The acid dissociation constant of triphenylethenethiol, a simple thioenol, and that of its oxygen-enol analog

Article abstract of DOI:10.1139/v98-027

The acidity constant, pQ(a)(E) = 8.49, for the stable thioenol, triphenylethenethiol, was determined by spectrophotometric titration, and that, pQ(a)(E) = 11.37, for its unstable oxygen analog, triphenylethenol, was determined by analysis of its ketonizat

Full text of DOI:10.1139/v98-027

6
57
The acid dissociation constant of
triphenylethenethiol, a simple thioenol, and that
of its oxygen-enol analog
Yvonne Chiang, A. Jerry Kresge, Norman P. Schepp, Vladimir V. Popik, Zvi
Rappoport, and Tzvia Selzer
E
Abstract: The acidity constant, pQa = 8.49, for the stable thioenol, triphenylethenethiol, was determined by
E
spectrophotometric titration, and that, pQa = 11.37, for its unstable oxygen analog, triphenylethenol, was determined by
analysis of its ketonization rate using enol generated flash photolytically from triphenylvinyl bromide. (These acidity
constants are concentration quotients applicable at 0.10 M ionic strength in 50 vol.% aqueous methanol.) This appears to be
the first determination of the acid strength of a simple thioenol. Triphenylethenethiol was unreactive in dilute acid or base, but
in concentrated perchloric acid solutions it was slowly transformed into diphenylacetophenone and
2
,3-diphenylbenzo[b]thiophene; the mechanisms of these reactions are believed to involve rapid equilibrium protonation of
the enol on the β-carbon followed by rate-determining capture of the cation so formed, either externally by solvent or
internally by one of the β-phenyl groups of the ion.
Key words: enol ketonization, thioenol, flash photolysis, vinyl halide.
E
Résumé : Utilisant un titrage spectrophotométrique, on a déterminé la constante d’acidité, pQ , du thioénol stable,
a
E
a
triphényléthènethiol, alors que l’on a déterminé celle, pQ , de son analogue oxygéné instable, le triphényléthénol, par analyse
de sa vitesse de cétonisation en utilisant l’énol généré par photolyse éclair à partir du bromure de triphénylvinyle. (Ces
constantes d’acidité sont les quotients de concentration applicables à une force ionique de 0,10 M, dans une solution aqueuse
de méthanol à 50% en volume.) Il semble que ce travail corresponde à la première détermination de la force acide d’un
thioénol simple. Le triphényléthènethiol ne réagit pas en milieux acide ou base faibles; toutefois, en solutions concentrées
d’acide perchlorique, il se transforme lentement en diphénylacétophénone et en 2,3-diphénylbenzo[b]thiophène. On croit que
les mécanismes de ces réactions impliquent une protonation à équilibre rapide de l’énol sur le carbone-β, suivie d’une capture
cinétiquement limitante du cation ainsi formé, d’une façon externe par le solvant ou externe par l’un des groupes β-phényles
de l’ion.
Mots clés : cétonisation des énols, thioénol, photolyse éclair, halogénure de vinyle.
[
Traduit par la rédaction]
keto isomer, diphenylthioacetophenone, 2, eq. [1], under con-
ditions much more drastic than usually required to equilibrate
oxygen keto–enol systems (3). This predominance of the enol
isomer in the system of eq. [1] is consistent with the estimate
Introduction
Simple enols are generally much less stable than their aldehyde
or ketone tautomers (1). The same, however, is not true of
simple thioenols: in the sulfur systems, enol and keto isomers
usually exist in comparable amounts, and the enol is some-
times even the dominant tautomer (2). A recently examined
example of the latter situation is provided by
triphenylethenethiol, 1, which could not be converted to its
2
pK = 2.7 for the oxygen system and the suggestion that the
E
Ph
S
Ph
Ph
SH
Ph
[1]
Ph
Ph
2
1
Received December 16, 1997.
ratio of the keto–enol equilibrium constant of a sulfur system
to that of its oxygen analog, K (S)/K (O), will generally be
equal to or greater than 10 (3).
We have taken advantage of this stability of
triphenylethenethiol to measure the equilibrium constant for
its ionization as an acid, eq. [2], using a conventional spectro-
scopic method based upon the difference in uv absorbance of
This paper is dedicated to Professor Erwin Buncel in
recognition of his contributions to Canadian chemistry.
E
E
6
1
Y. Chiang, A.J. Kresge, N.P. Schepp, and V.V. Popik.
Department of Chemistry, University of Toronto, Toronto,
ON M5S 3H6, Canada.
Z. Rappoport and T. Selzer. Department of Organic
Chemistry, The Hebrew University, Jerusalem 91904, Israel.
1
2
Author to whom correspondence may be addressed.
Telephone/Fax: (416) 978–7259. E-mail:
akresge@alchemy.chem.utoronto.ca
This estimate is based upon pK = 0.98 for diphenylacetaldehyde (4)
E
and the effect of adding another phenyl group, obtained from pK =
E
6.23 for acetaldehyde (5) and pK = 7.96 for acetophenone (6).
E
Can. J. Chem. 76: 657–661 (1998)
© 1998 NRC Canada

6
58
Can. J. Chem. Vol. 76, 1998
Ph
Ph
S –
Ph
at which the difference in absorption between
triphenylethenethiol and its thiolate anion is greatest. A Cary
Ph
Ph
SH
Ph
E
a
Q
+ H+
[
2]
2
2
200 spectrometer, with cell compartment thermostatted at
5.0 ± 0.05°C, was used. The absorbance of the acid, base, or
3
buffer solution without substrate was first recorded, a fixed
amount of an acetonitrile stock solution of substrate (0.02 M)
was then added (10.0 µL to 3.00 mL or 5.00 µL to 1.00 mL),
and the absorbance was recorded again. The difference, ∆A,
between such pairs of absorbance readings was then calcu-
lated, and the relationship between ∆A and the hydrogen ion
concentration of the solutions was analyzed by least-squares
fitting of eq. [5]; here A and A are the limiting absorbances
the enol (λmax = 285 nm) and its enolate ion, 3 (λmax = 315 nm).
This, to the best of our knowledge, is the first determination
of the acidity constant of a simple thioenol.
To assess the effect on enol acidity of replacing oxygen
with sulfur, we also determined the equilibrium constant for
the ionization of triphenylethenol, 4, eq. [3]. This enol, how-
ever, is not stable with respect to its keto isomer, and it reverts
to the latter too rapidly to allow measurements by the method
we employed for the thio analog. We therefore used a flash
B
A
of the basic and acidic forms of the substrate and Q is its acidity
3
constant.
E
+
A Q + A [H ]
B
a
A
O–
[5]
∆A =
Ph
QaE + [H+]
Ph
Ph
OH
Ph
E
Q
a
+ H+
[
3]
Hydrogen ion concentrations of the solutions in which the ab-
sorbances were recorded were determined by measuring their
pH using a Beckman Research pH meter, in conjunction with
a ROSS combination electrode model no. 81–02 (Orion Re-
Ph
Ph
4
photolytic technique instead: we generated the enol by pho-
toionization of triphenylvinyl bromide, 5, and subsequent hy-
dration of the ensuing vinyl cation, 6, eq. [4], and then
determined the acidity constant of the enol, 4, from the depend-
ence of its rate of ketonization to diphenylacetophenone, 7,
upon the acidity of the medium.
Because some of the substrates used were insufficiently
soluble in a purely aqueous solvent, this work was done in 50
vol.% aqueous methanol.
+
search Inc.), and then converting that reading into [H ] with
the aid of eq. [6] (8a).
+
[
6]
−log[H ] = pH
+ log γH
+
− δ = pHmeas − 0.205
meas
+
In this equation, γH
+
is the activity coefficient of H , which was
calculated using the Debye–Hückel equation (8b) with pa-
rameters appropriate to the medium employed, and δ is an
empirical parameter (8a). The validity of this method was
checked by measuring the pH of five perchloric acid solutions
of known concentration in the range 0.02–0.10 M. The results
+
obtained gave an average value of (–log[H ] – pHmeas) =
–0.2049 ± 0.0061, in excellent agreement with eq. [6].
Ph
Ph
Br
Ph
Ph
+
•
hn
Ph
[
4]
+
Br–
Flash photolysis
Ph
5
6
Flash photolytic rate determinations were made using a con-
ventional flash-lamp system with cell compartment thermo-
statted at 25.0 ± 0.05°C that has already been described (5).
Measurements were carried out in perchloric acid and sodium
hydroxide solutions whose ionic strength was maintained at
0.10 M; substrate (triphenylvinyl bromide) concentrations
Ph
Ph
O
Ph
Ph
OH
H O
2
H+
–
Ph
Ph
7
4
–
5
were (2–3) × 10 M. The lifetimes of the transient species
observed in basic solution were short enough to be analyzed by
the flash photolysis monitoring system, but those observed in
acidic solution were too long-lived (t = 15–40 min). For rate
measurements in acidic solutions, therefore, transients were
generated by a single flash in the flash system, and the reacting
solutions were then transferred to a Cary 118 spectrometer,
with cell compartment thermostatted at 25.0 ± 0.05°C, for rate
measurement there. Decay of the transients observed in both
kinds of experiment obeyed the first-order rate law accurately,
and observed first-order rate constants were calculated by
least-squares fitting of an exponential function.
Experimental section
Materials
Triphenylethenethiol was a sample that had been prepared be-
fore (3), and 2,3-diphenylbenzo[b]thiophene was kindly pro-
vided by Dr. T. Kitamura (7). All other materials were best
available commercial grades.
Spectrophotometric pK determination
a
The acid dissociation constant of triphenylethenethiol was de-
termined by monitoring the absorbance of this substance in
solutions of different acidity. The solvent was 50 vol.% aque-
ous methanol, and its acidity was controlled by using either
perchloric acid or sodium hydroxide, or acetic acid, biphos-
phate ion, tris(hydroxymethyl)methylamine, or ammonia buff-
ers; the ionic strength of these solutions was maintained at
Results
Spectrophotometric pK determination
a
The equilibrium constant for ionization of triphenylethenethiol
3
0.10 M by adding sodium perchlorate as required.
Measurements were made at λ = 340 nm, the wavelength
This is a concentration dissociation constant, applicable at ionic
strength = 0.10 M.
©
1998 NRC Canada

Chiang et al.
659
Fig. 2. Relationship between sodium hydroxide concentration and
rates of ketonization of triphenylethenol in 50 vol.% aqueous
Fig. 1. Spectrophotometric titration curve for the acid ionization of
triphenylethenethiol in 50 vol.% aqueous methanol at 25°C.
methanol solution at 25°C.
as an acid, eq. [2], was determined by measuring the
absorbance of this substance dissolved in 50 vol.% aqueous
methanol solutions of different acidity. Two series of experi-
ments were carried out, one at a fixed total concentration of
S
Ph
Ph
O
Ph
–5
–4
substrate of 7 × 10 M and another at 1 × 10 M. The data so
obtained are summarized in Table S1.
Ph
7
Ph
4
8
As Fig. 1 illustrates, these data describe the expected titra-
tion curve, eq. [5], well. Least-squares analysis gave the re-
Flash photolysis
sults Q = (3.48 ± 0.22) × 10 M, pQaE = 8.46 ± 0.03, and
E
–9
a
Flash photolysis of vinyl halides is known to produce vinyl
cations (11), which, in the presence of water, are hydrated to
enols; the enols then ketonize to their carbonyl isomers
E
–9
E
a
Q = (3.07 ± 0.21) × 10 M, pQ = 8.51 ± 0.03. The weighted
a
average of these gives Qa = (3.27 ± 0.15) × 10 M, pQaE
E
–9
=
8.49 ± 0.02, as the best value of this acidity constant.
(
11b,c). Evidence that this is the sequence of reactions that
occurred in the present study upon flash photolysis of
triphenylvinyl bromide, as depicted in eq. [4], is provided by
the uv spectral changes that took place. The uv spectrum of
triphenylvinyl bromide in 50 vol.% aqueous methanol consists
of a strong absorption band with λmax = 290 nm. Upon flash-
ing, this band immediately increased in intensity in a process
that was complete within the duration of the flash (50 ms).
Such a change in intensity without the appearance of any new
absorption bands is to be expected for the conversion of a vinyl
halide to a vinyl alcohol with no change in the other double-
bond substituents; the lifetime of vinyl cations in aqueous me-
dia, moreover, is known to be much shorter than 50 ms (11c),
and formation of the enol would be expected to be immediate
on the time scale of the present experiment. This increase in
absorbance at λ = 290 nm was followed by a decrease whose
velocity was catalyzed by both hydronium and hydroxide ions,
with saturation of the hydroxide ion catalysis at high basicities.
Such behavior is characteristic of the ketonization of simple
enols, where saturation of hydroxide ion catalysis represents
conversion of enol to enolate ion (12). The absorbance de-
crease at λ = 290 was also accompanied by an increase occur-
Reaction of triphenylethenethiol in strong acid
Except for the acid ionization described above,
triphenylethenethiol proved to be unreactive in dilute 50 vol.%
aqueous methanol solutions of acid or base. In concentrated
perchloric acid, however, its uv absorbance did decrease
slowly in an acid-catalyzed process, and some rate measure-
ments were made by monitoring this decay at λ = 300 nm. The
4
rate constants so obtained are summarized in Table S2. These
data were extrapolated down to dilute solution by the
Cox–Yates method (9) using the X acidity function (10) on the
0
assumption that values of X in 50 vol.% aqueous methanol are
0
the same as those in wholly aqueous solutions of the same acid
–
8
–1 –1
concentration. This gave kH
+
= (2.72 ± 0.41) × 10
M s as
the dilute solution rate constant.
A product study was performed by allowing 0.0025 M
triphenylethenethiol dissolved in 4 M perchloric acid in 2:1
methanol:water (vol/vol) to react for 36 h, and then subjecting
the resulting solution to HPLC analysis. This showed that two
products were formed in comparable amounts, and spiking
with authentic samples showed them to be diphenylacetophe-
none, 7, and 2,3-diphenylbenzo[b]thiophene, 8.
ring at the same rate at
λ
= 250 nm, where
diphenylacetophenone, the expected product of the present ke-
tonization reaction, has an absorption maximum. It appears
safe to conclude, therefore, that the process monitored here
was ketonization of triphenylethenol to diphenylacetophenone.
Rates of this ketonization reaction were measured in
4
Tables S1–S4 may be purchased from: The Depository of
Unpublished Data, Document Delivery, CISTI, National
Research Council Canada, Ottawa, Canada K1A OS2.
©
1998 NRC Canada

6
60
Can. J. Chem. Vol. 76, 1998
E
0
.02–0.10 M perchloric acid solutions. The data are summa-
k′ Qa
+
o
4
[9]
kobs = [
rized in Table S3. Observed rate constants proved to be accu-
rately proportional to acid concentration, and the data were
therefore analyzed by least-squares fitting of eq. [7]. This pro-
duced the result kH
acid-catalyzed reaction and kuc = (2.11 ± 0.30) × 10
an “uncatalyzed” process.
E
H ] + Qa
–
2
–1 –1
+
= (1.05 ± 0.04) × 10
M
s
for the
s
Discussion
–
4 –1
for
OEt
SEt
+
[
7]
kobs = kuc + kH
+
[H ]
Rates of ketonization of triphenylethenol were also measured
in sodium hydroxide solutions over the concentration range
9
10
4
Comparison of the rate constants obtained here for the acid-
catalyzed ketonization of triphenylethenol and the disappear-
ance of triphenylethenethiol in concentrated perchloric acid
0.002–0.10 M. These data are summarized in Table S4. As
Fig. 2 shows, observed first-order rate constants increased
with increasing basicity at low hydroxide ion concentrations,
but at higher basicity the effect diminished, and rate constants
finally reached a constant value independent of hydroxide ion
concentration. Such behavior is commonly found in enol ke-
tonization. It is known to be caused by a shift in the initial state
of the reaction from enol to enolate ion, plus the facts that
ketonization occurs by rate-determining protonation of the
substrate on the β-carbon and that enolate ions are much more
susceptible to such electrophilic attack than are enols (12). In
weakly basic solutions, therefore, where the substrate exists
mainly as un-ionized enol, ketonization will occur as shown in
eq. [8]: rapid equilibrium ionization of the enol will be fol-
lowed by rate-determining reaction of the more labile enolate
ion by proton transfer from the only available proton donor, the
hydroxylic solvent. Since hydrogen ions are produced in the
pre-equilibrium step but are not used up in the rate-determin-
ing step, the overall reaction rate will show an inverse depend-
ence on hydrogen ion concentration or a direct dependence on
hydroxide ion concentration. This apparent hydroxide ion ca-
talysis, however, will diminish as the hydroxide ion concentra-
tion increases and the position of the pre-equilibrium step
5
solutions shows that the oxygen enol is 3.8 × 10 times as
reactive as the thioenol. This is an unusually large oxygen-to-
sulfur rate ratio for rate-determining protonation of a carb-
on–carbon double bond activated by these atoms; for example,
the acid-catalyzed hydrolysis of ethyl vinyl ether, 9, and ethyl
vinyl sulfide, 10, both of which occur by rate-determining pro-
tonation of the substrate on its β-carbon atom, give an oxygen-
to-sulfur rate ratio of only 37 (13), some 10 000 times smaller
than the present value.
This striking difference suggests that the disappearance of
triphenylethenethiol in concentrated perchloric acid does not
represent rate-determining protonation of this substance on its
β-carbon atom, but rather that such protonation occurs rapidly
and reversibly and is then followed by some other, much
slower, reaction step. This hypothesis is consistent with the
conclusion reached before (3) that the enol in this system is
more stable than the thioketone, for the ketone, once formed,
would then revert to the enol. It is also consistent with our
observation that the oxygen ketone and a thiophene derivative,
rather than the thioketone, are the products of the concentrated
acid reaction. It is likely that the oxygen ketone is formed by
occasional capture, by water or methanol, of the cation pro-
duced by reversible carbon protonation of the thioenol, fol-
lowed by decomposition of that adduct, as shown in eq. [10],
and that the thiophene product is formed by intramolecular
capture of this cation by one of its β-phenyl groups, in an
electrophilic aromatic substitution reaction, followed by aro-
matization of the thiophene ring, as shown in eq. [11]; a similar
cyclization to a benzofuran has been observed in an analogous
stable oxygen-enol system (14).
Ph
OH
E
O–
Ph
Ph
Q
a
[
8]
+
H+
Ph
Ph
Ph
Ph
Ph
O
k'
o
Ph
These observations reinforce the conclusion reached before
3) that the enol is the more stable tautomer in the
(
shifts toward enolate ion, and that will produce the saturation
behavior shown in Fig. 2.
diphenylthioacetophenone keto–enol system. Additional evi-
dence that this is so is provided by the fact that we were able
to determine the acidity constant of this thioenol by a conven-
tional slow method. The limiting rate constant for ketonization
of the corresponding oxygen enol, triphenylethenol, deter-
The rate law that applies to the reaction scheme of eq. [8]
is shown in eq. [9]. Least-squares fitting of the data using this
–
1
E
equation gave k′ = 63.1 ± 0.07 s and Qa = (4.28 ± 0.22) ×
o
–
12
E
3
–1
10
M, pQa = 11.37 ± 0.02. Hydrogen ion concentrations
mined by our flash photolysis experiments, is 63 s . Applying
–
2
needed for this purpose were obtained from pH meter readings
with the aid of eq. [6].
a sulfur-to-oxygen rate ratio of ca. 10 to this, as expected, for
example, on the basis of the rates of reaction of ethyl vinyl
+
Ph
SH
Ph
Ph
SH
OR
Ph
Ph
O
Ph
Ph
SH
Ph
H+
–RSH
ROH
H+
[
10]
H
H
H
–
Ph
Ph
Ph
Ph
©
1998 NRC Canada

Chiang et al.
661
+
SH
SH
S
Ph
Ph
[
11]
Ph
Ph
Ph
Ph
3
4
. T. Selzer and Z. Rappoport. J. Org. Chem. 61, 5462 (1996).
. Y. Chiang, A.J. Kresge, and E.T. Krogh. J. Am. Chem. Soc. 110,
600 (1988).
ether and ethyl vinyl sulfide discussed above, puts the lifetime
of triphenylethenethiol at about 1 s; this short lifetime would
have made our conventional measurements impossible if the
enol were not the stable isomer.
We are aware of no previous determination of the acidity
constant of a simple thioenol, and we consequently cannot
compare the value we have obtained with other results. It is
2
5
6
7
8
9
. Y. Chiang, M. Hojatti, J.R. Keeffe, A.J. Kresge, N.P. Schepp,
and J. Wirz. J. Am. Chem. Soc. 109, 4000 (1987).
. J.R. Keeffe, A.J. Kresge, and J. Toullec. Can. J. Chem. 64, 1224
(
1986).
. T. Kitamura, S. Soda, K. Morizane, Y. Fujiwara, and
H. Taniguchi. J. Chem. Soc. Perkin Trans. 2, 473 (1996).
. R.G. Bates. Determination of pH theory and practice. Wiley–In-
terscience, New York. 1973. (a) pp. 243–245; (b) p. 216.
. R.A. Cox. Acc. Chem. Res. 20, 27 (1987).
interesting,
triphenylethenethiol, with pQa = 8.49, a considerably stronger
acid than its oxygen analog, triphenylethenol, with pQaE
1.37, for this is consistent with the fact that simple thiols are
however,
that
E
our
value
makes
=
1
considerably stronger acids than simple oxygen alcohols (15).
A quantitative comparison might be more appropriately based
upon phenols rather than alcohols, for phenols, like enols, have
their acidic groups attached to carbon–carbon double bonds,
and enol pK ’s are in fact quite similar to phenol pK ’s (12). It
1
1
0. R.A.Cox and K. Yates. Can. J. Chem. 59, 2116 (1981);
A.J. Kresge, H.J. Chen, G.L. Capen, and M.F. Powell. Can. J.
Chem. 61, 249 (1983).
1. (a) W. Schnabel, I. Naito, T. Kitamura, S. Kobayashi, and
H. Taniguchi. Tetrahedron, 36, 3229 (1980); S. Kobayashi,
T. Kitamura, H. Taniguchi, and W. Schnabel. Chem. Lett. 2101
(1983); S. Kobayashi, Q.Q. Zhu, and W. Schnabel. Z. Natur-
forsch. B: Chem. Sci. 43, 825 (1988); F.I.M. van Ginkel,
R.J. Wisser, C.A.G.O. Varma, and G. Lodder. J. Photochem. 30,
a
a
is significant, therefore, that the difference in acid strength
E
between the oxygen and sulfur enols examined here, ∆pQa
=
2.88, is not unlike the difference in acidity between phenol
(
3
pK = 10.00 (16)) and thiophenol (pK = 6.49 (17)), ∆pK =
a
.51.
a
a
4
53 (1985); R.A. McClelland, F. Cozens, and S. Steenken. Tet-
rahedron Lett. 31, 2821 (1990); (b) A.J. Kresge and
N.P. Schepp. J. Chem. Soc. Chem. Commun. 1548 (1989); (c)
Y. Chiang, R. Eliason, J. Jones, A.J. Kresge, K.L. Evans, and
R.D. Gandour. Can. J. Chem. 71, 1964 (1993).
Acknowledgements
We are grateful to Professor T. Kitamura for supplying 2,3-
diphenylbenzo[b]thiophene and to the Natural Sciences and
Engineering Research Council of Canada for financial support
of this research.
1
2. J.R. Keeffe and A.J. Kresge. In The chemistry of enols. Edited
by Z. Rappoport. Wiley–Interscience, New York. 1990.
pp. 399–480.
1
1
3. R.A.McClelland. Can. J. Chem. 55, 548 (1977).
4. M. Schmittel and U Baumann. Angew. Chem. Int. Ed. Engl. 29,
References
5
41 (1990).
1
5. R. Stewart. The proton: applications to organic chemistry. Aca-
demic Press, New York. 1985. pp. 45–46.
1
. J. Toullec. In The chemistry of enols. Edited by Z. Rappoport.
Wiley–Interscience, New York. 1990. pp. 323–398.
. F. Duus. In Comprehensive organic chemistry. Vol. 3. Edited by
D. Barton and W.D. Ollis. Pergamon Press, Exeter, U K. 1979.
pp. 385–388.
2
16. A.J. Biggs. Trans. Faraday Soc. 52, 35 (1956).
17. C.L. Liotta, E.M. Perdue, and H.P. Hopkins, Jr. J. Am. Chem.
Soc. 96, 7981 (1974).
©
1998 NRC Canada