Kinetics and mechanism of hydrolysis of 3-diazobenzofuran-2-one and its hydrolysis product (3-hydroxybenzofuran-2-one)

Article abstract of DOI:10.1139/v01-191

Rates of acid-catalyzed hydrolysis of 3-diazobenzofuran-2-one, measured in concentrated aqueous perchloric acid and hydrochloric acid solutions, were found to correlate well with the Cox-Yates X_o excess acidity function, giving k_H+ = 1.66 × 10^-4 M^-1 s^-1, m^? = 0.86 and k_H+/k_D+ = 2.04. The normal direction (k_H/k_D > 1) of this isotope effect indicates that hydrolysis occurs by rate-determining protonation of the substrate on its diazo-carbon atom. It was found previously that the next higher homolog of the present substrate, 4-diazoisochroman-3-one, also undergoes hydrolysis by this reaction mechanism but with a rate constant 15 times greater than that for the present substrate; this difference in reactivity can be understood in terms of the various resonance forms that contribute to the structures of these substrates. The product of the present hydrolysis reaction is 3-hydroxybenzofuran-2-one, which itself quickly undergoes subsequent acid-catalyzed hydrolysis to 2-hydroxymandelic acid. The acidity dependence of this subsequent hydrolysis is much shallower than that of the diazo compound precursor, and rates of reaction correlate as well with [H⁺] as with X_o. This is due in part to incursion of a nonproductive protonation on the hydroxy group of 3-hydroxybenzofuran-2-one that impedes hydrolysis and produces saturation of acid catalysis. Rates of hydrolysis of the hydroxy compound were also measured in dilute HClO₄ and NaOH solutions as well as in CH₃CO₂H, H₂PO₄^-, (CH₂OH)₃CNH₃^-, and NH₄^- buffers, and the rate profile constructed from these data showed the presence of uncatalyzed and hydroxide ion-catalyzed reactions. This hydroxide-ion catalysis became saturated at [NaOH] ? 0.05 M, implying occurrence of yet another nonproductive substrate ionization.

Full text of DOI:10.1139/v01-191

8
2
Kine tics and me chanis m of hydrolys is of 3-
diazobe nzofuran-2-one and its hydrolys is product
(3-hydroxybe nzofuran-2-one )
Y. Chia ng, A.J . Kre s ge , a nd Q. Me ng
Abstract: Rates of acid-catalyzed hydrolysis of 3-diazobenzofuran-2-one, measured in concentrated aqueous perchloric
acid and hydrochloric acid solutions, were found to correlate well with the Cox–Yates X excess acidity function, giv-
o
ing kH
+
= 1.66 × 10–4
M
–1
s , m = 0.86 and k /kD = 2.04. The normal direction (k /k > 1) of this isotope effect in-
–1
‡
H+
+
H D
dicates that hydrolysis occurs by rate-determining protonation of the substrate on its diazo-carbon atom. It was found
previously that the next higher homolog of the present substrate, 4-diazoisochroman-3-one, also undergoes hydrolysis
by this reaction mechanism but with a rate constant 15 times greater than that for the present substrate; this difference
in reactivity can be understood in terms of the various resonance forms that contribute to the structures of these sub-
strates. The product of the present hydrolysis reaction is 3-hydroxybenzofuran-2-one, which itself quickly undergoes
subsequent acid-catalyzed hydrolysis to 2-hydroxymandelic acid. The acidity dependence of this subsequent hydrolysis
+
is much shallower than that of the diazo compound precursor, and rates of reaction correlate as well with [H ] as with
Xo. This is due in part to incursion of a nonproductive protonation on the hydroxy group of 3-hydroxybenzofuran-2-one
that impedes hydrolysis and produces saturation of acid catalysis. Rates of hydrolysis of the hydroxy compound were
−
+
+
also measured in dilute HClO and NaOH solutions as well as in CH CO H, H PO , (CH OH) CNH , and NH buff-
4
3
2
2
4
2
3
3
4
ers, and the rate profile constructed from these data showed the presence of uncatalyzed and hydroxide ion-catalyzed
reactions. This hydroxide-ion catalysis became saturated at [NaOH] ≅ 0.05 M, implying occurrence of yet another
nonproductive substrate ionization.
Key words: diazo compound hydrolysis, lactone hydrolysis, Cox–Yates excess acidity, acid catalysis, alcohol
protonation.
Résumé : On a observé que les vitesses de réactions d’hydrolyse acidocatalysées de la 3-diazobenzofuran-2-one en so-
lutions aqueuses des acides perchlorique et chlorhydrique donnent une bonne corrélation avec la fonction d’acidité en
–
4
–1 –1
‡
excès de Cox–Yates, X , qui conduit à kH
+
= 1,66 × 10
M
s , m = 0,86 et k /kD
H+
+
= 2,04. La direction normale
o
(
k /k > 1) de cet effet isotopique indique que l’hydrolyse se produit par une étape de protonation du substrat cinéti-
H
D
quement déterminante sur l’atome de carbone du diazo. On avait observé antérieurement que l’homologue directement
supérieur au substrat étudié, la 4-diazoisochroman-4-one, subit aussi une hydrolyse par ce mécanisme réactionnel, mais
que la constante de vitesse de sa réaction est 15 fois supérieure à celle observée avec le substrat étudié ici; cette diffé-
rence de réactivité peut s’expliquer en termes de diverses formes de résonance qui contribuent aux structures de ces
substrats. Le produit d’hydrolyse du produit considéré ici est la 3-hydroxybenzofuran-2-one qui subit rapidement une
hydrolyse acidocatalysée subséquente conduisant à l’acide 2-hydroxymandélique. La dépendance de cette hydrolyse
subséquente sur l’acidité est beaucoup plus faible que celle sur le diazo précurseur et les constantes de vitesse donnent
+
aussi une bonne corrélation avec [H ] ainsi qu’avec X . Ce résultat est dû en partie à l’incursion d’une protonation non
o
productive sur le groupe hydroxyle de la 3-hydroxybenzofuran-3-one qui inhibe l’hydrolyse et qui produit une satura-
tion de la catalyse acide. Les constantes de vitesse du composé hydroxylé ont aussi été mesurées dans des solutions di-
−
+
+
luées de HClO et de NaOH ainsi que dans des tampons de CH CO H, H PO , (CH OH) CNH et NH4 et les profils
4
3
2
2
4
2
3
3
de vitesses construits à partir de ces données mettent en évidence l’existence de réactions non catalysées et d’autres ca-
talysées par l’ion hydroxyde. Cette catalyse par l’ion hydroxyde devient saturée à [NaOH] ≅ 0,05 M, ce qui implique
l’existence d’une autre ionisation non productive du substrat.
Mots clés : hydrolyse de composés diazo, hydrolyse de lactones, acidité en excès de Cox–Yates, catalyse acide, proto-
nation d’alcool.
[
Traduit par la Rédaction] Chiang et al. 88
Received 30 July 2001. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 17 January 2002.
1
Y. Chiang, A.J. Kresge, and Q. Meng. Department of Chemistry, University of Toronto, Toronto, ON M5S 3H6, Canada.
1Corresponding author (telephone and fax: (416) 978-7259; e-mail: akresge@chem.utoronto.ca).
Can. J. Chem. 80: 82–88 (2002)
DOI: 10.1139/V01-191
© 2002 NRC Canada

Chiang et al.
83
Fig. 1. Comparison of rates of hydrolysis of 3-diazobenzofuran-
2
-one in concentrated perchloric (O) and hydrochloric (∆) acids
with rates of hydrolysis of 3-hydroxybenzofuranone in concen-
trated hydrochloric acid (᭛).
Expe rim e nta l s e c tion
Materials
2
-Hydroxymandelic acid (2) (5) was made by sodium
borohydride reduction of 2-hydroxyphenylglyoxylic acid, it-
self prepared by selenium dioxide oxidation of 2-hydroxy-
acetophenone (2a). As reported (2), 2-hydroxymandelic acid
could not be obtained completely free of solvent, and it was
consequently used as an ethyl acetate solvate. 3-Hydroxy-
benzofuran-2-one (2a) (4) was synthesized by treating 2-
hydroxymandelic acid with dicyclohexylcarbodiimide. 3-
Diazobenzofuran-2-one (3) (1) was prepared by converting
2
-hydroxyphenylglyoxylic acid into benzofuran-2,3-dione,
using refluxing acetic anhydride to effect ring closure, and
then making the tosyl hydrazone and cleaving that with so-
dium hydroxide (3). The spectral and physical properties of
these substances agreed with literature values.
All other materials were best available commercial grades.
Introduc tion
It has been reported that irradiation of 3-diazobenzofuran-
-one (1) in low-temperature matrices produces the ketene 2
Kinetics
2
Rates of reaction were measured spectrophotometrically,
using a Cary 220 spectrometer whose cell compartment
was thermostatted at 25.0 ± 0.05°C. Rates of hydrolysis of
and the oxocumulene 3 (eq. [1]) (1). To learn whether these
interesting molecules are also formed upon irradiation of 3-
diazobenzofuran-2-one in aqueous solution at ambient tem-
perature, and, if so, how they react with water, we have un-
dertaken a study of the photolysis of this substance in
aqueous media.
3
-diazobenzofuran-2-one were monitored by following the
decrease of its absorbance at λ = 310 nm, where neither 3-
hydroxybenzofuran-2-one nor 2-hydroxymandelic acid ab-
sorb to any significant extent, and rates of hydrolysis of 3-
hydroxybenzofuran-2-one were monitored by following the
increase of 2-hydroxymandelic acid absorbance at λ =
2
78 nm. The data for both reactions conformed to the first-
order rate law well, and observed first-order rate constants
were determined by least-squares fitting of a single exponen-
tial function.
Diazo compounds such as 1 also undergo acid-catalyzed
hydrolysis in aqueous solution, and such hydrolysis could
interfere with our photolysis study. To determine whether
this would be so, we also examined the hydrolysis reaction.
The results proved to be interesting in their own right, and
we therefore report them separately here.
The acid-catalyzed hydrolysis of diazo compounds usually
produces the corresponding hydroxy derivative, which in the
present system would be the hydroxy lactone, 3-hydroxy-
benzofuran-2-one (4) (eq. [2]). We have found that this is in-
deed the case, but that this product is then quickly
hydrolyzed itself, giving 2-hydroxymandelic acid (5)
Product analysis
Product compositions were determined by HPLC using a
Varian Vista 5500 instrument with a NovaPak C18 reverse-
phase column and methanol–water (60/40 = v/v) as the
eluent. Products were identified by comparing retention
times and UV spectra with those of authentic samples.
Re s ults a nd dis c us s ion
Product analysis
Rate constants for the hydrolysis of 3-diazobenzofuran-2-
one and 3-hydroxybenzofuran-2-one, determined in concen-
trated acid solutions as described below, are displayed in
Fig. 1. It may be seen that both reactions are acid-catalyzed
but that the acidity dependences of the two processes are
quite different. Because the rates cross one another and have
(
eq. [3]) as the ultimate reaction product. We have measured
rates of this second hydrolysis reaction as well.
+
the same value in the vicinity of [H ] = 4 M, this medium
was chosen for product analysis. The reactions are also quite
fast at this concentration, with half-lives of the order of
2
min. Samples of reaction mixtures were, therefore, taken
©
2002 NRC Canada

8
4
Can. J. Chem. Vol. 80, 2002
Fig. 2. Rates of hydrolysis of 3-diazobenzofuran-2-one in con-
centrated perchloric acid (O) and hydrochloric acid (∆) solutions
plotted according to eq. [4].
measured in perchloric acid solutions over the concentration
range [HClO ] = 3.6–5.8 M and in hydrochloric acid solu-
4
tions over the range [HCl] = 0.5–6.4 M; the data obtained
2
are summarized in Table S1. Some rate measurements were
also made in D O solutions of hydrochloric acid-d over the
2
range [DCl] = 0.5–6.2 M, and these data are listed in Ta-
2
ble S1 as well.
The observed first-order rate constants determined in
these concentrated acid solutions increased with increasing
solution acidity more steeply than in direct proportion to
acid concentration. This is commonly the case for reaction
rates measured in concentrated acids, and the data in such
situations are generally analyzed using an acidity function
rather than acid concentration as a measure of acid strength.
The X scale applied as shown in eq. [4] appears to be the
o
best function available for this purpose (4),
+
) + m‡Xo
[
4]
log (kobs/[H ]) = log (kH
+
and the present data were therefore analyzed using this ex-
pression with X values for HCl and HClO solutions taken
0
4
at short time intervals, were quenched into sufficient sodium
acetate solution to give a buffer of pH ≅ 5, and were then
immediately analyzed by HPLC.
–
4
–1 –1
The reaction mixture formed upon a 1.0 min exposure of
3
-diazobenzofuran-2-one to 4 M perchloric acid consisted
of, in addition to unreacted substrate, both 3-hydroxybenzo-
furan-2-one and 2-hydroxymandelic acid in a roughly 2:1 ratio.
Another sample taken after 2.0 min exposure showed the
same two products but now in approximately comparable
amounts. Successive samples taken at later times gave pro-
gressively smaller amounts of 3-hydroxybenzofuran-2-one at
the expense of increasing amounts of 2-hydroxymandelic
acid, until after 15 min only the hydroxymandelic acid was
left. This is the behavior expected of two successive reactions
and it confirms that 3-hydroxybenzofuran-2-one is first
formed by the hydrolysis of 3-diazobenzofuran-2-one and that
it then quickly hydrolyzes further to 2-hydroxymandelic acid.
Product analyses were also conducted for the hydrolysis of
D O, but several other acidity functions have been shown to
2
have the same value in D O as in H O when the comparison
2
2
is made at the same molar acid concentration (6). It seems
likely that this would be so for the X scale as well, and the
o
rate constants determined here for hydrolysis of 3-
diazobenzofuran-2-one in DCl were therefore correlated us-
ing X values obtained on this basis. These data also con-
o
formed to the expression of eq. [4] well, and least-squares
–
5
–1 –1
‡
analysis gave kD+ = (8.15 ± 0.12) × 10
M
s
and m =
0.837 ± 0.008. Combination of the results obtained in H O
2
and D O provides the isotope effect kH+ /k + = 2.04 ± 0.05.
2
D
The normal direction of this isotope effect (k /k > 1) in-
H
D
dicates that proton transfer is rate-determining in this reac-
tion (7) and that the process occurs by the standard
mechanism for diazo compound hydrolysis in which rate-
determining protonation of the diazo-carbon atom is fol-
lowed by rapid displacement of nitrogen by a molecule of
water (eq. [5]) (8). Another mechanism that is sometimes
found for diazo compound hydrolysis consists of rapid and
reversible carbon protonation followed by rate-determining
loss of nitrogen, but such a scheme would give an inverse
3
-hydroxybenzofuran-2-one itself. This was done in 0.1 M
+
perchloric acid solution, in an acetic acid buffer of [H ] =
3
buffer of [H ] = 8 × 10 M, in an ammonium ion buffer of
–
5
× 10 M, in a tris-(hydroxymethyl)methylammonium ion
+
–9
+
–10
[
H ] = 5 × 10 M, and in 0.01 M sodium hydroxide solu-
tion. 2-Hydroxymandelic acid was the only product found in
the perchloric acid and sodium hydroxide solutions and in
the acetic acid buffer as well. 2-Hydroxymandelic acid was
also the principle product formed in the tris-(hydroxy-
methyl)methylammonium ion and the ammonium ion buff-
ers, but a minor amount of a second, unidentified product
was also formed in each of these buffers. These minor prod-
ucts could be the corresponding 2-hydroxymandelic acid
amides formed by aminolysis of the lactone starting material
by the amine buffer components.
isotope effect, kH+ /k + < 1 (7), contrary to the present result.
D
3
-Diazobenzofuran-2-one
Rates of hydrolysis of 3-diazobenzofuran-2-one were
The next higher homolog of the presently examined
substrate, 4-diazoisochroman-3-one (6), also undergoes
2
Tables of rate data (Tables S1–S5) have been deposited and may be purchased from the Depository of Unpublished Data, Document Deliv-
ery, CISTI, National Research Council Canada, Ottawa, ON K1A OS2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information
on ordering electronically).
©
2002 NRC Canada

Chiang et al.
85
Fig. 3. Rates of hydrolysis of 3-hydroxybenzofuran-2-one in con-
Fig. 4. Rate profile for the hydrolysis of 3-hydroxybenzofuran-2-
centrated perchloric acid solutions.
one in aqueous solution at 25°C.
acid-catalyzed hydrolysis by a rate-determining proton
transfer mechanism (9); its hydrogen-ion catalytic coeffi-
cient, however, is 15 times greater than that for the pres-
ently examined substrate. This difference in reactivity can
be understood in terms of the various resonance forms
that contribute to the structure of these diazo lactones.
One of the principle resonance forms of the diazo group
puts negative charge on the adjacent carbon atom, as in 7,
and, since this is the site to which the positively charged
proton is transferred, the rate of reaction will increase
with increasing contribution of this form. This negative
charge can also be shifted onto the oxygen atom of the
lactone carbonyl group, as in 8, and, since this removes
negative charge from the reactive position, the rate of re-
action will decrease with increasing contribution of this
form. The lactone carbonyl group, however, is also conju-
gated with the lactone ether oxygen atom, giving rise to
resonance form 9, which opposes conjugation of the car-
bonyl with the diazo group, thus raising the contribution
of form 7 and increasing the rate of reaction. This oppos-
ing conjugation will be stronger in the case of 4-diazoiso-
chroman-3-one than in the case of 3-diazobenzofuran-2-
one, because the latter is a phenolic lactone whose ether
oxygen atom is also conjugated with its benzene ring. In
the isochromanone system, therefore, as compared to the
benzofuranone system, the contribution of form 9 will be
greater, that of form 8 less, and that of form 7 greater,
leading to a greater rate of reaction.
This argument is supported by the fact that N,N-dimethyl-
phenyldiazoacetamide (10) is considerably more reactive to-
ward acid-catalyzed hydrolysis than is the corresponding
ester, methylphenyldiazoacetate (11) (10). This is in keeping
3-Hydroxybenzofuran-2-one (concentrated acids)
Rates of hydrolysis of 3-hydroxybenzofuran-2-one were
measured in concentrated perchloric acid solutions over the
concentration range [HClO ] = 0.28–6.3 M. The data ob-
4
with the greater electron-donating ability of NMe over OMe
2
2
tained are summarized in Table S2.
and the consequent greater contribution to the structure of
the amide made by the resonance form corresponding to 9.
Figure 1 shows that observed first-order rate constants for
this hydrolysis reaction in these concentrated acid solutions
increase with acidity much less steeply than do those for the
hydrolysis of 3-diazobenzofuran-2-one, and Fig. 3, where
the data are plotted on a nonlogarithmic scale, reveals that
this is because acid catalysis of the hydroxy substrate be-
comes saturated at the high concentration end of the range
investigated. This saturation of acid catalysis suggests that a
nonproductive protonation of 3-hydroxybenzofuran-2-one is
taking place, in addition to the productive protonation that
©
2002 NRC Canada

8
6
Can. J. Chem. Vol. 80, 2002
results in lactone hydrolysis; the molecule does, of course,
have such a nonproductive protonation site in its hydroxyl
group, as shown by the reaction scheme of eq. [6].
The presence of this nonproductive protonation requires
pKa ≅ –2.0 (12). This, however, is again as might be
expected, inasmuch as hydrogen bonding between the posi-
tively charged hydroxyl group and the adjacent carboxyl ox-
ygen atom in the present system will stabilize the
undissociated acid and thus, reduce its acid strength.
It is interesting that results similar to those produced by
the analysis described above may be obtained by treating the
data in a much simpler way, without recourse to an acidity
function, using the expression of eq. [9] as the correlating
equation.
modification of the simple X correlation function of eq. [4]; a
o
term must be added to take this additional reaction into ac-
count. A further modification of eq. [4] is required, moreover,
by the occurrence of an uncatalyzed lactone hydrolysis reac-
tion (eq. [7]) in addition to the acid-catalyzed process. The
presence of such an uncatalyzed reaction is evidenced by the
significant rate intercept at zero acid concentration shown by
Fig. 3 and, perhaps more convincingly, by the long central
horizontal plateau shown by the rate profile for this reaction
(
Fig. 4), constructed as detailed below. In addition, it seems
+
+
+
+
likely that the acid-catalyzed reaction occurs by a simple A2
mechanism in which water is an active reactant (11). It has
been shown in simpler systems that the reaction order in wa-
ter activity for such a process is two (11) and a term taking
this into account should also be added to eq. [4].
[9]
kobs = (kH
+
[H ] + kuc)Ka /(Ka + [H ])
Least-squares fitting of this expression gives the results
–
3
–1 –1
+
k
H+
= (5.07 ± 0.84) × 10
M
s , Ka = 1.44 ± 0.29 M,
+
–3 –1
pKa = –0.16 ± 0.09, and kuc = (1.12 ± 0.27) × 10 s ,
which are not significantly different from values obtained by
the X analysis. This simple treatment is successful in this
The final correlating function incorporating all of these
o
+
case because the dependence of Ka and kH
+
on X are both
o
quite shallow and approximate quite well a simple depend-
+
ence on [H ].
3
-Hydroxybenzofuran-2-one (dilute solutions)
Rates of hydrolysis of 3-hydroxybenzofuran-2-one were
also measured in dilute aqueous perchloric acid solutions
additional features is shown as eq. [8], in which the rate and
equilibrium constants are defined by eqs. [6] and [7],
over the concentration range [HClO ] = 0.0002–0.1 M, in di-
4
lute aqueous sodium hydroxide solutions over the concentra-
tion range [NaOH] = 0.003–0.1 M, and in acetic acid,
biphosphate ion, tris-(hydroxymethyl)methylammonium ion,
and ammonium ion buffers. The ionic strength of these solu-
tions was maintained at 0.10 M by the addition of sodium
perchlorate as required. These data are summarized in Ta-
+
[
8]
log {(kobs – k )/[H ]} = log k
H+
uc
‡
+
+
+
mX
o
+
m X + log {K /(K + [H ]10 } + 2 log a
o
a a H O
2
a
is the activity of water, supplied here on the mol frac-
tion scale, and m and m are the X slope parameters for the
rate and equilibrium processes, respectively. This expression
contains five dependent parameters to be determined by the
fit of experimental data (k , k , K , m , and m), which, un-
fortunately, proved to be too many: least-squares analysis
failed to produce a sensible result. The number of parame-
ters could be reduced to four, however, by supplying a value
of kuc (= 0.001 s ) based on the intercept at zero acid con-
centration of Fig. 3 and the central plateau of Fig. 4, and this
modification of eq. [8] did produce a sensible result. Least-
H O
2
‡
o
2
bles S3–S5.
The rate measurements in buffers were performed using a
series of solutions of varying buffer concentration but con-
stant buffer ratio and, since the ionic strength was constant,
also constant hydrogen ion concentration. Observed first-
order rate constants determined in each solution series in-
creased linearly with increasing buffer concentration, and
the data were, therefore, analyzed using the buffer dilution
expression shown in eq. [10]. Least-squares fitting produced
+
‡
H+
uc
a
–
1
–
3
–1 –1
squares analysis gave k
+
= (4.49 ± 0.17) × 10
K = 1.48 ± 0.17 M, pK = –0.17 ± 0.05, m = 0.68 ± 0.11,
M
s ,
H
zero-buffer-concentration intercepts (k ) which,
int
+
+
‡
a
a
and m = 0.04 ± 0.12.
[
10] kobs = kint + k [buffer]
cat
Very shallow acidity function dependences, similar to
the m value obtained here, have been found for the equi-
librium protonation of simple alcohols, whereas somewhat
steeper dependences, consistent with the present value of
together with the observed first-order rate constants deter-
mined in perchloric acid and sodium hydroxide solutions,
were used to construct the rate profile shown in Fig. 4. Hy-
drogen ion concentrations of the buffer solutions needed for
this purpose were obtained by calculation, using thermody-
namic acidity constants for the buffer acids from the litera-
ture and activity coefficients recommended by Bates (13).
‡
m , occur in the hydrolysis of simple esters (12). The re-
sult pK a+ = –0.17, on the other hand, makes the conjugate
acid of 3-hydroxybenzofuran-2-one somewhat less strongly
acidic than oxygen-protonated simple alcohols, for which
©
2002 NRC Canada

Chiang et al.
87
This rate profile shows the hydrogen-ion-catalyzed and uncatalyzed reactions plus saturation of hydrogen-ion catalysis
noted above, and it also shows a hydroxide-ion-catalyzed process with saturation of hydroxide-ion catalysis as well. This satu-
ration of hydroxide-ion catalysis implies incursion of yet another nonproductive ionization of the substrate, and a complete
o
rate law formulated by adding this feature to the rate expression of eq. [9] is shown as eq. [11]; Q is the equilibrium constant
a
for this nonproductive ionization and Q is the ionization constant of water.
w
+
+
+
+
o
+
[
11] kobs = (kH
+
[H ] + k )K /(K + [H ]) + k
HO-
Q /(Q + [H ])
w
a
uc
a
a
–
3
–1 –1
–3 –1
Least-squares analysis using this expression gave the results kH = (5.42 ± 0.43) × 10 M s , k = (1.01 ± 0.02) × 10 s ,
+
uc
+
+
3
–1 –1
o
–12
o
K = 1.33 ± 0.15 M, pK = –0.12 ± 0.05, k
-
= (6.37 ± 0.17) × 10 M s , and Q = (2.75 ± 0.10) × 10 M, pQ =
a
a
HO
a
a
3
+
1
1.56 ± 0.02. These values of k , k , and K agree well with determinations of these quantities obtained by the other treat-
H+
uc
a
ments of the experimental data described above; the weighted averages of all results are k = (4.63 ± 0.16) × 10–3
M
–1
s ,
–1
H+
kuc = (1.01 ± 0.02) × 10–3 s , and K = 1.40 ± 0.11 M, pK = –0.15 ± 0.03.
–1
+
+
a
a
There are two candidates for the nonproductive ionization of 3-hydroxybenzofuran-2-one that could be responsible for satu-
ration of hydroxide-ion catalysis of the hydrolysis of its lactone functional group: (i) ionization of the hydroxyl group to give
an alcoholate anion (eq. [12]); and (ii) ionization of the carbon—hydrogen bond next to the lactone function to give an enolate
ion (eq. [13]). Both of these reactions put negative charge near the lactone group and thus, hinder reaction of this group with a
negatively charged hydroxide ion.
It is difficult to choose between these two alternatives. A Hammett–Taft relation correlating acidity constants of secondary
alcohols (14) leads to the estimate pK = 12.0 for the hydroxyl group ionization of eq. [12], which is consistent with the ex-
a
o
perimentally determined value pQ = 11.6. On the other hand, pK = 12.3 has been reported for the carbon acid ionization of
a
a
benzofuran-2-one itself (15) (12) (eq. [14]), which again is not inconsistent with the presently determined value.
The carbon acid ionization reaction of eq. [13] raises the possibility of an alternative lactone hydrolysis route, replacing the
traditional attack of hydroxide ion on the lactone carbonyl carbon atom leading to a tetrahedral intermediate. This alternative
route (eq. [15]) has the enolate ion (13) breaking down to give a ketene (14), which then rapidly hydrates to a carboxylic acid
enolate ion (15), and that ketonizes to the final 2-hydroxymandelic acid product (16).
The generation of ketenes from ester enolates like this is a well-known reaction (16), and the hydration and ketonization of
analogs of 14 and 15 without the ortho-phenoxide group are known to be very fast reactions (17). On the other hand, although
3
-phenylbenzofuran-2-one (17) readily ionizes to its enolate ion (18) (eq. [16]), hydrolysis of this lactone is believed to occur
by the traditional tetrahedral intermediate route and not through the corresponding ketene (15).
3
This is a concentration dissociation constant applicable at the ionic strength (0.10 M) at which it was determined.
©
2002 NRC Canada

8
8
Can. J. Chem. Vol. 80, 2002
7
8
9
. J.R. Keefe and A.J. Kresge. Investigation of rates and mecha-
nisms of reactions. In Techniques of chemistry. Vol. VI. Edited
by C.F. Bernasconi. Wiley–Interscience, New York. 1986.
Chap. XI.
. (a) R.P. Bell. The proton in chemistry. Cornell Univ. Press,
Ithaca, New York. 1973. pp. 164–171; (b) M. Regitz and G.
Maas. Diazo compounds, properties and synthesis. Academic
Press, New York. 1986. pp. 122–134.
Ac knowle dge m e nt
We are grateful to the Natural Sciences and Engineering
Research Council of Canada (NSERC) and the United States
National Institutes of Health for financial support of this
work.
. E.A. Jefferson, A.J. Kresge, and S.W. Paine. Can. J. Chem. 74,
1
369 (1996).
Re fe re nc e s
1
0. M.-H. Bui-Nguyen, H. Dahn, and J.F. McGarrity. Helv. Chim.
1
2
. O.L. Chapman, C.-C. Chang, J. Kolc, N.R. Rosenquist, and H.
Tomioka. J. Am. Chem. Soc. 97, 6586 (1975).
. (a) R. Howe, B.S. Rao, and H. Heyneker. J. Chem. Soc. (C),
Acta, 63, 63 (1980).
11. R.A. Cox. Adv. Phys. Org. Chem. 35, 1 (2000).
12. A. Bagno, G. Scorrano, and R.A. More O’Ferrall. Rev. Chem.
Intermed. 7, 313 (1987).
2
510 (1967); (b) A.J. Hoefnagel, J.A. Peters, and H. van
Bekkum. Rec. Trav. Chim. Pays–Bas, 115, 353 (1996).
. W.M. Horspool and G.D. Khandelwal. J. Chem. Soc. (C), 3328
13. R.G. Bates. Determination of pH theory and practice. Wiley–
Interscience, New York. 1973. p. 49.
3
4
(
1971).
14. D.D. Perrin, B. Dempsey, and E.P. Serjeant. pK Prediction for
a
. A.J. Kresge, H.J. Chen, G.L. Capen, and M.F. Powell. Can. J.
Chem. 61, 249 (1983).
organic acids and bases. Chapman and Hall, New York. 1981.
p. 127.
5
6
. R.A. Cox and K. Yates. Can. J. Chem. 59, 2116 (1981).
. (a) E. Högfeldt and J. Bigeleisen. J. Am. Chem. Soc. 82, 15
15. D.M. Heathcote, G.A. DeBoose, J.H. Atherton, and M.I. Page.
J. Chem. Soc. Perkin Trans. 2, 535 (1998).
(
(
1960); (b) J. Sierra, M. Ojeda, and P.A.H. Wyatt. J. Chem. Soc.
B), 1570 (1970); (c) R.A. Cox, S.-O. Lam, R.A. McClelland,
16. T.T. Tidwell. Ketenes. Wiley, New York. 1995. p. 67.
17. Y. Chiang, A.J. Kresge, V.V. Popik, and N.P. Schepp. J. Am.
Chem. Soc. 119, 10 203 (1997).
and T.T. Tidwell. J. Chem. Soc. Perkin Trans. 2, 272 (1979).
©
2002 NRC Canada