Formation of a cyclic tetrahedral intermediate by the addition of water to 2-methyl-4H-3,1-benzoxazine followed by ring opening to 2-aminobenzyl acetate and 2-acetylaminobenzyl alcohol; pH-dependence of rate of reaction and product ratio

Article abstract of DOI:10.1139/v99-091

Kinetic studies have shown that addition of water to protonated 2-methyl-4H-3,1-benzoxazine occurs to give a cyclic tetrahedral carbonyl addition intermediate. At pH <5, the intermediate is protonated and reacts to 2-aminobenzyl acetate, whereas at pH >7.5, the unprotonated intermediate collapses to give 2-acetylaminobenzyl alcohol. The former reaction is catalysed by buffer base but the latter is uncatalysed. At pH 9-12, reaction of hydroxide ion with protonated 2-methyl-4H-3,1-benzoxazine to give 2-acetylaminobenzyl alcohol becomes important, and at pH >12, the same product is formed by reaction of hydroxide ion with unprotonated 2-methyl-4H-3,1-benzoxazine.

Full text of DOI:10.1139/v99-091

1
035
Formation of a cyclic tetrahedral intermediate
by the addition of water to 2-methyl-4H-
3
2
,1-benzoxazine followed by ring opening to
-aminobenzyl acetate and 2-acetylaminobenzyl
alcohol; pH-dependence of rate of reaction and
product ratio
Wendy J. Dixon and Frank Hibbert
Abstract: Kinetic studies have shown that addition of water to protonated 2-methyl-4H-3,1-benzoxazine occurs to give
a cyclic tetrahedral carbonyl addition intermediate. At pH <5, the intermediate is protonated and reacts to 2-
aminobenzyl acetate, whereas at pH >7.5, the unprotonated intermediate collapses to give 2-acetylaminobenzyl alcohol.
The former reaction is catalysed by buffer base but the latter is uncatalysed. At pH 9–12, reaction of hydroxide ion
with protonated 2-methyl-4H-3,1-benzoxazine to give 2-acetylaminobenzyl alcohol becomes important, and at pH >12,
the same product is formed by reaction of hydroxide ion with unprotonated 2-methyl-4H-3,1-benzoxazine.
Key words: mechanism, addition, tetrahedral intermediate, hydrolysis, pH profile.
Résumé : Des études cinétiques ont montré que l’addition d’eau à la 2-méthyl-4H-3,1-benzoxazine protonée se produit
pour donner un intermédiaire d’addition au carbonyle de nature tétraédrique et cyclique. À un pH <5, l’intermédiaire
est protoné et réagit vers l’acétate du 2-aminobenzyle alors qu’à pH >7,5, l’intermédiaire non protoné se décompose
pour donner l’alcool 2-acétylaminobenzylique. La première réaction est catalysée par une base tampon, mais la dernière
n’est pas catalysée. À des pH allant de 9–12, la réaction de l’ion hydroxyde avec la 2-méthyl-4H-3,1-benzoxazine
protonée conduisant à l’alcool 2-acétylaminobenzylique devient important et à un pH >12, le même produit se forme
par réaction de l’ion hydroxyde avec 2-méthyl-4H-3,1-benzoxazine non protonée.
Mots clés : mécanisme, addition, intermédiaire tétraédrique, hydrolyse, profil de pH.
[
Traduit par la Rédaction] Dixon and Hibbert 1041
Introduction
variation of the observed first-order rate coefficient with
concentration of buffer base is given by eq. [2]. At low
In the pH range 1–5, 2-methyl-4H-3,1-benzoxazine under-
goes addition of water and ring opening to give 2-amino-
benzyl acetate, eq. [1] (1). The form of the rate coefficient –
pH profile and the observation of rectilinear plots of rate
coefficient against the concentration of buffer base have
identified the mechanism shown in Scheme 1. The collapse
of the cyclic tetrahedral intermediate to product involves
uncatalysed and buffer base (B) catalysed pathways, and the
+
[H O ]
k (k + k [B])
1 2 B
3
[
2]
kobs
=
×
+
KSH
+
+ [H O ]
k–1 + k + k [B]
2 B
3
buffer concentrations, buffer base catalysed collapse of the
tetrahedral intermediate to 2-aminobenzyl acetate is rate-
limiting, but at high buffer concentrations, addition of water
to protonated 2-methyl-4H-3,1-benzoxazine becomes rate-
limiting. In previous studies, measurements were limited to
pH <5, under which conditions reaction occurs to give 2-
aminobenzyl acetate in greater than ca. 95% yield. At higher
pH values, it was noted (1) that increasing amounts of 2-
acetylaminobenzyl alcohol were formed, eq. [3]. It was also
found that at fixed pH the proportion of 2-aminobenzyl ace-
[
1]
Received November 11, 1998.
This paper is dedicated to Jerry Kresge in recognition of his
many achievements in chemistry.
1
W.J. Dixon and F. Hibbert. Department of Chemistry,
[3]
King’s College London, Strand, London WC2R 2LS, U.K.
1Author to whom correspondence may be addressed.
Telephone: 44 0171 873 2284. Fax: 44 0171 873 2810.
e-mail: frank.hibbert@kcl.ac.uk
Can. J. Chem. 77: 1035–1041 (1999)
© 1999 NRC Canada

1
036
Can. J. Chem. Vol. 77, 1999
Fig. 1. Reaction of 2-methyl-4H-3,1-benzoxazine to 2-
Scheme 1.
acetylaminobenzyl alcohol and 2-aminobenzylacetate; variation
with pH of the percentage of 2-acetylaminobenzyl alcohol.
tate formed as product compared with 2-acetylaminobenzyl
alcohol increased as the concentration of buffer base was in-
creased, implying that reaction to the amide is less suscepti-
ble to buffer catalysis than formation of the ester. We have
now investigated the factors influencing the formation of es-
ter and amide and the kinetics of the formation of the amide
in detail over the whole of the pH range. The reaction of
acyclic imidates is known to give amides and esters depend-
ing on the reaction conditions and evidence is available to
support the participation of tetrahedral carbonyl addition in-
termediates (2, 3). The tetrahedral carbonyl addition inter-
mediates involved in the present work are cyclic, and their
stability relative to reactants and products compared with
that of acyclic intermediates is of particular interest.
spectrum following reaction was found to be quantitatively
identical with the spectrum of an authentic sample of 2-
aminobenzyl acetate, and at pH 7.5–13, the spectrum of the
product was identical with that of 2-acetylaminobenzyl alco-
hol. At intermediate pH values of 5–7.5, hydrolysis of 2-
methyl-4H-3,1-benzoxazine results in a mixture of 2-amino-
benzyl acetate and 2-acetylaminobenzyl alcohol, depending
on the pH and buffer concentration. In acetate, pivalate, and
phosphate buffers, the exact composition of the product mix-
ture was calculated from the measured absorbance at 284 nm
using the known absorbance of 2-aminobenzyl acetate and
2-acetylaminobenzyl alcohol at this wavelength. Results for
the variation with pH of the fraction of 2-acetylaminobenzyl
alcohol compared with 2-aminobenzyl acetate formed as
product are shown in Fig. 1. The results were obtained at a
buffer concentration of ca. 0.01 mol dm in the basic com-
ponent of the buffer. The points in Fig. 1 are experimental
values, and the solid line is a best fit predicted from the
mechanism in Scheme 2 (see below). It was also found that
the percentage of 2-aminobenzyl acetate formed increased
with buffer base concentration, indicating that reaction to 2-
aminobenzyl acetate was more susceptible to catalysis by
buffer than reaction to 2-acetylaminobenzyl alcohol. This
was confirmed in kinetic studies (see below).
Experimental
Materials
The preparations of 2-methyl-4H-3,1-benzoxazine, 2-amino-
benzyl acetate, and 2-acetylaminobenzyl alcohol have been
described previously (1). Carboxylic acid buffer solutions
were made up by partial neutralization of a standardized
solution of the carboxylic acid with a standard solution of
potassium hydroxide. Buffer solutions of dihydrogen ortho-
phosphate monoanion and monohydrogen orthophosphate
dianion were made up by partial neutralization of a solution
of potassium dihydrogen orthophosphate with a standard
solution of potassium hydroxide. Buffer solutions of tris(hy-
droxymethyl)aminomethane and 2-amino-2-methyl-1,3-pro-
panediol with the corresponding protonated amine were made
up by partial neutralization of a standardized solution of the
amine with standard hydrochloric acid solution. The ionic
–
3
–
3
strength of all solutions was maintained at 0.25 mol dm by
addition of a weighed quantity of potassium chloride. Dou-
ble distilled water was used for the preparation of solutions.
Reactions under various conditions were begun by injecting
3
0
.01 cm of a concentrated solution of 2-methyl-4H-3,1-
3
benzoxazine in Me SO into the reaction medium (3 cm ).
Kinetic measurements
2
The methods used to study the kinetics of the ring open-
ing of 2-methyl-4H-3,1-benzoxazine to 2-aminobenzyl ace-
tate in the pH range 1–5 have been described previously (1).
In the present work, kinetic measurements of the reaction of
2-methyl-4H-3,1-benzoxazine to 2-aminobenzyl acetate and
Product analysis
The products of hydrolysis of 2-methyl-4H-3,1-benzoxa-
zine were identified from the changes in UV spectrum ac-
companying the reaction. In the range pH 1–5, the final
©
1999 NRC Canada

Dixon and Hibbert
1037
–
1
Scheme 2.
Fig. 2. Dependence on pH of the rate coefficient (log10 k/s )
for the ring opening of 2-methyl-4H-3,1-benzoxazine in aqueous
solution.
to 2-acetylaminobenzyl alcohol were made at pH values
–
3
5
–13 at 298.2 K and ionic strength 0.25 mol dm . Studies
in the range pH 6–9 were made in buffers of orthophosphate
2
monanion and dianion (pK 6.52) (4) and of tris(hydroxy-
a
3
methyl)aminomethane (pKa 8.07) (5) and 2-amino-2-
3
methyl-1,3-propanediol (pK 8.80) (5) with the correspond-
a
ing protonated amines at various buffer ratios. The depend-
ence of the rate of reaction on buffer concentration was also
investigated. Solutions of potassium hydroxide with concen-
trations in the range 0.001–0.25 mol dm were used for
studies at higher pH. Reactions were begun by injecting a
concentrated solution of 2-methyl-4H-3,1-benzoxazine in
–
3
typically 0.9999, and standard deviations of the calculated
first-order rate coefficients were typically less than 1% The
reactions occurred with half-lives in the range ca. 40 – 3.5 ×
3
1
0 s.
Me SO into the reaction mixture to give an initial concentra-
2
–
4
–3
tion of ca. 1.0 × 10 mol dm . The concentration of
2
1
-methyl-4H-3,1-benzoxazine was always kept in at least
0-fold deficit compared with buffer species or hydronium
Results and discussion
ion or hydroxide ion. Hydrolysis was followed by observing
the UV spectral change with time. In a buffer of tris(hy-
droxymethyl)aminomethane (buffer ratio 1.0, pH 8.07 at a
Rate–pH profile
The pH dependence of the rate coefficient (k) for the hy-
drolysis of 2-methyl-4H-3,1-benzoxazine is shown in Fig. 2.
The data were obtained from studies in solutions of hydro-
chloric acid and potassium hydroxide or in the presence of
buffers. The value of the rate coefficient (k) given for hydro-
lysis in buffer solutions was obtained by extrapolation of the
observed rate coefficient (kobs) to zero buffer concentration
and hence refers to reaction involving species derived from
the solvent (water). In the range pH 1–5 the product of reac-
tion at high buffer concentrations is exclusively 2-amino-
benzyl acetate, and at pH 7.5–13, 2-acetylaminobenzyl alco-
hol is the sole product. In the region 5–7.5, hydrolysis of 2-
–
3
buffer concentration of 0.25 mol dm in each buffer compo-
nent), the spectral changes with time occurred to give clean
isosbestic points at 242 and 308 nm. Reactions were usually
followed by measuring the decrease in absorbance at 261 nm
and were found to be accurately of first order. The value of
the first-order rate coefficient (kobs) was determined from an
exponential fit to the variation of absorbance with time over
at least three half lives. Reactions were followed for a suffi-
cient time to allow the determination of an absorbance value
corresponding to complete reaction except in the case of
some of the slowest reactions. Correlation coefficients were
methyl-4H-3,1-benzoxazine yields
a
mixture of 2-
2
–3
The pK value (7.20) (ref. 4) for dihydrogen orthophosphate monoanion at infinite dilution was adjusted to ionic strength (I) 0.25 mol dm
a
using the Debye–Huckel expression –log γ ± = 0.51Zi2 √I/(1 + √I).
3
–3
The value of pK at infinite dilution (5) was used uncorrected for the present conditions of ionic strength 0.25 mol dm .
a
©
1999 NRC Canada

1
038
Can. J. Chem. Vol. 77, 1999
Table 1. Values of equilibrium constants and rate coefficients for the best fit of eq. [12] to the rate coefficient – pH profile.
k k /(k + k2) k /k K
K
SH+
k1
1H+
3
k0
kOH
1
2
–1
2
–
3
–1
–1
(mol dm–3)
1.29 ± 0.1 × 106
–1
3
–1 –1
(
mol dm )
.4 ± 0.5 × 10
(s )
9.5 ± 1 × 10
(s )
0.17
(s )
(dm mol s )
5.3 ± 0.5 × 10
–
5
–3
–5
–4
2
5.1 ± 0.5 × 10
aminobenzyl acetate and 2-acetylaminobenzyl alcohol de-
pending on the reaction conditions. The results in each of
these pH regions will be considered in turn.
–1
Fig. 3. Variation of the rate coefficient (kobs/s ) with
concentration for buffers of 2-amino-2-methyl-l,3-propanediol at
different buffer ratios (r).
Reaction to 2-aminobenzyl acetate at pH 1–5
Data for hydrolysis in the range pH 1–5 have been pub-
lished (1). Reaction occurs by addition of water to the
protonated form of 2-methyl-4H-3,1-benzoxazine to give a
cyclic tetrahedral addition intermediate which collapses to 2-
aminobenzyl acetate, Scheme 1. Evidence for the involve-
ment of a tetrahedral addition intermediate on the reaction
pathway was provided by the observation of rectilinear plots
of the observed rate coefficient against buffer base concen-
tration as described earlier. According to the mechanism in
Scheme 1, the rate coefficient for hydrolysis of 2-methyl-
4
H-3,1-benzoxazine extrapolated to zero buffer concentra-
tion is given by eq. [4] in which KSH
+
is the acid dissociation
+
+
[
4]
k = k k [H O ]/(k + k )(K
SH+
+ [H O ])
1
2
3
–1
2
3
constant of protonated 2-methyl-4H-3,1-benzoxazine and the
rate coefficients k , k , and k refer to the steps in
1
–1
2
Scheme 1. The solid line in Fig. 2 through the data points in
the region pH 1–5 is a best-fit of eq. [4] to the experimental
results (1) using the values shown in Table 1 for KSH
+
, k1,
and k /(k + k ). The observed rectilinear dependence of the
2
–1
2
first-order rate coefficient (kobs) on buffer concentration,
eq. [2], was used to deduce values for the ratio k /(k + k2)
B
–1
where k is the value of the second-order rate coefficient for
B
catalysis of the collapse of the tetrahedral intermediate to
2
-aminobenzyl acetate by chloroacetate, formate, acetate,
and pivalate ions. The value of the rate coefficient ratio
kB/(k–1 + k ) was found to increase with increase in base
2
strength of the catalysing buffer, giving a Brønsted plot with
a value of β ca. 0.8 for the Brønsted exponent. It was also
methyl-1,3-propanediol buffer at different buffer ratios (r =
+
2
found that the value of the rate coefficient ratio k /(k–1 + k2)
[B]/[BH ]) are shown in Fig. 3. The increase in k with
obs
for spontaneous collapse of the intermediate with k cor-
2
buffer concentration up to a concentration of the buffer cat-
–
3
rected to a second-order rate coefficient for reaction with
water fitted the Brønsted plot for catalysis by buffers. This is
compatible with the solvent acting as a base in catalysing
the collapse of the tetrahedral intermediate.
ion of 0.125 mol dm amounts to ca. 10, 8, and 9% at
buffer ratios of r = 0.5, 1.0, and 2.0, respectively. Similar ef-
fects are observed in buffers of tris(hydroxymethyl)amino-
methane. The shallow dependence on buffer concentration
could be due to a salt effect as the inert salt (potassium chlo-
ride) is replaced by the buffer cation at constant ionic
strength. A similar effect was previously observed in the re-
action of 2-methylnaphth[1,8-de]-1,3-oxazine to 1-hydroxy-
8-acetylaminonaphthalene in acetate buffers and was attrib-
uted to a salt effect (6). However, the possibility of weak
buffer catalysis cannot be ruled out.
Reaction to 2-aminobenzyl acetate and
2
-acetylaminobenzyl alcohol at pH 5–9
In the present work, kinetic studies have been extended
above pH 5, and the hydrolysis of 2-methyl-4H-3,1-benz-
oxazine gives a mixture of 2-aminobenzyl acetate and 2-
acetylaminobenzyl alcohol, depending on the reaction condi-
tions. Measurements were made in pivalate, phosphate, and
in amine buffers.
In buffers of 2-amino-2-methyl-1,3-propanediol and of
tris(hydroxymethyl)aminomethane (pH 7.8–9.1) where the
reaction occurs to give exclusively 2-acetylaminobenzyl al-
cohol, there is a shallow dependence of kobs on buffer con-
centration. Data for the reaction in buffers of 2-amino-2-
In the lower pH range, measurements were made in the
presence of buffers of dihydrogen orthophosphate monoanion
and monohydrogen orthophosphate dianion at buffer ratios r
2
–
–
(HPO4 /H PO ) 0.5, 1.0, and 2.0, corresponding to pH val-
2
4
ues 6.22, 6.52, and 6.82. The change in kobs with buffer con-
centration was small. For example, at a buffer ratio r = 1.0,
–
1
the value of kobs increased from 0.114 to 0.230 s as the
©
1999 NRC Canada

Dixon and Hibbert
1039
5
–1
buffer base concentration was increased from 0.001 to 0.05
Fig. 4. Variation of the rate coefficient (10 k/s ) for ring
opening of 2-methyl-4H-3,1-benzoxazine with hydroxide ion
concentration.
–
3
mol dm . At buffer ratios of r = 0.5 and 2.0, the values of
–
1
kobs increased from 0.213 to 0.447 s and from 0.0671 to
–
1
0
.107 s , respectively. Extrapolation to zero buffer concen-
tration gave values of k that did not fit the rate–pH profile
determined in other buffers. At these pH values, the reaction
of 2-methyl-4H-3,1-benzoxazine occurs to give both 2-
aminobenzyl acetate and 2-acetylaminobenzyl alcohol. Al-
though the latter reaction is not susceptible to strong cataly-
sis by buffer, the reaction to 2-aminobenzyl acetate is
catalysed by buffer. Details of the catalysis of the reaction to
2
-aminobenzyl acetate by carboxylic acid buffers can be
used to predict the behaviour in orthophosphate buffers. The
variation of kobs with buffer concentration for reaction of 2-
methyl-4H-3,1-benzoxazine to 2-aminobenzyl acetate is
given by eq. [2]. Extrapolation of the Brønsted plot for the
variation of the value of k /(k–1 + k ) with base strength for
B
2
carboxylic acid buffers (1) gives the result k /(k + k ) =
B
–1
2
3
3
–1
1
.92 × 10 dm mol for catalysis by monohydrogen
orthophosphate dianion. This value can be used to predict
that for the reaction of 2-methyl-4H-3,1-benzoxazine to 2-
aminobenzyl acetate in a phosphate buffer, catalysis will be
close to the upper limiting value of the rectilinear depend-
ence of kobs against buffer concentration for most of the con-
centrations of monohydrogen orthophosphate dianion
studied. At a buffer ratio r = 1.0 and a buffer concentration
–
3
of 0.001 mol dm , catalysis will have reached ca. 66% of
the limiting rate. It follows that extrapolation of values of
2
–
–3
kobs obtained in the range [HPO4 ] 0.001–0.05 mol dm to
zero buffer concentration is unreliable, and the results are
not included in the rate–pH profile in Fig. 2.
form of eq. [6], and linear regression analysis gave a value
6
–1
3
for k /k K
1H+
of 1.29 × 10 mol dm . This result was used
2
3
in eq. [6] to calculate the solid line in Fig. 1 for the pH de-
pendence of the percentage of 2-acetylaminobenzyl alcohol
formed as product.
At pH <5 where 2-aminobenzyl acetate is the exclusive
product, the terms involving k can be neglected, and eq. [5]
3
reduces to eq. [4]. In the pH region 8–9 where 2-acetyl-
aminobenzyl alcohol is the only product, the terms involving
k can be neglected, and eq. [5] reduces to eq. [7]. If it is
2
+
further assumed that k K
1H+
> k [H O ], eq. [8] is obtained,
3
–1
3
which further reduces to eq. [9] on the assumption that K
SH+
+
>
[H O ]. Equation [9] predicts a direct dependence of k
against [H O ]. The experimental values of k in the range
pH 7.8–9.1 were plotted against [H O ]. However, a linear
and intercept 4.9
was obtained from linear regression analysis
rather than the direct dependence of k against [H O ] pre-
3
+
3
+
3
3
3
–1 –1
plot with gradient 7.1 × 10 dm mol
s
–
5
–1
×
10
s
+
3
dicted by eq. [9]. It was therefore necessary to modify the
rate expression to include a pH-independent rate term k0 as
in eq. [10]. Thus, eq. [10] is compatible with the experimen-
3
3
–1 –1
tal results with values of k /K
SH+
= 7.1 × 10 dm mol
s
1
–5 –1
and k = 4.9 × 10 s .
0
+
[
7]
k = k [H O][H O ]k K /(KSH
1H+
+
1
2
3
3
According to the mechanism in Scheme 2, the percentage of
-acetylaminobenzyl alcohol (A) formed as product com-
+
+
+
[H O ])(k K
1H+
+ k [H O ])
2
3
3
–1
3
pared with 2-aminobenzyl acetate (E) is given by the expres-
sion in eq. [6]. The results obtained for the percentage of 2-
acetylaminobenzyl alcohol formed as product compared
with 2-aminobenzyl acetate were plotted in the reciprocal
+
+
[
[
8]
9]
k = k [H O][H O ]/(K
SH+
+ [H O ])
1
2
3
3
+
k = (k /K )[H O][H O ]
SH+
1
2
3
©
1999 NRC Canada

1
040
Can. J. Chem. Vol. 77, 1999
Scheme 3.
protonated 2-methyl-4H-3,1-benzoxazine, the result k 5.1 ×
0
–
5 –1
1
0 s corresponds to a second-order rate coefficient for re-
4
3
–1 –1
action with hydroxide ion of 5.7 × 10 dm mol
s
using
2.4 × 10 mol dm for the acid dissocia-
tion constant of protonated 2-methyl-4H-3,1-benzoxazine
–
5
–3
the values of KSH
+
–
14
2
–6
and Kw = 2.19 × 10 mol dm for the ionic product of
water under the reaction conditions. This is to be compared
–
3
3
–1 –1
with the value k = 3.1 × 10 dm mol
s
obtained for the
1
reaction of water with protonated 2-methyl-4H-3,1-
7
benzoxazine. The 1.8 × 10 -fold difference seems reasonable
in view of the very large difference in nucleophilicity of hy-
droxide ion and water. If the pH-independent term arises
from rate-limiting attack of water on unprotonated 2-methyl-
–
5
–1
4
H-3,1-benzoxazine, the result k 5.1 × 10
s
corresponds
0
–
7
3
–1 –1
to a second-order rate coefficient of 9.2 × 10 dm mol
s
for reaction with water. This is to be compared with the
–
4
3
–1 –1
value kOH 5.3 × 10 dm mol
s
found for reaction of hy-
droxide ion with unprotonated 2-methyl-4H-3,1-benz-
oxazine. This 600-fold difference is smaller than would have
been anticipated from the very large difference in
nucleophilicity of water and hydroxide ion. For this reason,
the preferred explanation of the pH-independent term in the
rate law, eqs. [10] and [11], is a reaction involving addition
of hydroxide ion to protonated 2-methyl-4H-3,1-benzoxa-
zine.
+
[
SH+
10] k = k + (k /K )[H O][H O ]
0 1 2 3
Reaction to 2-acetylaminobenzyl alcohol at pH 10–13
In the range pH 10–13, the product of reaction is 2-
acetylaminobenzyl alcohol. Kinetic studies in this region
were made in solutions of potassium hydroxide at concentra-
Dependence of k in the range pH 0–13
The pH dependence of the rate coefficient (k) for hydroly-
sis of 2-methyl-4H-3,1-benzoxazine involving species
derived from the solvent over the whole of the pH range
0–13 is fitted by eq. [12], which is obtained by combining
eq. [5] with eq. [11]. The solid line in Fig. 2 is a plot of eq.
[12] using the values of the rate coefficients and equilibrium
–
3
tions in the range 0.001–0.25 mol dm . The dependence of
the measured first-order rate coefficient on hydroxide ion
concentration is linear as shown in Fig. 4 with linear regres-
–
4
3
–1 –1
sion gradient k = 5.3 × 10 dm mol
=
s
and intercept k0
OH
–1
–
5
5.2 × 10 s . Hence, a pH-independent term and a term
that is of first order in hydroxide ion contribute to the rate in
this region, eq. [11]. The pH-independent term with a slightly
+
+
k1[H2O][H3O ](k2[H3O ] + k3K1H )
+
[
12] k =
+
+
(
K
SH+
+ [H O ]){k K
1H+
+ (k–1 + k )[H O ]}
11] k = k + kOH[OH–]
0
3
3
2 3
[
different value k = 4.9 × 10 s–1 was also found from the
–5
+ k + kOH[OH– ]
0
0
+
variation of k with [H O ] in the range 7.8–9.1, eq. [10].
The average of these two values leads to k 5.1 × 10 s .
3
–
5
–1
constants given in Table 1. At low pH, reaction of pro-
tonated 2-methyl-4H-3,1-benzoxazine with solvent occurs
through a protonated tetrahedral intermediate, leading to 2-
aminobenzyl acetate as product. As the pH is raised, reac-
tion of protonated 2-methyl-4H-3,1-benzoxazine leads to 2-
acetylaminobenzyl alcohol, since the tetrahedral intermedi-
ate that is common to both pathways is present in the
unprotonated form. In the protonated form, the intermediate
leads to 2-aminobenzyl acetate because departure of an aro-
matic amine as leaving group is facile. Collapse of the
unprotonated intermediate to 2-aminobenzyl acetate would
involve departure of an amine anion, and reaction to 2-
acetylaminobenzyl alcohol involving an aliphatic alkoxide
as leaving group is preferred. The change in pathway that
occurs between pH ca. 6 and 7 leads to an inflexion point in
the rate coefficient against pH profile. The relative amount
of the two products is determined by the fraction of the tet-
rahedral intermediate that is protonated and the relative rates
of collapse of the protonated intermediate to 2-aminobenzyl
acetate and the unprotonated intermediate to 2-acetylamino-
benzyl alcohol, eq. [6]. At pH 9, reaction of protonated
o
A mechanism that is compatible with the formation of 2-
acetylaminobenzyl alcohol as the sole product and with the
experimentally observed dependence of k against [OH ]
–
shown in Fig. 4 involves addition of hydroxide ion to pro-
tonated and unprotonated forms of 2-methyl-4H-3,1-benzo-
xazine to give a tetrahedral intermediate which collapses to
product, Scheme 3.
In Scheme 3, it can be assumed that the tetrahedral inter-
mediates are present in low concentration. The pH-
independent term and the first-order term in hydroxide ion
are explained by reaction of hydroxide ion with the pro-
tonated and unprotonated forms of 2-methyl-4H-3,1-benzo-
xazine, respectively. An alternative mechanism that will
explain the pH-independent term involves reaction of the
unprotonated form of 2-methyl-4H-3,1-benzoxazine with
solvent. The relative importance of these two possibilities
for the pH-independent term can be assessed by comparison
of the value of the rate coefficient k with the values of the
0
other rate coefficients in the mechanism. If the pH-
independent term arises from reaction of hydroxide ion with
©
1999 NRC Canada

Dixon and Hibbert
1041
2
-methyl-4H-3,1-benzoxazine with hydroxide ion becomes
important, and above pH 12, reaction of unprotonated
-methyl-4H-3,1-benzoxazine with hydroxide ion predomi-
buffer catalysis would be undetectible compared with sol-
vent catalysed collapse.
2
nates. Both of these reactions result in the formation of 2-
acetylaminobenzyl alcohol, presumably through the unpro-
tonated form of the tetrahedral intermediate.
References
1
2
3
4
. W.J. Dixon, F. Hibbert, and J.F. Mills. J. Chem Soc. Perkin
Trans. 2, 1503 (1997).
. R.H. De Wolfe and D.G. Neilson. In The chemistry of amidines
and imidates. Edited by S. Patai. Wiley, New York. 1975.
. A.C. Satterthwait and W.P. Jencks. J. Am. Chem. Soc. 96,
It was found that collapse of the protonated intermediate
to 2-aminobenzyl acetate is catalysed by buffer base. This
presumably occurs by assistance of removal of the hydroxy
proton as the carbonyl group of the product is formed. In
contrast, collapse of the unprotonated tetrahedral intermedi-
ate to 2-acetylaminobenzyl alcohol is not assisted by buffer
base. The absence of catalysis in this case could indicate that
addition of water to protonated 2-methyl-4H-3,1-benzoxa-
zine is rate-limiting (that is, collapse of the intermediate to
product occurs more rapidly than collapse to protonated 2-
methyl-4H-3,1-benzoxazine). Alternatively, if the value of
the Brønsted exponent for base catalysis of the collapse of
the unprotonated tetrahedral intermediate was close to zero,
7
031 (1974).
. D.D. Perrin. In Dissociation constants of inorganic acids and
bases in aqueous solution. 2nd ed. Butterworths, London.
Pergamon Press, London. 1982.
5
6
. D.D. Perrin. In Dissociation constants of organic bases in
aqueous solution. Butterworths, London; Pergamon Press, Ox-
ford. (1st ed.) 1965; (supplement) 1972.
. W.J. Dixon and F. Hibbert. J. Chem. Soc. Perkin Trans. 2, 795
(
1994).
©
1999 NRC Canada