Structure - reactivity correlation in the reactions of pyrrolidine with O-ethyl S-aryl dithiocarbonates in aqueous ethanol

Article abstract of DOI:10.1002/(SICI)1097-4601(1997)29:2<113::AID-KIN4>3.0.CO;2-X

The reactions of pyrrolidine with O-ethyl S-(X-phenyl) dithiocarbonates (X = 4-methyl, 4-methoxy, H, 4-chloro, 4-nitro, 2,4-dinitro, and 2,4,6-trinitro) are subjected to a kinetic study in 44 wt% aqueous ethanol, 25.0°C, and ionic strength 0.2 M (maintained with KCl). Pseudo-first-order kinetics are found under amine excess. Linear plots of the pseudo-first-order rate coefficient against concentration of free-base pyrrolidine are obtained for all the reactions, nucleophilic rate coefficient (k_N) being the slope of such plots. The Bronsted-type plot (log k_N vs. pK_a for the leaving group) is linear with slope β_1g = -0.2, which is consistent with a mechanism through a tetrahedral intermediate (T^±) where its formation is rate determining. The β_1g value is very similar to that found in the same reactions in water. There is a great difference in the mechanism of the reactions of O-ethyl S-phenyl dithiocarbonate with pyrrolidine (order one in amine) and piperidine (complex order in amine) in aqueous ethanol, and this is attributed to a greater nucleofugality from T^± of piperidine rather than pyrrolidine.

Full text of DOI:10.1002/(SICI)1097-4601(1997)29:2<113::AID-KIN4>3.0.CO;2-X

Structure –Reactivity
Correlation in the Reactions
of Pyrrolidine with O-Ethyl
S-Aryl Dithiocarbonates in
Aqueous Ethanol
ENRIQUE A. CASTRO, MAURICIO CABRERA, and JOSE G. SANTOS
Facultad de Química (502), Pontificia Universidad Católica de Chile, Casilla 306, Santiago-22, Chile
ABSTRACT
The reactions of pyrrolidine with O-ethyl S-(X-phenyl) dithiocarbonates (X ϭ 4-methyl, 4-
methoxy, H, 4-chloro, 4-nitro, 2,4-dinitro, and 2,4,6-trinitro) are subjected to a kinetic study in
44 wt% aqueous ethanol, 25.0°C, and ionic strength 0.2 M (maintained with KCl). Pseudo-
first-order kinetics are found under amine excess. Linear plots of the pseudo-first-order rate
coefficient against concentration of free-base pyrrolidine are obtained for all the reactions,
the nucleophilic rate coefficient (kN ) being the slope of such plots. The Bronsted-type plot
(log kN vs. pKa for the leaving group) is linear with slope ␤lg ϭ Ϫ 0.2, which is consistent with
a mechanism through a tetrahedral intermediate (TϮ) where its formation is rate determin-
ing. The ␤lg value is very similar to that found in the same reactions in water. There is a great
difference in the mechanism of the reactions of O-ethyl S-phenyl dithiocarbonate with pyrro-
lidine (order one in amine) and piperidine (complex order in amine) in aqueous ethanol, and
this is attributed to a greater nucleofugality from TϮ of piperidine rather than pyrrolidine.
© 1997 John Wiley & Sons, Inc.
ethyl S-(2,4,6-trinitrophenyl) dithiocarbonate [1] and
a stepwise path through a tetrahedral intermediate in
the other reactions [2,3]. In the reaction of O-ethyl S-
phenyl dithiocarbonate the zwitterionic tetrahedral in-
termediate formed by the amine attack can be depro-
tonated to yield an anionic intermediate in a
kinetically significant step [2].
INTRODUCTION
We have studied the mechanism of the reactions of a
series of secondary alicyclic amines (HNR1R2 ϭ
piperidine, piperazine, 1-(2-hydroxyethyl)piperazine,
morpholine, 1-formylpiperazine, and piperazinium
ion) with O-ethyl S-(X-phenyl) dithiocarbonates
(X ϭ H, p-nitro, 2,4-dinitro, and 2,4,6-trinitro), see
eq. (1), in 44 wt% aqueous ethanol [1,2,3]. We have
found a concerted process in the aminolysis of O-
EtO9CS9S(PhX) ϩ HNR1R2 9:
EtO9C(S)9NR1R2 ϩ XPhSH (1)
We have also studied the above reactions in water [4]
and those of the same amines with the other three O-
ethyl S-(X-phenyl) dithiocarbonates (X ϭ p-chloro,
Received April 10, 1996; accepted July 1, 1996
International Journal of Chemical Kinetics, Vol. 29, 113–117 (1997)
© 1997 John Wiley & Sons, Inc.
CCC 0538-8066/97/020113-05

114
CASTRO, CABRERA, AND SANTOS
p-methyl, and p-methoxy) in water [5]. We have also
reported a kinetic investigation on the reactions of
pyrrolidine with all the above dithiocarbonates in wa-
ter [5].
and substituted benzenethiols (Sigma) were used as
purchased.
Kinetic Measurements
We now report a kinetic study of the reactions of
pyrrolidine with all the above substrates in 44 wt%
aqueous ethanol. Our aim is three fold: (i) to assess
the influence of the solvent on the kinetics and mech-
anism of the title reactions by comparison with the
same reactions in water; (ii) to compare the behavior
of pyrrolidine and the secondary alicyclic amines in
aqueous ethanol; and (iii) to determine the sensitivity
of the nucleofugality rate to leaving group basicity
(␤lg ) in aqueous ethanol (as the slope of a Bronsted-
type plot) and to compare it with that in water.
These were carried out by following spectrophoto-
metrically the corresponding benzenethiolate anion
release at 257 nm (substrates with X ϭ H, 4-chloro,
4-methyl, and 4-methoxy), 420 nm (X ϭ 4-nitro and
2,4-dinitro), and 390 nm (X ϭ 2,4,6-trinitro) through
the instruments and method described [1–5]. The re-
actions were carried out in 44 wt% aqueous ethanol
at 25.0 Ϯ 0.1°C and ionic strength 0.2 M (KCl). In
all cases the initial substrate concentration was c.a.
5
10Ϫ M. Under pyrrolidine excess, pseudo-first-order
rate constants (kobs ) were found for all reactions. The
experimental conditions for the reactions and the kobs
values are shown in Table I.
EXPERIMENTAL
Materials
Determination of pKa
The dithiocarbonates were prepared as reported
[4–6]. Pyrrolidine (Merck a.r.), ethanol (Merck a.r.),
The pKa values of p-chloro, p-methyl, and p-methoxy
benzenethiols and that of pyrrolidinium ion were ob-
Table I Experimental Conditions and Values of kobs and kN for the Present Reactionsa
Substrate
Substituent
(pKa ) b
c
d
102 [N]total
/M
103 kobs
No. of
Runs
kN
1
1
1
pH
/sϪ
/sϪ MϪ
4-CH3
(7.78)
10.70
11.07
11.67
10.77
11.07
11.37
10.70
11.20
11.70
10.77
11.07
11.37
10.77
11.07
11.37
10.70
10.32
11.07
10.32
10.70
10.95
11.07
1.0–25
0.4–14
3.9–68
2.2–63
5.3–54
2.8–29
2.6–29
2.8–34
0.4–15
0.5–16
0.8–64
2.6–46
2.1–108
4.1–105
2.5–16
1.2–21
2.5–16
1.7–19
2.3–20
1.6–20
1.3–11
1.3–10
1.4–12
1.6–8.1
8
8
5
6
4
6
9
7
6
0.9
1.0–8.0
1.0–8.0
0.5–5.0
0.5–5.0
0.2–8.0
0.1–2.7
0.1–8.0
0.9–10
4-CH3O
(7.77)
1.1
1.0
1.5
2.2
9.5
5.0
None
(7.70)e
4-Cl
(6.95)
5
0.5–14
0.6–11
10
10
6
6
6
6
6
6
6
4-NO2
(4.77)e
0.3–2.1
0.1–1.8
0.1–1.1
0.07–0.7
0.14–1.4
0.04–0.4
0.14–1.4
0.07–0.7
0.05–0.5
0.04–0.3
2,4-(NO2 )2
(3.50)e
2,4,6-(NO2 )3
6
5
5
a In 44 wt% aqueous ethanol, at 25.0°C, ionic strength 0.2 M (KCl).
b pKa of substituted benzenethiol, obtained under the kinetic conditions. The pKa of pyrrolidinium was 11.07.
c Concentration of total pyrrolidine (free amine ϩ conjugate acid).
d Value of the slope of a kobs vs. [free pyrrolidine] plot which includes all runs (at all the pH values).
e Values of pKa taken from ref. [3].

REACTIONS OF PYRROLIDINE
115
tained spectrophotometrically at 270 nm (the three
thiols) and 205 nm (pyrrolidinium), under the same
conditions of the kinetic measurements, as reported
[5,7,8]. The measurement of pH was made on a Ra-
diometer PHM-62 pH-meter. The pKa values of the
benzenethiols are shown in Table I. The value for
pyrrolidinium ion is 11.07. These are pKa “mixed”
values [8].
In the reactions of 6-SAA with 4-nitrophenyl and
2,4-dinitrophenyl O-ethyl dithiocarbonates in 44 wt%
aqueous ethanol, a first-order rate in free amine was
found [3]. This was explained by a faster nucleo-
fugality of XPhSϪ from T Ϯ compared to the deproto-
nation rate from the same intermediate, i.e.,
k2 ϾϾ k3 [HNR1R2 ] in Scheme I [3].
In contrast, a complex order in amine, intermedi-
ate between one and two, was found in the same reac-
tions with O-ethyl phenyl dithiocarbonate [2]. This
was attributed to a lower nucleofugality rate of PhSϪ,
compared to the 4-nitro and 2,4-dinitro substituted
anions, which makes possible a competition between
the deprotonation step and the expulsion of the amine
Product Studies
The presence of the substituted benzenethiolate ions
was determined spectrophotometrically by compari-
son of the UV-vis spectra at the end of the reactions
with those of the corresponding benzenethiolate ions
under the same experimental conditions.
from T Ϯ, such that k2 Ͻ k3 [HNR1R2 ] Ϸ kϪ for
1
these reactions [2].
The reaction of pyrrolidine with O-ethyl phenyl
dithiocarbonate in 44 wt% aqueous ethanol (this
work) exhibits a first-order dependence on the free
amine, in spite of that the concentration of pyrroli-
dine was ca. 1/10 of that of piperidine in its reaction
with the same substrate [2]. This means that for the
pyrrolidine reaction k3[pyrrolidine] ϽϽ k2 , and ac-
cording to the Bronsted-type plot obtained (see be-
low), k 1 ϽϽ k2 .
RESULTS AND DISCUSSION
The reactions of pyrrolidine with the title dithiocar-
bonates, depicted in eq. (2) where HN(CH2 )4 is
pyrrolidine, are governed by the rate law in eqs. (3)
and (4), under the experimental conditions.
EtO9CS9S(PhX) ϩ HN(CH2 )4 9:
EtO9CS9NϩH(CH2 )4 ϩ XPhSϪ (2)
TheϪse results indicate a much lower nucleofugality
from T Ϯ of pyrrolidine than piperidine (at least ten
fold), in spite of their similar basicities (pKa 11.1 and
10.8 in 44 wt% aqueous ethanol and 25°C, respec-
tively).
d[XPhSϪ]/dt ϭ kobs [EtOCSSPhX]
obs ϭ kN [HN(CH2 )4 ]
(3)
(4)
k
Plots of kobs vs. free pyrrolidine concentration were
linear with the slopes (kN ) independent of pH for all
the reactions.
Scheme I shows the general mechanistic behavior
exhibited by the reactions of six-member ring sec-
ondary alicyclic amines (6-SAA) with the title series
of dithiocarbonates in 44 wt% aqueous ethanol [2]. In
this scheme the k3 step is deprotonation of the tetra-
hedral intermediate T Ϯ by the free amine to yield the
anionic intermediate T Ϫ.
The different behavior of pyrrolidine and piperi-
dine has some precedents in the literature [9,10].
Bunnett et al. found a great rate difference between
pyrrolidine and piperidine in their reactions toward
substituted phenyl phenyl ethers in dioxane/water [9]
and 2,4-dinitro-1-naphthyl ethyl ether in dimethyl
sulfoxide solution [10]. They attributed the rate dif-
ference to the superior ability of the pyrrolidino moi-
ety in a Meisenheimer complex to expel the leaving
group from it, compared to the ability of a piperidino
moiety from a similar complex [9,10]. The smaller
ability of the latter group was explained by large
steric compressions in the Meisenheimer complex of
the piperidine reaction. Although the tetrahedral in-
termediate T Ϯ of Scheme I is different than the above
complex, this explanation could also be valid to ac-
count for the larger nucleofugality from T Ϯ of piperi-
dine compared to pyrrolidine, as found in this work.
With the kN values for the reactions of this study
(except that of the trinitro derivative) and the pKa of
the conjugate acids of the leaving groups of the sub-
strates (Table I), the Bronsted-type plot of Figure 1
was obtained. The slope is ␤lg ϭ Ϫ 0.20 Ϯ 0.05, and
the correlation coefficient 0.94. The reason for not in-
Scheme I

116
CASTRO, CABRERA, AND SANTOS
both the larger basicity and the smaller nucleofugality
of the latter relative to the former amine (smaller kϪ
1
for pyrrolidine). It is noteworthy that the k2 value
should not be much affected by the amine nature due
to the absence of an electron pair on the nitrogen
atom of the amino moiety in T Ϯ [11,12].
If the rate determining step is k1 for the reactions
of pyrrolidine with dithiocarbonates in aqueous
ethanol (k2 ϾϾ k 1 in Scheme I), it is reasonable that
the same step beϪrate-limiting in water [5] since: (i)
the kϪ value should decrease by the change from
1
aqueous ethanol to water in view that the transition
state for the k1 step is less polar than T Ϯ [12] and (ii)
the k2 value should not be significantly affected by
the solvent polarity, in view that both the transition
state for breakdown to products and T Ϯ are highly
polar [12].
Figure 1 Bronsted-type plot for the leaving group for the
reactions of pyrrolidine with O-ethyl S-(X-phenyl) dithio-
carbonates in 44 wt% aqueous ethanol, 25.0°C, and ionic
strength 0.2 M.
As expected, the kN ϭ k1 value in Table I in-
creases as the nucleofuge basicity decreases due to
increasing electron withdrawal from the phenyl sub-
stituents in the leaving group, which results in an in-
creasingly more positive thiocarbonyl carbon atom.
An exception to this behavior is presented by the re-
action of O-ethyl 2,4,6-trinitrophenyl dithiocarbonate
which shows a smaller k1 value than the 2,4-dinitro-
phenyl derivative. The same exception was found in
the reactions of these two substrates with piperidine
in both water [4b] and 44 wt% aqueous ethanol [1,3].
The above discrepancy regarding the k1 values for
the pyrrolidine reactions in 44 wt% aqueous ethanol
(Table I) could be explained by steric reasons due to
the extra ortho-nitro substituent in the trinitro deriva-
tive. Another explanation could be a change in the re-
action mechanism from stepwise in the reaction of
pyrrolidine with the dinitro substrate to concerted for
that of the trinitro compound. This is the mechanistic
change that was found in the reactions of piperidine
with these substrates in 44 wt% aqueous ethanol
[1,3]. This is the reason why the trinitro compound
was not included in the Bronsted-type plot of Fig-
ure 1.
cluding the trinitro substrate in the Bronsted plot will
be discussed later.
The above Bronsted slope is similar to the corre-
sponding ones found in the aminolyses of acetate es-
ters [11], diaryl carbonates [12], substituted triazin-
pyridinium ions [13], and O-ethyl dithiocarbonates in
water [5]. It is also in line with those obtained in the
reactions of piperidine with O-ethyl dithiocarbonates
in 44 wt% aqueous ethanol [3] and in the title reac-
tions in water [5]. In all these works the low ␤lg value
(between 0 and Ϫ0.3) has been associated with a
stepwise mechanism where the formation of the tetra-
hedral intermediate is the rate-determining step. The
fact that in the reactions of pyrrolidine with dithiocar-
bonates the ␤lg value does not change in going from
water (␤lg ϭ Ϫ 0.23 Ϯ 0.1) [5] to 44 wt% aqueous
ethanol, means that the effective charge development
on the leaving group from reactants to the transition
state of the k1 step is approximately the same in both
solvents [13,14].
Therefore, according to these arguments, the kN
values obtained in this work (Table I) correspond to
the k1 values of Scheme I. This means that for all the
present reactions k2 ϩ k3 [pyrrolidine] ϾϾ kϪ1 (with
the k2 path faster than that of k3 , see above).
A biphasic Bronsted-type plot was obtained in the
reactions of a series of 6-SAA with 4-nitrophenyl and
2,4-dinitrophenyl O-ethyl dithiocarbonates in 44 wt%
aqueous ethanol [3]. The experimental point for
piperidine lies on the linear portion at high pKa for
both reactions; i.e., for the reactions of piperidine the
rate-determining step is formation of the zwitterionic
tetrahedral intermediate. If this is so for piperidine, it
should be even more so for pyrrolidine in view of
Comparison of the k1 values obtained in the pres-
ent work with those found for the same reactions in
water [5] reveals that these values are ca. 3–5 times
larger in water. This should be due to the fact that the
transition state for the first step is more polar than re-
actants and therefore the former should be more sta-
bilized (relative to reactants) by the change to a more
polar solvent.
We thank FONDECYT of Chile for financial assistance for
this work.

REACTIONS OF PYRROLIDINE
117
7. N. Alvarado and J. G. Santos, J. Chem. Soc. Perkin
Trans. 2, 1061 (1989).
BIBLIOGRAPHY
1. E. A. Castro, M. Cubillos, G. Muñoz, and J. G. Santos,
Int. J. Chem. Kinet., 26, 571 (1994).
2. E. A. Castro, M. Cabrera, and J. G. Santos, Int. J.
Chem. Kinet., 27, 49 (1995).
8. A. Albert and E. P. Serjeant, The Determination of Ion-
ization Constants, Chapman and Hall, London, 1971.
9. J. F. Bunnett and A. V. Cartaño, J. Am. Chem. Soc.,
103, 4861 (1981).
3. E. A. Castro, G. Muñoz, M. Salas, and J. G. Santos,
Int. J. Chem. Kinet., 27, 987 (1995).
10. J. F. Bunnett, S. Sekiguchi, and L. A. Smith, J. Am.
Chem. Soc., 103, 4865 (1981).
4. (a) M. Cabrera, E. A. Castro, M. Salas, J. G. Santos,
and P. Sepúlveda, J. Org. Chem., 56, 5324 (1991); (b)
E. A. Castro, F. Ibáñez, M. Salas, J. G. Santos, and P.
Sepúlveda, J. Org. Chem., 58, 459 (1993).
11. A. C. Satterthwait and W. P. Jencks, J. Am. Chem. Soc.,
96, 7018 (1974).
12. M. J. Gresser and W. P. Jencks, J. Am. Chem. Soc., 99,
6963 (1977).
5. E. A. Castro, M. Cubillos, F. Ibáñez, I. Moraga, and
J. G. Santos, J. Org. Chem., 58, 5400 (1993).
6. J. R. Cox, C. L. Gladys, L. Field, and D. E. Pearson, J.
Org. Chem., 25, 1083 (1960).
13. M. R. Cullum, A. H. M. Renfrew, D. Rettura, J. A. Tay-
lor, J. M. J. Whitmore, and A. Williams, J. Am. Chem.
Soc., 117, 9200 (1995).
14. A. Williams, Acc. Chem. Res., 22, 387 (1989).