Mechanisms of acid decomposition of dithiocarbamates. 3. Aryldithiocarbamates and the torsional effect

Article abstract of DOI:10.1021/jo016190t

The acid decomposition of some p-substituted aryldithiocarbamates (arylDTCs) was observed in 20% aqueous ethanol at 25°C, μ = 1.0 (KCl, for pH > 0). The pH-rate profiles showed a dumbell shape with a plateau where the observed first-order rate constant k_obs was equal to k_o, the rate constant of the decomposition of the dithiocarbamic acid species. The acid dissociation constants of the dithiocarbamic acids (pK_a) and their conjugate acids (pK⁺) were calculated from the pH-rate profiles. Comparatively, k_o was more than 10⁴-fold faster than alkyldithiocarbamates (alkDTCs) with similar pK_N (the acid dissociation constant of the parent amine). It was observed that the values of pK_a and pK⁺ were 5 and 8 units of pK, respectively, higher than the expected values from the pK_N of alkylDTCs. The higher values were attributed to the inhibition of the delocalization of the nitrogen electron pair into the benzene ring because of the strong electron withdrawal effect of the thiocarbonyl group. Comparison of the activation parameters showed that the rate acceleration was due to a decrease in the enthalpy of activation. Proton inventory indicated the existence of a multiproton transition state, and it was consistent with an S to N proton transfer through a water molecule. There are two hydrogens contributing to a secondary SIE, and there are also two protons that are being transferred at the transition state to form a zwitterion followed by fast C-N bond cleavage. The mechanism could also be a concerted asynchronic process where the N-protonation is more advanced than the C-N bond breakdown. The kinetic barrier is similar to the torsional barrier of thioamides, suggesting that the driving force to reach the transition state is the needed torsion of the C-N bond that inhibits the resonance with the thiocarbonyl group and the aromatic moiety, increasing the basicity of the nitrogen and making the proton transfer thermodynamically favorable.

Full text of DOI:10.1021/jo016190t

3662
J . Org. Chem. 2002, 67, 3662-3667
Mech a n ism s of Acid Decom p osition of Dith ioca r ba m a tes. 3.
Ar yld ith ioca r ba m a tes a n d th e Tor sion a l Effect
Eduardo Humeres,*,†,‡ Nito A. Debacher,† J ose´ Dimas Franco,† Byung Sun Lee,† and
Adriano Martendal†
Department of Chemistry, Universidade Federal de Santa Catarina, 88040-900 Floriano´polis, SC, Brazil,
and Institute for Fundamental Research in Organic Chemistry, Kyushu University, Hakozaki,
Fukuoka, 812-8581 J apan
humeres@mbox1.ufsc.br
Received October 11, 2001
The acid decomposition of some p-substituted aryldithiocarbamates (arylDTCs) was observed in
20% aqueous ethanol at 25 °C, µ ) 1.0 (KCl, for pH > 0). The pH-rate profiles showed a dumbell
shape with a plateau where the observed first-order rate constant kobs was equal to ko, the rate
constant of the decomposition of the dithiocarbamic acid species. The acid dissociation constants of
the dithiocarbamic acids (pKa) and their conjugate acids (pK+) were calculated from the pH-rate
profiles. Comparatively, ko was more than 104-fold faster than alkyldithiocarbamates (alkDTCs)
with similar pKN (the acid dissociation constant of the parent amine). It was observed that the
values of pKa and pK+ were 5 and 8 units of pK, respectively, higher than the expected values from
the pKN of alkylDTCs. The higher values were attributed to the inhibition of the delocalization of
the nitrogen electron pair into the benzene ring because of the strong electron withdrawal effect of
the thiocarbonyl group. Comparison of the activation parameters showed that the rate acceleration
was due to a decrease in the enthalpy of activation. Proton inventory indicated the existence of a
multiproton transition state, and it was consistent with an S to N proton transfer through a water
molecule. There are two hydrogens contributing to a secondary SIE, and there are also two protons
that are being transferred at the transition state to form a zwitterion followed by fast C-N bond
cleavage. The mechanism could also be a concerted asynchronic process where the N-protonation
is more advanced than the C-N bond breakdown. The kinetic barrier is similar to the torsional
barrier of thioamides, suggesting that the driving force to reach the transition state is the needed
torsion of the C-N bond that inhibits the resonance with the thiocarbonyl group and the aromatic
moiety, increasing the basicity of the nitrogen and making the proton transfer thermodynamically
favorable.
In tr od u ction
secondary amines showed different structural relation-
ships, and consequently, it was not possible to draw
conclusions about their mechanisms.1 But recently, a
detailed study of the acid cleavage of alkyldithiocarbam-
ates5 concluded that the reaction occurs through two
mechanisms, depending on the pKN, the acid dissociation
constant of the parent amine. At pKN < 10.5, they
decompose through a zwitterionic intermediate SH(
(Scheme 1) whose N-protonation is slower than the C-N
bond breakdown when pKN < 9.2. At pKN > 9.2, the C-N
bond breakdown is the slowest step. General acid cataly-
sis was not detected. Because of the low basicity of the
nitrogen, only acids stronger than Hydron can transfer
the proton in the rate-determining step.6 For parent
amines with pKN > 10.5, a new mechanism emerges that
has been proposed to occur as a concerted intramolecular
proton transfer and breaking of the C-N bond.5
The acid cleavages of carbamates1-4 and monothiocar-
bamates1 (eq 1) have been studied in relation to the effect
of amine basicity on the reaction rate, the general acid
catalysis of the expulsion of weakly basic amines, and
the rate of protonation of the leaving nitrogen.
The specific acid catalysis of the decomposition of dithio-
carbamates depends on the amine structure. However,
dithiocarbamates derived from alkyl, aryl, and cyclic
The acid decomposition of aryldithiocarbamates
(arylDTCs) occurs much faster than that of alkyldithio-
carbamates (alkDTCs), but no cleavage mechanism has
been proposed to explain their reactivity. In this work,
* To whom correspondence should be addressed. Phone: +55 48 331
9219. Fax: +55 48 331 9711.
† Universidade Federal de Santa Catarina.
‡ Kyushu University.
(1) Ewing, S. P.; Lockshon, D.; J encks, W. P. J . Am. Chem. Soc.
1980, 102, 3072.
(2) Caplow, M.; Yager, M. J . Am. Chem. Soc. 1967, 89, 4513.
(3) Caplow, M. J . Am. Chem. Soc. 1968, 90, 6795.
(4) J ohnson, S. L.; Morrison, D. L. J . Am. Chem. Soc. 1972, 94, 1323.
(5) Humeres, E.; Debacher, N. A.; Sierra, M. M. de S.; Franco, J .
D.; Schutz, A. J . Org. Chem. 1998, 63, 1598.
(6) Humeres, E.; Debacher, N. A.; Sierra, M. M. de S. J . Org. Chem.
1999, 64, 1807.
10.1021/jo016190t CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/01/2002

Aryldithiocarbamates and the Torsional Effect
J . Org. Chem., Vol. 67, No. 11, 2002 3663
Sch em e 1
the acid decomposition of some p-substituted arylDTCs
was studied in the acidity range from pH 4 to Ho -5.
F igu r e 1. pH-rate profile of the acid decomposition of
phenyldithiocarbamate in 20% aqueous ethanol at 25 °C:
continuous line, calculated rate constant; O, sodium salt; b,
ammonium salt.
Exp er im en ta l Section
Ma ter ia ls. All reagents were of analytical grade and used
without further purification, except the amines that were
previously distilled or crystallized in ethanol and kept at low
temperature in amber bottles. Deuterium oxide (99.8% D),
deuterated sodium hydroxide (40%, in D2O), deuterated sul-
furic acid (99.5% D), and EtOD (99.5% D) were purchased
commercially.
KCl. The kinetics were started by rapidly adding, with the aid
of a Teflon plunger, 5 µL of the sodium salt solution prepared
in situ to a deoxygenated acidic solution of 20% v/v aqueous
ethanol to avoid the oxidation of the dithiocarbamate to
thiouram. The final concentration of the arylDTC was about
10-4 M. Similar values for the rate constants were obtained
from the decomposition of the ammonium salts.
p-Su bstitu ted Ar yldith iocar bam ates, Am m on iu m Salts.
These were obtained by adding 110 mmol of the corresponding
aniline to a solution of 25 mL of 8 M NH4OH, 110 mmol of
CS2, and 25 mL of acetone in an Erlenmeyer flask washed with
nitrogen and capped with a septum. The mixture was mag-
netically stirred until a maximum reading was obtained at 290
nm, and then it was allowed to crystallize. The solid was
filtered, washed with cold ethanol, and dried under vacuum
over phosphorus pentoxide.7
Activa tion P a r a m eter s. Activation parameters were cal-
culated from the ko values in the range of 15.0-30.0 °C by
least-squares fitting to the Eyring equation.
Kin etic Solven t Isotop e Effect a n d P r oton In ven tor y.
Deuterium oxide, deuterated ethanol, ethanol, and distilled
deionized water were all previously deoxygenated by boiling
under nitrogen and then kept in sealed bottles. The % D was
calculated for D2O and EtOD by 1H NMR using acetone as
the internal standard. The kinetic solvent isotope effect and
proton inventory were measured by mixing equal volumes of
20 v/v % EtOL in L2O containing the substrate and 0.2 M L2-
SO4 in 20% v/v EtOL in L2O, in a stopped-flow spectrometer.
Phenyl: mp 97-99 °C (lit.7 108-109 °C); λmax 252 and 292
nm in 20% ethanol in deoxygenated water. Anal. (calcd values
for C7H10N2S2 in parentheses): C, 45.0 (45.1); H, 5.5 (5.4); N,
15.0 (15.0); S, 34.1 (34.4). 13C NMR (D2O) δ: 210.7 (CS2), 137.5
(C-1), 125.8 (C-3,5), 123.7 (C-4), 123.0 (C-2,6).
p-Chlorophenyl: mp 93-95 °C; λmax 305 and 294 nm in
ethanol. Anal. (calcd values for C7H9N2S2Cl in parentheses):
C, 38.4 (38.4); H, 3.9 (4.1); N, 12.3 (12.7); S, 29.3 (29.1).
p-Methoxyphenyl: mp 62-64 °C; λmax 291 nm in deoxygen-
ated water. Anal. (calcd values for C8H12N2S2O in parenthe-
ses): C, 44.9 (44.4); H, 5.4 (5.6); N, 12.9 (13.0); S, 30.1 (29.7).
p-Toluyl; mp 82-84 °C; λmax 232 and 290 nm in 20% ethanol
in deoxygenated water. Anal. (calcd values for C8H12N2S2 in
parentheses): C, 48.0 (48.1); H, 6.0 (6.0); N, 13.9 (13.9); S, 32.0
(31.9).
Alternatively, the aryldithiocarbamates were obtained in
situ as sodium salts by adding a solution of 3 mmol of CS2 in
2 mL of dioxane and 2.74 mmol of the corresponding aniline
to 4.3 mL of 0.65 M NaOH (2.74 mmol) in a 10 mL Erlenmeyer
flask, following the same procedure used for the ammonium
salts.
Kin etics. Runs of UV spectra and kinetics measurements
for the pH-rate profiles were followed at 290 nm, allowing
the solution without the substrate to equilibrate thermically
for 10 min, and using thermostated quartz cells at the set
temperature (0.1 °C.
Resu lts a n d Discu ssion
p H-Ra te P r ofiles. The pH-rate profiles of the acid
cleavage of arylDTCs show a dumbell shape (Figure 1)
similar to that found for the alkylDTCs,5 where the
observed first-order rate constant can be expressed by
eq 2.
ko
kobs
)
(2)
Ka
aH
aH+
K+
+ 1 +
+
The acid dissociation constants of the dithiocarbamic
acids (pKa) and their conjugate acids (pK+) were calcu-
lated from the pH-rate profiles (Scheme 1). The value
of ko was obtained from the average value of kobs at the
plateau (Table 1). The value of ko represents the rate
constant of the decomposition through the free dithio-
carbamic acid (k1 in Scheme 1) and is kinetically equiva-
lent to kHKa, where kH is the specific acid catalysis rate
constant of the decomposition of the dithiocarbamate
anion through the zwitterion SH(. Comparatively, for
ArylDTCs, ko is about 104-fold faster than alkDTCs from
parent amines with pKN < 10.5
All runs gave first-order rate constants that were calculated
using appropriate software, from the average of at least three
runs with correlations no less than 0.99. The ionic strength of
the runs at pH > 0 was kept constant at 1.0 by the addition of
(7) Freund, M.; Bachrach, G. Ann. 1895, 285, 184.

3664 J . Org. Chem., Vol. 67, No. 11, 2002
Humeres et al.
Ta ble 1. Dissocia tion a n d Ra te Con sta n ts Rela ted to th e
Acid Clea va ge of p-Su bstitu ted Ar yld ith ioca r ba m a tes in
20% Aqu eou s Eth a n ol a t 25 °C
p-XC6H4NHC(S)SH
10-2kH,
a
X
pKN
pKa
pK+
ko, s-1
M-1 s-1b
MeO 5.29 3.0 ( 0.4
-2.7 ( 0.3
1.40 ( 0.14
14.0
13.5
14.0
7.7
Me
H
5.07 2.8 ( 0.04 -2.6 ( 0.05 2.14 ( 0.08
4.58 2.7 ( 0.08 - 3.1 ( 0.04 2.80 ( 0.04
3.81 2.4 ( 0.02 - 3.3 ( 0.09 3.06 ( 0.10
Cl
NO2
0.98 1.7c
-4.9c
9.0c
4.5d
a
Handbook of Biochemistry; Sober, H. A., Hart, R. A., Eds.;
CRC: Cleveland, 1968. Calculated from kH ) ko/Ka. c Extrapo-
lated values. From ref 1.
b
d
F igu r e 4. Brφnsted plot of the rate constants and the pKN of
the leaving amine for the acid decomposition of substituted
methyldithiocarbamates (O, 0) in water at 25 °C:5,6 k2 and k3,
as defined in Scheme 1; kH, specific acid rate constant; 2, kH
value for phenyldithiocarbamate; b, piperidinedithiocarbam-
ate.1
F igu r e 2. Dependence on the pKN of the parent amine of the
pKa of dithiocarbamic acids: b, this work; O, alkylDTC, ref 5.
The conjugate acid presents total inhibition of resonance
with the benzene ring, and consequently, 8 units of pK
represents the maximum increase in basicity.
The same reason can be ascribed to the low pKa values
observed for the NH dissociation of some dithiocarbamic
esters.8 The difference in pKa between N-benzyl and
N-phenyl esters is only 2.3 pK units, the same as the
value obtained for the NH dissociation of the correspond-
ing dithiocarbamic acids.9 The smaller value in compari-
son with the difference between benzylamine and aniline
of 4.8 pK units supports the assumption that aryldithio-
carbamates behave as derivatives of strong basic amines.
An increase in the basicity of the nitrogen by about
5-8 pK units places the arylDTCs in the pKN range
where a new mechanism of cleavage has been proposed
to occur.5 In Figure 4, it can be observed that the specific
acid rate constant kH of phenylDTC corresponds to an
alkDTC from a parent amine with a pKN about 7 pK units
higher than that of aniline.
F igu r e 3. Dependence on the pKN of the parent amine of the
pK+ of the conjugate dithiocarbamic acids: b, this work; O,
alkylDTC, ref 5.
Acid Dissocia tion Con sta n ts of Ar yld ith ioca r -
ba m a tes. From the plots of pKa (or pK+) vs pKN (Figures
2 and 3), it can be observed that the values for the
arylDTCs are much higher than those for the alkDTCs
with similar pKN values, corresponding to basicities of
an alkylic nitrogen of about 4.9 and 8.0 units of pK,
respectively, higher than the pKN of the parent anilines.
The higher values might be due to the inhibition of the
delocalization of the nitrogen electron pair into the
benzene ring on account of the strong effect of the
thiocarbonyl group that acts as a powerful electron sink.5
A further piece of support for this theory was obtained
from the Yukawa-Tsuno (Y-T) equation.10 This rela-
tionship combines linearly σo Hammett normal substitu-
(8) Drobnica, L.; Gemeiner, P. Collect. Czech. Comm. 1975, 40, 3346.
(9) Takami, F.; Wakahara, S.; Maeda, T. Tetrahedron Lett. 1971,
2645.
(10) (a) Yukawa, Y.; Tsuno, Y. Bull. Chem. Soc. J pn. 1959, 32, 965;
(b) 1959, 32, 971; (c) 1966, 39, 2274. (d) Tsuno, Y.; Fujio, M. Chem.
Soc. Rev. 1996, 129; (e) Adv. Phys. Org. Chem. 1999, 32, 267.

Aryldithiocarbamates and the Torsional Effect
J . Org. Chem., Vol. 67, No. 11, 2002 3665
Ta ble 2. Activa tion P a r a m eter s of P h en ylDTC a n d 2-Am m on iu m -Eth ylDTC
pKN
ko, s-1
∆G, kcal/mol
∆H, kcal/mol
∆S, cal mol-1 K-1
a
PhenylDTC
2-Am-EtDTC
4.58
7.52
2.80 ( 0.04
16.80 ( 0.01
21.16 ( 0.05
15.14
19.47
-5.6
-5.7
(20.9 ( 0.2) × 10-4b
H2SO4 (0.1 M), µ ) 1.0 (KCl), and 20% v/v aqueous ethanol, 25 °C. Water, 25 °C.5
a
b
incompatible with a Brφnsted slope and with the in-
creased stability of the zwitterion and slower C-N bond
cleavage. The new mechanism was proposed to be an
intramolecular S to N proton transfer of the dithiocar-
bamic acid SH in Scheme 1, concerted with the C-N
breakdown.5 The reaction should be pH independent and
corresponds to path 1 in Scheme 1, where k1 ) ko.
Kin etic Solven t Isotop e Effect a n d P r oton In ven -
tor y. Inverse solvent isotope effect for acid catalysis is
consistent with a partial transfer of the proton at a late
transition state of the rate-determining step. The pH-
independent rate constant ko of phenylDTC exhibits an
inverse SIE kDo /koH ) 2.51. If the aryldithiocarbamates
behave as derivatives of strong basic amines and decom-
pose through an intramolecular S to N proton-transfer
mechanism, proton inventory could give insight into the
structure of the transition state involved.
For phenylDTC, the kn vs n plot was bulged downward,
which always indicates the existence of a multiproton
transition state (Table 3, Figure 6).13,14 Consistently, the
plot of ln kn vs n was linear (r ) 0.998) and k0.5 ) (kok1)1/2
) 4.44 was very similar to the experimental value of 4.34.
However, there must be more than two protons undergo-
ing change upon activation because the plot of (kn/ko)1/2
vs n was not linear and (k1/ko)1/2) 1.59 is too high for the
fractionation factor of the transition state (assuming that
each proton contributes the same isotope effect). This
analysis suggests that the S to N transfer occurs through
a water molecule as shown in eq 4. Preliminary theoreti-
cal calculations carried out at the HF/3-21 G(d) level sup-
port this assumption because it was found that for phen-
yldithiocarbamate, the water-catalyzed mechanism is
strongly favored with respect to the direct intramolecular
proton transfer by 13 kcal/mol.15 According to this mech-
anism, the Gross-Butler equation has the form of eq 5.
F igu r e 5. Yukawa-Tsuno plot for the acid decomposition of
aryldithiocarbamates in 20% v/v aqueous ethanol at 25 °C: O,
Brown σ+ constant; b, σo Hammett constant; 0, σj Y-T
constant; F ) 0.68, r+ ) 0.27. For H and p-NO2, σ+ and σo are
the same.
ent constants, which do not involve π-electronic interac-
tions between the substituent and the reaction center,
and Brown σ+ constants, where a positive charge can be
delocalized into a π-arylic system. The substituent effects
can be more generally described by the Y-T equation in
the form
k
ko
log ) Fσj ) F(σo + r+∆σj+)
(3)
where ∆σj+ ) σ+ - σo measures the capability for
π-delocalization of π-electron donor substituents, and r+
is a constant characteristic of a reaction and measures
the extent of resonance interaction between the aryl
group and the reaction center in the rate-determining
transition state. For r+ ) 1, the resonance effect pre-
dominates, while when r+ ) 0, the effect does not involve
π-electronic interaction. The Yukawa-Tsuno plot for the
arylDTC series (Figure 5) showed F ) 0.68; that is, the
reaction presents very low sensitivity for p-substituents,
and because the low value of r+ ) 0.27, the effect is
predominantly inductive rather than resonant.
Activa tion P a r a m eter s. In Table 2, a comparison is
made between the activation parameters of phenylDTC
and 2-ammonium-ethylDTC. Despite the fact that the
pKN of the first is about 3 pK units lower, ko is about 3
orders of magnitude faster. The increase in reactivity is
mainly due to a decrease in the enthalpy of activation.
For alkylDTCs with pKN < 9.2, the slowest step is the
N-protonation of the dithiocarbamate anion that forms
a zwitterion followed by a fast C-N bond cleavage
(Scheme 1, Figure 4).5 At pKN > 10.5, a minimum of the
Brφnsted plot indicates a change of mechanism.11,12 The
fast increase in the rate constant at higher pKN is
(1 - n + nφTN1-L
)
(1 - n + nφTO2-L
(1 - n + nφO-L
)
)
kn ) k0
(
) (
)
(1 - n + nφNL
(1 - n + nφTS-3 L-O)(1 - n + nφOT4-L-N
)
sec
sec
)
(5)
(
)
(1 - n + nφS-L)(1 - n + nφO-L
)
pri
There are two hydrogens (N-L and O-L) experiencing
some change in bonding, contributing as a secondary SIE,
and two protons (S-L-O and O-L-N) being transferred
at the transition state, contributing as a primary SIE.
(12) Leffler, J . E.; Grunwald, E. Rates and Equilibria in Organic
Reactions; Wiley: New York, 1963; Chapter 7.
(13) Venkatasubban, K. S.; Schowen, R. L. Crit. Rev. Biochem. 1984,
17, 1.
(14) Schowen, K. B. J . Solvent Hydrogen Isotope Effects. In Transi-
tion States for Biochemical Processes; Gandour, R. D., Schowen, R. L.,
Eds.; Plenum Press: New York, 1978.
(11) Exner, O. In Advances in Linear Free Energy Relationships;
Chapman N. B., Shorter, J ., Eds.; Plenum: New York, 1972; Chapter
1.
(15) Abboud, J . L.; Notario, R., personal communication.

3666 J . Org. Chem., Vol. 67, No. 11, 2002
Humeres et al.
Ta ble 3. P r oton In ven tor y of th e Acid Decom p osition of
Am m on iu m P h en yld ith ioca r ba m a te a t 25.0 °Ca
ÌD
kenxp, s-1
(7.04)e
6.62 ( 0.04
6.06 ( 0.02
5.13 ( 0.04
4.08 ( 0.03
3.56 ( 0.02
2.90 ( 0.03
2.80 ( 0.04
(RSC)b
(kn/ko)N-Lc
kcnalcd, s-1d
sec
1.00
0.40
1.049
1.046
1.041
1.031
1.022
1.013
1.004
1.000
7.34
6.72
5.95
4.78
3.98
3.41
2.97
2.80
0.94 ( 0.01
0.85 ( 0.01
0.66 ( 0.01
0.47 ( 0.01
0.28 ( 0.00
0.09 ( 0.00
0.0
0.436
0.490
0.604
0.718
0.832
0.946
1.000
a
b
In L2SO4 0.1 M and 20% v/v aqueous EtOL. (RSC) ) (1 - n
+ nφS-L); φS-L ) 0.40. c From eq 6; φN-L ) 0.92; φN+-L ) 0.97; ø
) 0.9. From eq 7; (TSC) ) 1. e Extrapolated.
d
A late transition state would look very similar to a
zwitterion, where the S to O and O to N transfers are
almost complete. The charge on the oxygen of the water
molecule would be very small, and consequently, the
secondary effect due to the O-L bond would be very close
to unity. The secondary effect due to the N-L bond was
estimated from eq 6 where ø is the weighting factor
describing the TS structure.
F igu r e 6. Proton inventory of the acid decomposition of
ammonium phenyldithiocarbamate at 25.0 °C in L2SO4 0.1 M
and 20% v/v aqueous EtOL. Continuous curve was drawn
according to eq 5, where φO-L ) 1.0, φS-L ) 0.40, φN-L ) 0.92;
φTO-L) 1.0, φTN4-L) 0.97.
Wa ter Ca ta lysis. Water acts as an acid-base catalyst,
but the main effect is that in order to reach the transition
state, a torsion is imposed on the C-N bond that
decreases its double-bond character and has a weakening
effect. As observed in Table 1, electron-withdrawing
groups accelerate the decomposition because they de-
crease the partial C-N double bond. The extent of the
torsional angle θ (I)
N-L
ø
kn
(1 - n + nφN
)
+-L
)
(6)
( )
(
)
ko
(1 - n + nφN-L
)
sec
sec
Since it is a late TS, the calculation of this effect was
made for ø ) 0.9. This value seems reasonable because
the calculated value was only 0.4% lower than the
equilibrium value (ø ) 1) for n ) 1.
Therefore, eq 5 can be simplified as eq 7 where (TSC)
and (RSC) are the
N-L
kn
TSC
RSC
kn ) ko
(7)
( )
(
)
ko
sec
TS and reagent state contributions to the primary SIE,
respectively, given by eq 8 (considering that φO-L ) 1.0).
(TSC) ) (1 - n + nφTS-3 L-O)(1 - n + nφOT4-L-N
(RSC) ) (1 - n + nφS-L
)
(8a)
(8b)
is also related to the decrease in resonance of the
aromatic ring and the nitrogen, increasing its basicity.
Several examples in the literature show the double-
bond character and rotational hindrance around the C-N
bond for amides and thioamides16 and thiocarbamic
esters.17,18 In normal amides, the carbonyl group becomes
deactivated by the resonance of the N-lone pair with the
CdO group. However, for acylimidazols, the charges on
N-1 and N-3 show that the electron pair is pulled in the
opposite direction,19 activating the carbonyl and increas-
ing the reactivity of these compounds (II and III).20
)
Calculation of TSC, considering the SIE from N-L
secondary isotope effects, gave an average value of 1.02
( 0.03.
The maximum equilibrium value (MEV) for the cleav-
age of the dithiocarbamic acid by concerted intramolecu-
lar N-protonation and C-N bond breakdown (SH S N
in Scheme 1) is only 2.30. However, when considering a
slow intramolecular N-protonation to form a zwitterion
(SH S SH(), followed by fast C-N bond cleavage, the
MEV increases to 2.56. This value is very reasonable
when compared to the observed value of 2.51.
The mechanism could also be concerted: an asynchro-
nic process where the N-protonation is more advanced
than the C-N bond breakdown. Nonetheless, this mech-
anism implies an intramolecular general acid catalysis
by an acid of low pKa value (2.7), which has been shown
to be incapable of acting as an effective intramolecular
catalyst.6 Therefore, the structure of the TS should
explain why the intramolecular N-protonation occurs so
easily and the C-N bond breaks so fast for a species
whose N-basicity corresponds to a pKN higher than 10
(Figure 4).
This is similar to the effect observed in the aryldithio-
carbamates, where the electron pair is shifted from the
(16) Krueger, P. J .; Fulea, A. O. Tetrahedron 1975, 31, 1813.
(17) Bauman, R. A. J . Org. Chem. 1967, 32, 4129.
(18) Lee, C. M.; Kumler, W. D. J . Am. Chem. Soc. 1961, 83, 4596.
(19) Pullman, B.; Pullman, A. Quantum Biochemistry; Wiley-Inter-
science: New York, 1963.
(20) Fife, T. H. Acc. Chem. Res. 1993, 26, 325.

Aryldithiocarbamates and the Torsional Effect
J . Org. Chem., Vol. 67, No. 11, 2002 3667
nitrogen to the thiocarbonyl group, thus decreasing the
resonance with the benzene ring.
amide bond diverge from this linear correlation. Thus,
3,5,7-trimethyl-1-azaadamantan-2-one, which is the most
twisted amide (the N-electron lone-pair is coplanar with
the CdO group), presents a strong inhibition of the n(N)-
π*(CdO) interaction that significantly stabilizes n(O).
The value of IP[n(O)] falls well off of the straight line,
and 3,5,7-trimethyl-1-azaadamantan-2-one is better de-
scribed as an aminoketone than as an amide. There is
very little decrease in the bond length of the CdO group
with distortion, but the increase in the CO-N bond
proves to be substantial, indicating that inhibition of the
amide resonance leads to a normal C-N single bond
length. For the arylDTCs, twisting the CS-N bond at
the transition state involves the pyramidalization of the
amino group and possibly the thiocarbonyl carbon atom.
The overall distortion is measured by the angular pa-
rameter θ (I).
The high double-bond character of the C-N bond in
N,N-dialkylthioamides prevents the free rotation16,17 with
a rotational barrier in the range of 20-23 kcal/mol.21 This
barrier is near the activation barrier of the arylDTCs and
suggests that the driving force to reach the transition
state is the needed torsion of the C-N bond that inhibits
the resonance with the thiocarbonyl group and the
aromatic moiety, increasing the basicity of the nitrogen
and making the proton transfer thermodynamically
favorable. The increase in basicity must be assisted by
the lack of coplanarity of the benzene ring and the
dithiocarbamic group.
Am id e Gr ou p . There is a relationship between these
results and certain properties of the amide group that is
particularly relevant to the chemistry of peptides and
proteins and the reactions involved in their transforma-
tions, especially those that are catalyzed by enzymes. It
has been suggested that an important feature of the
action of the peptidase enzyme might be the twisting
effect on the amide moiety as it binds to the active site,
restricting the resonance and enhancing the ease of C-N
bond breakage.22
The conjugation of the nitrogen lone-pair is optimum
when the CO-N unit is planar, leading to the partial
double-bond character of the C-N bond that hinders the
internal rotation at the same time that it weakens the
CdO bond, lowering the carbonyl stretching frequency.
Recently, Kirby et al.23 have shown that there is a linear
correlation between the ionization potential IP[n(O)]
related to the oxygen lone-pair orbital n(O) and the
frequency that was corrected for angle strain ν120(CdO)
of the carbonyl vibration. Compounds with a twisted
Con clu sion s
The spontaneous acid decomposition of aryldithiocar-
bamates is more than 104-fold faster than that of alkyl-
dithiocarbamates with similar pKN values. The reaction
is catalyzed by a water molecule that acts as an acid-
base catalyst. The mechanism could be a slow intramo-
lecular N-protonation to form a zwitterion, followed by
fast C-N bond cleavage, or concerted where, at the TS,
the N-protonation is more advanced than the C-N bond
breakdown. The main driving force to reach the TS is
the torsional effect on the C-N bond that inhibits the
resonance with the thiocarbonyl group.
Ack n ow led gm en t. This work was written in part
during the stay of E.H. at the Institute of Fundamental
Research of Organic Chemistry at Kyushu University,
as Visiting Professor. E.H. thanks Dr. Shinjiro Koba-
yashi for the invitation. We also thank Profs. Cesar
Zucco and Omar El Seoud for allowing us to use their
stopped-flow spectrometers for some of the kinetics
measurements.
(21) (a) Neuman, R. C., J r.; J onas, V. J . Phys. Chem. 1971, 75, 3532;
(b) J . Org. Chem. 1974, 39, 292. (c) Walter, W.; Schaumann, E. Chem.
Ber. 1971, 104, 3361. (d) Walter, W.; Franzen-Sleveking, M.; Schau-
mann, E. J . Chem. Soc., Perkin Trans. 2 1975, 528.
(22) Lipscomb, W. N. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 3875.
(23) Kirby, A. J .; Komarov, I. V.; Kowski, K.; Rademacher, P. J .
Chem. Soc., Perkin Trans. 2 1999, 1313.
J O016190T