The base-catalysed cyclisation of phenyl N-(2-hydroxybenzyl)-N-methylcarbamates is concerted

Article abstract of DOI:10.1039/b209323p

The kinetics of the cyclisation in aqueous solution of phenyl-(2-hydroxybenzyl) -N-methylcarbamates to 3-methyl-3,4-dihydrobenzol[e][1,3]oxazin-2-ones and phenolate ions fit the rate law: k_obs = k_c/(1 + [H₃O⁺]/ k_a) The values of k_c and pK_a fit Broonsted equations agains the pK_a's of the corresponding free phenols but the system does not conform to the reactivity-selectiviy hypothesis. The values of the Broonsted parameters β_Y and βX vary as a function of Y and X according to the equations: β_X = -0.179pK_a^HY + 0.87 β_Y = -0.179pK_a^HX + 2.30 The magnitude and sign of the Cordes-Thornton cross-interaction coefficient p_XY(-0.179) rule out a stepwise mechanism involving a tetrahedral intermediate and is consistent with a concerted displacement mechanism. A similar concerted mechanism is proposed for the base-catalysed cyclisation of phenyl-N-(2-hydroxyphenyl)-N-methylcarbamate esters to benzoxazol-2-ones.

Full text of DOI:10.1039/b209323p

View Article Online / Journal Homepage / Table of Contents for this issue
The base-catalysed cyclisation of phenyl N-(2-hydroxybenzyl)-N-
methylcarbamates is concerted
a
a
a
a
ˇ
Vojeslav Steˇrba,* Oldrˇich Hrabík, Jaromír Kaválek, Jaromír Mindl and Andrew Williams
a Department of Organic Chemistry, University of Pardubice, 532 10 Pardubice, Czech Republic.
E-mail: vojeslav.sterba@upce.cz. E-mail: jaromir.kavalek@upce.cz.
E-mail: jaromir.mindl@upce.cz
b University Chemical Laboratory, Canterbury, Kent, UK CT2 7NH. E-mail: aw@ukc.ac.uk
Received 23rd September 2002, Accepted 23rd October 2002
First published as an Advance Article on the web 11th December 2002
The kinetics of the cyclisation in aqueous solution of phenyl-(2-hydroxybenzyl)-N-methylcarbamates to 3-methyl-
3,4-dihydrobenzo[e][1,3]oxazin-2-ones and phenolate ions fit the rate law:
kobs = kc/(1 ϩ [H3Oϩ]/Ka)
The values of kc and pKa fit Brønsted equations against the pKa’s of the corresponding free phenols but the system
does not conform to the reactivity–selectivity hypothesis. The values of the Brønsted parameters βY and βX vary as a
function of Y and X according to the equations:
βX = Ϫ0.179pKaHY ϩ 0.87
βY = Ϫ0.179pKaHX ϩ 2.30
The magnitude and sign of the Cordes–Thornton cross-interaction coefficient pXY (Ϫ0.179) rule out a stepwise
mechanism involving a tetrahedral intermediate and is consistent with a concerted displacement mechanism. A
similar concerted mechanism is proposed for the base-catalysed cyclisation of phenyl-N-(2-hydroxyphenyl)-N-
methylcarbamate esters to benzoxazol-2-ones.
Introduction
Experimental
The study of intramolecular reactions provides a useful tool to
investigate the mechanisms of fundamental reactions because
the structure and geometry of the interaction between func-
tional groups in intramolecular reactions can be defined more
precisely than in the corresponding intermolecular processes.
Intramolecular reactions, especially those involving nucleo-
philic displacements, are usually faster than the corresponding
intermolecular ones by many orders of magnitude;1 because of
this it is possible to study bimolecular processes which would
otherwise not be possible due to their poor reactivity.
Materials
Buffer components were of analytical reagent grade and were
used without further purification. Both NaOH and buffer
solutions were prepared with water distilled from glass and free
of CO2. All solutions for kinetics were degassed before use.
Compounds (Ia–v) and (IIa–e) (Table 1 and Scheme 1) were
from previous studies.5
Carbamate esters possess acyl groups highly deactivated by
resonance with the adjoining nitrogen; hydroxide ion reacts
with 4-nitrophenyl-N-methyl-N-phenylcarbamate some 2×104
times more slowly than with 4-nitrophenyl acetate.2,3 A fast
intramolecular reaction is therefore an ideal vehicle for the
study of the carbamoyl group transfer process.4 The rapid
intramolecular reaction would not be complicated by competi-
tive intermolecular reaction with hydroxide ion. The present
study exploits the cyclisation of X-(substituted)phenyl-N-
[2-hydroxy-Y-(substituted)benzyl]-N-methylcarbamate esters
(Ia–v) in alkaline solution followed by hydrolysis of the
intermediate benzoxazinone (II) to give (III).
The cyclisation reaction has the advantage that the charge on
both forming and breaking bonds can be studied by variation
of Y or X respectively. The methylene group between benzene
nucleus and the nitrogen reduces the influence of the substitu-
ents Y on the electrophilic reactivity of the carbamate group
(via the nitrogen atom) which would otherwise complicate
the interpretation of substituent effects on the displacement
reaction.
Scheme 1 (see Table 1 for the identities of the substrates).
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
415

View Article Online
Table 1 Reactivity of I and II as a function of X and Y
f
X i
Y i
X
Y
103k/sϪ1
pKa
Ϫlogkobs
pHag
λ/nmk
Nh
pKa
pKa
Ia
Ib
Ic
Id
Ie
If
Ig
Ih
Ii
H
4-Cl
3-Cl
3-NO2
4-NO2
H
4-Cl
3-Cl
H
H
H
H
1.26
6.31
15.8
70.8
275
9.98
9.90
9.86
9.78
9.70
9.46
9.40
9.39
9.26
9.02
9.17
9.05
8.91
8.76
8.42
8.23
8.00
7.36
7.31
7.24
7.16
7.01
2.82–3.92c
2.15–3.10
1.79–3.22
1.12–2.45
0.53–2.47
3.32–4.18
2.93–3.77
2.48–3.43
1.67–5.51
1.02–2.37
3.57–4.28
2.87–3.37
2.08–2.53
1.36–2.22
3.92–4.50
2.66–3.45
1.87–2.51
4.23–4.63
4.08–4.46
3.94–4.54
3.76–4.20
3.23–3.62
8.49–13.30bd
8.62–11.70
8.32–12.39b
8.32–12.87b
7.53–12.87b
8.49–11.70
8.32–11.70
8.32–11.70
7.53–11.62
7.53–11.13
8.32–11.13
8.32–11.13
8.32–11.13
7.53–11.13
6.98–11.13
6.98–10.05
7.25–10.05
6.70–10.05
6.70–10.05
6.50–10.05
6.52–10.05
6.70–10.30
297
294
294
393
400
301
303
302
395
401
296
275
244
401
400
262
401
414
412
416
409
406
10
10
8
10
9
9
10
9
9
9
8
8
8
9
8
9
8
8
8
8
9.99
9.38
9.02
8.35
7.14
9.99
9.38
9.02
8.35
7.14
9.99
9.02
8.35
7.14
9.99
8.35
7.14
9.99
9.38
9.02
8.35
7.14
9.99
9.99
9.99
9.99
9.99
9.34
9.34
9.34
9.34
9.34
9.02
9.02
9.02
9.02
8.35
8.35
8.35
7.14
7.14
7.14
7.14
7.14
H
5-Br
5-Br
5-Br
5-Br
5-Br
4-Cl
4-Cl
4-Cl
4-Cl
4-NO2
4-NO2
4-NO2
5-NO2
5-NO2
5-NO2
5-NO2
5-NO2
0.40
1.26
3.16
19.9
89.1
0.25
1.26
7.94
39.8
0.126
2.24
12.6
0.063e
0.080e
0.11e
0.158
0.56
3-NO2
4-NO2
H
Ij
Ik
Il
3-Cl
Im
In
Io
Ip
Iq
Ir
Is
It
3-NO2
4-NO2
H
3-NO2
4-NO2
H
4-Cl
3-Cl
3-NO2
4-NO2
Iu
Iv
8
8
ArOH j
kOH/MϪ1 sϪ1
0.033
0.055
0.035
0.125
pKa
9.99
9.34
9.02
8.35
7.14
IIa
IIb
IIc
IId
IIe
H
290
301
297
392
400
6-Br
7-Cl
7-NO2
6-NO2
0.235
a Buffers employed in the pH-ranges were: 0.05 M HPO42Ϫ, 0.025 M borate, 0.1 M 2-amino-2-(hydroxymethyl)propane-1,3-diol buffer and 0.05 M
Ϫ
H2PO4
.
b NaOH solutions employed for the higher pH’s. c Values of rate constants in NaOH solution were calculated from data using the
consecutive two step equation. d Values of pH in NaOH were calculated from log[HOϪ]. e Extrapolated from rate constants measured in the range
47–67 ЊC. f Range of observed rate constants. g Range of pH values employed in the kinetics. h Number of data points. i pKa of the phenol
corresponding to the substituent X or Y. j pKa of the phenol corresponding to the leaving group in (II). k Wavelength for kinetic studies.
Methods
shown in Table 1 except in the case where the leaving group is
phenolate or 4-chlorophenolate ion (see later). The rate con-
stants are independent of the nature and concentration of the
buffer components (excepting the concentration of HOϪ). The
benzoxazinones (IIa–e) decompose slowly in NaOH solutions
to give the carbamate salts (III) (Scheme 1). The hydrolysis of
(II) did not interfere with the determination of the cyclisation
rate constant except for X = H or 4-Cl when in some cases a
biphasic progress curve was observed diagnostic of a two-step
Ultraviolet-visible spectra were scanned with a Hewlett-
Packard 8453 Diode Array Spectrometer. Buffer pH values
were determined with an MV 870 Apparatus (VEB Pröcitronic)
and a combined glass and silver electrode calibrated with
standard aqueous buffers. The ionisation constants of esters (I)
with reactivities less than 5×10Ϫ3 sϪ1 were determined spectro-
scopically6 in phosphate, borate and carbonate buffers.
Kinetics were followed by injecting solutions of substrate
(20 µL) in methanol into 2.5 mL buffer or NaOH solution in 1
cm silica cuvettes in the thermostatted cell compartment of
the spectrometer. Initial substrate concentrations were between
2×10Ϫ5 M and 3×10Ϫ5 M in the cuvette . The spectrum was
scanned repetitively to obtain the optimum wavelength for
kinetic study. Absorbance at constant wavelength was meas-
ured as a function of time and was analysed according to a first
order rate law (At = A∞ Ϫ (A∞ Ϫ A0)exp(Ϫkt)). Data were fitted
to theoretical equations by use of standard software.
mechanism (Fig. 1). In those cases the rate constants for I
and II III were obtained from the biexponential equation:
II
Product analysis was carried out by comparing the UV
spectra (after ca. 5–7 half lives) with the spectra of authentic
samples (II) under the same conditions.
Results
Product analysis showed that the reaction involves formation of
intermediate benzoxazinones (II) followed by their slow
hydrolysis to (III) as in Scheme 1. In cases where the intermedi-
ate decomposes to final acid product at rates commensurate
with its formation the rate of its decomposition was demon-
strated to be the same as that of an authentic sample of the
benzoxazinone (II) under the same conditions of pH.
Fig. 1 Progress graph at single wavelength (297 nm) showing
formation and decay of the intermediate (IIa) at pH 13.34 from Ia.
(example shown is calculated from parameters in Table 1 and assuming
A0 is 0.146, A∞ is 0.217, AB is Ϫ0.0549, k1 = 0.00126 sϪ1 and k2 = 0.0134
The kinetics of cyclisation of carbamates (Ia–v) to (II) exhib-
ited excellent pseudo-first-order kinetics over the ranges of pH
s
Ϫ1).
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
416

View Article Online
Table 2 Brønsted and Leffler parameters for the cyclisation of I.a
b
Y d
βX(Lg)
αX
logkcH
rc
Y
pKa
Ϫ0.82 0.10
Ϫ0.85 0.08
Ϫ0.79 0.07
Ϫ0.705 0.040 0.39
Ϫ0.34 0.03
0.46
0.47
0.44
5.34 0.84
0.9801
H
9.99
9.34
9.02
8.35
7.14
5.15 0.69
4.32 0.61
3.17 0.34
0.9876 5-Br
0.9923 4-Cl
0.9984 4-NO2
0.19 Ϫ0.908 0.30 0.9850 5-NO2
b
Y d
βY(Nuc)
αY
logkcH
rc
X
H
pKa
0.44 0.07
0.635 0.10
0.723 0.10
0.932 0.02
0.945 0.05
0.24
0.35
0.40
Ϫ7.46 0.61 0.9645
Ϫ8.67 0.88 0.9882 4-Cl
Ϫ9.21 0.89 0.9816 3-Cl
9.99
9.38
9.02
8.35
7.14
0.52 Ϫ10.45 0.18 0.9993 3-NO2
0.53
Ϫ9.92 0.42 0.9963 4-NO2
ArOH
a Parameters are for logkc = βpKa
ϩ logkcH b α = β/βeq (see text).
.
c r = correlation coefficient. d pKa of the appropriately substituted
phenol.
At = A∞ ϩ {[(k2 Ϫ k1)A0 ϩ k1(AB Ϫ A∞)]/(k2 Ϫ k1)}exp(Ϫk1t) Ϫ
[k1(AB Ϫ A∞)/(k2 Ϫ k1)]exp(Ϫk2t).7
The pH-dependence of the cyclisation reaction is illustrated in
Fig. 2 for 1a–e. The cyclisation rate constants fit a rate eqn. (1)
and the values of kc and pKa for (Ia–v) are recorded in Table 1.
Fig. 3 Brønsted dependences of kc on the pKa of (A) X-substituted
phenols (Y = constant) and (B) Y-substituted phenols (X = constant);
data points are from Table 1 and lines are calculated from parameters in
Table 2.
Fig. 2 pH-Dependence of the rate constants of cyclisation Ia–e and of
hydrolysis of IIa. Lines are theoretical from parameters in Table 1 and
eqn. (1) for Ia–e and a bimolecular equation for IIa. (Example shown is
calculated from parameters in Table 1 and assuming A0 is 0.146, A∞ is
0.217, AB is Ϫ0.0549, k1 = 0.00126 sϪ1 and k2 = 0.0134 secϪ1).
kobs = kc/(1 ϩ [H3Oϩ]/Ka)
(1)
The kinetics of hydrolysis of benzoxazinones (II) has pseudo-
first-order rate constants linear up to 0.5M concentrations of
HOϪ (see Fig. 2) and the corresponding kOH values, evaluated
from the linear relationship, are recorded in Table 1. The rate
constants, kc, fit Brønsted relationships illustrated in Fig. 3.
The values of the Brønsted parameters, βX and βY recorded in
Table 2, are dependent on Y and X respectively (Fig. 4) accord-
ing to the eqns. (2) and (3). The Cordes–Thornton coefficient
(pXY),8 is Ϫ0.179 0.03.
Nuc
Lg
Fig. 4 Dependence of βX and βY on pKa
and pKa respectively;
data are from Table 2 and the lines are calculated from eqns. (2) and (3).
pKaY,NO2 = 0.948 0.034pKaArOH(Y) ϩ
0.19 0.30 (r = 0.9981) (5)
The rate constants for the alkaline hydrolysis of the benz-
oxazinones (II) obey a Brønsted eqn. (4) as do the pKa values
for the ionisation of the carbamate esters (I) (Ia–e, Eqn. 5) and
(Ie, Ij, In, Iq and Iv, Eqn. 6). The superscripts Y and X refer to
the substituents in the two rings as indicated in Scheme 1.
βX = pXYpKaY ϩ 0.87 0.25
βY = pXYpKaX ϩ 2.30 0.26
(2)
(3)
logkOH = Ϫ0.318 0.072pKaArOH ϩ
1.64 0.62 (r = 0.9316) (4)
pKaH,X = 0.098 0.009pKaArOH(X) ϩ
8.985 0.078 (r = 0.9879) (6)
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
417

View Article Online
Y,NO2
The Brønsted β value for the effect of changing Y on pKa
In addition to reporting the extent of bond change the
effect of substituents Y and X can also monitor the inter-
action between these bond changes. This property is termed the
cross-interaction effect and is alternately the effect of X on
is 0.948 close to unity as expected for substituents directly
attached to the nucleus nucleus of the ionising phenol. The
H,X
value of β (0.098) for the effect of changing X on pKa
is
X
Y
consistent with the substituent change being insulated from the
ionising phenol hydroxyl group. It is possible that the residual
effect of X is due to some sort of field interaction because
inductive transmission from the X-substituted phenyl through
six atoms is likely to be vanishingly small even though there
is a possibility of π-electron polarisation via the benzene and
carbamate ester functions.
βY or Y on βX. The parameter pXY = ∂βY/∂pKa = ∂βX/∂pKa
X
Y
where pKa and pKa refer to the phenols corresponding
to X and Y. The existence of a substantial cross-interaction
effect in a displacement reaction is evidence of a concerted
process.9
In the present case the variations of βY and βX with X and
Y respectively predict a substantial cross-interaction effect
(Ϫ0.179) which is prima facie evidence of a concerted process
(Scheme 4). By considering the direction of the observed
changes in βX and β (Table 2) as a function of cross-substituent
X or Y it is possible to exclude the stepwise process of Scheme
Discussion
The pH-dependence and the observation of an intermediate in
the hydrolysis of the carbamate esters (I) are consistent with a
mechanism involving ionisation to the conjugate base followed
by its rate limiting breakdown (kc) to benzoxazinone (II) and a
slower decomposition of II to III (Scheme 2).
Y
3. According to Scheme 3 increasing the value of pKa will
decrease kϪ1 while leaving k2 relatively unchanged. This should
alter the rate limiting step from k2 to k1 so that k2 (which is
X-dependent) becomes less predominant in kc. Thus βX should
become numerically smaller in contrast to the experimentally
observed increase in its numerical magnitude (Ϫ0.34
Ϫ0.82,
X
see Table 2). When pKa increases, the rate constant k2 should
decrease, whereas kϪ1 should be relatively unchanged under
these conditions; in this case the rate limiting step should
change from k1 to k2 so that bond formation (which is moni-
tored by βY) is fully complete in the transition structure of the
rate limiting step. Thus βY should increase towards unity
contrary to the experimentally observed decrease (0.945
0.44, see Table 2).
Scheme 2
The changes in βY as a function of change in X are not con-
sistent with the More O’Ferrall–Jencks map for the stepwise
mechanism. The k1 step would be rate limiting in the putative
stepwise mechanism because k2 is likely to be fast due to the
Cross-interaction effects
The formation of benzoxazinone from the conjugate base of I
could be either stepwise (Scheme 3) or concerted (Scheme 4).
X
stability of the cyclic intermediate (II). Thus an increase in pKa
should increase βY and shift the position of the transition struc-
ture South to North in Fig. 5 (ie according to the Hammond
Scheme 3 Stepwise mechanism for the kc step.
Fig. 5 More O’Ferrall–Jencks diagram for the putative stepwise
cyclisation mechanism (Scheme 3).
postulate). An increase in pKaY should not substantially change
bond fission so that βX should remain constant (and relatively
small); the position of the transition structure would only suffer
a small West–East change as illustrated in Fig. 5. Another way
of expressing this argument is as follows: when the k1 step is
rate limiting and the transition structure is close to the western
edge of the diagram the shape of the surface will constrain any
movement of the transition structure largely in a North–South
direction and be essentially controlled by Hammond’s
postulate.10
Scheme 4 Concerted mechanism for the kc step.
The substituent effect βY or βX is a measure of the extent of
bond formation or fission respectively in the transition struc-
ture of the rate limiting step.9 Thus the effect of Y on the cyclis-
ation rate constant refers to bond formation between oxyanion
and carbamyl carbon. The effect of changing X refers to bond
fission between leaving oxygen and the carbamyl carbon.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
418

View Article Online
Scheme 5 Effective charge map for concerted cyclisation of Iv (NO2,NO2) and Ia. (H,H). Numbers in brackets are values of effective charge for Ia.
Numbers on reactants and products refer to both Iv and Ia.
The sign and magnitude of the observed changes in βX and
Y
X
8
βY as a function of pKa and pKa respectively (pXY
) can be
accomodated by a concerted mechanism (Scheme 4).11 This
conclusion is in agreement with concerted displacement mech-
anisms discovered for other acyl group transfer reactions12,13
where the concerted mechanism of acyl group transfer is
favoured by weakly basic nucleophiles and leaving groups such
as phenolate ions or pyridines.14 Decreased stability of the
tetrahedral intermediate of the putative stepwise mechanism
is also important in favouring a concerted process.13 In the
carbamate case the greater electron donating power of the
nitrogen would destabilise the putative tetrahedral adduct (IV)
compared to that of the methyl group (V). Since the reaction of
phenolate ions with phenyl acetates is concerted it follows that
the phenolate ion displacements at the carbamoyl function
should also be concerted because the structural factors are even
more favourable for a concerted process.
Fig. 6 More O’Ferrall–Jencks diagram for the concerted cyclisation
reaction I II illustrating the movement of the position of the
transition structure as a function of increasing electron donating power
of X and Y.
The consequences of variation of X and Y on the position of
the transition structure in the More O’Ferrall–Jencks dia-
gram13,15 are illustrated in Fig. 6. It is convenient at this stage to
An effective charge map17 can be constructed for the cyclis-
ation I
II and the effective charge distribution in the
X
use normalised β values (Leffler parameters αX = βX/βeq and
transition structure will vary according to the identity of the
substituents X and Y. The map is illustrated for the cyclisation
of Iv (4-NO2, 5-NO2) and Ia (H,H) in Scheme 5.
αY = βY/βeqY) as these are used conventionally as x and y
coordinates for the diagrams (Fig. 6). We make the reasonable
X
Y
assumption that the parameters βeq (Ϫ1.8) and βeq (ϩ1.8)
are the same as that for transfer of the NH2CO– group between
phenolate ion nucleophiles.16 The derived values of αX and αY
are recorded in Table 2.
In the case of Iv the change in effective charge on the
carbonyl oxygen, obtained by difference assuming additivity of
effective charge, is (Ϫ1.405 Ϫ Ϫ0.8 = Ϫ0.605); together with the
changes in charge on nucleophilic and leaving oxygen (ϩ0.945)
and (Ϫ0.34) respectively it is deduced that the transition struc-
ture has some of the character of a tetrahedral adduct. In the
case of Ia the slight increase in positive effective charge on the
carbonyl oxygen (Ϫ0.42 Ϫ Ϫ0.8 = ϩ0.38) is consistent with a
transition structure with some acylium ion character.
The hydrolysis of the conjugate base of phenyl-N-(2-
hydroxyphenyl)-N-methylcarbamate (VI), which involves cyclis-
ation to 3-methyl-2,3-dihydrobenzo[1,3]oxazol-2-one (VII)
(Scheme 6), is analogous to that of the present study.18 The
cyclisation rate constant is almost three orders of magnitude
larger than that of Ia in agreement with the observation that
cyclisation by intramolecular reaction of oxygen nucleophiles
with a carbonyl group to five-membered rings is invariably
faster than the formation of six-membered rings.1,19–23 The
Expressed in terms of Leffler α parameters the x,y coordin-
ates of the transition structure for 1v are (0.19,0.53) and for 1a
are (0.46,0.24). These coordinates represent two extremes of
structure and the transition structures for all the esters (1a–v)
studied here lie within the envelope bounded by the two
reaction coordinates illustrated in Fig. 6. Changing the Y-sub-
stituent in I (for example Ir
a West–East direction (with no North–South movement)
whereas change in the X-substituent (for example Iv Ir)
Ie) alters the coordinates in
moves the transition structure in a North–South direction with
no West–East movement.
Since the putative stepwise mechanism has been ruled out the
large changes in transition structure can only be due to changes
on an energy surface corresponding to a concerted mechanism.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
419

View Article Online
hydrolysis of the oxazolone (VII) (kOH = 4.8.10Ϫ3MϪ1secϪ1) is
ten times smaller than that of Ia. The cyclisation reactions of
HO–CH2CH2NPhCO2Ph and of HO–CH2CH2NCH3CO2Ph24
are equally reactive indicating that the replacement of N–CH2–
by NPh is unlikely to be the cause of the enhanced rate of
cyclisation to benzoxalone (VII) compared with that to ben-
zoxazinones (II).
The effective molarity for the reaction of Scheme 6 (Y = X =
H) is ca. 108 M compared with the effective molarity for the
cyclisation to benzoxazinone (IIa from Ia) of 2.105M using the
calculated intermolecular rate constant of Hutchins and Fife.18
The substantial value of the Hammett ρ obtained for Scheme 6
when Y = H3 translates to a βLg of Ϫ0.9 and αLg = 0.53 consist-
ent with advanced fission of the ArO–C bond in the transition
structure of the rate limiting step. Moreover Hegarty et al.25
showed that bond formation is substantially advanced in the
transition structure. The results fit both stepwise and concerted
processes. We propose a concerted displacement mechanism for
the formation of the benzoxazolone (VIII) on the basis of the
present data for the benzoxazinone formation.
ions and substituted phenyl acetate esters has a positive
pXY coefficient (ϩ0.17) and the structure reactivity variation
conforms with the reactivity–selectivity hypothesis. There is no
cogent reason why the reactivity–selectivity hypothesis should
be adhered to except in a reaction where only one bond is
undergoing a major change and there are no concurrent
changes such as solvation or resonance interactions.10 The
results for the benzoxazinone formation contrast with that for
the phenolysis of phenylacetate esters where increasing the
nucleophilicity of nucleophile and leaving group moves the
transition structure towards that of the putative tetrahedral
structure (0,1); the same change in nucleophilicity for the
cyclisation I
II shifts the structure towards the putative
acylium ion structure (1,0).30
It is interesting to speculate on the reason why electron
donating substituents cause the motion of the transition struc-
ture towards the acylium ion structure (1,0). In this case the
acylium ion structure is more stable than that of the acetylium
ion (CH3COϩ) in phenolysis of phenyl acetate esters due to the
adjacent nitrogen. The More O’Ferrall–Jencks map is therefore
more skewed towards the (1,0) coordinate in the benzoxazinone
formation. It is unlikely that the acylium ion structure would be
stabilised by electron donating substituents X or Y. However,
these substituents would decrease the stability of the putative
adduct at (0,1) and thus move the transition structure in a
South–East direction.
Conclusions
This study shows that the base-catalysed cyclisation (Scheme 1)
does not conform to the reactivity–selectivity hypothesis and
involves a concerted displacement reaction at the carbamoyl
centre. Arguments show that a similar concerted mechanism
probably operates for the formation of the benzoxazolone
(Scheme 6). The results add to the substantial body of evidence
that a concerted displacement mechanism is possible at a
carbonyl centre under certain conditions of nucleophile, leaving
group and acyl function.
The kinetics of hydrolysis of esters (I),26 where a nitro group
is substituted para to an hydroxyl function, show enhanced
resonance interaction due to extensive bond formation and
fission in the transition structure. The data indicate that
rehybridisation and bond formation or fission are not syn-
chronous.27 Studies with more than one group capable of
undergoing resonance interactions will be necessary before
further comment is possible on the relative progress of delocal-
isation and bond formation in this reaction.
Acknowledgements
We are grateful to a referee for suggesting this collaboration
between the English and Czech laboratories. The work was
carried out with the support of the Ministry of Education of
the Czech Republic, project CI MSM 2531001.
References and notes
1 A. J. Kirby, Adv. Phys. Org. Chem., 1980, 17, 183.
2 T. H. Fife, J. E. C. Hutchins and M. S. Wang, J. Am. Chem. Soc.,
1975, 97, 5878.
3 A. F. Hegarty and L. N. Frost, J. Chem. Soc., Perkin Trans. 2, 1973,
1719.
4 T. H. Fife and J. E. C. Hutchins, J. Am. Chem. Soc., 1981, 103, 4194.
5 (a) J. Mindl, O. Hrabik, V. Sterba and J. Kavalek, Coll. Czech. Chem.
Commun., 2000, 65, 1262; (b) O. Hrabik, P. Simunek, J. Mindl and
J. Kavalek, Molecules, 2002, 7, 200.
6 A. Albert, The Determination of Ionization Constants: A Practical
Manual, Chapman Hall, London, 2nd edn., 1971.
7 k1 and k2 are the velocity constants of the first and second step of the
reaction, A∞, A0 are the absorbances at the end and at the beginning
of the reaction, AB is the absorbance that characterizes the
intermediate. (a) W. G. Jackson, J. M. Harrowfield and P. D. Vowles,
Int. J. Chem. Kinet., 1977, 9, 535; (b) O. Pytela: Chemometrie,
University Press, Pardubice, 2000, p. 108.
Scheme 6
8 D. A. Jencks and W. P. Jencks, J. Am. Chem. Soc., 1977, 99, 7948.
9 A. Williams, Concerted Organic and Bio-Organic Mechanisms,
CRC Press, Boca Raton, FL, 2000.
10 G. S. Hammond, J. Am. Chem. Soc., 1955, 77, 334.
11 (a) I. Lee, Chem. Soc. Rev., 1990, 19, 317; (b) I. Lee, Adv. Phys. Org.
Chem., 1992, 27, 57.
12 (a) S. A. Ba-Saif, K. Luthra and A. Williams, J. Am. Chem. Soc.,
1989, 111, 2647; (b) S. A. Ba-Saif, K. Luthra and A. Williams,
J. Am. Chem. Soc., 1987, 109, 6362; (c) S. A. Ba-Saif, M. Colthurst,
Reactivity–selectivity hypothesis
The increase in reactivity as a function of X or Y is associated
with an increase in selectivity (Fig. 3) and this is manifested in
the sign of the Cordes–Thornton coefficient, Ϫ0.179).8,11,28 The
variation of reactivity with structure does not conform with the
reactivity–selectivity hypothesis.10,29 In contrast the inter-
molecular displacement reaction between substituted phenolate
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
420

View Article Online
M. A. Waring and A. Williams, J. Chem. Soc., Perkin Trans. 2, 1991,
1901; (d ) A. C. Hengge and R. A. Hess, J. Am. Chem. Soc., 1994,
116, 11256; (e) J. P. Guthrie, J. Am. Chem. Soc., 1991, 113, 3941.
13 A. Williams, Acc. Chem. Res., 1989, 22, 387.
14 A. K. Luthra, S. A. Ba-Saif, E. Chrystiuk and A. Williams,
Bull. Soc. Chim. France, 1988, 391.
23 T. C. Bruice and F. C. Lightstone, Acc. Chem. Res., 1999, 32, 127.
24 O. Hrabik, J. Mindl, M. Kavalek and V. Sterba, Chem. Listy, 2000,
94, 1049.
25 A. F. Hegarty, L. N. Frost and D. Cremin, J. Chem. Soc., Perkin
Trans. 2, 1974, 1249.
26 Ir, Is, It, Iu, Iv and Ie, Ij, In and Iq.
15 M. I. Page & A. Williams, Organic and Bio-Organic Mechanisms,
Addison Wesley Longman, Harlow, 1997, 118–119.
16 H. Al-Rawi and A. Williams, J. Am. Chem. Soc., 1977, 99, 2671.
17 A. Williams, Acc. Chem. Res., 1984, 17, 425.
18 J. E. C. Hutchins and T. H. Fife, J. Am. Chem. Soc., 1973, 95, 2282.
19 F. C. Lightstone and T. C. Bruice, J. Am. Chem. Soc., 1996, 118,
2595.
27 (a) C. F. Bernasconi, Acc. Chem. Res., 1987, 20, 301; (b) C. F.
Bernasconi, Adv. Phys. Org. Chem., 1992, 27, 119.
28 W. P. Jencks, Chem. Rev., 1985, 85, 511.
29 (a) E. Buncel, H. Wilson and C. Chuaqui, J. Am. Chem. Soc., 1982,
104, 4896; (b) E. Buncel, J. Chem. Educ., 1987, 64, 475.
30 The More O’Ferrall–Jencks coordinates obtained from parameters
in reference 12 for reaction of Y-substituted phenolate ions with
X-substituted phenyl acetate esters are: X,Y, (αX,αY): NO2,NO2,
(0.39,0.44); H,H (0.19,0.76); H,NO2, (0.76,0.39); NO2,H,
(0.44,0.19). Thus changing from X,Y = NO2,NO2 to H,H moves the
coordinates from (0.39,0.44) to (0.19,0.76).
20 F. C. Lightstone and T. C. Bruice, J. Am. Chem. Soc., 1997, 119,
9103.
21 F. M. Menger, Acc. Chem. Res., 1985, 18, 128.
22 M. I. Page, Philos. Trans. R. Soc. London, Ser. B, 1991, 332, 149.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 1 5 – 4 2 1
421