Carboxylic acid amide enols: Keto-enol/enolate interconversion of N-methylindoline-2-one and its 2-thione and 2-selone analogs in aqueous solution

Article abstract of DOI:10.1139/v99-169

Carbon-acid ionization constants, Q(a)(K) (concentration quotiem at ionic strength = 0.10 M), were determined by spectrophotometric titration in aqueous solution for N- methylindoline-2-one, pQ(a)(K) = 15.70, N-methylindoline-2-thione, pQ(a)(K) = 8.93, and N-methylindoline-2-selone, pQ(a)(K) = 7.25. Rate profiles were also constructed for the thione and selone. These were interpreted, with support from the form of acid-base catalysis as well as solvent and substrate isotope effects, as representing keto - enol/enolate ion interconversion. That led to the enol acidity constant pQ(a)(E) = 4.05 and the keto-enol equilibrium constant pK(E) = 4.88 for the thione and estimates of the limits on these quantities, pQ(a)(E) < 3 and pK(E) > 4, for the selone.

Full text of DOI:10.1139/v99-169

1528
Carboxylic acid amide enols: keto–enol/enolate
interconversion of N-methylindoline-2-one and
its 2-thione and 2-selone analogs in aqueous
solution
A. Jerry Kresge and Qingshui Meng
K
Abstract: Carbon-acid ionization constants, Qa (concentration quotient at ionic strength = 0.10 M), were determined
K
by spectrophotometric titration in aqueous solution for N-methylindoline-2-one, pQa = 15.70, N-methylindoline-2-
K
K
thione, pQa = 8.93, and N-methylindoline-2-selone, pQa = 7.25. Rate profiles were also constructed for the thione
and selone. These were interpreted, with support from the form of acid–base catalysis as well as solvent and substrate
E
isotope effects, as representing keto – enol/enolate ion interconversion. That led to the enol acidity constant pQa
=
4.05 and the keto–enol equilibrium constant pKE = 4.88 for the thione and estimates of the limits on these quantities,
E
pQa < 3 and pKE > 4, for the selone.
Key words: lactam, thiolactam, selenolactam, keto–enol equilibria, carbon-acid ionization constants.
Résumé : Opérant en solution aqueuse et faisant appel à des titrages spectrophotométriques, on a mesuré les constantes
K
d’ionisation d’acide carboné, Qa (quotient des concentrations à une force ionique = 0,10 M), de la N-méthylindoline-
K
K
K
2-one, pQa = 15,70, de la N-méthylindoline-2-thione, pQa = 8,93 et de la N-méthylindoline-2-sélénone, pQa = 7,25.
On a aussi élaboré les profils de vitesse de la thione et de la sélénone. En se basant sur la forme de la catalyse acide–
base ainsi que sur les effets isotopiques du solvant et du substrat, on peut interpréter ces résultats sur la base d’une
E
interconversion cétoénolique/ion énolate. Sur cette base, la constante d’acidité énolique, pQa = 4,05, et la constante
d’équilibre cétoénolique, pKE = 4,88, pour la thione alors que les évaluations des limites de ces grandeurs pour la
E
sélénone sont pQa < 3 et pKE > 4.
Mots clés : lactame, thiolactame, sélénolactame, équilibre cétoénolique, constantes d’ionisation d’acide carboné.
[Traduit par la Rédaction] Kresge and Meng 1536
Enol isomers of simple carboxyic acids and their func-
where the enol is stabilized by some special structural
features. Bulky groups in the β-position of enols can accom-
plish this by blocking access to the enol double bond (5),
through whose protonation ketonization must take place (6),
and this strategy has recently been used to make stable enols
of bis(2,4,6-triisopropylphenyl)acetic acid amides (7).
In the present study of amide enols, we have used a some-
what different approach: we have examined systems in
which the enol or enolate ion is stabilized by formation of a
new aromatic ring, eq. [1], and also by replacement of the
carbonyl oxygen atom with sulfur or selenium. Replacement
of oxygen by sulfur has recently been shown to increase
enol content by a large factor (8), and it seemed likely that
selenium would have a similar effect.
tional derivatives, such as esters and amides, are very labile
and thermodynamically unstable substances, and their life-
times in aqueous solution are expected to be exceedingly
short. For example, the keto–enol equilibrium constant for
acetic acid in aqueous solution, though not yet determined
experimentally, has been reliably estimated as pKE = 20 (1),
and the hydronium-ion catalytic coefficient for conversion
of acetic acid enol to its keto form, also not yet measured
+
experimentally, is believed to lie in the range kH = 108–109
M–1 s–1 (2). Recent high-level calculations (3) suggest that
keto–enol equilibrium constants for esters and amides will
be not very different from those for carboxylic acids, and
this is borne out by the estimate pKE = 19 for ethyl acetate
(4), similar to pKE = 20 for acetic acid.
We have found that the enol isomers (2a, b, and c) and
enolate ions (3a, b, and c) of N-methylindoline-2-one (1a),
N-methylindoline-2-thione (1b), and N-methylindoline-2-
Daunting realities such as this have driven the study of
enols of carboxylic acids and their derivatives to systems
XH
X– + H+
X
–
–
–
N
N
N
[1]
2
3
1
(a: X = O, b: X = S, c: X = Se)
Received June 25, 1999.
A.J. Kresge1 and Q. Meng. Department of Chemistry, University of Toronto, Toronto, ON M5S 3H6, Canada.
1Author to whom correspondence may be addressed. Telephone: (416) 978-7259. Fax: (416) 978-7259.
e-mail: akresge@chem.utoronto.ca
Can. J. Chem. 77: 1528–1536 (1999)
© 1999 NRC Canada

Kresge and Meng
1529
selone (1c) are stable enough to exist in aqueous solution.
This has allowed us to determine acidity constants for these
substances ionizing as carbon acids in that medium. We
have also found that rates of keto–enol/enolate interconver-
sion in the thione and selone systems can be measured by
stopped flow methods, and this has enabled us to determine
keto–enol and enol acidity constants for the thione system
and to set limits on these quantities for the selone system.
25.0 ± 0.05°C. Constant stoichiometric substrate concentra-
tions (2 × 10–5 M) were used, and the data were analyzed by
nonlinear least-squares fitting of the titration curve expres-
K
sion shown as eq. [2], where Qa is the carbon-acid ioniza-
tion constant of the substrate, and AHA and AB are the
limiting absorbances of its acidic and basic forms, respec-
tively.
K
K
[2]
A = (ABQa + AHA[H+])/(Qa + [H+])
Experimental section
Kinetics
Rates of approach to keto–enol/enolate equilibrium in
Materials
the N-methylindoline-2-thione and -selone systems were
determined using a Hi-Tech Scientific Model SF-S1
stopped-flow spectrometer operating at 25.0 ± 0.1°C. For
studies on solutions more basic than the carbon-acid disso-
N-Methylindoline-2-one was prepared by methylating oxindole
(Aldrich) using a published procedure (9).
N-Methylindoline-2-thione was prepared by treating N-
methylindole-2-one (1.00 g) with Lawesson’s reagent
(10) (1.64 g) in dry toluene at 100°C for 2 h under an argon
atmosphere. The solvent was removed, and the residue was
purified by flash chromatography on silica gel with 20%
ethyl acetate in pentane as the eluent. Recrystallization from
ethyl acetate:pentane gave a solid, mp 109–110°C (lit. (9)
106.5–108°C) whose NMR spectra were consistent with lit-
erature values (9, 11).
K
ciation constant of the substrate, Qa , kinetic runs were ini-
tiated by mixing a solution of substrate in water contained
in one syringe with a solution of sodium hydroxide or
buffer contained in the other syringe, and for studies in so-
K
lutions more acidic than Qa , runs were initiated by mixing
a solution of initially deprotonated substrate in 0.0002 M
sodium hydroxide solution in one syringe with a solution
of perchloric acid or buffer in the other syringe. Reactions
were monitored by following changes in enol and enolate
absorbance at λ = 240 nm or keto form absorbance at λ =
310 nm for N-methylindoline-2-thione and changes in keto
form absorbance at λ = 330 nm for N-methylindoline-2-
selone. Substrate concentrations in the final reaction mix-
tures were 5 × 10–6 M. The rate data conformed to the first-
order rate law well, and observed first-order rate constants
were obtained by least-squares fitting of an exponential
function.
N-Methylindoline-2-selone was prepared by adapting a
method used to synthesize the parent N–methyl-2-pyrroline-
2-selone (12). To a solution of 2.0 g N-methylindole in
30 mL tetrahydrofuran maintained at 0°C in an argon atmo-
sphere, 8.5 mL of 1.8 M phenyllithium dissolved in 70:30
cyclohexane:ether was added. When the addition was com-
plete, the temperature was raised to 40°C, and the reaction
mixture was stirred for 2 h. It was then cooled to 0°C, 1.2 g
of selenium was added, and stirring was continued for an-
other hour. The resulting reaction mixture was quenched
with cold aqueous acetic acid, was extracted with chloro-
form, and the extract was washed with water and brine and
was dried over magnesium sulfate. The residue obtained
upon evaporation of the solvent was purified by flash chro-
matography on silica gel using 5% ethyl acetate in hexane as
the eluent. Repeated recrystallization from chloroform:hex-
ane gave 100 mg (3% yield) of a yellow solid; mp: 131–
Results
Carbon-acid ionization constants
For N-methylindoline-2-thione and -selone, reversible UV
spectral changes characteristic of carbon-acid ionization
took place in aqueous solution in the pH range, and mea-
surements could consequently be made in dilute sodium hy-
droxide and buffer solutions. For N-methylindoline-2-one,
on the other hand, these spectral changes occurred only in
more basic solutions, and measurements for this substrate
were therefore made in concentrated aqueous potassium hy-
droxide solution.
1
133°C; H NMR (CDCl3), δ/ppm: 7.02–7.36 (m, 4 H), 3.81
(s, 2 H), 3.70 (s, 3 H); 13C NMR (CDCl3), δ/ppm: 203.25,
146.79, 131.17, 128.08, 124.72, 124.03, 110.01, 53.63,
34.03; HRMS, m/e: 210.9895 (calc.), 210.9900 (obs.). Only
part of the starting material had reacted, suggesting that a
higher yield would have been obtained if a stronger base,
such as butyllithium, had been used.
–
+
+
For the thione and selone, HCO3 , NH4 , (CH2OH)3CNH3 ,
–
H2PO4 , and CH3CO2H buffers were used. Two to four dif-
All other materials were best available commercial grades.
ferent buffer ratios were employed for each buffer system,
and several different sodium hydroxide concentrations were
used as well; triplicate absorbance measurements were made
with each solution. The ionic strength of all solutions was
maintained at 0.10 M using NaClO4 as required. Hydrogen
ion concentrations were obtained by calculation using litera-
ture pKa values and activity coefficients recommended by
Bates (13). These data are summarized in Table S1.2
Acidity constants
Carbon-acid ionization constants of N-methylindoline-2-
one, -thione, and -selone were determined by monitoring the
reversible UV spectral changes that these substances un-
dergo upon ionization. Absorbance (A) measurements were
made at λ = 310 nm for N-methylindoline-2-one, at λ =
240 nm for N-methylindoline-2-thione, and at λ = 330 nm
for N-methylindoline-2-selone, using a Cary Model 2200
spectrometer whose cell compartment was thermostatted at
As Fig. 1 illustrates, the data so obtained conformed to
the expected titration curve expression of eq. [2] quite well.
Least-squares fitting gave Qa = (1.18 ± 0.04) × 10–9 M,
K
2 Tables of equilibrium and rate data (Tables S1–S8) have been deposited and may be purchased from: The Depository of Unpublished Data,
Document Delivery, CISTI, National Research Council Canada, Ottawa, Canada, K1A 0S2.
© 1999 NRC Canada

1530
Can. J. Chem. Vol. 77, 1999
Fig. 1. Spectrophotometric titration curve for N-methylindoline-
2-thione in aqueous solution at 25°C.
Fig. 2. Rate profile for keto–enol/enolate interconversion in the
N-methylindoline-2-thione system in aqueous solution at 25°C;
circles and solid line: rate of approach to equilibrium, dashed
line: rate of enolization, dotted line: rate of ketonization.
pQa = 8.93 ± 0.023(see footnote 3) for N-methylindoline-2-
K
K
K
thione, and Qa = (5.66 ± 0.29) × 10–8 M, pQa = 7.25 ±
Kinetics: N-methylindoline-2-thione
0.02 (see footnote 3) for N-methylindoline-2-selone.
Rates of approach to keto–enol/enolate ion equilibrium for
N-methylindoline-2-thione were measured in aqueous per-
chloric acid solutions over the concentration range 0.001–
0.05 M, and in aqueous sodium hydroxide solutions over the
concentration range 0.0004–0.1 M. The ionic strength of
these solutions was maintained at 0.10 M using NaClO4 as
required. These data are summarized in Tables S3 and S4.2
Rate measurements for this substrate were also made in
Absorbance measurements for N-methylindoline-2-one
were made in KOH solutions over the concentration range
0.1–11 M. The data so obtained, summarized in Table S2,2
were transformed into values of log I (= [enolate ion]/[keto
form]), and the fit of log I to different basicity functions was
examined. Three functions were tried: an H_ scale based on
neutral indole indicators (14), an H= scale based on indoles
bearing carboxylate groups (14), and an H= scale based on
aromatic amines bearing carboxylate and sulfonate groups
(15). The H_ scale was found to be the most appropriate: it
gave a plot of log I vs. H_ whose slope was near unity (0.95
± 0.02), whereas the other basicity functions produced
slopes significantly farther from unity (0.77 ± 0.01 and 0.78
± 0.01). The data were therefore analyzed using the H_
scale, i.e., were fitted using the titration curve expression of
–
+
+
aqueous CH3CO2H, H2PO4 , (CH2OH)3CNH3 , and NH4
buffer solutions, using series of buffer solutions of different
total buffer concentration but constant buffer ratio and con-
–
stant ionic strength (0.10 M). For the H2PO4 and
+
(CH2OH)3CNH3 buffers, this served to hold [H+] constant
along a given buffer series, but in some of the CH3CO2H
+
and NH4 buffers “buffer failure” (18) occurred; i.e., [H+]
K
eq. [2] with 10–H_ replacing [H+]. This gave Ka = (1.40 ±
changed systematically somewhat as the buffer was made
more dilute. In these cases, observed rate constants were ad-
justed to values they would have had had [H+] been con-
stant; these adjustments were made using required rate and
equilibrium constants obtained from the rate profile (vide in-
fra). Such adjustments were minor — never more than a few
percent. The data are summarized in Table S5.2
Buffer catalysis was strong, with rate constants increasing
with increasing buffer concentration in a linear fashion. The
data were therefore analyzed by least-squares fitting of
eq. [3] in which kexp represents either observed or adjusted
rates of approach to equilibrium.
K
0.08) × 10–16 M, pKa = 15.86 ± 0.03.
Since basicity functions are referenced to an infinitely di-
lute standard state whose ionic strength is zero, this value is
a thermodynamic ionization constant, expressed in terms of
activities, unlike the concentration constants determined
above for N-methylindoline-2-thione and -selone. This ther-
modynamic value may be converted into a concentration
ionization constant by applying appropriate activity coeffi-
cients. The result for ionic strength = 0.10 M, obtained using
the activity coefficient for the hydrogen ion recommended
by Bates (13) and that for picrate ion (16) as a surrogate for
the unavailable activity coefficient of the present enolate
[3]
kexp = ko + kBuff[Buffer]
K
ion, is pQa = 15.70 ± 0.03.
Carbon-acid ionization constants for two of the substrates
studied here, N-methylindoline-2-one and N-methylindoline-
2-thione, have been measured in DMSO solution (9). The re-
sults obtained, pKa = 18.5 and 10.0, respectively, show these
substances to be stronger acids in H2O than in DMSO by
relatively small amounts, as expected for acids whose ion-
ization produces highly delocalized anions (17).
Values of ko obtained in this way, together with the rate
constants measured in perchloric acid and sodium hydroxide
solutions, were used to construct the rate profile shown in
Fig. 2. This profile describes the rate of approach to keto–
enol/enolate equilibrium by reaction through solvent-related
species. It can be analyzed as the sum of rate constants
for enolization occurring through rate-determining proton
3 This is a concentration ionization constant, applicable at the ionic strength of its determination.
© 1999 NRC Canada

Kresge and Meng
1531
removal from the keto isomer by either hydroxide ion or a water molecule to give an equilibrium mixture of enol and enolate,
eq. [4], plus rate constants for ketonization occurring through rate-determining carbon protonation of enolate ion, in equilib-
rium with enol, by either hydronium ion or a water molecule, eq. [5].
HO–
H2O
HO–
–
kE
HO
S–
SH +
+
S
+
[4]
[5]
kE
N
N
N
H3O+
H2O
o
H2O
H2O
H3O+
H2O
H2O
+
kK
H
S–
+
+
S
SH +
kK
N
N
N
o
HO–
HO–
This interpretation is consistent with the established reaction
mechanisms for enolization and ketonization (6), and it is
supported in the present instance by isotope effects and the
form of acid–base catalysis by buffer species (vide infra).
The rate law for this reaction scheme is given in eq. [6],
analogous expression for kK may be derived using the
process shown in eq. [9] whose equilibrium constant, the
o
kK
o
S–
S
+ H2O
+ HO–
[9]
kE
–
N
N
E
HO
whose rate constants are defined by eqs. [4] and [5], and Qa
[6]
kexp = kE + kE [HO–]
HO−
o
K
ionization constant of water Qw divided by Qa , is also equal
+ (kK + kK [H+]){Qa /(Qa + [H+])}
to the rate constant ratio kKo/kE ; this leads to kK
=
E
E
H+
HO−
o
o
K
QwkE /Qa . Combination of these expressions then gives
HO−
is the acidity constant of the enol ionizing as an oxygen
acid. Least-squares fitting of this expression as it stands
the relationship shown in eq. [10]. Since all the quantities on
[10] kE /kK = kK (QaK)2/ kE Qw
could not be effected, because the fraction modifying kK ,
H+
HO−
o
o
o
E
E
Qa / (Qa + [H+]), is essentially unity in the region where
the right side of eq. [10] are known, the ratio kE /kK can be
this rate constant makes a significant contribution: [H+] =
o
o
evaluated, and combination of that with the sum kE + kK
E
10–8–10–9 M (Qa
10–4 M, vide infra). That makes the rela-
o
o
obtained from the rate profile produces the results kE
=
o
tionship between this rate constant and solvent acidity effec-
(4.72 ± 0.87) × 10–3 s–1 and kKo = (3.27 ± 0.65) × 10–2 s–1.
tively the same as that of kE ; these two rates are therefore
o
These results, coupled with values of kE and kK ob-
HO−
H+
strongly correlated and cannot be evaluated separately by
analysis of the rate profile. They are consequently combined
as kEo + kKo = kuc, giving eq. [7]. Least-squares fitting of this
tained from the rate profile, provide complete rate laws for
the enolization and ketonization reactions effected by sol-
vent-related species over the range of acidity studied. This
allows separation of observed rates of equilibration into
rates of enolization and rates of ketonization, as illustrated
in Fig. 2. Such dissection shows that ketonization dominates
observed rates of equilibration in acidic solutions down to
about [H+] = 10–8 M and that enolization dominates in basic
solutions beginning at about [H+] = 10–9 M.
These dominances are consistent with the nature of buffer
catalysis in the buffer solutions examined. The buffer cata-
lytic coefficients of eq. [3] may be separated into general
acid, kHA, and general base, kB, contributions with the aid of
eq. [11], in which fA is the fraction of buffer present in the
[7]
kexp = kE [HO–] + kuc
HO−
+ kK [H+]{Qa /(Qa + [H+])}
E
E
H+
expression produced kE
= (3.42 ± 0.06) × 103 M–1 s–1,
HO−
kuc = (3.75 ± 0.65) × 10–2 s–1, kK = (5.62 ± 0.27) ×
H+
E
E
106 M–1 s–1, and Qa = (8.85 ± 0.54) × 10–5 M, pQa = 4.05
K
± 0.03.3 Combination of the latter constant with Qa accord-
ing to the relationship Qa /QaE = KE then gave the keto–enol
K
equilibrium constant KE = (1.33 ± 0.10) × 10–5, pKE = 4.88
± 0.03.
Although only the sum of the rate constants kE and kK
o
o
can be evaluated from the rate profile, individual values may
[11] kBuff = kB + (kHA – kB)fA
be obtained by combining this sum with the ratio kE /kK de-
o
rived in the following way. The equilibrium constant foor the
acid form. The data obtained obeyed this relationship well;
an example is shown in Fig. 3. Least-squares analysis gave
the results listed in Table 1.4 This analysis showed that only
general acid catalysis was significant in CH3CO2H and
process shown in eq. [8], QK , is also equal to the ratio of
a
K
forward to reverse rate constants for this reaction: Qa
=
kE /kK ; this leads to the expression kE = Qa kK . An
K
H+
H+
o
o
–
H2PO4 buffers, which is consistent with the fact that these
buffers gave solutions in the region [H+] = 1 × 10–4 to 6 ×
10–8 M where ketonization dominates; ketonization, being a
reaction involving rate-determining proton transfer from
acid catalyst to substrate, can be expected to show only gen-
kE
kK
o
+ H3O+
S–
+ H2O
S
[8]
+
N
N
H
4 Data obtained in acetic acid buffers were adjusted to allow for the fact that the enolate ion exists in equilibrium with significant amounts of
enol in these solutions.
© 1999 NRC Canada

1532
Can. J. Chem. Vol. 77, 1999
Table 1. General-acid and general-base catalytic coefficients for enolization and ketonization in the N-methylindoline-2-thione
and -selone systems in aqueous solution at 25°C.a
Catalyst
Thione
kHA/102 M–1 s–1
Selone
kHA/102 M–1 s–1
Acid
Base
kB/102 M–1 s–1
kB/102 M–1 s–1
–
CH3CO2H
H2PO4
CH3CO2
HPO4
327.
16.0
3.06
0.114
—
—
0.460
0.247
44.6
2.21
0.335
—
—
–
=
0.551
2.31
1.38
+
(CH2OH)3CNH3
(CH3OH)2CCNH2
NH3
+
NH4
a
Ionic strength = 0.10 M.
+
+
Fig. 3. Relationship between buffer catalytic coefficients and
fraction of buffer present in the acid form for keto–enol/enolate
interconversion of the N-methylindoline-2-thione system in
eral acid catalysis. The data for (CH2OH)3CNH3 and NH4
buffers, on the other hand, gave significant general base as
well as general acid catalysis, which is consistent with the
fact that these buffers gave solutions in the region [H+] = 3 ×
10–8 to 1 × 10–10 M where both enolization and ketonization
make important contributions to observed rates; enolization,
being a reaction that involves rate-determining proton re-
moval from the substrate by a basic catalyst will show gen-
eral base catalysis.
–
=
aqueous H2PO4 /HPO4 buffer solutions at 25°C.
The numerical values of the general acid and general base
catalytic coefficients obtained by this analysis offer further
support for the interpretation made. The equilibrium con-
stant for keto–enolate interconversion catalyzed by a general
K
acid–base pair, shown in eq. [12], has the value Qa /QaHA, in
kE
B
S–
[12]
+ HA
+ B
S
kK
N
N
HA
HA
which Qa is the acidity constant of the catalyzing acid.
This equilibrium constant is also equal to the rate constant
Further support for the interpretation made comes from
isotope effects. These are provided by rate measurements
performed in D2O solutions of DClO4, summarized in Ta-
ble S3,2 and in CH3CO2D buffers, summarized in Table S5.2
The measurements in buffers were done in series of solu-
tions of constant buffer ratio and ionic strength, just like the
measurements in H2O solutions of CH3CO2H, and the data
were treated in an analogous fashion. Only general acid ca-
talysis was found, again just like in H2O, and combination
ratio kE /kKHA, and equating these two values leads to eq. [13],
B
K
which shows that an estimate of Qa may be obtained from
the rate constant pair and QaHA. Data for (CH2OH)3CNH3
+
K
[13] Qa = QaHAkEB/kK
HA
K
+
buffers give pQa = 8.91 ± 0.12, and data for NH4 buffers
K
give pQa = 8.95 ± 0.04; both of these results are in very
good agreement with pQa = 8.93 ± 0.02 obtained by spec-
K
trophotometric titration of this substrate.
Additional support for the interpretation made may be
had from estimates of KE based on keto–enol interconver-
of the general acid catalytic coefficient determined, kDA
=
(3.27 ± 0.11) × 103 M–1 s–1, with its H2O counterpart gives
the isotope effect kHA/kDA = 9.98 ± 0.65. This is a large ki-
netic isotope effect, fully consistent with the dominance of
ketonization in this region of acidity and the expectation that
ketonization occurs by rate-determining hydron transfer
from acetic acid to the enolate substrate.
sion by the two routes of eqs. [14] and [15]. The route
E
of eq. [14] gives KE = kE /kK Qa , and the route of eq. [15]
H+
o
gives KE = kE Qw/kK Qa . Insertion of the quantities re-
E
HO−
quired into the first ofothese expressions produces pKE
=
5.02 ± 0.09 and into the second, pKE = 4.73 ± 0.09; both of
these results agree well with pKE = 4.88 ± 0.03 obtained by
combining QaK determined by the spectrophtometric titration
The intercepts of the D2O buffer dilution plots (ko, eq. [3]),
together with the rate constants measured in DClO4 solu-
tions, are compared with their H2O counterparts in Fig. 4. It
E
with Qa based upon the rate profile.
kE
o
[14]
[15]
+ H3O+
S–
+ H2O
+ HO–
S
SH
+ H2O
+ HO–
E
N
N
N
N
kK
+
Qa
H
kE
–
HO
Qw
S–
+ H2O
S
SH
kKo
E
N
N
Qa
© 1999 NRC Canada

Kresge and Meng
1533
Fig. 5. Rate profile for keto–enol/enolate interconversion in the
N-methylindoline-2-selone system in aqueous solutions at 25°C;
circles and solid line: rate of approach to equilibrium, dashed
line: rate of enolization, dotted line: rate of ketonization.
Fig. 4. Comparison of rates of ketonization of N-methylindoline-
2-thione enol/enolate in acidic H2O (᭺) and D2O (᭝) solutions
at 25°C.
+
kK
H
E
Qa
+ H+
S–
SH
S
N
[16]
N
N
secondary component in addition to a normal (kH/kD > 1)
primary component (21); the result obtained here, in fact, is
a near-maximum value (22), consistent with the large iso-
tope effect found for ketonization of this substrate catalyzed
by acetic acid (vide supra).
may be seen that reactions in H2O are faster than those in
D2O throughout the range of acidity examined, as expected
for ketonization occurring by rate-determining hydron transfer
to carbon. The difference between rates in H2O and those in
D2O, however, is greater in the horizontal plateau region at
higher acidity than it is in the diagonal region at lower acid-
ity. This again is consistent with the general ketonization
mechanism assigned in eq. [5], shown in the specific form
that applies here as eq. [16]. Enolate ions are known to be
orders of magnitude more reactive to protonation on carbon
than are the corresponding unionized enols (19), and keto-
nization consequently occurs by rate-determining reaction of
the enolate ion throughout the region of acidity examined
here. In the more acidic solutions, however, the process
starts with unionized enol as the initial state, and this species
must first undergo equilibrium ionization to provide the
more reactive enolate. In this region, therefore, the isotope
effect is on the enol equilibrium ionization constant as well
as on the rate-determining rate constant. In the region of
lower acidity where enolate ion is the initial state, on the
other hand, the isotope effect is on the rate constant alone.
Analysis of the D2O data by least-squares fitting of the
acidic portion of the rate law of eq. [7] gave the results
A substrate isotope effect was also determined by allow-
ing N-methylindoline-2-thione to undergo deuteration at the
3-position in 0.001 M DClO4–D2O for 30 min and then add-
ing enough NaOD to give [OD–] = 0.01 M and measuring
the rate of the subsequent enolization reaction. The data so
obtained are summarized in Table S42; when combined with
rates of reaction of the undeuterated substrate in H2O solu-
tions of NaOH, they give the isotope effect kH/kD = 4.35 ±
0.07. This isotope effect contains, in addition to its primary
component, an inverse secondary component on the con-
sumption of hydroxide ion (21). It consequently is a strong
effect, as expected for enolization effected by rate-
determining removal of a hydron from the substrate by hy-
droxide ion. This reinforces the conclusion that the basic
portion of the rate profile corresponds to an enolization process.
Kinetics: N-methylindoline-2-selone
Rates of approach to keto–enol/enolate equilibrium for
this substrate were measured, as for N-methylindoline-2-
thione, in aqueous perchloric acid and sodium hydroxide so-
E
E
(Qa )D O = (5.16 ± 0.12) × 10–5 M, p(Qa )D O = 4.29 ± 0.01
2
2
(see footnote 3), and kK = (1.81 ± 0.03) × 10–6 M–1 s–1,
D+
–
+
lutions and in CH3CO2H, H2PO4 , (CH2OH)3CNH3 , and
NH4+ buffers. These data are summarized in Tables S6–S8.2
The results obtained in buffer solution were analyzed in a
manner similar to that applied to buffer data for the thione
(vide supra). The zero buffer concentration intercepts ob-
tained in this way were then used, together with the rate
constants determined in perchloric acid and sodium hydrox-
ide solutions, to construct the rate profile shown in Fig. 5. It
may be seen that this rate profile differs from that for the
thione system (Fig. 2) in that it lacks a bend in acid solution.
and combination of these with their H2O counterparts gives
E
E
(Qa )H O/(Qa )D O = 1.90 ± 0.09 and kK /kK = 2.83 ±
H+
D+
2
2
0.11. Both of these isotope effects are entirely consistent
E
with expectation. The equilibrium constant Qa refers to the
ionization of a thiol acid, and solvent isotope effects on the
ionization of a number of thiols have been found to be
KH O/KD O 2 (20). The rate constant kK refers to hydron
H+
2
2
transfer from the hydronium ion to substrate, a process
whose solvent isotope effect includes an inverse (kH/kD < 1)
© 1999 NRC Canada

1534
Can. J. Chem. Vol. 77, 1999
Table 2. Summary of rate and equilibrium constants for the N-methylindoline-2-thione and -selone
systems at 25°C.a
aAqueous solution; ionic strength = 0.10 M; acidity constants are concentration quotients applicable at ionic
strength = 0.10 M.
Such bends are produced by acid ionization of the enol, and
absence of this bend here means that ionization of the selone
enol occurs at acidities that produce rates beyond the capa-
bility of our stopped-flow method (upper limit: k 500 s–1).
We were therefore unable to determine this acidity constant
for the selone enol. We could, however, set a lower limit of
QEa > 5 × 10–4 M, pQEa < 3.3, based upon the highest acidity
we were able to use, [H+] = 5 × 10–4 M.
Resolution of kuc into its components with the aid of
eq. [10] gave kE = (7.01 ± 0.68) × 10–2 s–1 and kK = (4.28
o
± 0.54) × 10–3 s–1. This result, coupled with values oof kE
HO−
and kK , then provides complete rate laws for enolization
H+
and ketonization promoted by solvent-related species, and
that in turn allows separation of observed rates of equilibra-
tion into rates of enolization and ketonization, as shown in
Fig. 5. It may be seen that the two rates become equal at
about [H+] = 10–7 M, and that ketonization dominates at
acidities greater than this while enolization dominates at
acidities lower than this.
Once again, the rate constants for uncatalyzed enolization,
kE , and ketonization, kK , could not be separated, and the
o
o
rate–profile data were analyzed by least-squares fitting of a
version, eq. [17], of the general rate law of eq. [6] that com-
bines these rate constants as kuc; this rate law is also without
The nature of buffer catalysis was once again consistent
with these dominances: CH3CO2H buffers gave only general
+
acid catalysis, and NH4 buffers gave only general base ca-
[17] kobs = kE [HO–] + kuc + kK [H+]
HO−
H+
–
+
talysis, whereas H2PO4 and (CH2OH)3NH3 buffers showed
both general acid and general base catalysis. These catalytic
coefficients are summarized in Table 1. Combination of the
Qa because Qa > [H+] throughout the region examined.
E
E
The results obtained are kE
= (1.33 ± 0.03) × 104 M–1 s–1,
HO−
HA
K
kuc = (7.44 ± 0.72) × 10–2 s–1, and kK = (1.08 ± 0.02) ×
results with Qa according to eq. [13] gave pQa = 7.38 ±
H+
–
K
106 M–1 s–1.
0.06 for H2PO4 buffers and pQa = 7.25 ± 0.04 for
© 1999 NRC Canada

Kresge and Meng
1535
Fig. 6. Comparison of rate profiles for keto–enol/enolate inter-
conversion in the N-methylindoline-2-thione (᭺) and N-methyl-
indoline-2-selone (᭝) systems in aqueous solution at 25°C.
+
K
(CH2OH) 3NH3 buffers, consistent with pQa = 7.25 ± 0.02
obtained by spectrophotometric titration.
Rate measurements in the N-methylindoline-2-selone sys-
tem were also made in D2O solutions of DClO4 and of
NaOD, the latter using deuterated substrates generated as de-
scribed above for the thione system. These data are summa-
rized in Tables S6 and S7.2 When combined with the
appropriate H2O data they give the isotope effect kK /kK
H+
D+
= 2.61 ± 0.10, as expected for the ketonization reaction that
dominates in acid solutions, and kE /kE = 4.12 ± 0.03,
HO−
DO−
again as expected for the enolization that dominates in basic
solution.
N-Methylindoline-2-one
Measurement of rates of keto–enol/enolate interconver-
sion by the methods used in the present study require the
keto form to be a sufficiently strong carbon acid to allow
significant amounts of enolate ion to be present at equilib-
rium in dilute aqueous solution. This is necessary, both for
direct observation of enolate formation from the keto form,
and also for preparation of enolate in dilute base for quench-
ing into acid to generate greater than equilibrium amounts of
enol. Since the carbon-acid ionization of N-methylindoline-
ring upon enolization of N-methylindoline-2-thione, as illus-
trated in eq. [1]. Formation of such a new aromatic ring has
been shown to have a strong influence upon the ionization of
indoline-2-ones and related substances in DMSO solution
(9). It is likely, however, that reduced conjugation between
the carbonyl group and amide nitrogen in N-methylindoline-
2-thione will play a role as well. It is conjugative stabiliza-
tion of the keto form that makes enol contents of simple ox-
ygen amides so low, and in thioamides this conjugation will
be weaker than in oxygen amides because of mismatch of
orbital size. In an aromatic amide such as N-methylindoline-
2-thione, moreover, conjugation of carbonyl and nitrogen
will be reduced further by offsetting conjugation of nitrogen
with the aromatic ring.
2-one occurs only in concentrated base solution (pQK
=
a
15.86), this requirement is not met, and rate measurements
could not be made with this substrate. We hope, however, to
be able to make these measurements by another technique at
some future time.
Discussion
Keto–enol equilibria
The keto–enol equilibrium constant determined here for
N-methylindoline-2-thione, pKE = 4.88, is remarkably large
for a carboxylic acid amide: it is nearly as great as that for
2-indanone, 4, (pKE = 3.84) (23), a structurally similar sim-
ple ketone. On the assumption that the keto–enol equilib-
rium constants for carboxylic acid amides are similar to
those of carboxylic acid esters (3), and using the value pKE
= 19 for ethyl acetate (4), the present result makes the enol
content of N-methylindoline-2-thione some 14 orders of
magnitude greater than that of a simple acyclic carboxylic
acid amide such as acetamide, 5.
Determination of a keto–enol equilibrium constant by the
method employed here requires knowledge of the acidity
E
constant of the enol, Qa , which in turn is evaluated from a
bend in the rate profile for ketonization in acidic solution.
Unfortunately, rate constants in the N-methylindoline-2-
selone system became too fast to measure before this bend
could be reached (see Fig. 5), and we consequently could
not determine a keto–enol equilibrium constant for this sub-
E
strate. We were, however, able to set the limit pQa < 3.3,
CH3
K
E
and use of this in the relationship pKE = pQa – pQa , cou-
pled with pQaK = 7.25, then gives the lower limit pKE > 4.0.
This limit is consistent with an estimate that can be made
based on the fact that selenols are generally more acidic than
the corresponding thiols by a roughly constant 3pK units
O
S
CH3
O
CH3
CH3CNH2
CH3
5
6
4
E
(25). Application of this difference to pQa = 4.05 for the
E
A good portion of this large difference may be attributed
to replacement of carbonyl oxygen by sulfur in going from
acetamide to N-methylindoline-2-thione. Thioketones are
well known to have considerably greater enol contents than
oxygen ketones, a phenomenon generally attributed to the
relative weakness of the sulfur–carbon π-bond (24), and a re-
cent comparison of keto–enol equilibrium constants for
2,4,6-trimethylthioacetophenone, 6, with that of its oxygen
analog puts this effect at six orders of magnitude (8). That
leaves eight orders of magnitude for other effects, the stron-
gest of which is likely to be formation of a new aromatic
enol of N-methylindoline-2-thione then gives pQa = 1 for
the selone, and that leads to pKE = 6 for the selone system.
This result is not very different from pKE = 4.88 for N-
methylindoline-2-thione, and that suggests that, whereas re-
placing carbonyl oxygen by sulfur produces a marked in-
crease in enol content, further replacement of sulfur by
selenium has little effect.
Reaction rates
The comparison of rate profiles provided by Fig. 6 shows
that ketonization of the enol of N-methylindoline-2-thione in
© 1999 NRC Canada

1536
Can. J. Chem. Vol. 77, 1999
8. A.J. Kresge and Q. Meng. J. Am. Chem. Soc. 120, 11 830
(1998).
acid solution is more rapid than that of N-methylindoline-2-
selone, whereas for enolization in basic solution this order is
reversed. The greater rate of ketonization of the thione enol
is consistent with the fact that aryl vinyl sulfides undergo
acid-catalyzed hydrolysis faster than the corresponding aryl
vinyl selenides (26); both of the hydrolyses are cation-forming
reactions that occur by rate-determining hydron transfer to
carbon, just like enol ketonization, and the difference in re-
activity indicates that sulfur is better able to stabilize an ad-
jacent positive charge than is selenium. Enolization, on the
other hand, is an anion-generating reaction, and the superior-
ity of the selone here indicates that selenium is better than
sulfur at stabilizing negative charge; this is consistent with
9. F.G. Bordwell and H.E. Fried. J. Org. Chem. 56, 4218 (1991).
10. B.S. Pedersen, S. Scheibye, N.H. Nilsson, and S.-O.
Lawesson. Bull. Soc. Chim. Belg. 87, 223 (1978); M.P. Cava
and M.I. Levinson. Tetrahedron, 41, 5061 (1985).
11. J.-B. Baudin, M. Bekhazi, S.A. Julia, O. Ruel, R.L.P. de Jong,
and L. Brandsma. Synthesis, 956 (1985).
12. V.Y. Vvedensky, L. Brandsma, E.D. Sheftan, and B.A.
Trofimov. Tetrahedron, 53, 16 783 (1997).
13. R.G. Bates. Determination of pH theory and practice. John
Wiley & Sons, New York. 1973. p. 49.
14. G. Yagil..J. Phys. Chem. 71, 1034 (1967).
15. J.-C. Halle, F. Terrier, and R. Schall. Bull. Soc. Chim. Fr. 4569
(1969).
16. J. Kielland. J. Am. Chem. Soc. 59, 1675 (1937).
17. F.G. Bordwell. Acc. Chem. Res. 21, 456 (1988).
18. J.R. Keeffe and A.J. Kresge. In Techniques of chemistry. Vol.
VI. Investigation of rates and mechanisms of reactions. Part 1.
Edited by C.F. Bernasconi. John Wiley & Sons, New York.
1986. p. 776.
19. P. Pruszynski, Y. Chiang, A.J. Kresge, N.P. Schepp, and P.A.
Walsh. J. Phys. Chem. 90, 3760 (1986); Y. Chiang, A.J.
Kresge, J.A. Santabella, and J. Wirz. J. Am. Chem. Soc. 110,
5506 (1988).
the greater C-H acidity of N-methylindoline-2-selone (pQK
a
= 7.25) over N-methylindoline-2-thione (pQK = 8.93) as
a
well as the greater acid strength of selenols over thiols (25).
Acknowledgments
We are grateful to the Natural Sciences and Engineering
Research Council of Canada for financial support of this
work.
20. W.P. Jencks and K. Salvesen. J. Am. Chem. Soc. 93, 4433
(1971).
21. A.J. Kresge, R.A. More O’Ferrall, and M.F. Powell. In Iso-
topes in organic chemistry. Edited by E. Buncel and C.C. Lee.
Elsevier, New York. 1987. Chap. 4.
22. A.J. Kresge, D.S. Sagatys, and H.L. Chen. J. Am. Chem. Soc.
99, 7228 (1977).
23. J.R. Keeffe, A.J. Kresge, and Y. Yin. J. Am. Chem. Soc. 110,
8201 (1988).
24. T. Selzer and Z. Rappoport. J. Org. Chem. 61, 5462 (1996).
25. F. Fringuellim and A. Taticchi. In The chemistry of organic se-
lenium and tellurium compounds. Vol. 1. Edited by S. Patai
and Z. Rappoport. John Wiley & Sons, New York. 1986.
p. 600.
References
1. J.P. Guthrie. Can. J. Chem. 71, 2123 (1993); J.P. Guthrie and
Z. Liu. Can. J. Chem. 73, 1395 (1995).
2. A.J. Kresge. Chem. Soc. Rev. 25, 275 (1996).
3. S. Sklenak, Y. Apeloig, and Z. Rappoport. J. Am. Chem. Soc.
120, 10 359 (1998).
4. T.L. Amyes and J.P. Richard. J. Am. Chem. Soc. 118, 3129
(1996).
5. H. Hart, Z. Rappoport, and S.E. Biali. In The chemistry of
enols. Edited by Z. Rappoport. Wiley, New York. 1990. Chap. 8.
6. J.R. Keeffe and A.J. Kresge. In The chemistry of enols. Edited
by Z. Rappoport. Wiley, New York. 1990. Chap. 7.
7. J. Frey and Z. Rappoport. J. Am. Chem. Soc. 118, 3994
(1996); Z. Rappoport, J. Frey, M. Sigalov, and E. Rochlin.
Pure Appl. Chem. 69, 1933 (1997).
26. R.A. McClelland and M. Leung. J. Org. Chem. 45, 187 (1980).
© 1999 NRC Canada