Mechanisms of acid decomposition of dithiocarbamates. 2. Efficiency of the intramolecular general acid catalysis

Article abstract of DOI:10.1021/jo981382i

The acid decomposition of ethylenebis(dithiocarbamate) (EbisDTC) and glycinedithiocarboxylate (glyDTC) was studied in water at 25 °C in the range of rio -5 to pH 5. The acid dissociation constants of all species involved were calculated from LFER and from the pH-rate profiles. According to the pK(a) of the parent amine of the reactive species, both compounds decompose through the dithiocarbamate anion and a zwitterion intermediate. The intermolecular N-protonation rate constant of the carboxylic conjugate acid of glyDTC anion is 12.6 M^-1 s^-1, slower than the C-N breakdown. This species also cleaves through an intramolecular general acid-catalyzed mechanism where the rate constant for the N-protonation is (7.1 ± 4.2) x 10³ s^-1 and the efficiency of the proton-transfer step as measured by the effective molarity is (5.6 ± 3.3) x 10² M. The acid decomposition of the dithiocarbamic conjugate acid of EbisDTC anion proceeds through a fast N- protonation and a slower C-N breakdown. The intramolecular general acid catalysis rate constant is (8.2 ± 2.8) x 10⁶ s^-1, but the efficiency of this fast proton transfer is only (14.3 ± 4.9) M. The intramolecular general acid catalysis of the free acid forms of the carboxylic and dithiocarbamic groups is unfavorable for about 4 kcal mol^-1 with respect to the protonation of the external hydron, and consequently, no external buffer catalysis is expected to be observed for dithiocarbamates that decompose through a zwitterion intermediate. The difference between the pK(b) of the proton acceptor and the pK(a) of the donor follows the order of the proton efficiency. Estimation of the strength of the hydrogen bonding in the reagent and product supports the assumption that a thermodynamically favorable change of hydrogen bonding from reagent to product increases the efficiency of proton transfer.

Full text of DOI:10.1021/jo981382i

J . Org. Chem. 1999, 64, 1807-1813
1807
Mech a n ism s of Acid Decom p osition of Dith ioca r ba m a tes. 2.
Efficien cy of th e In tr a m olecu la r Gen er a l Acid Ca ta lysis
Eduardo Humeres,*,† Nito A. Debacher, and M. Marta de S. Sierra
Department of Chemistry, Universidade Federal de Santa Catarina, 88040-900 Florianopolis-SC, Brazil
Received J uly 15, 1998
The acid decomposition of ethylenebis(dithiocarbamate) (EbisDTC) and glycinedithiocarboxylate
(glyDTC) was studied in water at 25 °C in the range of Ho -5 to pH 5. The acid dissociation constants
of all species involved were calculated from LFER and from the pH-rate profiles. According to the
pKa of the parent amine of the reactive species, both compounds decompose through the
dithiocarbamate anion and a zwitterion intermediate. The intermolecular N-protonation rate
constant of the carboxylic conjugate acid of glyDTC anion is 12.6 M-1 s-1, slower than the C-N
breakdown. This species also cleaves through an intramolecular general acid-catalyzed mechanism
where the rate constant for the N-protonation is (7.1 ( 4.2) × 103 s-1 and the efficiency of the
proton-transfer step as measured by the effective molarity is (5.6 ( 3.3) × 102 M. The acid
decomposition of the dithiocarbamic conjugate acid of EbisDTC anion proceeds through a fast
N-protonation and a slower C-N breakdown. The intramolecular general acid catalysis rate constant
is (8.2 ( 2.8) × 106 s-1, but the efficiency of this fast proton transfer is only (14.3 ( 4.9) M. The
intramolecular general acid catalysis of the free acid forms of the carboxylic and dithiocarbamic
groups is unfavorable for about 4 kcal mol-1 with respect to the protonation of the external hydron,
and consequently, no external buffer catalysis is expected to be observed for dithiocarbamates that
decompose through a zwitterion intermediate. The difference between the pKb of the proton acceptor
and the pKa of the donor follows the order of the proton efficiency. Estimation of the strength of
the hydrogen bonding in the reagent and product supports the assumption that a thermodynamically
favorable change of hydrogen bonding from reagent to product increases the efficiency of proton
transfer.
In tr od u ction
General acid catalysis is not observed in the acid
cleavage of alkyldithiocarbamates,1,7-9 although some
previous publications have reported the presence of
general catalysis for this reaction.3,10 For alkyldithiocar-
bamates that decompose through the zwitterion, the pKa
of the intermediate is much lower than the solvated
proton, making the proton transfer for N-protonation
thermodynamically unfavorable, and a general acid
catalyst with pKa higher than the hydron is not expected
to be effective, as was confirmed experimentally.1
Alkyldithiocarbamate anions undergo acid decomposi-
tion through a zwitterion intermediate when the pKa of
the parent amine is less than ca. 10.1 The mechanism
explains how the hydron is transferred to the nitrogen
and how the transition state adjusts itself to the change
of about 14 pKa units of the nitrogen as the C-N bond is
broken.
At pKa values of the parent amine higher than ca. 10,
the alkyldithiocarbamates decompose through another
mechanism that has been proposed to occur by a water-
catalyzed S to N proton transfer concerted with the C-N
bond cleavage.1 In this mechanism, to reach the transi-
tion state, the C-N bond has to twist, inhibiting the
resonance of the nitrogen with the thiocarbonyl group,
increasing its basicity, and making the transfer of the
proton from the thiocarboxylic sulfur to the nitrogen
thermodynamically favorable. These two mechanisms,
through the zwitterion or by intramolecular proton
transfer, agree with those that have been proposed
previously2-7 and show that the basicity of the parent
amine is determinant for the change of one to the other.
Dithiocarbamates, in particular ethylenebis(dithiocar-
bamate) metallic salts, are widely used as fungicides.
However, their mechanism of action upon fungi is not
well understood.11 The results of a preliminary study of
the acid cleavage of ethylenebis(dithiocarbamate) are
insufficient for the drawing of any conclusions regarding
the mechanisms involved.12
(5) Takami, F.; Tokuyama, K.; Wakahara, W.; Maeda, T. Chem.
Pharm. Bull. 1973, 21, 1311.
(6) Takami, F.; Tokuyama, K.; Wakahara, W.; Maeda, T. Chem.
Pharm. Bull. 1973, 21, 594.
(7) Ewing, S. P.; Lockshon, D.; J encks, W. P. J . Am. Chem. Soc.
1980, 102, 3072.
(8) Hallaway, M. Biochim. Biophys. Acta 1959, 36, 538.
(9) (a) Zahradnik, R.; Zuman, P. Collect. Czech. Chem. Commun.
1959, 24, 1132. (b) Zahradnik, R. Collect. Czech. Chem. Commun. 1958,
23, 1435.
† Tel.: 55 48 331 9219. Fax: 55 48 331 9711. E-mail: humeres@
mbox1.ufsc.br or humeres@qmc.ufsc.br.
(1) Humeres, E.; Debacher, N. A.; Sierra, M. M. de S.; Franco, J .
D.; Schutz, A. J . Org. Chem. 1998, 63, 1598.
(2) J oris, S. J .; Aspila, K. I.; Chakrabarti, C. L. Anal. Chem. 1970,
42, 647.
(10) Zahradnik, R. Collect. Czech. Chem. Commun. 1956, 21,
1111.
(3) De Filippo, D.; Deplano, P.; Devillanova, F.; Trugu, E. F.; Verani,
G. J . Org. Chem. 1973, 38, 560.
(4) Takami, F.; Tokuyama, K.; Wakahara, S.; Maeda, W. Chem.
Pharm. Bull. 1973, 21, 329.
(11) (a) Cremlyn, R. J . Pesticides. Preparation and Mode of Action;
Wiley: Chichester, 1978. (b) Corbett, J . R.; Wright, K.; Baillie, A. C.
The Biochemical Mode of Action of Pesticides; Academic: London, 1984.
(12) Miller, D. M.; Latimer, R. A. Can. J . Chem. 1962, 40, 246.
10.1021/jo981382i CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/19/1999

1808 J . Org. Chem., Vol. 64, No. 6, 1999
Humeres et al.
Sch em e 1. Va lu es in P a r en th eses Ar e th e p Ka
In this work, we studied the pH-rate profiles of the
acid decomposition of ethylenebis(dithiocarbamate),
EbisDTC, and glycinedithiocarboxylate, glyDTC, eq 1, to
Va lu es
characterize the mechanisms involved and the possible
intramolecular catalysis by the carboxylic and dithiocar-
bamic groups and their conjugate acids. When properly
oriented, catalytic groups become highly efficient in
intramolecular reactions.13
Exp er im en ta l Section
Ma ter ia ls. Reagents were all of analytical grade. Ethyl-
enediamine was distilled before use, and dioxane was distilled
and filtered through an alumina column. Carbon disulfide was
purified over metallic mercury and over phosphorus pentoxide.
It was then distilled.
Disodium ethylenebis(dithiocarbamate)14 was prepared by
slowly adding carbon disulfide (61.0 g, 0.8 mol) to a solution
of ethylenediamine (31 g, 0.69 mol) in 150 mL of 5.3 M sodium
hydroxide (32 g, 0.8 mol) and allowing the mixture to react
for 3 h at room temperature. The product was precipitated,
recrystallized in acetone, and dried under vacuum: mp > 260
°C (dec);15 λmax (water) 257 and 284 nm.16
A solution of disodium glycine-N-dithiocarboxylate17-19 was
prepared from glycine (0.25 g, 3.3 mmol) added to a solution
of sodium hydroxide (0.13 g, 3.3 mmol) in water (7 mL), and
then stirring the solution until the glycine was dissolved. This
solution was treated with carbon disulfide (0.25 g, 3.3 mmol)
dissolved in dioxane (7 mL) and was allowed to react with
stirring until the glycine was completely consumed as shown
by a UV spectrum. The mixture was stripped of half the
solvent, under reduced pressure, and 7 mL of ethanol was
added. The product was not crystallized, but the solution was
kept in the refrigerator until examination, using aliquots of
the reaction mixture: λmax (water) 253 and 286 nm.
Acid Dissocia tion Con sta n ts of EbisDTC. Ethylenebis-
(dithiocarbamic) acids dissociate according to the mechanism
shown in Scheme 1, depending on the acidity of the medium.
The dissociation constant K1 of the conjugate acid 1a was
determined from a series of measurements of the absorbance
at 270 nm, between 4 and 14 M sulfuric acid.20 The mean value
of pK1 was -4.71 ( 0.07, and the statistically corrected value
Kin etics. The rate of decomposition of the dithiocarbamates
in water was followed spectrophotometrically in the region of
270-280 nm at 25 °C in a Cary 219 spectrophotometer. The
acidity ranged between Ho -5 to pH 5. The ionic strength of
solutions less acid than 1 M HCl was kept equal to unity with
KCl. Between pH 2 and 5, the solutions were buffered at least
at two buffer concentrations, and no general acid catalysis was
observed. For instance, at pH 3.1, when the concentration of
formate buffer was increased from 0.2 to 1.0 M, no general
catalysis resulted for EbisDTC.
Kinetics were all first order, except for EbisDTC between
pH 2.0 and 4.3, where the plot of log(A∞ - At) vs time was
biphasic, corresponding to two consecutive reactions. The rate
constant of the first step was calculated by the Guggenheim
method.21 The rate constant of the second step was calculated
from the straight line of the second part of the plot. This value
was similar to that obtained from the decomposition of the
N-2-aminoethyldithiocarbamate.1
of -4.41 is used in this text. For the monoanion 1e, pK6
)
3.15 ( 0.05 was calculated from the variation of absorbance
at 284 nm in the range of pH 1-5 and was also statistically
corrected to 2.85. The other pKa values of EbisDTC were
calculated from LFER1 and adjusted by an iterative program
(see below) to the pH-rate profile within a deviation of (0.04
pK unit, taking into consideration the thermodynamic con-
straint. The pKa values related to N-2-aminoethyldithiocar-
bamate had been determined previously.1
(13) (a) Menger, F. M. Acc. Chem. Res. 1985, 18, 128. (b) Menger,
F. M. Atual. Fis.-Quim. Org. 1989, 136.
(14) Klopping, H. L.; Van der Kerk, G. J . M. Recl. Trav. Chim. Pays-
Bas 1951, 70, 949.
(15) Frank, A. W. Appl. Spectrosc. 1982, 36, 282.
(16) Vandebeek, R. R.; J oris, S. J .; Aspila, K. I.; Chakrabarti, C. L.
Can. J . Chem. 1970, 48, 2201.
(17) Frank, A. W. Phosphorus, Sulfur Silicon Relat. Elem.1990, 54,
109.
(18) Andreasch, R. Monatsh. 1910, 31, 785.
(19) Budevsky, O.; Russeva, E.; Mesrob, B. Talanta 1966, 13, 277.
(20) Albert, A.; Sergeant, E. P. The determination of Ionization
Constants; Chapman and Hall: London, 1971.
A generalized minimum least-squares program was used to
fit the experimental points to the theoretical curve of the pH-
rate profiles.
(21) Guggenheim, E. A. Philos. Mag. 1926, 2, 528.

Mechanisms of Acid Decomposition of Dithiocarbamates
J . Org. Chem., Vol. 64, No. 6, 1999 1809
Ta ble 1. Dissocia tion a n d Ra te Con sta n ts Rela ted to th e Acid Clea va ge of Eth ylen ebis(d ith ioca r ba m a te) a n d
Glycin ed ith ioca r boxyla te in Wa ter a t 25 °Ca
no. order
R-CH2-NHC(S)SH, R )
σI
pK+
pKa
pKN
104ko, s-1
kH, M-1 s-1
1d
1e
1f
2
3g
3h
3i
CH2NH+C(SH)2
CH2NHC(S)SH
CH2NHC(S)S-
0.05b
0.04b
0.04b
0.36g
0.56
-4.41
-1.50
-1.12
-5.60d
-3.7
2.49
2.87b
2.85
9.86c
9.95c
9.95c
7.52h
5.2
20.1 ( 4.8
4.5 ( 0.6
0.62 ( 0.08d
0.33 ( 0.02e
0.66 ( 0.05e,f
0.21 ( 0.01
9.3 ( 1.6
CH2NH+
2.00d
-0.07
2.1
20.9 ( 0.2d
352 ( 126
22.2 ( 2.8
1.70 ( 0.22
3
COOH+
0.03 ( 0.007d
0.28 ( 0.008e
0.34 ( 0.01e
2
COOH
0.34i
-3.8
7.8
COO-
-0.17g
-1.8
3.3j
9.91j
This work unless indicated; pKN, pKa of the parent amine. Extrapolated from LFER.1 c Extrapolated from LFER for substituted
a
b
methylammonium ions.28 From the pH-rate profile. e Extrapolated from Figure 2. f Statistically corrected for two basic sites. g Reference
28a. Reference 25. Exner, O. In Advances in Linear Free Energy Relationships; Chapman, N. B., Shorter, J ., Eds.; Plenum: New York,
1972; Chapter 1. Reference 20.
d
h
i
j
Sch em e 2. Va lu es in P a r en th eses Ar e th e p Ka
Resu lts a n d Discu ssion
Va lu es
Su b st it u en t Con st a n t s of t h e Dit h ioca r b a m ic
Gr ou p . The dithiocarbamic acids dissociate according to
eq 2. The values of σI for the different forms of the
methylenedithiocarbamic group (R ) CH2NH+C(SH)2,
CH2NHC(S)SH, CH2NHCS2-) were obtained from the
LFER of pKa (or pKa+) vs σI1 (Table 1).
The substituent constants of the dithiocarbamic groups
were calculated from the series X-CH2CH2NHC(S)SH,
(X ) NH3+, OMe, OH, H, Et).1 For NH+C(SH)2, NHC-
(S)SH, and NHC(S)S-, σI was found to be 0.15, 0.13, and
0.14, respectively. The close similarity of σI for the
different forms of the dithiocarbamic group, despite the
charge on the sulfur atom, suggests that there is little
difference of the positive charge on the nitrogen between
the conjugate acid (I), the free acid (II and III), and the
anion (IV and V) because there extensive delocalization
of the nitrogen electron pair to the thiocarbonyl group
with predominance of forms III and V. This conclusion
is in agreement with results from UV22 and infrared23
spectra, i.e., that there is a strong character of double
Acid Dissocia tion Con sta n ts of Glycin ed ith ioca r -
boxyla te. The microscopic pKa values of glycine (Scheme
2) were obtained assuming that pK22 is equal to that of
bond between carbon and nitrogen atoms in the dithio-
carbamic moiety.
The large decrease of about 14 units in the pKa
the methyl ester,24 and taking into account the values of
calculated for the protonation of nitrogen as a conse-
the experimental macroscopic constants pKA ) 2.36 )
quence of the dithiocarboxylate group in alkyldithiocar-
20,25-27
pK25 and pKB ) 9.91 ) pK23.
The thermodynamic
bamates has been ascribed to this strong electron with-
constraint permitted the calculation of pK20. The correla-
drawing effect of the thiocarbonyl group.1
tion for the pKN of substituted methylammonium ions28
(22) (a) J ansen, M. J . Recl. Trav. Chim. Pays-Bas 1960, 79, 454. (b)
J ansen, M. J . Recl. Trav. Chim. Pays-Bas 1960, 79, 464. (c) J ansen,
M. J . Recl. Trav. Chim. Pays-Bas 1960, 79, 1068.
(23) (a) Thorn, G. D.; Ludwig, R. A.Can. J . Chem. 1954, 32, 872. (b)
Thorn, G. D. Can. J . Chem. 1960, 38, 1439. (c) Chatt, J .; Duncanson,
L. A.; Venanzi, L. M. Suom. Kemistil. 1956, B29, 75. (d) Chatt, J .;
Duncanson, L. A.; Venanzi, L. M. Nature 1956, 177, 1042. (e) Wahlberg,
A. Acta Chem. Scand., Ser. A 1976, 30, 433. (f) Cox, B. G.; de Maria,
P. J . Chem. Soc., Perkins Trans. 2 1977, 1385.
(24) Sober, H. A., Hart, R. A., Eds. Handbook of Biochemistry; CRC
Press: Cleveland, 1968.
(25) Kortum, G., Vogel, W., Andrussow, K., Eds. Dissociation
Constants of Organic Acids in Aqueous Solution; IUPAC-Butter-
worth: London, 1961.
(26) Adams, E. Q. J . Am. Chem. Soc. 1916, 38, 1503.
(27) Fleck, G. M. Equilibria in Solution; Holt, Rinehart, Winston:
New York, 1966; Chapter 5.

1810 J . Org. Chem., Vol. 64, No. 6, 1999
Humeres et al.
F igu r e 2. Brønsted plot of the specific acid catalysis rate
constants and the pKN of the leaving amine for the acid
decomposition of substituted methyldithiocarbamates in water
at 25 °C: O, ref 1; b, ref 7; 0, this work.
F igu r e 1. pH-rate profile of the acid decomposition of
ethylenebis(dithiocarbamate) in water at 25 °C: O, cleavage
of the first group; b, cleavage of the second group; 0, N-2-
aminoethyldithiocarbamate; s, calculated rate constant; ---,
contribution of the kinetic terms.
The pKN values of the parent amine of species 1b,d , 1c,e,
and 1f are 9.86, 9.95, and 9.95, respectively (Table 1,
Scheme 1). Therefore, the decomposition should occur
through the dithiocarbamate anion and the zwitterion
species with specific acid catalysis on 1d , 1e, and 1f and
was used to calculate pK21, assuming that the substituent
+
constant of the COOH2 group is similar to that of the
NH3+ group. The pKa values of the substituted conjugate
acetic acids were estimated from eq 329,30 adjusted for the
rate constants kH(, kH-, and kH respectively,1 plus the
)
possible intramolecular general acid catalysis on ld and
1e by the acid forms of the other dithiocarbamic moiety,
with rate constants ki( and ki- as shown in eq 5a, where
X1d , X1e, and X1f are the corresponding molar fractions.
Equation 5a is kinetically equivalent to eq 5b.
pKa+ ) -7.5σI - 6.8 (n ) 3, r ) 0.710)
(3)
thermodynamic constraint with a standard deviation of
(0.5 pK unit.
The set of equilibrium constants of glyDTC (Scheme
2) was obtained as a first approximation from LFER1 and
eq 3. The pKa’s of the substituted acetic acids were
calculated from eq 4.28
kobs ) [kH(aH+ + k (]X + [kH-aH+ + k -]X
+
1e
i
1d
i
kH)aH+X (5a)
1f
pKa ) -3.95σI + 4.71 (n ) 13, r ) 0.996)
(4)
1c K2
aH
1e K2K4
1b
kobs ) k0 + k0
+ k0
X1b
(5b)
2
[
]
+
(aH
)
+
Then, the set of values was optimized and adjusted by
the iterative method and the thermodynamic cycle to the
pH-rate profile of the acid decomposition, with a devia-
tion of (0.04 pK unit. The values of pK13, pK15, pK16, and
pK18 coincided with those calculated from LFER. The pKa
values of the dithiocarbamic and carboxylic conjugate
acids were higher than those calculated from LFER,
possibly due to intramolecular hydrogen bonding.
where
K4
K5
k0 ) kH(K3; k0 ) kH-K4 + ki(
;
1b
1c
1e
k0 ) kH)K6 + ki-
p H-Ra te P r ofile of Eth ylen ebis(d ith ioca r ba m -
a te). The pH-rate profile of the acid cleavage of the first
dithiocarbamic group of EbisDTC is shown in Figure 1.
The curve of the pH-rate profile in Figure 1 was
obtained for k0 ) (2.01 ( 0.04) × 10-3 s-1, k0 ) (1.40
1b
1c
1e
-1
( 0.03) × 10-3
s
-1, and k0 ) (1.42 ( 0.17) × 10-3
s
.
(
The specific acid catalysis rate constant kH of the
(28) (a) Charton, M. J . Org. Chem. 1964, 29, 1222. (b) Charton, M.
Prog. Phys. Org. Chem. 1981, 13, 119.
(29) Arnett, E. M. Prog. Phys. Org. Chem. 1963, 1, 223.
(30) Goldfarb, A. R.; Mele, A.; Gutstein, N. J . Am. Chem. Soc. 1955,
77, 6194.
1b
decomposition of species 1d (Scheme 1) is equal to k0
/
K3 ) (0.62 ( 0.08) M-1 s-1 and is the expected value from
the Brønsted plot (Figure 2) considering that from Table

Mechanisms of Acid Decomposition of Dithiocarbamates
J . Org. Chem., Vol. 64, No. 6, 1999 1811
corresponding parent amines of species 3d ,g, 3e,h , and
3f,i are 5.2, 7.8, and 9.9, respectively.1 The total experi-
mental first-order rate constant is given by eq 6a, where
3i
kH3g, kH3h , and kH are the specific acid-catalyzed rate
constants
kobs ) [kH3gaH+ + k 3g]X3g + [kH3h aH+ + k 3h ]X3h
+
i
i
kH3iaH+X (6a)
3i
k0′′′
kobs ) k0′ + k0′′aH
+
X3e
(6b)
+
+
(
)
aH
where
ki3gK14
3h
k0′ )
+ kH K
15
K12
3g
kH
K
14
k0′′ )
K12
3i
k0′′′ ) ki3h K15 + kH K K1
16
of the species with molar fractions X3g, X3h , and X3i,
respectively. Species 3g and 3h can also undergo in-
tramolecular general acid-catalyzed cleavage, with first-
order rate constants ki3g and ki3h . The molar fractions can
be related to X3e, and eq 6a becomes kinetically equiva-
lent to eq 6b.
The curve of the pH-rate profile shown in Figure 3
was obtained for k0′ ) (5.35 ( 0.16) × 10-3, k0′′ ) (5.56
( 0.17) × 10-6, and k0′′′ ) (1.61 ( 0.05) × 10-6. The
F igu r e 3. pH-rate profile of the acid decomposition of
glycinedithiocarboxylate in water at 25 °C: O, observed rate
constants; s, calculated rate constant; ---, contribution of the
kinetic terms.
calculated value of kH ) k0′′K12/K14) (3.0 ( 0.7) × 10-2
3g
M-1 s-1 is close to the expected value for the decomposi-
tion of species 3g as can be observed in Figure 2, and it
is another piece of evidence that eq 6 properly describes
the mechanism. From the Brønsted plot (Figure 2), the
extrapolated values of the specific acid catalysis rate
1, pKN ) 9.86 for species 1d . This provides strong
evidence that eq 5b correctly describes the reaction.
1c
The rate constant k0 contains the specific acid cataly-
3h
3i
constants for 3h and 3i are kH ) 0.28 and kH ) 0.34
M-1 s-1, respectively, and like EbisDTC, the intramo-
lecular catalytic rate constants must be different from
zero to obtain the pH-rate profile curve, with values of
sis rate constant kH- of the cleavage of species 1e (Scheme
1) with pKN ) 9.95 (Table 1). The extrapolated value of
-
kH is 0.33 ( 0.02 M-1 s-1. Since pK4 ) 2.87 ( 0.03
(Scheme 1), the term kH-K4 ) (0.45 ( 0.06) × 10-3 s-1 is
too small compared to k01c, and consequently ki(K4/K5
must be different from zero with an intramolecular rate
constant ki( equal to 9.3 ( 2.1 s-1. It is not possible to
obtain the pH-rate profile curve without considering this
intramolecular catalytic term.
ki3g ) (16.8 ( 5.4) s-1 and ki3h ) (2,0 ( 0.3) × 10-4 s-1
,
calculated from eq 6b.
In tr a m olecu la r Gen er a l Acid Ca ta lysis. Proton
transfer between electronegative centers are intrinsically
rapid, and they can become rate determining in reactions
involving high energy intermediates. However, intramo-
lecular proton transfer in general acid catalysis is typi-
cally inefficient as measured by the effective molarity.31
Although the rate of an intramolecular reaction depends
on a critical distance between the reacting groups,13 for
intramolecular proton transfer reactions this is not the
only parameter involved, and it has been proposed that
the efficiency of the proton transfer increases when the
transferred proton ends up in a thermodynamically
favored intramolecular hydrogen bond in the reaction
product.32 In Table 2 are shown the main features of the
intramolecular acid catalysis of some of the reactive
species of EbisDTC and glyDTC according to Scheme 3.
We note that the calculation of the effective molarity is
not strictly correct because the intermolecular reaction
is between the reactant and the hydron. Nonetheless the
Following the same reasoning, the specific acid rate
)
constant kH of the decomposition of 1f (statistically
corrected for the two dithiocarbamate groups) is 0.66 (
0.05 M-1 s-1, since pKN ) 9.95 (Table 1). Considering the
value of pK6 ) 2.85 ( 0.04, the intramolecular catalytic
term ki- is (4.9 ( 3.3) × 10-4 s-1. Consequently, species
1d and 1e in Scheme 1 decompose through specific acid
catalysis and also through an intramolecular general
acid-catalyzed mechanism.
p H-Ra te P r ofile of Glycin ed ith ioca r boxyla te.
Glycinedithiocarboxylic acids provide the opportunity to
study the effect of a neighboring carboxylic group as
intramolecular general acid catalyst on the rate of acid
cleavage in the carboxylic free acid form or as the
conjugate acid COOH2+. The pH-rate profile of the acid
decomposition of glycinedithiocarboxylate is shown in
Figure 3. It was considered that it cleaves through species
3g, 3h , and 3i (Scheme 2) because the pKN of the
(31) Kirby, A. J . Adv. Phys. Org. Chem. 1980, 17, 183.
(32) Kirby, A. J . Acc. Chem. Res. 1997, 30, 290.

1812 J . Org. Chem., Vol. 64, No. 6, 1999
Humeres et al.
Ta ble 2. In tr a m olecu la r Gen er a l Acid Ca ta lysis of th e Acid Clea va ge of Eth ylen ebis(d ith ioca r ba m a te) a n d
Glycin ed ith ioca r boxyla te in Wa ter a t 25 °Ca
c
d
d
AH
species
EMb
k2i, s-1
pKA
pKB
pKC
pKD
pKps
CH2NH+C(SH)2
CH2NHC(S)SH
COOH2
1d
1e
3g
3h
14.3 ( 4.9
(8.2 ( 2.8) × 106
(1.1 ( 0.8) × 103
(7.1 ( 4.2) × 103
3.6 ( 0.6
-1.12
2.85
-1.6
5.2
-4.24
-4.15
-8.90
-6.30
-4.15
-4.15
-6.30
-4.20
-1.21
2.85
-4.20
3.10
3.03
7.00
4.70
9.40
(1.5 ( 1.1) × 10-3
(5.6 ( 3.3) × 102
(7.1 ( 1.1) × 10-4
+
COOH
a
b
d
From Scheme 3. Effective molarity. c From Schemes 1 and 2. From ref 1.
Sch em e 3
The acid decomposition of species 1d proceeds through
a fast N-protonation and a slower C-N breakdown (eq
10).1 The intramolecular general acid catalysis rate
pKa values of the conjugated acids (dithiocarbamic and
carboxylic) are quite close to that of the hydron. It can
be observed that the rate constant for the intramolecular
proton transfer k2i does not determine the efficiency. The
intramolecular proton switch constant Kps is thermody-
namically unfavorable, and for the same type of general
acid, the efficiency increases when pKps decreases for the
same pair of electronegative centers. There is also an
apparent relationship between the difference the pKb of
the proton acceptor and the pKa of the donor that follows
the order 24.5 > 19.4 > 15.3 > 15.1 for 3g, 1d , 1e, and
3h , respectively, the same as the proton efficiency.
The acid decomposition of alkyldithiocarbamates from
parent amines with pKN < 10 occurs through the dithio-
carbamate anion and a zwitterion intermediate (eq 7).1
constant k2i is (8.2 ( 2.8) × 106 s-1, but the efficiency
1d
of this fast S to N proton transfer is low (EM ) 14.3 (
4.9 M). The value of pK( does not change after the proton
transfer, and the thiocarbonyl sulfur is known to be a
poor hydrogen bond acceptor.
The negligible EM for both the dithiocarbamic and
carboxylic free acid groups indicates that the intramo-
lecular catalysis is about 4 kcal mol-1 unfavorable
regarding the protonation by the external hydron. This
result might be a consequence of the fact that high values
of pKa increase the strength of the hydrogen bonding in
the reagent. Also, it indicates that intermolecular general
acid catalysis by acids with higher pKa values must be
negligible for dithiocarbamates that decompose through
a zwitterion intermediate, as was found experimentally.1
It is difficult to evaluate the strength of the hydrogen
bonding in the reagent and product, but a rough estima-
tion can be made considering the difference between the
calculated pKa from the pH-rate profile and that from
LFER. As mentioned above, in some cases a difference
of 2-3 pK units was observed and was attributed to
intramolecular hydrogen bonding. For the reagent species
2a , 2b, 4a , and 4b, the ∆pK value due to N‚‚‚H bonding
is low, in the range of -1.2 to 0.3. For the dithiocarbamic
species (1d , 1e, 3g, and 3h ), the difference is much
higher, suggesting that the dithiocarbamic group is part
of the main group involved in the hydrogen bonding. The
weak N‚‚‚H bond estimated for the amino species must
be much weaker for the dithiocarbamic species because
of the strong electron-withdrawing effect of the dithio-
carboxylate group.
The specific acid catalysis rate constant is given by eq 8,
where Ka is the acid dissociation constant of the dithio-
carbamic acid.
k2k3Ka
kH
)
(8)
k3 + k-2
The intermolecular N-protonation of species 3g is
slower than the C-N breakdown and k23g ) 12.6 M-1 s-1 1
.
For the intramolecular general acid-catalyzed cleavage,
the effective molarity (EM) ) (5.6 ( 3.3) × 102
M
represents the efficiency of the N-protonation step,
3g
3g
because according to eq 8, ki3g/kH ) k2i3g/k2 and
consequently k2i ) (7.1 ( 4.2) × 103 s-1. In eq 9 it is
3g
assumed that the O to N proton transfer occurs through
a water molecule, which does not change the type of the
hydrogen bonds but improves the position of the heavy
atoms involved for the proton transfer. The acid dissocia-
tion constant of the zwitterion (pK() increases 2.6 pK
units upon the proton transfer because the donor and
acceptor groups are not electronically independent.
The hydrogen bonding strength in the product can be
estimated from the difference in the pK( values, those

Mechanisms of Acid Decomposition of Dithiocarbamates
J . Org. Chem., Vol. 64, No. 6, 1999 1813
for 1e, 1f, 3h , and 3i species are 1.1, -2.8, 0.6, and -7.2,
respectively. Therefore, for the conjugate acid group pair
there is an increase in the hydrogen bonding strength
from reagent to product of ∆∆pK 2.3 and 1.0 for 1d and
3g, while for the free acid group pair, 1e and 3h , there
is a decrease of -2.1 and -7.5, respectively. These results
support the assumption that a thermodynamically favor-
able change of hydrogen bonding from reagent to product
would increase the efficiency of proton transfer.
values must be negligible for alkyldithiocarbamates that
decompose through a zitterion intermediate, as was found
experimentally.
The rate constant for the intramolecular proton trans-
fer does not determine the efficiency. The intramolecular
proton switch constant Kps is thermodynamically unfa-
vorable, and for the same type of general acid, the
efficiency increases when pKps decreases for the same pair
of electronegative centers. The difference between the pKb
of the proton acceptor and the pKa of the donor follows
the order of the proton’s efficiency. Estimation of the
strength of the hydrogen bonding in the reagent and
product supports the assumption that a thermodynami-
cally favorable change of hydrogen bonding from reagent
to product increases the efficiency of proton transfer.
Con clu sion s
The acid decomposition of ethylenebis(dithiocarbam-
ate) and glycinedithiocarboxylate anions is intramolecu-
larly general acid catalyzed by the neighbor conjugate
acid group. For the carboxylic group, the effective mo-
larity is (5.6 ( 3.3) × 102 M, and for the dithiocarbamic
group it is 14.3 ( 4.9 M. For both the dithiocarbamic and
carboxylic free acid groups, the intramolecular catalysis
is about 4 kcal mol-1 unfavorable with respect to the
protonation by the external hydron. Accordingly, inter-
molecular general acid catalysis by acids with higher pKa
Ack n ow led gm en t. This paper was written in part
during the stay of E.H. at the Chemistry Department
of the University College London, as an Academic
Visitor. We thank CNPq for the travel grant and
Professor Brian Capon for his comments.
J O981382I