Kinetics and mechanism of the base-catalysed cyclisation of 2-(substituted benzoylamino)benzamides giving quinazolin-4-one and quinazolin-4-thione derivatives

Article abstract of DOI:10.1002/1099-0690(200206)2002:11<1855::AID-EJOC1855>3.0.CO;2-A

Acylation of 2-aminobenzamide and 2-(methylamino)benzamide with substituted benzoyl chlorides in acetone has been used to prepare the respective 2-(substituted benzoylamino)benzamides 1a-i, which were then subjected to sodium methoxide catalysed ring closure to give the respective 2-(substituted phenyl)quinazolin-4 -ones 2a-i. The kinetics of the cyclisation reactions were monitored by UV/Vis spectroscopy at 25 °C in methanolic solutions of sodium methoxide. In the case of 2-(substituted benzoylamino)benzamides 1a-i and 2-(substituted benzoylamino)thiobenzamides 3a-j, non-linear dependences of observed rate constants κ_obs on the sodium methoxide concentrations were obtained, the shape of them being typical of a reaction with rapid pre- equilibrium. All the cyclisation reactions satisfactorily obeyed the Hammett correlation. In the case of 2 - [(benzoyl) (methyl) amino] benzamides 1e-i, increasing sodium methoxide concentration resulted in a progressive increase in κ_obs values which is probably due to formation of dianion. In the case of 2-(substituted benzoylamino)thiobenzamides 3b and 3h, which differ in the presence of a methyl group on the nitrogen atom the values of the activation Gibbs energy ΔG^? _{25 °C}, activation enthalpy ΔH^? _{25 °C}, and activation entropy ΔS^? _{25 °C} for their respective cyclisations to 2-(substituted phenyl)quinazoline-4-thiones 4b and 4h were determined. Whereas the ΔG^? _{25 °C} values were very close for the two substances, the ΔH^? _{25 °C} and ΔS^? _{25 °C} distinctly differed. This was interpreted by the enthalpy and entropy factors operating against each other in the cyclisation. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002).

Full text of DOI:10.1002/1099-0690(200206)2002:11<1855::AID-EJOC1855>3.0.CO;2-A

FULL PAPER
Kinetics and Mechanism of the Base-Catalysed Cyclisation of 2-(Substituted
benzoylamino)benzamides Giving Quinazolin-4-one and Quinazolin-4-thione
Derivatives
[a]
[a]
[a]
[a]
ˇ
ˇ
ˇ
ˇ´
ˇ
´
˚
Jirı Hanusek,* Milos Sedlak, Petr Simunek, and Vojeslav Sterba
´
´
Dedicated to Professor JaromırKavalek on the occasion of his 65th birthday
Keywords: Kinetics / Cyclisations / Sulfur / Nitrogen heterocycles
Acylation of 2-aminobenzamide and 2-(methylamino)benza- ides 1e−i, increasing sodium methoxide concentration re-
mide with substituted benzoyl chlorides in acetone has been sulted in a progressive increase in kobs values which is prob-
used to prepare the respective 2-(substituted benzoylamino)- ably due to formation of dianion. In the case of 2-(substituted
benzamides 1a−i, which were then subjected to sodium me-
benzoylamino)thiobenzamides 3b and 3h, which differ in the
presence of a methyl group on the nitrogen atom the values
thoxide catalysed ring closure to give the respective 2-(sub-
stituted phenyl)quinazolin-4-ones 2a−i. The kinetics of the of the activation Gibbs energy ∆G‡25 °C, activation enthalpy
cyclisation reactions were monitored by UV/Vis spectroscopy ∆H‡25 °C, and activation entropy ∆S‡
for their respective
25 °C
at 25 °C in methanolic solutions of sodium methoxide. In the
case of 2-(substituted benzoylamino)benzamides 1a−i and 2-
(substituted benzoylamino)thiobenzamides 3a−j, non-linear
dependences of observed rate constants kobs on the sodium
methoxide concentrations were obtained, the shape of them
being typical of a reaction with rapid pre-equilibrium. All the
cyclisation reactions satisfactorily obeyed the Hammett cor-
cyclisations to 2-(substituted phenyl)quinazoline-4-thiones
4b and 4h were determined. Whereas the ∆G‡
values
and
25 °C
were very close for the two substances, the ∆H‡
25 °C
∆S‡
distinctly differed. This was interpreted by the en-
25 °C
thalpy and entropy factors operating against each other in
the cyclisation.
( Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
relation. In the case of 2-[(benzoyl)(methyl)amino]benzam- 2002)
Introduction
The derivatives of quinazolin-4-one and -4-thione are po-
tential medical drugs.[1] They started to attract attention
after the discovery of the antimalarial activity of the natural
alkaloid Febrifugin. Along with the development of new
antimalarials it was found that quinazolin-4-one derivatives
show a general calming effect on the CNS and possess hyp-
notic,[2] sedative, analgesic, antipyretic,[3] and diuretic[4]
properties. At present, about 25 preparations based on qui-
nazolin-4-one are registered.[5]
In our previous paper[6] we dealt with the synthesis and
structure of 2-(benzoylamino)thiobenzamides and products
of their cyclisation in basic medium, i.e. 2-phenylquinazol-
ine-4-thiones.
Scheme 1
The aim of the present paper is to study the kinetics and
mechanism of base-catalysed ring closure of substituted 2-
(benzoylamino)thiobenzamides (Scheme 1) and to compare
their reactivity with that of their oxygen analogues.
Results and Discussion
The cyclisation reaction was studied in detail using meth-
anolic solutions of sodium methoxide in the concentration
range of 0.01Ϫ1.6 mol·LϪ1. The cyclisation rate has been
found to depend nonlinearly on the concentration of the
base used, i.e. sodium methoxide, which applies to various
[a]
University of Pardubice, Faculty of Chemical Technology,
Department of Organic Chemistry
ˇ
´
´
Nam. Cs. Legiı 565, 53210 Pardubice, Czech Republic
Eur. J. Org. Chem. 2002, 1855Ϫ1863  WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0611Ϫ1855 $ 20.00ϩ.50/0
1855

ˇ
ˇ
ˇ
´
˚
J. Hanusek, M. Sedlak, P. Simunek, V. Sterba
FULL PAPER
substituents both at the nitrogen atom (R1 ϭ H, CH3) and amount of 18-crown-6 ether, which is an efficient complex
at the benzene ring of the starting 2-(benzoylamino)benza- former for the Naϩ cation, the cyclisation reaction was
mide or 2-(benzoylamino)thiobenzamide. Scheme 2 can be slowed down (Figure 1) but the trend of kobs vs. [CH3ONa]
suggested for the said ring closure
dependence remained unchanged. In addition to that, fur-
ther additions of 18-crown-6 ether caused further decelera-
tion of the reaction, which can be due to the change of
the medium. When the Naϩ concentration was increased by
addition of sodium perchlorate at a constant concentration
of sodium methoxide, the observed rate constant increased
but slightly (Table 1). These findings support the idea that
the catalytic effect of Naϩ is small, which is connected with
the impossibility of effective stabilisation of the transition
state of the rate-limiting step. The catalysis by the sodium
ion in methanol is only significant if Naϩ is strongly bound
by chelate formation in the transition state.[8aϪ8c]
Figure 1. Observed rate constant of the cyclisation (kobs/sϪ1) of the
2-[(benzoyl)(methyl)amino]benzamides 1eϪi vs. the concentration
of sodium methoxide (c/mol·LϪ1) at 25 °C
Scheme 2
In the sodium methoxide solutions used, a fast pre-equi-
librium deprotonation takes place first: a proton is split off
from both the nitrogen atom adjacent to benzene ring (if
R1 ϭ H; the so-called blind alley) and the terminal nitrogen
atom (CXNH2 group). The nucleophilic terminal anion
formed in the latter case attacks the carbonyl group of the
benzoyl moiety to give a tetrahedral intermediate, which
undergoes a proton transfer from nitrogen to oxygen atom
Another possibility lies in the participation of two
CH3ONa molecules in the formation of the transition state
of the rate-limiting step. That means that the formation of
؊
In1 is preceded by a pre-equilibrium, and the rate-limiting
؊
step consists of a reaction of this In1 with one CH3ONa
molecule giving the dianion In2؊. In principle, the interme-
؊
diate In2 can be formed by three reaction pathways
(Scheme 3).
؊
(In1 Ǟ In2؊) and then splits off a hydroxide anion to pro-
The first possibility is a solvent-mediated proton trans-
fer[9] from the nitrogen to the oxygen atom mediated by one
or two methanol molecules either through a synchronous
mechanism or through a cyclic transition state or via inter-
mediate InH. The second possibility consists of the forma-
tion of the dianion[10,11] of the tetrahedral intermediate
duce 2-phenylquinazolin-4-one or 2-phenylquinazoline-4-
thione, respectively.
1. Cyclisation of 2-[(Benzoyl)(methyl)amino]benzamides
1e؊i and -thiobenzamides 3g؊j
In the case of the oxygen derivatives 1eϪi (Figure 1), a In2؊, which on further reaction with methanol will give
gradual increase in slope of the dependence of kobs vs. In2؊. On the basis of Schemes 2 and 3 and using the Bod-
[CH3ONa] is observed. This can be due to a catalytic effect enstein approximation, we derived Equation (1) for the ob-
of the sodium cation.[7aϪ7c] Therefore, we studied the effect served rate constant, where k3 is the rate constant of the
of Naϩ on the cyclisation rate and/or the trend of the kobs formation of In2؊ by the reaction with CH3ONa, and kS is
vs. [CH3ONa] dependence. In the presence of an equimolar the rate constant of the solvent-mediated proton transfer
1856
Eur. J. Org. Chem. 2002, 1855Ϫ1863

Base-Catalysed Cyclisation of 2-(Substituted benzoylamino)benzamides
FULL PAPER
Table 1. Cyclisation of 2-[(4-methoxybenzoyl)(methyl)amino]benzamide (1g) in 0.20 mol·LϪ1 sodium methoxide at 25 °C at varying
concentrations of sodium perchlorate
c(NaClO4) [mol·LϪ1
]
102·kpoz [sϪ1
1g
]
c(NaClO4) [mol·LϪ1
]
102·kpoz [sϪ1
]
1g
0.10
0.20
0.30
1.55
1.55
1.59
0.40
0.50
0.60
1.73
1.79
1.90
Scheme 3
؊
from the nitrogen to the oxygen atom through the cyclic formation of In1 would become rate-limiting, and the de-
mechanism and/or via the InH intermediate.
pendence again should be linear with a zero slope.
Figure 3 presents the dependence of the observed rate
constant kobs on the methoxide concentration for the indi-
vidual sulfur derivatives 3gϪj. From the course of experi-
mentally found dependences it follows that in this case the
reaction order in methoxide is equal to one at low
[CH3ONa] values and gradually drops to zero at higher
[CH3ONa] values, which is characteristic of reactions with
a fast pre-equilibrium followed by the rate-limiting step. In
the compounds of the type Ϫ(CϭS)ϪNHϪR the proton at
the nitrogen atom is much more acidic than that in the type
Ϫ(CϭO)ϪNHϪR. For instance, the pKa value of 4-nitro-
phenylthiourea is four orders of magnitude lower than that
of 4-nitrophenylurea.[12] In thermodynamically unfavour-
k
obs ϭ {kc·K2·[CH3OϪ]·(k3·[CH3OϪ] ϩ
kS)}/(kϪc ϩ k3·[CH3OϪ] ϩ kS)
(1)
؊
When In1 is formed in a fast pre-equilibrium (kϪc ϾϾ
k3·[CH3ONa] ϩ kS), Equation (1) will be simplified after
dividing it through [CH3ONa] to give Equation (2).
kobs/[CH3OϪ] ϭ Kc·K2·(k3·[CH3OϪ] ϩ kS)
(2)
The dependence of kobs/[CH3ONa] vs. [CH3ONa] is pre- able proton-transfers between oxygen and nitrogen atoms,
sented in Figure 2. The intercepts at the y axis correspond the change in logk becomes comparable[13] with that in
to the solvent-mediated proton transfer, and the slope of logKa, which means that the reaction of In1؊ with CH3ONa
the linear dependence (for R2 ϭ OCH3, CH3, and H) corre- is much faster for the sulfur derivatives than for the oxygen
؊
sponds to the reaction of In1؊ with CH3ONa. With derivat- derivatives, and the formation of In1 becomes rate-limit-
ive 1h (R2 ϭ Cl) and still more with 1i (R2 ϭ NO2), increas- ing. For kobs we can suggest Equation (3), which describes
ing sodium methoxide concentration causes the slope to the reacting system very well, and the optimisation leads to
drop because the value of k3[CH3ONa] ϩ kS approaches to values of parameters kc (the cyclisation rate constant) and
k
Ϫc. On further increasing the CH3ONa concentration, the K2 (equilibrium constant of the pre-equilibrium).
Eur. J. Org. Chem. 2002, 1855Ϫ1863
1857

ˇ
ˇ
ˇ
´
˚
J. Hanusek, M. Sedlak, P. Simunek, V. Sterba
FULL PAPER
the basicity of medium is not expressed by [CH3ONa] and
the HM function[14a,14b] defined by Equation (4) has to be
used instead. However, the available literature does not give
the HM function for thiobenzamides and, therefore, we ad-
opted the HM function for formanilides[14c] and, using it
and Equation (5), we calculated the corresponding
(apparent Ϫ cЈ) concentration of CH3ONa.
Table 2. The parameters of cyclisation reaction obtained by op-
timising the data measured for derivatives 3gϪj using the sodium
methoxide concentration (c/mol·LϪ1) and the apparent concentra-
tion (cЈ/mol·LϪ1) calculated from Equation (5)
Com- R2
pound
kc [sϪ1] K2 [L·molϪ1] kcЈ [sϪ1] K2Ј [L·molϪ1
]
3g
3h
3i
H
0.448 1.86
0.419
0.202
0.312
0.259
2.06
1.86
2.58
2.71
4-N(CH3)2 0.277 1.14
4-CH3
4-OCH3
0.362 2.00
0.312 1.96
3j
Figure 2. Observed rate constant of the cyclisation divided by the
sodium methoxide concentration (kobs/c) of 2-[(benzoyl)(methyl)a-
mino]thiobenzamides 3gϪj vs. the concentration of sodium me-
thoxide (c/mol·LϪ1) at 25 °C
H
M ϭ pKS(CH3OH) ϩ logcCH ONa Ϫ
3
(4)
(5)
Ϫ
logaCH OH Ϫ log(γBϪ/γBHγCH O
)
3
3
HM ϭ pKS(CH3OH) ϩ logcЈCH ONa
3
With the sulfur derivatives the rate-limiting step consists
of the ring closure of the substrate anion. As the tetrahedral
؊
intermediate In1 formed is very unstable, the structure of
the transition state will be close to that of this intermediate
(the Hammond postulate), so the In1؊ structure can be used
for a qualitative evaluation of the effect of the medium on
the transition state, and the effect of the medium on the
dissociation of formanilides can be compared with the
؊
formation of In1 in methanol. In the case of formanilides,
the very strongly solvated CH3OϪ anion and relatively
weakly solvated amide are transformed into a strongly
solvated amide anion with delocalised charge. In the other
case, the weakly solvated amide and thioamide group and
strongly solvated CH3OϪ anion will produce the strongly
solvated In1؊, whose charge is concentrated at the oxygen
atom, hence the net change in solvation will be much
smaller in this case. An analogous situation can be expected
with the transition state, which is similar to the interme-
diate. Therefrom it can be concluded that the basicity of
medium for this reaction will lie between the limit cases
expressed by the CH3ONa concentration and the HM func-
tion (or apparent concentration cЈ) for the amides.
Figure 3. Observed rate constant of the cyclisation (kobs/sϪ1) of the
2-[(benzoyl)(methyl)amino]thiobenzamides 3gϪj vs. the concentra-
tion c (solid symbols and solid lines) and vs. the apparent concen-
tration cЈ (open symbols and dashed lines) of sodium methoxide
(c/mol·LϪ1) calculated by means of the HM function and Equation
(5) at 25 °C
k
obs ϭ (kc·K2·[CH3OϪ])/(1 ϩ K2·[CH3OϪ])
(3)
The values presented in Table 2 show that the cyclisation
constant decreases with increasing ability of the given sub-
The cyclisation is very fast (the reaction half-life values stituent to donate electrons to the reaction centre. The value
are about 2 s at the highest sodium methoxide concentra- of constant K2 depends only little on the substitution, and
tions used) and that is why only the parameters kc and K2 amounts to about 2 L·molϪ1. From the K2 values and
(Table 2) could be found for the deactivated substrates Equation (6) it is possible to calculate also pKa of 2-
3gϪj. Since the pKa value of thiobenzamide group is very [(benzoyl)(methyl)amino]thiobenzamides 3gϪj, which var-
high, the measurements had to be carried out in sodium ies from 16.6 to 16.9. The autoprotolysis constant value[15]
methoxide solutions with concentrations above 0.6 , where of methanol at 25 °C used is pKS(CH3OH) ϭ 16.916.
1858
Eur. J. Org. Chem. 2002, 1855Ϫ1863

Base-Catalysed Cyclisation of 2-(Substituted benzoylamino)benzamides
FULL PAPER
stabilised by resonance with both carbonyl group and ben-
zene ring). This can be expressed by K1 ϾϾ K2, and Equa-
tion (7) can be modified to Equation (8).
pKa ϭ pKS(CH3OH) ϩ pK2
(6)
2. Cyclisation of 2-(Benzoylamino)benzamides 1a؊d and -
thiobenzamides 3a؊f
k
obs ϭ (kc·K2·[CH3OϪ])/(1 ϩ K1·[CH3OϪ])
(8)
For both groups of derivatives Equation (7) was derived
for the observed rate constant in the way analogous to that
in the previous case.
Optimisation of measured data provided values of prod-
ucts kc·K2 and values of K1. Application of Equation (6) (in
which the value of pK2 is replaced by pK1) gave the values
of pKa1 (Table 4).
k
obs ϭ (kc·K2·[CH3OϪ])/{1 ϩ (K1 ϩ K2)·[CH3OϪ]}
(7)
In contrast to the N-methyl derivatives (R1 ϭ CH3), two
protons can be split off in the starting compounds 1aϪd
and 3aϪf. The proton easier to split off is at the nitrogen
atom adjacent to the benzene ring (the equilibrium constant
K1), but the anion thus formed [NϪ] cannot enter cyclis-
ation, so this is the so-called blind alley. The concentration
of the reactive anion [NHϪ] formed by deprotonation of
the terminal CXNH2 group is expressed by the equilibrium
constant K2.
Table 4. Parameters of cyclisation reaction obtained by optimis-
ation of data measured for derivatives 1aϪd
Compound
(X ϭ O)
103·kcK2
K1
pKa1
R2
[L·molϪ1·sϪ1
]
[L·molϪ1
]
1a
1b
1c
1d
4-H
4-CH3
4-Cl
1.55
0.97
4.11
22.9
0.7
0.45
3.0
17.1
17.3
16.4
15.7
4-NO2
15.4
The cyclisation of the oxygen derivatives 1aϪg has al-
ready been described,[16] but it was performed in water, not
in methanol. The cyclisation rate in methanol is markedly
lower than that in water, which is connected with differences
in the solvating ability of the starting compound and activ-
ated complex. Both the oxygen derivatives 1aϪd and the
sulfur analogues 3aϪf obey the given Equation (7) very
well. The value of K2 of the sulfur NH derivatives 3aϪf is
probably close to those found for the N-methyl derivatives
3gϪj, K2 ഠ 2. Under this presumption, K2 can be intro-
duced in Equation (7), which is valid for 2-(benzoylamino)-
thiobenzamides 3gϪj. In this way, Equation (7) is reduced
to a two-parameter equation.
However, the above said presumption is an approxi-
mation because in some cases an even small variation of the
basic skeleton can cause a distinct change in pKa, as was
observed, e.g., with 2,2-dioxo-2,3-dihydro-1H-2λ6-benzo[1,-
2,6]thiadiazin-4(3H)-ones[17a] and acylthioureas.[17b]
The simplified Equation (7) allows to optimise the values
(Table 3) of the remaining parameters (kc and K1). How-
The measured data and the curves obtained by means of
the optimised parameters for the sulfur derivatives 3aϪf are
presented in Figure 4. The values of the optimised parameters
kc and K1 fulfil the Hammett correlation very well (Figures 5
and 6). The only deviating point in the Hammett correlation
of K1 (Figure 6) is that of the 4-N(CH3)2 deriva-
tive, which exhibits a rather small extent of direct conjugation
with the reaction centre, and that is why the application of
σ0p constant is more appropriate. The reaction constant ρ ϭ
1.54Ϯ0.04, which is not an unusual value for a nucleophilic
attack at a benzoyl group.[18,19] The Hammett correlation for
the product kc·K2 (the K2 parameter refers to the dissociation
at a distant reaction centre that is not affected by the substitu-
ent) for the four said derivatives leads to ρ ϭ 1.44Ϯ0.08. The
same reaction in water[16] had a value of ρ ϭ 0.67. The given
increase in ρ value can be ascribed to the change in solvent
polarity,[20,21] as the substituent effects are expected to be-
come more pronounced in a solvent of lower polarity.
ever, this procedure cannot be applied to oxygen derivatives 3. Activation Enthalpy and Entropy of the Cyclisation
1aϪd, because the K2 value of the N-methyl derivatives Reactions
1eϪi is inaccessible from kinetic data. However, the
In order to quantify the effect of the methyl group (R1 ϭ
CONH2 group is substantially less acidic than the NH
H, CH3) in the basic skeleton of cyclising thiobenzamides
group adjacent to the benzene ring (the anion formed is
we tried to find the respective thermodynamic activation
parameters. In the case of 2-(benzoylamino)thiobenzamides
Table 3. Parameters of the cyclisation reaction obtained by optimis-
ation of data measured for derivatives 3aϪf
and/or 2-[(benzoyl)(methyl)amino]thiobenzamides we se-
lected the derivatives with R2 ϭ 4-N(CH3)2 [R1 ϭ H (3b),
CH3 (3h)] whose ring closures are measurable within the
whole range of temperatures and concentrations, and we
have determined their thermodynamic activation para-
meters. With both derivatives we carried out the kinetic
measurements at the temperatures of 10, 15, 19, 25, and 30
°C. With the help of Equations (3) and (7) and using the
respective dependences of the observed rate constant on the
sodium methoxide concentration we determined the values
kc and the values K1, K2 for the given temperatures. Then
Compound
(X ϭ S)
K1
kc
pKa1
R2
[L·molϪ1
]
[sϪ1
]
3a
3b
3c
3d
3e
3f
4-H
16.2
7.6
11.6
9.6
24.9
55.5
0.145
0.014
0.087
0.060
0.351
1.663
15.7
16.0
15.8
15.9
15.5
15.2
4-N(CH3)2
4-CH3
4-OCH3
4-Cl
3-NO2
Eur. J. Org. Chem. 2002, 1855Ϫ1863
1859

ˇ
ˇ
ˇ
´
˚
J. Hanusek, M. Sedlak, P. Simunek, V. Sterba
FULL PAPER
tivation energy Ea, activation enthalpy ∆H‡25 °C, Gibbs en-
ergy ∆G‡25 °C, and activation entropy ∆S‡25 °C, which are
summarised in Table 5.
Figure 7. ln kc vs. 1/T for the 2-(benzoylamino)thiobenzamides 3b
and 3h
Figure 4. Observed rate constant of the cyclisation (kobs/sϪ1) of the
2-(benzoylamino)thiobenzamides 3aϪf vs. the sodium methoxide
concentration (c/mol·LϪ1) at 25 °C
Table 5. Thermodynamic activation parameters of the cyclisation
reactions 3b Ǟ 4b and 3h Ǟ 4h
Compound Ea
[kJ·molϪ1
∆H‡
∆S‡
∆G‡
25 °C
25 °C
25 °C
]
[kJ·molϪ1
]
[J·molϪ1·KϪ1
]
[kJ·molϪ1
]
3b
3h
63Ϯ1
41Ϯ2
60Ϯ1
39Ϯ2
Ϫ70Ϯ4
Ϫ126Ϯ7
81Ϯ2
76Ϯ3
Table 5 shows that the ∆G‡
values are similar for the
25 °C
two compounds, but the ∆H‡25 °C and ∆S‡25 °C values differ
considerably. Apparently, the enthalpy and entropy factors
operate against each other, and their effects are partially
cancelled. From the values of the activation entropy of the
two substances it can be concluded that the difference in
the degree of ordering between the starting anion and the
activated complex is much higher for compound 3h than
for compound 3b.
Figure 5. Hammett correlation of the cyclisation rate constant (kc)
of the 2-(benzoylamino)thiobenzamides 3aϪf
The cyclisation of the non-methylated derivatives can be
negatively affected also by intramolecular hydrogen bond
formation, which stabilises the starting species (Figure 8).
The proton in the hydrogen bond is bound in a relatively
rigid six-membered ring, which is transformed into interme-
؊
diate In1 with a relatively low entropy loss. On the other
hand, in the case of the methyl derivative, where the hydro-
gen bond cannot be formed and the system thus has a
higher number of degrees of freedom, the ring closure giv-
؊
ing intermediate In1 is connected with a substantially
larger entropy drop.
Figure 6. Hammett correlation of the equilibrium constant of the
deprotonation (K1) of the 2-(benzoylamino)thiobenzamides 3aϪf
the log kc values were plotted against the reciprocal value of
the thermodynamic temperature (Figure 7). The slope and
intercept of the dependence obtained gave the values of ac- Figure 8. Intramolecular hydrogen bond
1860 Eur. J. Org. Chem. 2002, 1855Ϫ1863

Base-Catalysed Cyclisation of 2-(Substituted benzoylamino)benzamides
FULL PAPER
Table 6. Survey of the synthesised 2-(benzoylamino)benzamides 1fϪi
Compound
R1
R2
M.p. [°C]
Yield: g (%)
Empirical formula
Formula mass
calcd./found
C
H
Cl
N
1f
1g
1h
1i
CH3
CH3
CH3
CH3
4-CH3
4-OCH3
4-Cl
155Ϫ158
1.45 (54)
127Ϫ129
1.1 (45)
147Ϫ148
1.2 (43)
C16H16N2O2
268.31
C16H16N2O3
284.31
C15H13ClN2O2
288.73
C15H13N3O4
299.28
71.62/
71.68
67.59/
67.62
62.40/
62.50
60.20/
60.28
6.01/
6.09
5.67/
5.75
4.54/
4.60
4.38/
4.32
10.44/
10.54
9.85/
9.76
9.70/
9.59
12.28/
12.35
4-NO2
163Ϫ166
1.8 (60)
14.04/
14.01
1
Table 7. H NMR shifts (δ in ppm) of 2-benzoylaminobenzamides 1aϪi
2-H
3-H
4-H
5-H
o-H
m-H
NH2
R1
R2
1a
1b
1c
1d
1e
1f
8.77
m, 1 H
8.76
m, 1 H
8.73
m, 1 H
8.71
m, 1 H
7.53
m, 1 H
7.51
m, 1 H
7.52
m, 1 H
7.51
7.23
m, 1 H
7.23
m, 1 H
7.23
m, 1 H
7.28
m, 1 H
ǟ 7.24Ϫ7.32 Ǟ
m, 2 H
ǟ 7.25Ϫ7.32 Ǟ
m, 2 H
ǟ 7.27Ϫ7.43 Ǟ
m, 2 H
7.64
m, 4 H
7.61
m, 1 H
7.62
m, 1 H
7.65
7.96
m, 1 H
7.95
m, 1 H
7.97
m, 1 H
7.98
m, 2 H
7.08
m, 1 H
7.05
m, 1 H
7.07
m, 1 H
7.14
7.64
m, 4 H
7.42
m, 1 H
7.99
m, 2 H
8.47
m, 2 H
7.46
m, 2 H
6.99
m, 2 H
7.41
m, 2 H
7.45
8.01
m, 2 H
7.89
m, 2 H
7.69
m, 2 H
8.21
m, 2 H
7.19
m, 2 H
7.33
m, 2 H
6.74
m, 2 H
7.27
7.91 ϩ 8.49
2 br. s, 2 H
7.94 ϩ 8.49
2 br. s, 2 H
7.92 ϩ 8.50
2 br. s, 2 H
7.98 ϩ 8.54
2 br. s, 2 H
7.59 ϩ 7.89
2 br. s, 2 H
7.56 ϩ 7.85
2 br. s, 2 H
7.54 ϩ 7.84
2 br. s, 2 H
7.57 ϩ 7.87
2 br. s, 2 H
7.58 ϩ 7.87
2 br. s, 2 H
13.02
br. s, 1 H
12.98
br. s,1 H
13.08
br. s,1H
13.26
br. s, 1 H
3.33
s, 3 H
3.30
s, 3 H
3.30
s, 3 H
3.32
7.64
m, 4 H
2.43
s, 3 H
m, 1 H
7.24-
7.32
2.29
s, 3 H
3.72
s, 3 H
1g
1h
1i
ǟ 7.30Ϫ7.40 Ǟ
m, 2 H
ǟ 7.32Ϫ7.35 Ǟ
m, 2 H
m, 1 H
7.49
m, 1 H
m, 1 H
7.27
m, 1 H
m, 2 H
8.06
m, 2 H
m, 2 H
7.68
m, 2 H
s, 3 H
3.34
s, 3 H
tion range investigated, and the reaction order gradually
decreases from one to zero.
Conclusion
In the ring closure reactions of the oxygen derivatives
1eϪi (X ϭ O), the replacement of a hydrogen atom by a
methyl group at the nitrogen atom adjacent to benzene nuc-
leus (R1 ϭ CH3) leads to a gradual change in the rate-limit-
ing step. At the beginning the reaction is second order in
methoxide and the formation of the intermediate In2؊ is
rate-limiting. However, an increase of the sodium methox-
ide concentration changes the rate-limiting step to the
formation of intermediate In1؊, and the reaction order
gradually decreases. This decrease occurs earlier with deriv-
atives 1hϪi containing electron-withdrawing substituents
(4-Cl and 4-NO2). Replacement of the oxygen by a sulfur
Experimental Section
General Remarks: 2-(Benzoylamino)thiobenzamides 3aϪj and the
corresponding 2-phenylquinazoline-4-thiones 4aϪj were prepared
by the procedure given in ref.[6] 2-(Benzoylamino)benzamides 1aϪi
were prepared by acylating 2-aminobenzamide or 2-(methylamino)-
benzamide with substituted benzoyl chlorides, in contrast to the
method[16] where compounds 1aϪe were obtained by hydrolysis of
2-(benzoylamino)benzonitriles. The yields of acylation reactions
ranged from 40 to 80%, and the melting points agreed with those
in ref.[16] Tables 7, 8, 10, 11 summarise the NMR spectra of com-
atom in the benzamide group (X ϭ S) causes a distinct acid- pounds 1fϪi yet unpublished. The characterisation of compounds
1fϪi so far not described is summed up in Tables 6, 7, and 8. Quin-
azolin-4-ones 2aϪi were also prepared by known procedure,[16] and
the characterisation of newly prepared quinazolin-4-ones 2fϪi is
presented in Tables 9, 10, and 11.
ification of the proton at the terminal nitrogen atom in both
sets of derivatives 3aϪj (R1 ϭ H, CH3) and that is why,
؊
irrespective of ring substitution, the formation of In1 is
rate-limiting within the entire sodium methoxide concentra-
Eur. J. Org. Chem. 2002, 1855Ϫ1863
1861

ˇ
ˇ
ˇ
´
˚
J. Hanusek, M. Sedlak, P. Simunek, V. Sterba
FULL PAPER
Table 8. 13C NMR shifts (δ in ppm) of the 2-(benzoylamino)benzamides 1aϪi [the 13C NMR shifts of carbon atoms with similar values
of chemical shifts (C-2 and C-4) can be interchanged]
C-1
C-2
C-3
C-4
C-5
C-6
CϭO
(a)
CϭO
(b)
C-i
C-o
C-m
C-p
R1
R2
1a
1b
1c
1d
1e
1f
1g
1h
1i
119.3
119.1
119.3
119.4
134.0
133.6
133.9
133.8
134.0
132.2
131.9
132.7
128.9
130.2
129.9
130.4
129.9
130.0
122.8
122.6
122.9
123.3
128.9
128.4
128.7
128.7
128.7
132.7
132.7
133.5
132.7
130.8
130.6
130.6
130.8
130.9
120.2
120.1
120.2
120.3
128.5
128.4
128.7
127.4
127.2
140.2
140.3
140.1
140.2
142.6
142.6
142.8
142.1
141.4
171.4
171.3
171.3
171.2
169.8
169.5
169.1
169.0
168.9
164.6
164.4
163.4
162.6
169.3
169.0
169.1
168.4
167.6
134.8
128.8
128.8
139.8
136.7
133.8
128.5
134.0
142.8
127.1
127.1
128.9
128.5
127.4
127.1
127.0
127.7
129.4
129.1
129.5
129.1
124.1
127.8
128.6
112.9
130.1
122.8
128.9
142.2
137.0
149.4
129.5
138.9
160.0
135.4
147.4
21.1
37.9
37.8
38.0
37.7
37.6
21.0
55.1
Table 9. Survey of the synthesised 2-phenylquinazolin-4-ones 2fϪi
Compound
R1
R2
M.p. [°C]
Yield: g (%)
Empirical formula
Formula mass
calcd./found
Cl
C
H
N
2f
2g
2h
2i
CH3
CH3
CH3
CH3
4-CH3
4-OCH3
4-Cl
170Ϫ172
1.1 (45)
167Ϫ170
1.3 (48)
219Ϫ220
1.6 (69)
276Ϫ278
1.7 (61)
C16H14N2O
250.29
C16H14N2O2
266.29
C15H11ClN2O
270.71
C15H11N3O3
281.27
76.78/
76.82
72.17/
72.20
66.55/
66.60
64.05/
64.12
5.64/
5.70
5.30/
5.34
4.10/
4.12
3.94/
4.01
11.19/
11.25
10.52/
10.58
10.35/
10.40
14.94/
14.88
13.10/
13.16
4-NO2
1
Table 10. H NMR shifts (δ in ppm) of the 2-phenylquinazolin-4-ones 2aϪi
2-H
3-H
4-H
5-H
o-H
m-H
R1
R2
2a
2b
2c
2d
2e
2f
8.20
m, 1 H
8.19
m, 1 H
8.20
m, 1 H
8.23
m, 1 H
8.17
m, 1 H
8.17
m, 1 H
8.15
m, 1 H
8.17
7.52Ϫ7.64
m, 4 H
7.55
m, 1 H
7.59
m, 1 H
7.61
m, 1 H
7.57Ϫ7.63
m, 4 H
7.60
m, 1 H
7.58
m, 1 H
7.61
m, 1 H
7.64
m, 1 H
7.85
m, 1 H
7.87
m, 1 H
7.90
m, 1 H
7.90
m, 1 H
7.91
m, 1 H
7.92
m, 1 H
7.91
m, 1 H
7.93
7.77
m, 1 H
7.77
m, 1 H
7.79
m, 1 H
7.83
m, 1 H
7.78
m, 1 H
7.79
m, 1 H
7.77
m, 1 H
7.80
8.25
m, 2 H
8.14
m, 2 H
8.25
m, 2 H
8.48
7.52Ϫ7.64
m, 4 H
7.39
m, 2 H
7.68
m, 2 H
8.43
m, 2 H
7.73
m, 2 H
7.63
m, 2 H
7.14
m, 2 H
7.68
7.52Ϫ7.64
m, 4 H
12.49 2.43
bs. 1 Hs, 3 H
12.58
s, 1 H
m, 2 H
7.57Ϫ7.63
m, 4 H
7.41
m, 2 H
7.72
m, 2 H
7.78
m, 2 H
8.45
3.68 7.57Ϫ7.63
s, 3 H m, 4 H
3.69 2.45
s, 3 H s, 3 H
3.72 3.89
s, 3 H s, 3 H
3.68
2g
2h
2i
m, 1 H
8.20
m, 1 H
m, 1 H
7.96
m, 1 H
m, 1 H
7.84
m, 1 H
m, 2 H
8.03
m, 2 H
s, 3 H
3.66
s, 3 H
m, 2 H
1862
Eur. J. Org. Chem. 2002, 1855Ϫ1863

Base-Catalysed Cyclisation of 2-(Substituted benzoylamino)benzamides
FULL PAPER
Table 11. 13C NMR shifts (δ in ppm) of the 2-phenylquinazolin-4-ones 2aϪi [the 13C NMR shifts of carbon with similar values of
chemical shifts (C-2 and C-3) can be interchanged]
C-1
C-2
C-3
C-4
C-5
C-6
CϭO
C-7
C-i
C-o
C-m
C-p
R1
R2
2a
2b
2c
2d
2e
2f
2g
2h
2i
121.1
121.0
121.1
121.3
119.8
119.8
119.8
119.9
119.9
127.5
127.4
127.5
127.8
127.1
127.1
127.0
127.2
127.2
126.5
126.4
126.8
127.3
126.0
125.9
125.8
126.2
126.4
134.5
134.6
134.8
134.8
133.9
133.9
133.8
134.1
134.2
125.9
125.9
125.9
126.0
116.8
116.9
116.9
117.0
116.9
148.9
148.8
148.6
150.8
141.8
141.9
142.0
141.8
141.7
162.7
162.4
162.3
162.3
167.7
167.7
167.8
167.7
167.5
152.8
152.4
151.5
151.0
162.1
162.1
160.9
161.2
160.4
133.1
130.0
131.7
138.8
135.3
140.2
127.3
134.3
141.3
128.6
127.8
128.8
129.4
128.8
128.9
131.0
130.8
130.3
127.9
129.3
129.7
123.7
128.5
129.0
113.8
128.6
123.7
131.4
141.5
136.4
149.0
130.3
132.4
161.9
135.2
148.3
21.1
37.9
38.0
38.3
38.0
37.8
21.1
55.5
[4]
[5]
[6]
Measurements: 1H and 13C NMR spectra were measured at 360.14
and 90.57 MHz, respectively, with a Bruker AMX 360 spectrometer
at 25 °C. Compounds 1aϪe and 2aϪe were measured as saturated
solutions in hexadeuteriodimethyl sulfoxide, and the chemical shifts
are referenced to tetramethylsilane [δ(1H) ϭ 0 ppm] and the solvent
signal [δ(13C) ϭ 39.6 ppm]. The CH, CH3, and Cquart groups in the
13C NMR spectra were distinguished by the APT method.
J. R. Cummings, E. H. Stokey, Arzneimittel-Forsch. 1963, 13,
661Ϫ664.
M. Negwer, Organic-Chemical Drugs and Their Synonyms, 7th
revised and enlarged ed., Akademie Verlag, Berlin, 1994.
´
´
´
´
J. Hanusek, L. Hejtmankova, L. Kubicova, M. Sedlak, Mole-
cules 2001, 6, 323Ϫ337.
[7] [7a]
M. G. Pregel, E. J. Dunn, E. Buncel, Can. J. Chem. 1990,
68, 1846Ϫ1858. [7b] K. J. Msayib, C. I. F. Watt, Chem. Soc. Rev.
1992, 21, 237Ϫ243. [7c] L. Dusek, J. Kavalek, V. Sterba, J. Phys.
Org. Chem. 1999, 12, 626Ϫ634.
ˇ
ˇ
´
ˇ
Kinetic Measurements: These were carried out with an HP UV/Vis
8453 Diode Array apparatus. A 1-cm quartz cell was charged with
2 mL of methanolic sodium methoxide. At 25 °C 50 µL of a
methanolic solution of a substrate was injected and the absorbance
was measured at a selected wavelength. In the experiment with 18-
crown-6 ether, the concentration of the latter was always 2·10Ϫ3
mol·LϪ1 higher than that of sodium methoxide. The stability con-
stant[22] for complexation of 18-crown-6 ether with Naϩ at 25 °C
is logK ϭ 4.36.
[8] [8a]
ˇ
ˇ
ˇ
´
L. Dusek, J. Kavalek, V. Sterba, Collect. Czech. Chem. Com-
[8b]
mun. 1999, 64, 265Ϫ287.
D. S. Kwon, H. S. Parks, I. H.
Um, Bull. Korean Chem. Soc. 1991, 12, 93Ϫ97. [8c] D. S. Kwon,
H. S. Parks, I. H. Um, Bull. Korean Chem. Soc. 1994, 15,
654Ϫ658.
[9]
W. P. Jencks, Acc. Chem. Res. 1976, 9, 425Ϫ432.
K. Bowden, K. Bromley, J. Chem. Soc., Perkin Trans. 2 1990,
2103Ϫ2109.
[10]
[11]
[12]
[13]
ˇ
ˇ
´
ˇ
M. Sedlak, J. Hanusek, M. Holcapek, V. Sterba, J. Phys. Org.
General Method for the Preparation of the 2-(Benzoylamino)benz-
amides 1a؊i: A 100-ml three-necked flask, equipped with magnetic
stirrer and dropping funnel, was charged with 10 mmol of benz-
amide derivative, 40 mL of dry acetone and 10 mmol of triethyl-
amine. 10 mmol of benzoyl chloride derivative, dissolved in 10 mL
of acetone, was added dropwise during ca. 5 min. The reaction mix-
ture was stirred for 2 h at room temperature. The separated triethyl-
ammonium chloride (TEA·HCl) together with a part of the product
was separated by filtration through a sintered glass filter, and
TEA·HCl was removed by washing with water to obtain the first
portion of product. The mother liquor was concentrated at room
temperature, and the residue was recrystallised from toluene or
benzene to provide the second portion of product.
Chem. 2001, 14, 187Ϫ195.
ˇ
´
ˇ
J. Kavalek, V. Sterba, S. El Bahaie, Collect. Czech. Chem. Com-
mun. 1982, 48, 1430Ϫ1434.
M. Eigen, Angew. Chem. Int. Ed. Engl. 1964, 3, 1Ϫ19.
[14] [14a]
C. H. Rochester, Acidity Functions, Academic Press, Lon-
don, 1970. [14b]A. Bagno, G. Scorrano, F. Terrier, J. Chem. Soc.,
Perkin Trans. 2 1990, 1017Ϫ1027. [14c] J. Kavalek, F. Krampera,
´
ˇ
ˇ
V. S terba, Collect. Czech. Chem. Commun. 1976, 41,
1685Ϫ1691.
[15]
[16]
C. H. Rochester, B. Rossal, J. Chem. Soc. B 1967, 743Ϫ748.
B. Gardner, A. J. S. Kanagasooriam, R. M. Smyth, A. Willi-
ams, J. Org. Chem. 1994, 59, 6245Ϫ6250.
[17] [17a]
ˇ
ˇ
´
´ˇ
J. Kavalek, J. Jirman, V. Machacek, V. Sterba, Coll. Czech.
Chem. Commun. 1987, 52, 1992Ϫ1998. [17b] M. Sedlak, J. Kava-
´
´
ˇ
´ˇ
ˇ
lek, V. Machacek, V. Sterba, Molecules 1996, 1, 170Ϫ174.
S. L. Shames, L. D. Byers, J. Am. Chem. Soc. 1981, 103,
6170Ϫ6177.
[18]
[19]
[20]
[21]
[22]
Acknowledgments
J. F. Kirsch, W. Clewell, A. Simon, J. Org. Chem. 1968, 33,
127Ϫ132.
O. Exner, Correlation Analysis of Chemical Data, Plenum Press,
New York, 1988.
W. Walter, R. F. Becker:, Justus Liebigs Ann. Chem. 1969,
727, 71Ϫ80.
O. A. Raevsky, V. P. Solovev, A. F. Solotnov, H.-J. Schneider,
V. Rüdiger, J. Org. Chem. 1996, 61, 8113Ϫ8116.
Received September 12, 2001
The authors acknowledge the financial support provided by the
Grant Agency of the Czech Republic (Grant No. 203/01/0227).
[1]
A. H. Amin, D. R. Mehta, S. S. Samarth, Fortschr. Arzneimit-
telforsch. 1970, 14, 218Ϫ268.
M. L. Gujral, P. N. Saxena, R. S. Tiwari, Indian J. Med. Res.
[2]
1955, 43, 637Ϫ641.
J. N. Leonard, W. V. Ruyle, J. Org. Chem. 1948, 13, 903Ϫ903.
[3]
[O01438]
Eur. J. Org. Chem. 2002, 1855Ϫ1863
1863