevedo and Jose G. Santos
The reactions of bis(4-nitrophenyl), 3-chlorophenyl 4-nitrophenyl, and 3-methoxyphenyl 4-nitrophenyl thionocar-
bonates (1, 2, and 3, respectively) with a series of anilines are subjected to a kinetic investigation in 44 wt.% ethanol–
water, at 25.0 -C and an ionic strength of 0.2 M. Under aniline excess, pseudo-first-order rate coefficients (kobs) are
found. Plots of kobs versus aniline concentration are linear, with the slopes (kN) pH independent, kN being the rate
coefficient for the anilinolysis of the thionocarbonates. The Brønsted plot (log kN vs. pKa of anilinium ions) for
thionocarbonate 1 is linear, with slope (b) 0.62, which is consistent with a concerted mechanism. The Brønsted plots
for thionocarbonates 2 and 3 are curved, with slopes 0.1 at high pKa for both reaction series and slopes 0.84 and 0.79
at low pKa for the reactions of 2 and 3, respectively. The latter plots are in accordance to stepwise mechanisms,
through a zwitterionic tetrahedral intermediate (TW) and its anionic analogue (TS), the latter being formed by
deprotonation of TW by the basic form of the buffer (HPO24S). The Brønsted curves are explained by a change in the
rate-limiting step, from deprotonation of TW at low pKa, to its formation at high pKa. The influence of the amine nature
and the non-leaving and electrophilic groups of the substrate on the kinetics and mechanism is also discussed.
Copyright ß 2010 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: anilines; Brønsted plots; kinetics; mechanisms; thionocarbonates
with SA amines and other similar reactions, we evaluate the
INTRODUCTION
effects of the amine nature and the non-leaving and electrophilic
groups of the substrate on the kinetics and mechanisms of these
reactions.
Although the mechanisms of the aminolysis of esters and
carbonates have been extensively investigated,[1–3] there have
been relatively few studies on the kinetics and mechanisms of the
aminolysis of alkyl aryl and diaryl thionocarbonates.[4–10] Most of
the reactions of these compounds with pyridines and secondary
alicyclic (SA) amines follow a stepwise mechanism.[6–9] For the
reactions of some of these thionocarbonates with SA amines two
tetrahedral intermediates have been proposed, one zwitterionic
(Tꢀ), formed by the attack of the amine to the thiocarbonyl group,
and the other anionic (Tꢁ), formed by a proton transfer from Tꢀ
to the SA amine.[6–8] Nevertheless, when the thionocarbonate
possesses one very good nucleofuge, as in the case of ethyl 2,4-
dinitrophenyl thionocarbonate or phenyl 2,4-dinitrophenyl
thionocarbonate,[9] its reaction with SA amines proceeds through
only one tetrahedral intermediate (the zwitterionic Tꢀ). On the
other hand, when the above type of compound has two good
leaving groups, such as in bis(4-nitrophenyl) thionocarbonate,
the mechanism becomes concerted.[10]
It is of interest to know why some reactions of thionocarbo-
nates proceed through the intermediates Tꢀ and Tꢁ, whereas
other reactions occur with only one intermediate (Tꢀ). In order to
answer this question and shed more light on the mechanism of
the aminolysis of diaryl thionocarbonates, in this work we report
a kinetic study of the reactions of bis(4-nitrophenyl), 3-
chlorophenyl 4-nitrophenyl, and 3-methoxyphenyl 4-nitrophenyl
thionocarbonates (1, 2, and 3, respectively) with a series of
anilines. By a comparison between the kinetic results obtained in
this work and those for the reactions of the same compounds
RESULTS AND DISCUSSION
The reactions were followed by the appearance of 4-nitrophen-
oxide. Quantitative yields of this leaving group were found in all
cases. The pseudo-first-order rate constants (kobs), obtained
*
Correspondence to: J. G. Santos, Facultad de Qu´ımica, Pontificia Universidad
´
Catolica de Chile, Casilla 306, Santiago 6094411, Chile.
E-mail: jgsantos@uc.cl
a
E. A. Castro, R. Acevedo, J. G. Santos
Facultad de Qu´ımica, Pontificia Universidad Catolica de Chile, Casilla 306,
Santiago 6094411, Chile
´
J. Phys. Org. Chem. 2011, 24 603–610
Copyright ß 2010 John Wiley & Sons, Ltd.

E. A. CASTRO, R. ACEVEDO AND J. G. SANTOS
Table 1. Experimental conditions and kobs values for the reactions of anilines with bis(4-nitrophenyl) thionocarbonate (1)a
b
Aniline substituent
4-Amino
pH
FN
103[N]tot/Mc
104kobs/sꢁ1
No. of runs
7.2
7.5
7.8
6.5
7.0
7.5
6.5
7.0
7.5
6.5
7.0
7.5
6.5
7.0
7.5
7.2
7.5
0.8490
0.9182
0.9572
0.9394
0.9800
0.9936
0.9755
0.9921
0.9975
0.9910
0.9971
0.9991
0.9943
0.9982
0.9994
0.9999
1
0.411–1.78
0.382–1.15
0.358–1.05
20.9–105
29.9–90.4
20.1–73.7
13.3–63.3
29.9–79.7
13.3–66.3
26.6–115
28.9–125
29.1–146
16.9–84.5
12.5–62.3
20.8–62.3
0.574–2.87
0.352–1.52
8.53–35.9
8.89–25.8
7.62–24.2
91.9–434
83.6–377
92.7–340
30.2–128
58.4–174
28.2–140
28.7–126
34.5–130
37.0–179
9.31–46.0
7.12–32.1
11.0–31.4
0.424–0.811
0.300–0.418
6
6
6
7
6
5
7
6
7
6
6
7
7
6
6
6
5
4-Methoxy
4-Methyl
None
3-Methoxy
3-Acetyl
a In 44 wt.% ethanol–water, at 25.0 8C, ionic strength 0.2 M (KCl), 0.01 M phosphate buffer.
b Free amine fraction.
c Concentration of total amine (free base plus protonated forms).
under aniline excess, and the experimental conditions for the
studied reactions are detailed in the Supplementary Materials.
The ranges of amine concentration and kobs values are
summarized in Tables 1–3.
The kinetics of all the studied reactions obey Eqn (1), where k0
and kN are the rate coefficients for solvolysis and anilinolysis of
the substrates, respectively. The values of k0 and kN showed no
dependence on pH within the pH range employed. The pH was
maintained by phosphate buffer (H2POꢁ4 þ HPO42ꢁ), using a
constant concentration of 0.01 M. The values of k0 and kN were
obtained as the intercept and slope, respectively, of linear plots of
kobs against free aniline concentration, at constant pH.
kobs ¼ k0 þ kN½free anilineꢂ
(1)
Table 2. Experimental conditions and kobs values for the reactions of anilines with 3-chlorophenyl 4-nitrophenyl thionocarbonate (2)a
b
Aniline substituent
4-Amino
pH
FN
103[N]tot/Mc
104kobs/sꢁ1
No. of runs
7.2
7.5
7.8
6.5
7.0
7.5
6.5
7.0
7.5
6.5
7.0
7.5
6.5
7.0
7.5
7.5
7.8
0.8490
0.9182
0.9572
0.9394
0.9800
0.9936
0.9755
0.9921
0.9975
0.9910
0.9971
0.9991
0.9943
0.9982
0.9994
1
1.32–5.72
1.29–5.58
1.25–5.25
20.9–92.2
20.9–90.4
19.9–86.3
30.0–130
29.6–128
29.9–110
30.5–132
49.1–128
29.6–148
15.8–68.3
23.3–101
30.2–111
2.43–6.32
2.61–7.82
5.78–21.8
5.84–23.3
5.38–21.4
18.4–112
19.5–106
21.1–103
21.2–99.2
18.2–99.7
21.8–77.1
8.95–43.4
15.6–43.7
8.86–47.7
3.73–10.6
3.71–15.8
5.04–17.6
0.191–0.323
0.200–0.366
6
6
6
6
6
6
7
6
6
6
5
7
6
6
5
5
6
4-Methoxy
4-Methyl
None
3-Methoxy
3-Acetyl
1
a In 44 wt.% ethanol–water, at 25.0 8C, ionic strength 0.2 M (KCl), 0.01 M phosphate buffer.
b Free amine fraction.
c Concentration of total amine (free base plus protonated forms).
wileyonlinelibrary.com/journal/poc
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 603–610

KINETICS OF ANILINOLYSIS OF THIONOCARBONATES
Table 3. Experimental conditions and kobs values for the reactions of anilines with 3-methoxyphenyl 4-nitrophenyl thionocarbonate
(3)a
b
Aniline substituent
4-Amino
pH
FN
103[N]tot/Mc
104kobs/sꢁ1
No. of runs
7.2
7.5
7.8
7.2
7.5
7.8
6.5
7.0
7.5
6.5
7.0
7.5
6.5
7.0
7.5
7.5
0.8490
0.9182
0.9572
0.9873
0.9936
0.9968
0.9755
0.9921
0.9975
0.9910
0.9971
0.9991
0.9943
0.9982
0.9994
1
1.39–6.01
1.34–5.89
1.31–5.70
16.5–49.5
16.5–49.5
16.3–48.9
30.0–130
36.8–110
32.7–98.1
30.5–132
29.6–128
29.6–148
18.7–81.0
23.3–101
30.2–111
2.43–6.32
2.51–9.74
2.10–9.69
2.26–10.1
5.72–19.5
5.59–20.2
7.10–23.5
7.71–36.6
8.98–30.5
7.46–25.7
3.26–17.4
4.12–17.5
3.11–20.7
1.68–5.19
1.36–6.26
2.13–6.87
0.0795–0.139
6
6
6
6
6
6
6
6
6
6
5
7
6
5
5
5
4-Methoxy
4-Methyl
None
3-Methoxy
3-Acetyl
a In 44 wt.% ethanol–water, at 25.0 8C, ionic strength 0.2 M (KCl), 0.01 M phosphate buffer.
b Free amine fraction.
c Concentration of total amine (free base plus protonated forms).
For the studied reactions, the k0 values were much smaller than
the kN[free aniline] term in Eqn (1). The values of kN for the
reactions of thiocarbonates 1, 2, and 3 with anilines are shown in
Table 4.
The curved Brønsted plots obtained for the reactions of 2 and 3
could be explained by the mechanism shown in Scheme 1, where
a zwitterionic tetrahedral intermediate (Tꢀ) is formed and there is
a change in the rate-determining step, from that for its
breakdown (k2 step) to that for its formation (k1 step), as the
aniline becomes more basic.[7–9] Nevertheless, although this is
the mechanism found for numerous aminolysis of carbonates
and their thio derivatives,[7–9] as argued below we think the
mechanism for these reactions is not that of Scheme 1 but one
with a small difference.
In contrast to the reactions of thionocarbonates 2 and 3 with
anilines in 44 wt.% ethanol–water, which show linear plots of kobs
versus aniline concentration (this work), those of the same
compounds with weakly basic SA amines in the same solvents
mixture show curved upwards plots of kobs versus free SA amine
The kN values for the anilinolysis of thionocarbonates 1–3, as
well as those of the pKa of the conjugate acids of the anilines,
were statistically corrected with q ¼ 2 for 4-phenylendiamine
(q ¼ 1 for all the other anilines) and p ¼ 3 for all the conjugate
acids of the anilines. The parameter q is the number of equivalent
basic sites on the free amine and p is the number of equivalent
dissociable protons on the conjugate acid of the amine.[11]
After statistical correction of the data in Table 4 the Brønsted-
type plots of Fig. 1 were obtained. As observed, that for the
reaction of 1 is linear and those for the reactions of 2 and 3 are
curved.
Table 4. Values of pKa for the conjugate acids of anilines and kN values for the reactions of anilines with bis(4-nitrophenyl)
thionocarbonate (1), 3-chlorophenyl 4-nitrophenyl thionocarbonate (2), and 3-methoxyphenyl 4-nitrophenyl thionocarbonate (3)a
102kN/sꢁ1Mꢁ1
Aniline
pKa
1
2
3
4-Phenylendiamine
4-Methoxyaniline
4-Methylaniline
Aniline
3-Methoxyaniline
3-Aminoacetophenone
6.45
5.31
4.90
4.46
4.26
3.10
240 ꢀ 2
43 ꢀ 2
17.6 ꢀ 0.5
4.33 ꢀ 0.09
2.95 ꢀ 0.08
1.41 ꢀ 0.04
0.61 ꢀ 0.02
0.143 ꢀ 0.007
44 ꢀ 1
13.2 ꢀ 0.4
19.3 ꢀ 0.4
11.4 ꢀ 0.3
5.5 ꢀ 0.1
1.65 ꢀ 0.06
7.54 ꢀ 0.03
3.46 ꢀ 0.01
1.48 ꢀ 0.01
0.323 ꢀ 0.001
a Both the pKa and kN values were determined in 44 wt.% ethanol–water, at 25.0 8C, ionic strength 0.2 M (KCl), phosphate buffer
0.01 M.
J. Phys. Org. Chem. 2011, 24 603–610
Copyright ß 2010 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

E. A. CASTRO, R. ACEVEDO AND J. G. SANTOS
concentration.[12] This was explained by the formation of two
tetrahedral intermediates, one zwitterionic (Tꢀ) and other
anionic (Tꢁ) (Scheme 2).[12] The latter intermediate is formed
by deprotonation of Tꢀ by the corresponding SA amine in a
diffusion controlled pathway (k3 step). For these reactions the k3
step competes with the k2 step, that is, k2ꢃk3[amine].[12]
Therefore, the relevant question is: why for the reactions of 2
and 3 with weakly basic SA amines the plots of kobs versus
[amine] are curved, that is, the k3 step is as fast as the k2 step,
whereas for the reactions of the same compounds with anilines
(this work) the plots are linear? A first hypothesis is that for the
reactions of anilines the k3 step is much slower than that of k2.
In order to answer this question it is necessary first to find out
whether the k3 step for the reactions with anilines is also diffusion
controlled. The pKa values of the zwitterionic tetrahedral
intermediates derived from the reactions of 2 and 3 with SA
amines have been estimated to be lower than those of the
conjugate acids of the corresponding SA amines by 8.1 and
7.6 units, respectively.[12] Therefore, deprotonation of these
intermediates by the corresponding SA amine are thermodyna-
mically favorable and diffusion controlled.[12,13] On this basis a k3
value of 4 ꢄ 109 sꢁ1 Mꢁ1 was estimated in 44 wt.% ethanol–
water.[12]
Figure 1. Brønsted-type plots (statistically corrected, see text) for the
reactions of anilines with (*) bis(4-nitrophenyl) thionocarbonate (1), (*)
3-chlorophenyl 4-nitrophenyl thionocarbonate (2), and (&) 3-methoxy-
phenyl 4-nitrophenyl thionocarbonate (3), in 44 wt.% ethanol–water, at
25.0 8C, ionic strength 0.2 M (KCl).
Concerning the reactions with anilines, the difference between
the pKa value of the intermediate Tꢀ and that of the conjugate
Scheme 1.
Scheme 2.
wileyonlinelibrary.com/journal/poc
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 603–610

KINETICS OF ANILINOLYSIS OF THIONOCARBONATES
acid of the corresponding aniline (DpKa) should be the same as
that estimated for the reactions with SA amines because the
same amine is involved as amino moiety in Tꢀ and as a base
capturing the proton. Namely, the DpKa value should be
independent of both the nature and basicity of the amine.
Therefore, the pKa of the intermediates Tꢀ formed in the
reactions of dithiocarbonates 2 and 3 with anilines should also be
lower than those of the conjugate acids of the corresponding
anilines by 8.1 and 7.6 units, respectively. According to these, the
proton transfers involved should also be diffusion controlled.
Since the sizes of SA amines and anilines are similar, a value of ca.
4 ꢄ 109 sꢁ1 Mꢁ1 can also be estimated for k3 for the reactions with
anilines in 44 wt.% ethanol–water (this work). On the other hand,
the free amine concentration, [amine], used is similar for the
reactions of SA amines and anilines and the k2 value should be
the same for both reactions since it is known that this value is
independent of the amine nature and basicity.[14] Therefore, it is
clear that for both reactions, k2ꢃk3[amine]. Therefore, the
hypothesis that for the reactions with anilines k2 > k3[amine] is
not correct.
controlled.[13] Furthermore, due to the lower size of HPO24ꢁ
compared to anilines and the fact that the proton transfer is
between the cationic moiety of Tꢀ and a dianion, it is reasonable
that the rate constant involved (k4) be greater than that for the
proton transfer from Tꢀ to the corresponding aniline (k3). On the
other hand, it can be calculated that for most reactions the
concentrations of HPO24ꢁ and free anilines are similar. Therefore, it
follows that k4[HPO24ꢁ] should be larger than k3[aniline] (and also
larger than k2) and the third hypothesis seems to be valid.
Therefore, the most likely mechanism for the reactions of
thionocarbonates 2 and 3 with anilines is that depicted by
Scheme 3.
According to Scheme 3, the curved Brønsted plots in Fig. 1 can
be explained by a similar argument employed for Scheme 1 (see
above).[7–9] The difference is that the change in the rate-
determining step is from the k4 step in Scheme 3, instead of the k2
step in Scheme 1, for weakly basic anilines, to the formation of the
Tꢀ intermediate (k1 step for both schemes) for the most basic
anilines.
The curved Brønsted plots in Fig. 1 were calculated by means of
a semi-empirical equation, Eqn (2), deduced on the basis of
Scheme 1.[15] A similar equation has been reported by Gresser
and Jencks.[14] Although we assume that the reactions of
thionocarbonates 2 and 3 are driven by the mechanism shown in
Scheme 3, Eqn (2) can equally be applied to both schemes (see
below).
Equation (2) contains four parameters: b1 and b2, which are the
Brønsted slopes at high and low pKa, respectively, and kN0 and pKa0,
which are the corresponding values at the center of the Brønsted
curvature.[15]
To explain the different mechanisms found for the reactions of
2 and 3 with anilines (this work), relative to those with weakly
basic SA amines,[12] a second hypothesis is that the rate-
determining steps measured are different for the two reactions.
For those with weakly basic SA amines (pKa 5.4–8.5),
k
ꢁ1 ꢅ k2 þ k3[amine]. Since most of the anilines used in the
present work are less basic than the least basic SA amine (with the
exception of 4-phenylendiamine, pKa 6.45), but, on the other
hand, anilines are worse nucleofuges from the Tꢀ intermediate
than isobasic SA amines,[7,8] it is reasonable that for the reactions
with anilines, kꢁ1 ꢅ k2 þ k3[amine]. Therefore, the hypothesis that
the reactions of weakly basic SA amines and anilines have
different rate-limiting steps is also incorrect.
logðkN=kN0 Þ ¼ b2ðpKaꢁpK0Þꢁlogðð1 þ aÞ=2Þ
log a ¼ ðb2ꢁb1ÞðapKaꢁpK0Þ
(2)
a
A third hypothesis is that for the reactions of thionocarbonates
2 and 3 with anilines there is a proton transfer from Tꢀ to the
buffer base, which is faster than k2 or k3[amine]. Phosphate was
used as buffer for these reactions, in contrast to those with SA
amines, where there was no external buffer and partial
protonation of the amine acted as buffer. Since the basic form
of the phosphate buffer (HPO24ꢁ) is more basic (pKa 7.5 in 44 wt.%
ethanol–water) than the most basic aniline used in this work (pKa
6.45 at the same conditions), the proton transfer from Tꢀ to
HPO24ꢁ should also be thermodynamically favorable and diffusion
The curved Brønsted curves in Fig. 1 were calculated by means
of the following parameters: pKa0 ¼ 5:9, log kN0 ¼ ꢁ0:87, b1 ¼ 0.1,
and b2 ¼ 0.84 for the reactions of thionocarbonate 2, and
pKa0 ¼ 6:1, log kN0 ¼ ꢁ1:24, b1 ¼ 0.1, and b2 ¼ 0.79 for thiono-
carbonate 3. The errors of the slopes are ꢀ0.1, and those of pKa0
and log kN0 are ꢀ0.3 and ꢀ0.2, respectively. The values of b1 and
b2 are in accordance with those reported for other aminolysis
governed by stepwise mechanisms: b1 ¼ 0.1–0.3 and b2 ¼ 0.8–
1.1.[7–9,14,15]
Scheme 3.
J. Phys. Org. Chem. 2011, 24 603–610
Copyright ß 2010 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

E. A. CASTRO, R. ACEVEDO AND J. G. SANTOS
Equation (2) can equally be applied to Schemes 1 and 3 for the
following reasons: By application of the steady state treatment to
the Tꢀ intermediate, Eqns (3) and (4) can be deduced for
Schemes 1 and 3, respectively.
thiocarbonyl carbon of the former compound should be more
positive and, therefore, more prone to nucleophilic attack by
the amine.
In Fig. 1 it can be observed that the curved Brønsted-type plots
for the reactions of 2 and 3 are almost parallel and the curve for 2
is above that for 3. Namely, thionocarbonate 2 shows a greater
reactivity toward anilines relative to 3, irrespective of the rate-
determining step. For the reactions of anilines of pKa < pKa0 the
rate-limiting step is the proton transfer from the intermediate Tꢀ
to HPO24ꢁ (k4 step in Scheme 3). For the reactions of these amines,
kN is given by Eqn (6). Since k1 for 2 is larger than that for 3 (see
above) and the kN (2)/kN (3) ratio is fairly constant along the pKa
axis, the larger kN value for the reaction of 2 can be explained
solely by the larger k1 value for this reaction compared with that
k1k2
kN
¼
(3)
(4)
k
ꢁ1 þ k2
k1k4½HPO42ꢁꢂ
kN
¼
k
ꢁ1 þ k4½HPO42ꢁꢂ
For very basic anilines,
k
ꢁ1 ꢆ k2 (Scheme 1) and
k
ꢁ1 ꢆ k4[HPO24ꢁ] (Scheme 3). Therefore, kN ¼ k1 for both schemes
and formation of the Tꢀ intermediate is rate limiting. On the other
hand, for weakly basic anilines, kꢁ1 ꢇ k2 (Scheme 1) and
k
ꢁ1 ꢇ k4[HPO2ꢁ] (Scheme 3). In these cases, kN ¼ k1k2/k
for 3. Namely, the k4/k ratio should be similar or the same for
ꢁ1
ꢁ1
4
(Scheme 1) and kN ¼ k1k4[HPO24ꢁ]/kꢁ1 (Scheme 3) and breakdown
of Tꢀ to products is the rate-determining step in Scheme 1 (k2
step) and proton transfer from Tꢀ to HPO42ꢁ is rate limiting in
Scheme 3 (k4 step).
the reaction of a given aniline with 2 and 3. Since for the
reactions following Scheme 1 the kꢁ1/k2 ratio is related to the pKa0
value, Eqn (7),[17] a similar equation can be derived for Scheme 3,
Eqn (8). This means that the pKa at the center of the Brønsted
curvature should be similar or the same for both reactions series.
In fact the experimental pKa0 values obtained for the anilinolysis
of 2 and 3 are similar and their difference is within the
experimental error (see above). This indicates that the substi-
tution of 3-Cl by 3-MeO as the non-leaving group of the substrate
is not such a substantial change as to affect the pKa0 value.
Important pKa0 shifts have been observed when very different
substituents are placed in the non-leaving group. This is the case
of the pyridinolysis of bis(2,4-dinitrophenyl) and phenyl (2,4-
dinitrophenyl) carbonates, which show pKa0 values of 7.2 and 8.0,
respectively.[18] Namely, the addition of two nitro groups to the
non-leaving group of the latter compound lowers the pKa0 value
by 0.8 units.
According to the above arguments, the Brønsted slope for rate-
determining formation of the Tꢀ intermediate (b1) should be the
same for Schemes 1 and 3. On the other hand, the Brønsted slope
for rate-determining k2 or k4 steps (b2) should also be the same
for both schemes since the rate constants k2 and k4 are both
independent of the aniline basicity.[7,8,12,14] Namely, b2 is a
measure of the sensitivity of log K1 (K1 ¼ k1/kꢁ1) to the amine
basicity and is independent of k2 and k4 because the derivative of
log k2 or log k4 to the amine pKa is zero.[14]
Therefore, for all the reasons mentioned above, Eqn (2) can be
derived from Scheme 1 as well as from Scheme 3. Nevertheless,
the interpretation of pKa0 is different: for Scheme 1 this is the pKa
of an amine for which k ¼ k2,[15] whereas for Scheme 3 it is the
ꢁ1
pKa of an amine for which k ¼ k4[HPO2ꢁ].
ꢁ1
4
ꢀ
ꢁ
k
Since in the present work the plots of kobs against [aniline] were
performed at constant pH and three pH values were used for the
reactions of each aniline, maintaining constant the total buffer
concentration, Eqn (4) can yield Eqn (5). In this equation [B]tot
ꢁ1
log
¼ ðb2ꢁb1ÞðpKa0ꢁpKaÞ
(7)
(8)
k2
ꢀ
ꢁ
k
ꢁ1
log
¼ ðb2ꢁb1ÞðpKa0ꢁpKaÞ
represents the concentration of total buffer (H2POꢁ4 þ HPO42ꢁ
)
k4½HPO42ꢁꢂ
and FB ¼ KaB=ðKaB þ ½HþꢂÞ, where FB is the dianionic fraction of the
buffer and KaB is the dissociation constant of H2POꢁ4 . For anilines of
pKa < pKa0 (for which kꢁ1 > k4FB[B]tot), Eqn (5) simplifies to Eqn (6).
According to Scheme 3, the plots of kobs versus [aniline] should be
linear, as found. Nonetheless, since the slopes of the plots (kN) are
given by Eqns (5) and (6), these slopes should be pH dependent,
which was not found. Nevertheless, we think that the pH range
used (7.2–7.8 in some cases and 6.5–7.5 in others) was too small
to detect this effect.
As seen in Fig. 1, the Brønsted plot for the anilinolysis of
thionocarbonate 1 is linear. The value of the slope is b ¼ 0.62,
which is consistent with a concerted mechanism (shown in
Scheme 4) since most of these processes show Brønsted slopes
within the range 0.5–0.7.[7,8] Nonetheless, it is known that the b
value alone is not enough to conclude that a mechanism is
concerted. It is also necessary to make sure that the expected pKa
value at the center of the Brønsted curvature ðpKa0Þ for a
hypothetical stepwise mechanism is within the pKa range
used.[2,19]
If the mechanism of the anilinolysis of 1 were stepwise and the
curvature center were located at a pKa value greater than 6, the
experimental b value (0.62) would correspond to b2 (the Brønsted
slope when the k2 step is rate determining).[7,8,14] Nevertheless,
this value is too small to be considered as b2 since these values
are usually in the range 0.8–1.1.[7,8,14] On the other hand, if for the
hypothetical stepwise mechanism the value of pKa0 were lower
than pKa ca. 3 (that of the least basic aniline), the k1 step would be
rate-limiting and the Brønsted slope would be b1. The
experimental value (b ¼ 0.62) is too large in this case in
comparison with those obtained in stepwise aminolysis when
formation of the zwitterionic tetrahedral intermediate is rate-
determining (b1 ¼ 0.1–0.3).[7,8,14] Therefore, the stepwise mech-
k1k4FB½Bꢂtot
kN
¼
(5)
(6)
k
ꢁ1 þ k4FB½Bꢂtot
k1k4FB½Bꢂtot
kN
¼
k
ꢁ1
Greater values of kN were found for the reactions of anilines
with 2 compared to those with 3 (Table 4 and Fig. 1). For
the reactions of both compounds with the most basic aniline
(4-phenylendiamine) the formation of the intermediate Tꢀ is
rate-limiting (k1 step in Scheme 3). The larger kN (¼k1) value
for the reaction of this amine with 2, relative to that with 3, can
be explained by considering the inductive and resonance
effects of the substituents in the thionocarbonates. Since 3-Cl
in 2 is more electron-withdrawing than 3-MeO in 3,[16] the
wileyonlinelibrary.com/journal/poc
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 603–610

KINETICS OF ANILINOLYSIS OF THIONOCARBONATES
Scheme 4.
anism can be ruled out for the anilinolysis of thionocarbonate 1
and the concerted process (Scheme 4) seems more probable.
Effect of the amine nature
The reactions of SA amines with thionocarbonate 1 in water
proceed by a concerted mechanism.[10] On the other hand, it is
known that anilines are worse nucleofuges than isobasic SA
amines from a zwitterionic tetrahedral intermediate (Tꢀ).[7,8]
Therefore, the hypothetical intermediate Tꢀ formed with anilines
should be less unstable than that formed with SA amines.
Nevertheless, this intermediate should be more unstable in
aqueous ethanol (this work) than in water.[14] Therefore, it seems
reasonable that the stability of the intermediate Tꢀ be similar in
both reaction series and the anilinolysis of 1 in 44 wt.% ethanol–
water be concerted, as is the SA aminolysis of the same
compound in water.
On the other hand, the reactions of thionocarbonate 1 with
pyridines in water are driven by a stepwise mechanism.[10] The
fact that the reactions of this thionocarbonate with anilines in
aqueous ethanol are concerted (this work) is in accordance with
the findings that both anilines and aqueous ethanol destabilize
the intermediate Tꢀ in comparison with isobasic pyridines and
water, respectively.[7,8] The destabilization of this intermediate
caused by anilines, relative to isobasic pyridines, has been
ascribed to the superior nucleofugality from Tꢀ of anilines in
comparison with isobasic pyridines.[7,8]
Figure 2. Brønsted-type plots (statistically corrected) for the reactions of
anilines with (*) bis(4-nitrophenyl) thionocarbonate (1) and (~) bis(4-
nitrophenyl) carbonate (4), in 44 wt.% ethanol–water, at 25.0 8C, ionic
strength 0.2 M (KCl).
—
because of the stronger p-bond energy of C O, by 40 kcal,
—
[21]
—
relative to that of C S.
—
Figure 2 shows the Brønsted plots obtained for the anilinolysis
of thionocarbonate 1 (this work) and carbonate 4,[20] both in
44 wt.% ethanol–water. It can be observed that the thiono
compound is more reactive toward anilines (by one order of
magnitude) than the corresponding carbonate. The greater value
of kN for the reactions with 1, relative to 4, can be explained on
the basis of the ‘‘hard and soft acids and bases’’ (HSAB)
principle:[22] anilines, which are considered relatively soft bases,
should react faster with the relatively soft thiocarbonyl group of
the thionocarbonate compared with the relatively hard carbonyl
group of the carbonate.[22,23]
Effect of the non-leaving group
As discussed above, for the reactions of thionocarbonates 2 and 3
with anilines the change of 3-MeO by 3-Cl as the substituent in
the non-leaving group does not change significantly the pKa
value at the curvature center ðpKa0Þ, nor the reaction mechanism.
Effect of the electrophilic group
The reactions of anilines with bis(4-nitrophenyl) carbonate (4) in
44 wt.% ethanol–water take place by a concerted mechanism.[20]
Since the reactions of anilines with thionocarbonate 1 in the
same solvent mixture are also concerted, this means that
substitution of CO by CS as the electrophilic center of the
substrate does not change the mechanism. This is despite CS is
known to stabilize the intermediate Tꢀ relative to CO.[21] The
destabilization by CO has been explained by the superior ability
of Oꢁ, compared with Sꢁ, in the intermediate Tꢀ to form a double
bond and expel both the amine and leaving group. This is
EXPERIMENTAL
Materials
Anilines were purified by recrystallization or distillation. Thiono-
carbonates 1, 2, and 3 were prepared as described.[10,12] These
compounds were identified by their spectroscopic properties and
mp. The latter are: 195–196, lit[24] 196–197 8C for 1, 97.5–98.1 8C,
lit[12] 97.2–98.0 8C for 2 and 112.5–113.2 8C, lit[12] 112.3–113.2 8C
for 3.
J. Phys. Org. Chem. 2011, 24 603–610
Copyright ß 2010 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

E. A. CASTRO, R. ACEVEDO AND J. G. SANTOS
Kinetic measurements
REFERENCES
The kinetics of the reactions were analyzed through a diode array
spectrophotometer in 44 wt.% ethanol–water, at 25.0 ꢀ 0.18C and
an ionic strength of 0.2 M (maintained with KCl). The reactions
were followed at the 300–500 nm wavelength range.
[1] W. P. Jencks, Catalysis in Chemistry and Enzymology, McGraw-Hill, New
York, 1969.
[2] A. Williams, Chem. Soc. Rev. 1994, 23, 93–100.
[3] A. Williams, Free Energy Relationships in Organic and Bio-Organic
Chemistry, The Royal Society of Chemistry, Cambridge, 2003.
[4] D. D. Sung, I. S. Han, I. Lee, J. Sulfur Chem. 2007, 28, 483–491.
[5] I. H. Um, E. Y. Kim, H. R. Park, S. E. Jeon, J. Org. Chem. 2006, 71, 2302–
2306.
[6] I. H. Um, S. Yoon, H. R. Park, H. J. Han, Org. Biomol. Chem. 2008, 6,
1618–1624.
[7] E. A. Castro, Chem. Rev. 1999, 99, 3505–3524.
[8] E. A. Castro, J. Sulfur Chem. 2007, 28, 401–429.
[9] E. A. Castro, M. Cubillos, M. Aliaga, S. Evangelisti, J. G. Santos, J. Org.
Chem. 2004, 69, 2411–2416.
˜
[10] E. A. Castro, J. G. Santos, J. Tellez, M. I. Umana, J. Org. Chem. 1997, 62,
6568–6574.
[11] R. P. Bell, The Proton in Chemistry, Methuen, London, UK, 1959.
p 159.
[12] E. A. Castro, L. Leandro, N. Quesieh, J. G. Santos, J. Org. Chem. 2001,
The reactions were studied under at least 10-fold excess of
aniline over the substrate, the initial concentration of the latter
being 2.5 ꢄ 10ꢁ5 M. Under these conditions pseudo-first-order
rate coefficients (kobs) were found throughout, the kinetics being
measured for at least five half-lives at 400 nm, following 4-
nitrophenoxide ion formation. In all the reactions the pH was
maintained constant by using 0.01 M phosphate buffer.
The kobs values obtained and the experimental conditions for
the studied reactions are detailed in the Supplementary Material.
The ranges of amine concentration and kobs values are
summarized in Tables 1–3.
66, 6130–6135.
[13] M. Eigen, Angew. Chem. Int. Ed. Engl. 1964, 3, 1–19.
Product studies
[14] M. J. Gresser, W. P. Jencks, J. Am. Chem. Soc. 1977, 99, 6963–
6970.
[15] E. A. Castro, R. B. Moodie, J. Chem. Soc. Chem. Commun. 1973, 828–
829.
[16] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195.
[17] E. A. Castro, C. Araneda, J. G. Santos, J. Org. Chem. 1997, 62, 126–
129.
For all reactions 4-nitrophenoxide anion was identified as one of
the products. This was achieved by comparison of the UV–vis
spectra after completion of these reactions with those of an
authentic sample of 4-nitrophenol, under the same reaction
conditions. For the reactions of thionocarbonate 1, quantitative
UV–vis analysis showed that 4-nitrophenol concentration at the
end of the kinetic measurements was the same as that of the
initial substrate, that is, only one 4-nitrophenol was produced for
every substrate molecule.
˜
[18] E. A. Castro, M. Acuna, C. Soto, C. Trujillo, B. Vasquez, J. G. Santos, J.
Phys. Org. Chem. 2008, 21, 816–822.
[19] A. Williams, Acc. Chem. Res. 1989, 22, 387–392.
[20] E. A. Castro, M. Gazitua, J. G. Santos, J. Org. Chem. 2005, 70, 8088–
8092.
[21] S. V. Hill, S. Thea, A. Williams, J. Chem. Soc. Perkin Trans 2 1983, 437–
446.
Acknowledgements
[22] R. G. Pearson, J. Chem. Educ. 1968, 45, 643–648.
[23] R. G. Pearson, J. Org. Chem. 1989, 54, 1423–1430.
[24] H. R. Al-Kazimi, D. S. Tarbell, D. Plant, J. Am. Chem. Soc. 1955, 77,
2479–2482.
The authors thank MECESUP of Chile (project RED QUIMICA UCH-
0408) and FONDECYTof Chile (projects 1060594 and 1095145) for
financial assistance.
wileyonlinelibrary.com/journal/poc
Copyright ß 2010 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 603–610