Kinetics of cyclization of methyl s-(2,4,6-trinitrophenyl)-mercaptoacetate to 2-methoxycarbonyl-5,7-dinitro-benzo[d]thiazol-3-oxide

Article abstract of DOI:10.1135/cccc19971429

The cyclization kinetics of methyl S-(2,4,6-trinitrophenyl)mercaptoacetate to 2-methoxycarbonyl-5,7-dinitrobenzo[d]thiazol-3-oxide have been studied in acetate, methoxyacetate or N-methylmorpholine buffers. In the acetate and methoxyacetate buffers, the c

Full text of DOI:10.1135/cccc19971429

Kinetics of Cyclization
1429
KINETICS OF CYCLIZATION OF METHYL S-(2,4,6-TRINITROPHENYL)-
MERCAPTOACETATE TO 2-METHOXYCARBONYL-5,7-DINITRO-
BENZO[d]THIAZOL-3-OXIDE
1
2
3
Marek JANIK , Vladimir MACHACEK and Oldrich PYTELA
Department of Organic Chemistry, University of Pardubice, 532 10 Pardubice, Czech Republic;
e-mail: 1 janik@hlb.upce.cz, 2 machacev@hlb.upce.cz, 3 pytela@hlb.upce.cz
Received February 25, 1997
Accepted April 4, 1997
The cyclization kinetics of methyl S-(2,4,6-trinitrophenyl)mercaptoacetate to 2-methoxycarbonyl-5,7-
dinitrobenzo[d]thiazol-3-oxide have been studied in acetate, methoxyacetate or N-methylmorpholine
buffers. In the acetate and methoxyacetate buffers, the cyclization obeys the rate equation v =
[SH](k M′ eO[CH3O–] + k B′ [B–] + k ′B,MeO[B–][CH3O–]) and goes by two reaction paths differing in the
order of their reaction steps, the splitting off of the proton from C–H group being the rate-limiting
step in either path. In the N-methylmorpholine buffers, increasing concentration of the base results in
gradual decrease of reaction order in the base and change in the rate-limiting step of cyclization.
Methyl S-(2,4-dinitrophenyl)mercaptoacetate undergoes cyclization neither in the given buffers nor in
methoxide solution.
Key words: Cyclization kinetics; Reaction mechanism; 2-Methoxycarbonyl-5,7-dinitrobenzo[d]thiazol-3-
oxide.
The reactions of methyl N-(2,4,6-trinitrophenyl)aminoacetate1 and methyl N-methyl-N-
(2,4,6-trinitrophenyl)aminoacetate2 with methoxide in methanol produce 2-nitroso-4,6-
dinitroaniline and N-methyl-2-nitroso-4,6-dinitroaniline, respectively. The kinetics of
these reactions were studied1,2, and the mechanism suggested for them presumes an
intramolecular attack of carbonyl group by nitrogen atom of the amino group and for-
mation of substituted aziridinone as the key intermediate of reaction. Sulfur is more
nucleophilic than nitrogen and, in addition to it, tends to form thiirane rings in reactions
involving neighbouring group participation3. However, the base catalyzed reactions of
S-(2-nitro-4,6-disubstituted phenyl)mercaptoacetate esters do not produce nitroso com-
pounds but 2-alkoxycarbonyl-5,7-disubstituted benzo[d]thiazol-3-oxides (ref.4). The
aim of the present work is to study the kinetics of base catalyzed cyclization of methyl
S-(2,4,6-trinitrophenyl)mercaptoacetate leading to 2-methoxycarbonyl-5,7-dinitro-
benzo[d]thiazol-3-oxide.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1430
Janik, Machacek, Pytela:
EXPERIMENTAL
The 1H and 13C NMR spectra were measured on an AMX 360 Bruker apparatus at 25 °C at 360.14
and 90.57 MHz, respectively, using ca 5% solutions of the substances in hexadeuteriodimethyl sul-
foxide. The chemical shifts are referenced to the solvent signals δ(1H) 2.55 and δ(13C) 39.60, respec-
tively.
The mass spectra (EI) were measured on an MS 25 RFA (Kratos) spectrometer at the ionisation
voltage of 70 eV and ionic current 100 µA. Temperature of ion source was 250 °C, resolution R10% = 600.
The electron spectra were measured on an HP 8452 A Diode Array Spectrometer at 25 °C.
Methyl S-(2,4,6-Trinitrophenyl)mercaptoacetate (1)
2,4,6-Trinitrochlorobenzene (0.5 g, 2 mmol) was dissolved in dry benzene (30 ml) in a three-necked
flask with continuous stirring, whereupon methyl mercaptoacetate (0.18 ml, 0.21 g, 2 mmol) and
triethylamine (0.27 ml, 0.19 g, 0.19 mmol) were added at once. The mixture was stirred under argon
atmosphere at 10–20 °C for 1 h, triethylamine hydrochloride was filtered off, and the filtrate was
poured in dilute hydrochloric acid (100 ml ca 5%), thoroughly shaken, the organic layer was sepa-
rated and the aqueous layer extracted with chloroform (2 × 50 ml). Combined organic portions were
dried with Na2SO4 and the solvent was distilled off in vacuum. According to its 1H NMR spectrum
the raw evaporation residue contained 95% product 1 beside 5% 2. The evaporation residue was
separated by column chromatography (silica gel, acetone–chloroform 1 : 5 v/v). The substance ob-
tained from the first fraction was recrystallized from a benzene–cyclohexane mixture with addition of
alumina for chromatography. Yield 0.30 g (47%), m.p. 72–73 °C. 1H NMR spectrum: 9.14 s, 2 H (Ar); 3.91 s,
2 H (CH2); 3.62 s, 3 H (OCH3). 13C NMR spectrum: 154.18 (C-2,6); 147.89 (C-4); 127.76 (C-1); 122.33
(C-3,5-phenyl); 168.03 (CO); 52.71 (OCH3); 37.36 (CH2). For C9H7N3O8S (317.2) calculated: 34.08% C,
2.22% H, 13.25% N, 10.11% S; found: 34.26% C, 2.19% H, 13.34% N, 10.06% S.
2-Methoxycarbonyl-5,7-dinitrobenzo[d]thiazol-3-oxide (2)
The same procedure as above with triethylamine (0.31 ml, 2.2 mmol) and the reaction time of 2 h
gave, after column chromatography (silica gel, acetone–chloroform 1 : 10 v/v) and recrystallization
1
(toluene, alumina), product 2 (0.20 g, 33%) with m.p. 197–199 °C. H NMR spectrum: 9.26 and 9.12 AB,
2 H, 4J = 2.06 Hz (Ar); 4.02 s, 3 H (OCH3). 13C NMR spectrum: 156.88 (CO); 147.03, 146.13,
142.51, 137.38, 128.50 (5 × Cq); 121.19, 119.53 (2 × CH); 53.66 (OCH3). EI-MS, m/z (%): 299 (M+, 16);
283 (M+ – O, 20); 252 (M+ – O – OCH3, 35); 225 (M+ – O – COOCH3 + H, 100). For C9H5N3O7S
(299.2) calculated: 36.13% C, 1.68% H, 14.04% N, 10.72% S; found: 36.44% C, 1.70% H, 14.16% N,
10.87% S.
2-Methoxycarbonyl-5,7-dinitrobenzo[d]thiazole (3)
A mixture of compound 2 (200 mg, 7 mmol) and PCl3 (1.2 ml, 14 mmol) in chloroform (15 ml) was
heated to boiling 1 h, whereupon the solvent was distilled off in vacuum, and the evaporation residue
was submitted to column chromatography (silica gel, benzene–acetone 5 : 1 v/v) to give product 3
(100 mg, 50%) with m.p. 109–111 °C. 1H NMR spectrum: 9.55 and 9.18 AB, 2 H, 4J = 2.04 Hz
(Ar); 4.10 s, 3 H (OCH3). 13C NMR spectrum: 162.91 (CO); 159.48, 153.64, 146.86, 142.41, 136.26
(5 × Cq); 126.22, 118.29 (2 × CH); 54.14 (OCH3). EI-MS, m/z (%): 283 (M+, 48); 252 (M+ – OCH3, 18);
225 (M+ – COOCH3 + H, 100).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1431
Methyl S-(2,4-Dinitrophenyl)mercaptoacetate (4)
A solution of methyl mercaptoacetate (1.1 ml, 12 mmol) and triethylamine (1.4 ml, 10 mmol) in
chloroform (10 ml) was added drop by drop to a solution of 2,4-dinitrofluorobenzene (1.86 g, 10
mmol) in methanol (15 ml) at room temperature during 1 h. After 3 h stirring under argon atmos-
phere at room temperature, the solvents were distilled off in vacuum, the evaporation residue was
dissolved in chloroform, and the unreacted methyl mercaptoacetate was removed by extraction with
a 5% solution of sodium carbonate. Chloroform was distilled off, and the residue was recrystallized
from benzene with addition of Al2O3 to give compound 4 (1.6 g, 60%) with m.p. 94.5–95.5 °C (ref.4
1
4
3
gives m.p. 93–94 °C). H NMR spectrum: 9.10 d, 1 H, J = 2.50 Hz (H-3); 8.40 dd, 1 H, J = 9.02 Hz
(H-5); 7.69 d, 1 H (H-6); 3.85 s, 2 H (CH2); 3.79 s, 3 H (CH3). 13C NMR spectrum: 168.26 (CO);
145.07, 144.76, 144.40 (3 × Cq); 127.43, 127.14, 121.64 (3 × CH); 53.21 (OCH3); 34.66 (CH2).
Determination of pKA Differences
The ∆pKA values between acetic acid and methoxyacetic acid or N-methylmorpholine in methanol at the
ionic strength of 0.1 mol l–1 at 25 °C were determined spectrophotometrically from the absorbances of 2-
chloro-4-nitrophenoxide at the wavelength λ = 316 nm in the following buffers: [CH3COONa] =
[CH3COOH] = 0.1 mol l–1 (a); [CH3OCH2COONa] = 0.1 mol l–1, [CH3OCH2COOH] = 0.025 mol l–1 (b);
[morpholine] = 0.1 mol l–1, [morpholine . HCl] = 0.1 mol l–1 (c). The buffer (2 ml) was placed in a
1 cm quartz cell and after adjusting its temperature in thermostat a solution of 2-chloro-4-nitrophenol
in tetrahydrofuran (30 µl, 5 . 10–3 mol l–1) was added from a microdoser and the spectrum was
measured in the wavelength range from 240 to 470 nm. The spectrum of completely dissociated 2-chloro-
4-nitrophenol (AB) was measured in methanolic 0.1 mol l–1 sodium acetate and that of nondissociated
2-chloro-4-nitrophenol (ABH) in methanolic 0.1 mol l–1 acetic acid at the ionic strength I = 0.1 mol l–1
adjusted by addition of NaClO4. For the calculation of ∆pKA values from Eq. (1) for acetic (Ac) and
methoxyacetic (MAc) acids we used the absorbance values A, AB, and ABH measured at 316 nm. AAc
and AMAc are the absorbances of 2-chloro-4-nitrophenoxide in acetate and methoxyacetate buffers,
respectively.
∆pKA = pK AAc – pK AMAc = log ([CH3COOH]/[CH3COONa]) –
– log ([CH3OCH2COOH]/[CH3OCH2COONa]) + log [(AAc – ABH)/(AB – AAc)] –
– log [(AMAc – ABH)/(AB – AMAc)] = 1.16
(1)
For the N-methylmorpholine buffer we found the value ∆pKA = pK AAc – pK AMPh = 0.40. 2-Chloro-4-
nitrophenol was recrystallized from benzene, m.p. 107.5–108.5 °C, ref.5 gives m.p. 109.5–110.5 °C.
Kinetic Measurements
For the kinetic measurements we used solutions of compounds 1 and 2 (5 . 10–3 mol l–1) in dry
tetrahydrofuran and solutions of the following acids and bases in preboiled p.a. methanol: acetic acid,
anhydrous sodium acetate, methoxyacetic acid, N-methylmorpholine (1 mol l–1). The methanolic solution
of sodium methoxyacetate (1 mol l–1) was prepared from 4.25 mol l–1 sodium methoxide and 2 mol l–1
methoxyacetic acid. The solution of N-methylmorpholine hydrochloride (0.5 mol l–1) was prepared
by dissolving the crystalline hydrochloride in methanol. The hydrochloride was prepared by satura-
tion of N-methylmorpholine solution in dry ether with dry hydrogen chloride. The precipitated hydro-
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1432
Janik, Machacek, Pytela:
chloride was collected by suction in argon atmosphere, washed with dry ether, and dried at room
temperature under argon.
Sodium perchlorate p.a. for adjusting the ionic strength was dried at 110 °C/2.6 kPa for 4 h.
The stock solutions thus prepared were then used to prepare the following buffer stock solutions
with the ionic strength I = 0.1 mol l–1: Acetate buffers with [CH3COONa]/[CH3COOH] = 8 : 1; 6 : 1;
4 : 1; 2 : 1; 1 : 1; 1 : 2; 1 : 4; methoxyacetate buffers with [CH3OCH2COONa]/[CH3OCH2COOH] =
4 : 1; 2 : 1; 1 : 1; N-methylmorpholine buffers with [N-methylmorpholine]/[N-methylmorpholine
hydrochloride] = 4 : 1; 2 : 1; 1 : 1. The concentration of ionic component of buffer in the stock
solution was always 0.1 mol l–1. From these stock solutions we prepared series of buffers with
various ratios of components and various concentrations. The ionic strength of these solutions was
adjusted at I = 0.1 mol l–1 by adding methanolic 1 mol l–1 sodium perchlorate. For the kinetic experi-
ments, the stock solutions prepared (2 ml) were placed in 1 cm quartz cells and kept at 25 °C in the
closed cell. At the time t = 0, 20 µl stock solution of compound 1 was added and the absorbance was
monitored at 292 nm. The rate constants were calculated with the help of the OPKINA programme6.
RESULTS AND DISCUSSION
Substitued 2-alkoxycarbonylbenzo[b]thiazol-3-oxides are formed by base catalyzed re-
actions of substituted 2-nitrochlorobenzenes (which are further activated for nucleo-
philic substitutions) with mercaptoacetates4. Intermediates of these reactions –
substituted alkyl S-phenylmercaptoacetates – were only isolated in two cases, hence the
reaction described in Scheme 1 is preparatively carried out as a one-step reaction4,7.
Cl
SCH COOR
Y
2
Y
NO
Y
NO
2
2
S
HSCH COOR
2
B
+ HCl
+ H O
COOR
2
B
+
-
N
O
X
X
X
SCHEME 1
Methyl S-(2,4-dinitrophenyl)mercaptoacetate (4) was prepared in high yield4 since it
does not undergo a subsequent cyclization. Standing in 1 mol l–1 sodium methoxide
does not result in cyclization but in splitting off of the side chain and formation of a
mixture of products. This unwillingness to react in cyclization reaction – in contrast to
2-nitro-6-substituted sulfides of the type 4 – is obviously due to entropic effects: the
conformation suitable for cyclization is energetically unfavourable in compound 4.
Ethyl S-(4-cyano-2,6-dinitrophenyl)mercaptoacetate was prepared by reaction of
4-chloro-3,5-dinitrobenzonitrile and ethyl mercaptoacetate in benzene in the presence
of less than equimolar amount of triethylamine4. The reaction of 2,4,6-trinitrochloro-
benzene with methyl mercaptoacetate at similar conditions gave a mixture of methyl
S-(2,4,6-trinitrophenyl)mercaptoacetate and 2-methoxycarbonyl-5,7-dinitrobenzo[d]-
thiazol-3-oxide in practically quantitative yields (the ratio of the two substances was ca
20 : 1 after 1 h reaction). The mixture of both compounds is also formed by the reaction
in benzene in the presence of KHCO3, however, this reaction is slow and gives only
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1433
39% yield of compound 1 after 7 h reaction time. The anhydrous sodium acetate cata-
lyzed reaction of 2,4,6-trinitrochlorobenzene with ethyl mercaptoacetate4 only gave
2-ethoxycarbonyl-5,7-dinitrobenzo[d]thiazol-3-oxide.
1
Compounds 1 and 2 were identified by means of H and 13C NMR spectra, com-
pound 2 also by means of mass spectrum as was also its deoxygenation product –
2-methoxycarbonyl-5,7-dinitrobenzo[d]thiazol (3). The mass spectrum of compound 2
shows the typical fragment M+ – 16 which is absent after the deoxygenation.
The base catalyzed reaction of methyl S-(2,4,6-trinitrophenyl)mercaptoacetate leads
to products completely different from those obtained from the base catalyzed reaction
of methyl N-(2,4,6-trinitrophenyl)aminoacetate1 and methyl O-(2,4,6-trinitrophenyl)-
glycolate8 at similar conditions. With the sulfur derivative 1 no nitroso product is
formed, whereas the nitrogen derivative gives the nitroso compound as the only pro-
duct. The sulfur derivative gives its product by the attack of nitrogen atom of ortho-
standing nitro group by the carbanion –S–CH(–)–COOCH3 in spite of the fact that the
sulfur atom in compound 1 must be much more nucleophilic than the nitrogen atom in
the aminoacetates mentioned. Such an attack, on the other hand, was observed neither
with nitrogen1,2 nor oxygen8 derivatives. The different behaviour of the sulfur deriva-
tive is obviously due to substantial increase in the acidity of –SCH2COOCH3 group as
compared with >NCH2COOCH3 or –OCH2COOCH3 groups caused by the fact that sulfur
atom is better able to stabilize the carbanion. The effect of sulfur on acidity of CH
groups – as compared with the effect of oxygen – is obvious from the comparison of
pairs of 9-X-fluorenes9: X (pKA in DMSO at 25 °C) SC6H5 (15.4), OC6H5 (19.9), SC2H5
(18.0), OC2H5 (21.1).
The reaction of compound 1 with bases in methanol was studied spectrophotometri-
cally at 25 °C. The reaction with methanolic sodium methoxide of concentrations above
5 . 10–3 mol l–1 is very fast: the reaction course is complicated by formation of Meisen-
heimer adducts and the character of absorbance–time dependences at various wave-
lengths indicates that there occur consecutive reactions. Compound 2 is not the final
product of the reaction. Further decrease in methoxide concentration can be achieved
by application of buffers. In acetate, methoxyacetate and N-methylmorpholine buffers
the cyclization of compound 1 obeyed the 1st order rate equation, no intermediates
accumulated in the reaction course, and the reaction product showed the electronic
spectrum identical with that of compound 2 in the same buffer at the same concentra-
tion. The half-lives of cyclization of compound 1 to compound 2 ranged from about 30
to 7 200 s.
The mechanism of cyclization of compound 1 to compound 2 can be expressed by
Scheme 2. The cyclization can take two paths differing by the order of events: inter-
mediate In2(−) can be formed from intermediate In(1−) either via intermediate In(12−) after
primary deprotonation of CH group and subsequent protonation of oxygen, or on the
other hand, the protonation of oxygen can come first and be followed by deprotonation
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1434
Janik, Machacek, Pytela:
of CH group in intermediate In1(+/−). The resulting intermediate In(2−) splits off hydroxyl
ion to give product 2. The rate-limiting step can be deprotonation of CH group in
compound 1 or in some of the intermediates In(1−) or In1(+/−) or hydroxide ion splitting off
from the intermediate In(2−). The steps involving the protonation of oxygen in Scheme 2
must be fast equilibria. On the basis of kinetic studies it is possible to decide which
path is really adopted by the cyclization.
Acetate and methoxyacetate buffers. Tables I and II summarize the values of ob-
served rate constants kobs (s–1) of cyclization of compound 1 for buffers with various
ratios of components and various concentrations. In the acetate and methoxyacetate
buffers we found linear dependences of kobs vs buffer concentration in series of buffers
with constant ratios of components (Figs 1 and 2). The increasing ratio [B–]/[BH] of the
buffer components increases both the intercepts obtained by extrapolation to zero buf-
fer concentration and the slopes of dependences. Such behaviour is compatible with
rate equation (2):
k
obs = k0 + k ′MeO[CH3O–] + k B′ [B–] + k B′ ,MeO[B–][CH3O–] .
(2)
K
NO
1
NO
2
2
-
CH O
3
-
O N
2
S
CH
COOCH
O N
2
S
CH COOCH
2
3
3
CH OH
3
NO
NO
c
2
2
(-)
S
SH
k
-c
k
K
c
-1
NO
NO
2
2
K
2
-
S
CH O
S
COOCH
3
COOCH
3
3
+
N
+
N
CH OH
3
H
OH
H
-
O N
2
O N
2
-
-
O
O
O
(-)
In
(+/-)
In
1
1
-
k
[BH]
-
[B ]
B,MeO
Σ k [B ]
-B,MeO
k
Σ k
BH
[BH]
B
-1
NO
NO
K
2
2
3
-
CH O
S
-
+
3
S
-
k
2
COOCH
COOCH
3
3
2
+
CH OH
3
N
N
O N
2
O N
2
-
-
-
O
O
O
OH
(2-)
1
(-)
In
In
2
SCHEME 2
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1435
TABLE I
Observed (kobs) and calculated (kcal, from Eq. (2)) values (s–1) of rate constants of cyclization of com-
pound 1 to 2 in methanolic acetate buffers at 25 °C and I = 0.1 mol l–1
[B–]/[BH]
8 : 1
[B], mol l–1
kobs . 103, s–1
kcal . 103, s–1
0.10
0.05
0.01
0.10
0.05
0.01
0.10
0.07
0.05
0.01
0.10
0.05
0.01
0.10
0.07
0.05
0.01
0.10
0.07
0.05
0.02
0.01
0.10
0.085
0.07
0.05
0.035
0.02
0.01
13.41
11.91
10.51
10.36
9.02
7.78
7.22
6.71
6.27
5.55
4.11
3.35
2.79
2.53
2.14
1.92
1.48
1.62
1.34
1.15
0.86
0.75
1.04
0.97
0.90
0.75
0.66
0.52
0.42
13.51
11.86
10.55
10.36
9.02
7.95
7.21
6.59
6.18
5.36
4.06
3.34
2.76
2.49
2.14
1.92
1.46
1.70
1.40
1.21
0.91
0.81
–
6 : 1
4 : 1
2 : 1
1 : 1
1 : 2
1 : 4
–
–
–
–
–
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1436
Janik, Machacek, Pytela:
The dependence of kobs vs [B–] is not quite linear but slightly curved for the most acidic
acetate buffer (4 : 1), which resembles – even though to a much less extent – the
dependences of kobs vs [B] in the N-methylmorpholine buffers (Fig. 7) and could indi-
cate the same kinetic behaviour. As, however, in the still more acidic methoxyacetate
buffers the dependences of kobs vs [B–] are linear, the curvature in the acidic acetate
buffer is obviously due to a change in character of medium at the highest concentra-
tions of acetic acid. The term k ′MeO [CH3O–] corresponds to the cyclization catalyzed by
methoxide and its value was determined by extrapolation of linear dependences to the
zero buffer concentration. The k ′MeO values found in the series of acetate buffers are
constant and equal to (1.268 ± 0.012) . 10–3 s–1 (for acetate buffer with ratios of compo-
nents 1 : 1). Identical or very close values of k ′MeO were found in methoxyacetate and
morpholine buffers after recalculation to pH values of the acetate buffers with the help
of the ∆pKA values found: k M′ eO = 1.268 . 10–3 s–1 (for methoxyacetate buffers) and
k ′MeO = 1.304 . 10–3 s–1 (for morpholine buffers).
TABLE II
Observed (kobs) and calculated (kcal, from Eq. (2)) values (s–1) of rate constants of cyclization of com-
pound 1 to 2 in methanolic methoxyacetate buffers at 25 °C and I = 0.1 mol l–1
[B–]/[BH]
4 : 1
[B], mol l–1
kobs . 104, s–1
kcal . 104, s–1
0.100
0.085
0.070
0.050
0.020
0.010
0.100
0.070
0.050
0.030
0.010
0.100
0.070
0.050
0.030
0.010
5.20
4.82
4.52
4.22
3.77
3.60
3.07
2.58
2.31
2.01
1.73
1.92
1.54
1.34
1.05
0.97
5.12
4.86
4.60
4.25
3.73
3.56
2.99
2.58
2.31
2.04
1.77
1.93
1.58
1.34
1.11
0.87
2 : 1
1 : 1
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1437
As the dissociation constant KA of acid buffer component in methanol at the ionic
strength I = 0.1 mol l–1 is not known, the concentration of methoxide in Eq. (2) was
expressed by means of the concentration ratio [B–]/[BH] of the buffer components. The
methoxide concentration and the ratio of buffer components are interrelated by Eq. (3)
where KMeOH is the ion product of methanol (KMeOH = 10–16.92, ref.10). By introducing
the relation for methoxide concentration from Eq. (3) into Eq. (2) we get Eq. (4).
[CH3O–] = KMeOH[B–]/KA[BH]
(3)
kobs = k 0+ k M′ eO KMeOH[B–]/KA[BH] + k B′ [B–] + k ′B,MeO(KMeOH[B–]/KA[BH])[B–] (4)
–
[B ] : [BH]
8 : 1
12
6 : 1
3
k
. 10
obs
8
4 : 1
2 : 1
1 : 1
4
0
FIG. 1
Dependence of observed rate constants
kobs (s–1) of transformation of 1 into 2 upon
concentration of acetate ion [B–] (mol l–1) in
acetate buffers with varying ratio of buffer
components at I = 0.1 mol l–1 at 25 °C
1 : 2
–
0.00
0.05
0.10
[B ]
–
6
4
[B ] : [BH]
4 : 1
k
. 10
obs
4
2 : 1
1 : 1
2
0
FIG. 2
Dependence of observed rate constants
kobs (s–1) of transformation of 1 into 2
upon concentration of methoxyacetate ion
[B–] (mol l–1) in methoxyacetate buffers
with varying ratio of buffer components at
I = 0.1 mol l–1 at 25 °C
0.00
0.05
0.10
–
[B ]
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1438
Janik, Machacek, Pytela:
The dependences of kobs on concentration [B–] of the basic buffer component are
linear. The intercepts Di for a given ratio of buffer components are expressed as the
sum Di = k 0 + k M′ eO (KMeOH/KA)([B–]/[BH]) and correspond to the cyclization catalyzed
by methoxide. The slopes Ai of these dependences are expressed as the sum
k ′B + k ′B,MeO (KMeOH/KA)([B–]/[BH]). The values of intercepts and slopes are presented
in Table III.
The dependence of slopes Ai on the ratio [B–]/[BH] of the buffer components is
linear with the intercept corresponding to k ′B and the slope corresponding to
k ′B,MeO KMeOH /KA. These dependences for acetate and methoxyacetate buffers are re-
presented graphically in Figs 3 and 4. The slope values are loaded with large error in the
case of the methoxyacetate buffers and the justification for the presumption of linear
dependence in this case (Fig. 4) is supported by the results of the multiple linear re-
gression carried out. The intercepts at the y-axis have the values k ′B = (80.1 ± 8.6) . 10–4
300
4
A . 10
i
200
FIG. 3
Dependence of slopes Ai = ∆kexp/∆[B–]
100
(l mol–1 s–1) upon ratio [B–]/[BH] of buffer
components in acetate buffers at I = 0.1
mol l–1 at 25 °C
–
0
4
8
[B ]/[BH]
18
4
A . 10
i
14
10
FIG. 4
Dependence of slopes Ai = ∆kexp/∆[B–]
(l mol–1 s–1) upon ratio [B–]/[BH] of buffer
components in methoxyacetate buffers at I =
0.1 mol l–1 at 25 °C
0
2
4
–
[B ]/[BH]
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1439
and (9.7 ± 1.9) . 10–4 l mol–1 s–1 for the acetate and methoxyacetate buffers, respectively.
The respective slopes k ′B,MeO = kB,MeOKcK1KA/KMeOH have the values (31.8 ± 1.9) . 10–4
and (1.97 ± 0.71) . 10–4 l2 mol–2 s–1.
On the basis of the Bodenstein steady state approximation it is possible to formulate
the rate equations (5) or (6) for the system of consecutive reactions described by
Scheme 2 and going from starting compound 1 to product 2 via the intermediate In1(2−)
or In(1+/−), respectively.
v
(2–) = k o(2b−s)[SH] = kcK1kB,MeO[CH3O–][B–][SH]/(k–c + kB,MeO[B–])
(5)
(6)
v(+/–) = k
[SH] = KcK1k2kB[B–][SH]/K2(kBH[BH] + k2)
(+/−)
obs
The key intermediates for deriving the approximation are In(1−) and In2(−) for the paths via
In(12−) and In1(+/−), respectively.
The cyclization of compound 1 in acetate and methoxyacetate buffers follows fully
neither of the rate equations given because linear dependences were found experimentally
between kobs and concentrations of bases (Eq. (2)). If k–c >> kB,MeO[B–], then Eq. (5) is
reduced to Eq. (7). If k2 >> kBH[BH], then Eq. (6) is reduced to Eq. (8).
v(2–) = kobs[SH] = KcK1kB,MeO[CH3O–][B–][SH]
(+/–) = kobs[SH] = KcK1kB[B–][SH]/K2
(7)
(8)
v
TABLE III
Values of intercepts Di (s–1) = (k M′ eO (KMeOH/KA)([B–]/[BH])) and slopes Ai (l mol–1 s–1) =
k ′B + k B′ ,MeO (KMeOH/KA)([B–]/[BH]) of dependences of kobs vs base concentration in acetate (Ac) and
methoxyacetate (MAc) buffers determined by linear regression
Ac
MAc
[B–]/[BH]
Di . 104
Ai . 104
Di . 104
Ai . 104
1 : 2
1 : 1
2 : 1
4 : 1
6 : 1
8 : 1
6.63 ± 0.06
13.51 ± 0.24
26.33 ± 0.22
53.60 ± 0.35
75.29 ± 0.77
102.30 ± 0.92
96.30 ± 1.05
116.00 ± 3.71
146.90 ± 3.41
187.30 ± 5.29
285.90 ± 11.90
321.30 ± 14.20
–
–
0.80 ± 0.05
1.57 ± 0.02
3.34 ± 0.05
–
10.93 ± 0.87
14.82 ± 0.29
17.07 ± 0.84
–
–
–
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1440
Janik, Machacek, Pytela:
The right-hand sides of Eqs (7) and (8) are terms of Eq. (2). This means that the cycli-
zation of compound 1 in acetate and methoxyacetate buffers goes by both paths in
Scheme 2, i.e. via both the intermediates In(12−) and In1(+/−). The third term in Eq. (2),
k ′MeO[CH3O–], corresponds to specific base catalyzed reaction. It cannot be decided
which reaction path is adopted by the cyclization. The cyclizations obeying the rate
equations (7) and (8) can be described by the sequences of steps in (A) and (B). In both
paths the deprotonation of CH group of the intermediates is rate limiting. The Gibbs
energy changes along the reaction coordinate in the cyclizations going via the inter-
mediates In(12−) and In1(+/−) are represented qualitatively in Figs 5 and 6 (graph a), respec-
tively.
Kc
K1
K1
kB
SH
SH
S(–)
In(1−)
In(1−)
In(12−)
In(2−)
In(2−)
P
(A)
(B)
RLS
Kc
K2
kB
S(–)
In(1+/−)
P
RLS
Experimentally accessible are only the values of stoichiometric rate constants k B′ ,MeO
=
kB,MeOKcK1KA/KMeOH and k B′ = kBKcK1/K2 because the values of corresponding equili-
brium constants are not known.
The parameters of rate equation (4) were also evaluated by multiple linear re-
gression. kobs was chosen for the dependent variable, and the interpreting variables in-
volved [B–], [B–]/[BH] and [B–][B–]/[BH]. The values of rate constants thus obtained
for acetate and methoxyacetate buffers are summarized in Table IV. Tables I and II
give beside the kobs values also the kcal values recalculated from the rate constant values
for each buffer and each base concentration. We also tested the statistical significance
k [B]
B
∆G
(2−)
In
1
k
–1
(−)
In
2
(−)
In + B
1
(−)
(−)
(−)
S
+ B
FIG. 5
2
(–)
(–)
Gibbs energy profile of cyclization of 1 to 2 with
SH + CH O + B
3
the rate-limiting formation of In(12−) intermediate
reaction coordinate
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1441
of the k 0 constant, i.e. of the possibility of the cyclization catalyzed by methanol. The
value of test criterion is t = 1.470, t0.975(18) = 2.101 in the acetate buffers and t = 1.990,
t0.975(12) = 2.179 in the methoxyacetate buffers. The k 0 value is statistically insignifi-
cant wherefrom it follows that the solvent-catalyzed cyclization takes place neither in
acetate nor in methoxyacetate buffers.
N-Methylmorpholine buffers. In contrast to the acetate and methoxyacetate buffers,
the dependences of kobs vs [B] are not linear (Fig. 7, Table V), the slopes being de-
creased with increasing base concentration. This indicates a decreasing reaction order
in base with its increasing concentration and a change in the rate-limiting step. The
dependences of kobs vs [B] extrapolated to zero buffer concentration form intersects at
y-axis whose magnitude is proportional to the methoxide concentration in the buffers.
This indicates the existence of the k M′ eO[CH3O–] term in the rate equation for the cycli-
zation. As long as the cyclization in the N-methylmorpholine buffers follows the same
TABLE IV
Rate constants ki, their standard deviations, residual standard deviation s, and correlation coefficient
R of multiple correlation
Parameter
k0 . 105, s–1
k′MeO . 104, s–1
k′B . 104, l mol–1 s–1
k′B,MeO . 104, l2 mol–2 s–1
s . 105
Ac
MAc
7.742 ± 5.256
12.680 ± 0.122
83.540 ± 8.340
30.610 ± 1.918
8.446
1.228 ± 0.615
0.877 ± 0.023
9.858 ± 1.004
1.878 ± 0.358
0.585
R
0.9998
0.9993
k
[BH]
BH
∆G
a
k
2
b
(−)
2
In
(+ /−)
(−)
FIG. 6
In
+ B
1
(−)
(−)
Gibbs energy profile of cyclization of 1 to 2 with the
rate-limiting formation of In(2−) intermediate (a) or
with comparable velocities of decomposition of
In(2−) intermediate into In(1+/−) and transformation of
In(2−) to 2 (b)
In + B
1
(−)
(−)
S
+ B
2
(–)
(–)
SH + CH O + B
3
reaction coordinate
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1442
Janik, Machacek, Pytela:
mechanism as that in the acetate and methoxyacetate buffers, it should obey the rate
equation (9) or (10).
k
obs = k ′MeO[CH3O–] + (KcK1/K2)(k2kB[B]/(kBH[BH+] + k2))
(9)
kobs = k ′MeO[CH3O–] + kckB,MeOK1[CH3O–][B]/(k–c + kB,MeO[B])
(10)
Both equations express straight lines in the coordinates 1/(kobs – k ′MeO[CH3O–]) and
1/[B] (Eqs (11) and (12)). Parameters a1, a2, a3, a4 are combinations of rate and equili-
brium constants. For a series of buffers with various concentrations and various ratios
of components the solution of Eq. (11) represents a series of parallel straight lines with
the slope of 1/a1 = K2/KcK1kB.
(kobs – k M′ eO[CH3O–])–1 = (a1[B])–1 + (a2[CH3O–])–1
(11)
(12)
(kobs – k ′MeO[CH3O–])–1 = (a3[CH3O–][B])–1 + (a4[CH3O–])–1
In the case of Eq. (12) the slopes depend on the methoxide concentration in N-methyl-
morpholine buffer (the slope (a3[CH3O–])–1). The cyclization of compound 1 in
N-methylmorpholine buffers obeys rate equation (9) (Fig. 8), hence it goes through the
In(1+/−) intermediate. Again it is impossible to decide about the reaction path taken by the
20
3
k
. 10
obs
+
[B] : [BH ]
15
4 : 1
10
5
2 : 1
FIG. 7
Dependence of observed rate constants kobs (s–1
)
1 : 1
of transformation of 1 into 2 upon concentration
of N-methylmorpholine [B] (mol l–1) in N-methyl-
morpholine–N-methylmorpholinium chloride buffers
with varying ratio of buffer components at I = 0.1
mol l–1 at 25 °C
0
0.0
0.1
0.2
0.3
0.4
[B]
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1443
specific base catalyzed reaction. The graph in Fig. 6 depicts qualitatively the Gibbs
energy changes along the reaction coordinate. At low concentrations of N-methylmor-
pholine buffer the reaction is 1st order in base and the rate-limiting step is the formation
of the In(2−) intermediate. Both the formation and the reverse decomposition of inter-
mediate In(2−) are accelerated by increasing buffer concentration, and the reaction order
in N-methylmorpholine is decreased. The above-mentioned changes in the rate-limiting
step are reflected in the difference between graphs a and b in Fig. 6.
TABLE V
Observed (kobs) and calculated (kcal) values (s–1) of rate constants of cyclization of compound 1 to 2 in
methanolic N-methylmorpholine–N-methylmorpholinium chloride buffers at 25 °C at I = 0.1 mol l–1
[B]/[BH+]
4 : 1
[B], mol l–1
kobs . 103, s–1
kcal . 103,a, s–1
kcal . 103,b, s–1
0.400
0.300
0.200
0.100
0.075
0.050
0.030
0.010
0.200
0.100
0.075
0.050
0.030
0.010
0.100
0.085
0.070
0.050
0.035
0.020
0.010
19.72
18.52
15.55
10.60
9.32
7.21
5.62
3.30
9.64
7.51
6.68
5.36
4.05
2.19
4.76
4.59
4.38
3.80
3.10
2.33
1.59
19.42
17.75
15.22
10.92
9.33
7.40
5.54
3.30
9.73
7.55
6.61
5.36
4.02
2.19
4.81
4.57
4.27
3.74
3.17
2.36
1.58
19.54
17.77
15.13
10.75
9.16
7.26
5.46
3.31
9.77
7.57
6.62
5.37
4.03
2.21
4.89
4.64
4.34
3.78
3.20
2.37
1.60
2 : 1
1 : 1
a
b
Calculated on the basis of the results obtained by linearization. Calculated on the basis of the
results obtained by multiple nonlinear regression.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1444
Janik, Machacek, Pytela:
The rate constants were obtained from both the linearized form (Eq. (11)) and nonli-
near regression. Equation (9) was modified to Eq. (13).
k
obs = k ′MeO([B]/[BH+]) + ((KcK1/K2)kB[B])/((kBH/k2)[BH+] + 1)
(13)
With regard to the broad range of the dependent variable, the ln kobs values were used
in the optimization. The following values of rate constants and their combinations were
obtained from the calculation:
k ′MeO = (5.19 ± 0.20) . 10–4 s–1, KCK1kB/K2 = 0.129 ± 0.004 l mol–1 s–1,
k
BH/k2 = 19.56 ± 0.89 l mol–1.
The residual standard deviation determined for the ln kobs values was s = 1.946 . 10–2,
which indicates a very good fit.
+
[B] : [BH ]
800
1 : 1
2 : 1
4 : 1
k
400
k
FIG. 8
Dependence of 1/(kobs – k ′MeO[CH3O–]) (s) upon
0
0
40
80
1/[B] in N-methylmorpholine buffers
1/[B]
The authors are indebted to Prof. V. Sterba for valuable discussion. The research was financially
supported by the Grant Agency of the Czech Republic, Grant No. 203/97/0545.
REFERENCES
1. Machacek V., Hassanien M. M. M., Sterba V.: J. Chem. Soc., Perkin Trans. 2 1987, 867.
2. Kavalek J., Machacek V., Sterba V., Lycka A.: Collect. Czech. Chem. Commun. 53, 1033
(1988).
3. Capon B., McManus S. P.: Neighboring Group Participation, Vol. 1, Chap. 5, p. 195. Plenum
Press, New York 1976.
4. Wagner K., Heitzer H., Oehlmann L.: Chem. Ber. 106, 640 (1973).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Kinetics of Cyclization
1445
5. Simeson H. N., Hancock C. K., Meyers E. A.: J. Org. Chem. 30, 2678 (1965).
6. Pytela O., Vecera M., Vetesnik P.: Chem. Listy 73, 754 (1979).
7. Wagner K., Roos E. (Farbenfabriken Bayer AG): DOS 1 904 653; Chem. Abstr. 73, 131992
(1970).
8. Janik M., Machacek V., Sterba V.: Unpublished results.
9. Bordwell F. G.: J. Pure Appl. Chem. 49, 963 (1977).
10. Rochester C. H., Rossall B.: J. Chem. Soc. 1967, 743.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)