Protonated benzofuran, anthracene, naphthalene, benzene, ethene, and ethyne: Measurements and estimates of pKa and pKR

Article abstract of DOI:10.1021/ja012613x

Aqueous solvolyses of acyl derivatives of hydrates (water adducts) of anthracene and benzofuran yield carbocations which undergo competitive deprotonation to form the aromatic molecules and nucleophilic reaction with water to give the aromatic hydrates. Trapping experiments with azide ions yield rate constants k_p for the deprotonation and k_H2O for the nucleophilic reaction based on the azide clock . Combining these with rate constants for (a) the H⁺-catalyzed reaction of the hydrate to form the carbocation and (b) hydrogen isotope exchange of the aromatic molecule (from the literature) yields pK_R = -6.0 and -9.4 and pK_a = -13.5 and -16.3 for the protonated anthracene and protonated benzofuran, respectively. These pK values may be compared with pK_R = -6.7 for naphthalene hydrate (1-hydroxy-1,2-dihydronaphthalene), extrapolated to water from measurements by Pirinccioglu and Thibblin for acetonitrile-water mixtures, and pK_a = -20.4 for the 2-protonated naphthalene from combining k_p with an exchange rate constant. The differences between pK_R and pK_a correspond to pK_H2O, the equilibrium constant for hydration of the aromatic molecule (pK_H2O = pK_R - pk_a). For naphthalene and anthracene values of pK_H2O = +13.7 and +7.5 compare with independent estimates of +14.2 and +7.4. For benzene, pK_a = -24.3 is derived from an exchange rate constant and an assigned value for the reverse rate constant close to the limit for solvent relaxation. Combining this pK_a with calculated values of pK_H2O gives pK_R = -2.4 and -2.1 for protonated benzenes forming 1,2- and 1,4-hydrates, respectively. Coincidentally, the rate constant for protonation of benzene is similar to those for protonation of ethylene and acetylene (Lucchini, V.; Modena, G. J. Am. Chem Soc. 1990, 112, 6291). Values of pK_a for the ethyl and vinyl cations (-24.8) may thus be derived in the same way as that for the benzenonium ion. Combining these with appropriate values of pK_H2O then yields pKR = -39.8 and -29.6 for the vinyl and ethyl cations, respectively.

Full text of DOI:10.1021/ja012613x

Published on Web 06/27/2002
Protonated Benzofuran, Anthracene, Naphthalene, Benzene,
Ethene, and Ethyne: Measurements and Estimates of
pKa and pKR
Aoife C. McCormack, Claire M. McDonnell, Rory A. More O’Ferrall,*
AnnMarie C. O’Donoghue, and S. Nagaraja Rao
Contribution from the Department of Chemistry and Centre for Synthesis and Chemical Biology,
UniVersity College Dublin, Belfield, Dublin 4, Ireland
Received November 28, 2001
Abstract: Aqueous solvolyses of acyl derivatives of hydrates (water adducts) of anthracene and benzofuran
yield carbocations which undergo competitive deprotonation to form the aromatic molecules and nucleophilic
reaction with water to give the aromatic hydrates. Trapping experiments with azide ions yield rate constants
kp for the deprotonation and kH O for the nucleophilic reaction based on the “azide clock”. Combining these
2
with rate constants for (a) the H+-catalyzed reaction of the hydrate to form the carbocation and (b) hydrogen
isotope exchange of the aromatic molecule (from the literature) yields pKR ) -6.0 and -9.4 and pKa )
-13.5 and -16.3 for the protonated anthracene and protonated benzofuran, respectively. These pK values
may be compared with pKR ) -6.7 for naphthalene hydrate (1-hydroxy-1,2-dihydronaphthalene),
extrapolated to water from measurements by Pirinccioglu and Thibblin for acetonitrile-water mixtures, and
pKa ) -20.4 for the 2-protonated naphthalene from combining kp with an exchange rate constant. The
differences between pKR and pKa correspond to pKH O, the equilibrium constant for hydration of the aromatic
2
molecule (pKH O ) pKR - pKa). For naphthalene and anthracene values of pKH O ) +13.7 and +7.5 compare
2
2
with independent estimates of +14.2 and +7.4. For benzene, pKa ) -24.3 is derived from an exchange
rate constant and an assigned value for the reverse rate constant close to the limit for solvent relaxation.
Combining this pKa with calculated values of pKH O gives pKR ) -2.4 and -2.1 for protonated benzenes
2
forming 1,2- and 1,4-hydrates, respectively. Coincidentally, the rate constant for protonation of benzene is
similar to those for protonation of ethylene and acetylene (Lucchini, V.; Modena, G. J. Am. Chem Soc.
1990, 112, 6291). Values of pKa for the ethyl and vinyl cations (-24.8) may thus be derived in the same
way as that for the benzenonium ion. Combining these with appropriate values of pKH O then yields pKR
)
2
-39.8 and -29.6 for the vinyl and ethyl cations, respectively.
Scheme 1
Introduction
This paper reports measurements of equilibrium constants for
the conversion of aromatic molecules and their water adducts
(hydrates) to carbocations. The constants of interest, expressed
as their negative logarithms, are pKa and pKR for the protonated
aromatic species. These are related to the equilibrium constant
pKH O for hydration of the aromatic molecule by a thermody-
2
namic cycle which is illustrated for benzene, benzene hydrate
(1), and the benzenonium carbocation (2) in Scheme 1. As in
the previous paper, the equilibria are represented by single
arrows rather than double arrows to indicate the directions of
reaction to which they refer.1
The paper describes experimental measurements of pKR for
hydrates of benzofuran and anthracene. It draws on data from
the literature2-7 and the previous paper1 to obtain pKa values
and to establish the corresponding equilibrium constants for two
isomeric hydrates of naphthalene and, more speculatively, the
isomeric hydrates of benzene. A pKR value for one of the
naphthalene hydrates, 2-hydroxy-1,2-dihydronaphthalene, has
already been reported for 90% aqueous acetonitrile by Pirinc-
cioglu and Thibblin.2 An extension of these studies leads to
estimates of pKa and pKR for alkyl carbocations and the vinyl
cation.
(2) Pirinccioglu, N.; Thibblin, A. J. Am. Chem. Soc. 1998, 120, 6512.
(3) Kelly, S. C.; McDonnell, C. M.; More O’Ferrall, R. A.; Rao, S. N.; Boyd,
D. R.; Brannigan, I. N.; Sharma, N. D. Gazz. Chim. Ital. 1996, 126, 747.
(4) Katritzky, A. R.; Taylor, R. AdV. Heterocycl. Chem. 1992, 47, 1.
(5) Cox, R. A. J. Phys. Org. Chem. 1991 4, 233.
(6) Ansell, H. V.; Hirschler, M. M.; Taylor, R. J. Chem. Soc., Perkin Trans.
2 1977, 353.
(7) Eaborn, C.; Golborn, P.; Spillett, R. E.; Taylor, R. J. Chem. Soc. B 1968,
1112.
* Corresponding author. E-mail: rmof@ucd.ie.
(1) Dey, J.; O’Donoghue, A. C.; More O’Ferrall, R. A. J. Am. Chem. Soc.
2002, 124, 8561 (preceding paper in this issue).
9
10.1021/ja012613x CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 8575-8583
8575

A R T I C L E S
McCormack et al.
Scheme 2
this product was not isolated, HPLC analysis showed the
appearance of a new peak with the expected difference in
retention time between the peak assigned to the benzofuranyl
azide and that of the alcohol (hydrate) based on studies of similar
substrates.8,11 Moreover, the dependence of the intensity of the
azide peak upon the concentration of azide ion was consistent
with its formation in the competing pathways shown in Scheme
2. In addition, the derived product ratios of azide to hydrate
and azide to benzofuran were consistent with an independent
measurement of the suppression of benzofuran product formed
in the presence of azide ion determined spectrophotometrically.
Details of the HPLC product analysis for hydrate and azide
and the derived azide/hydrate product ratios are given in the
Experimental Section. From the dependence of the product ratios
upon the concentration of sodium azide, we obtain the ratio of
rate constants for attack of azide ion and water molecules on
The pK values in question are derived by combining rate
constants for forward and reverse reactions in a manner that is
now well-established for the formation of carbocations from
alcohols.2,8-14 For benzofuran hydrate (5), for example, a rate
constant kH for acid-catalyzed racemization of the hydrate3 is
taken as a measure of the rate of carbocation formation and
combined with a rate constant for the reverse attack of water
on the carbocation. The latter is measured by the azide clock
method.8-10 As shown in Scheme 2, solvolysis of a carboxylate
ester of the hydrate 3 is presumed to yield a carbocation 4 which
is partitioned between pathways leading to hydration, trapping
by azide ion, and deprotonation. From the ratio of product
concentrations of hydrate 5 and azide 6, the rate constant for
benzofuranyl carbocation 4 as kaz/kH O ) 320 M by use of eq 1
2
which is based on Scheme 2. The rate constant kH O is first order
2
with units s-1 at an ionic strength of 0.5 M.
-
[RN3] kaz[N3 ]
)
(1)
k
[ROH]
H2O
The independent measurement of kaz/kH O was based on
2
spectrophotometric determination of the intensity of the ben-
zofuran chromophore formed in the products of solvolysis of
dihydrobenzofuranyl chloroacetate in water in the presence and
absence of added azide ion. For the same concentrations of
chloroacetate reactant, the absorbance of the benzofuran product
decreased with increasing concentration of azide ion (after
correction for the absorbance of azide ion). The dependence of
the absorbance on azide concentration may be expressed by eq
2, in which ∆A is the absorbance corresponding to a concentra-
tion of azide ion [N3-], ∆Ao is the absorbance in the absence
of azide ion, and the rate constants again correspond to those
in Scheme 2. Both ∆A and ∆Ao are corrected for the absorbance
of the reactant as described in the Experimental Section.
attack of water on the carbocation (kH O) is derived by assuming
2
that the reaction with azide ion approaches diffusion control
(kaz ) 5 × 109 M-1 s-1). Combining this rate constant with
that for racemization (kH) yields a value of pKR.
Deprotonation of the carbocation 4 yields benzofuran 7. A
rate constant kp for the deprotonation may be derived in the
same manner as kH O. This may be combined with a rate constant
2
kA for protonation of the aromatic molecule to form the
carbocation. Since this carbocation is the reactive intermediate
in aromatic hydrogen isotope exchange, kA is obtained as the
rate constant for acid-catalyzed exchange of 2-tritiobenzofuran
corrected for a tritium isotope effect.
-
kaz[N3 ]
Results
1
1
)
+
(2)
∆A ∆Ao
∆Ao(kH O + kp)
Benzofuran-2,3-hydrate. Rate constants for acid-catalyzed
dehydration of benzofuran hydrate 5 and for racemization of
the chiral hydrate have been reported in an earlier paper as 1.5
× 10-3 and 4.5 × 10-3 M-1 s-1 respectively.3 The smaller value
for the dehydration indicates that formation of the carbocation
is not fully rate-determining in this reaction. In this work
measurements of azide trapping have been carried out using
solvolysis of the chloroacetate of hydrate 3 as a source of
carbocation 4. In the absence of azide ion, benzofuran 7 and its
hydrate are formed in a concentration ratio corresponding to
2
From the slope divided by the intercept of the plot of 1/∆A
against [N3-] the value of kaz/(kH O + kp) is derived as 135.
2
This may be converted to kaz/kH O ) 214 by taking the
previously determined ratio kH O /kp ) 1.7 based on either (a)
a ratio of rate constants for acid-catalyzed racemization and
dehydration of the benzofuran hydrate3 or (b) the increase in
the fraction of benzofuran product from further reaction with
acid of the initially formed mixture of hydrate and benzofuran
from solvolysis of the benzofuranyl chloroacetate. The value is
2
2
the ratio of rate constants kp/kH O (Scheme 2). In the presence
of azide ion, dihydrobenzofuranyl azide is also formed. Although
2
in satisfactory agreement with the HPLC measurement (kaz/kH O
2
) 320 M) considering the lower ionic strength (0.01 M
compared with 0.5 M) and the fact that there is interference
from absorption by azide ion with the spectrophotometric
measurements. Giving greater weight to the measurements at
(8) Richard, J. P.; Rothenburg, M. E.; Jencks, W. P. J. Am. Chem. Soc. 1984,
106, 1361.
(9) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1984, 106, 1373.
(10) McClelland, R. A.; Banait, N. A.; Steenken, S. J. Am. Chem. Soc. 1986,
108, 7023.
lower ionic strength the value kaz/kH O is taken as 240.
(11) McClelland, R. A., Kanagasabapathy, V. M.; Banait, N. S.; Steenken, S.
J. Am. Chem. Soc. 1989, 111, 3966
2
The ratio kaz/kH O may be converted to a rate constant kH O
2
2
(12) McClelland, R. A.; Cozens, F. C.; Steenken, S.; Amyes, T. L.; Richard, J.
P. J. Chem. Soc., Perkin Trans. 2 1993, 1717.
for attack of water on the carbocation if the rate constant kaz
(13) Amyes, T. L.; Richard, J. P.; Novak, M. J. Am. Chem. Soc. 1992, 114,
8032.
can be assigned. In the azide clock analysis usually this is
(14) O’Donoghue, A. C.; More O’Ferrall, R. A. Unpublished results.
assumed to have the value for diffusion, 5 × 109 M-1 s-1
.
9
8576 J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002

Aromatic Hydrates Yielding pK and pKR
A R T I C L E S
a
However for the benzofuranyl carbocation probably a better
value is kaz ) 2.6 × 109 directly determined by McClelland et
al. for reaction of azide ion with the 2-(4-methoxyphenyl)ethyl
cation 8 generated by flash photosolvolysis from the alkyl
chloride precursor.12 We will see that the pKR value (-9.4)
derived for the benzofuranyl cation is close to that for 8, which
was measured as -8.6 in a 50:50 (v/v) trifluoroethanol-water
mixture (and is expected to be about one unit more negative in
pure water13).
Figure 1. Plots of reciprocals of ∆A (final - initial absorbances) against
concentration of azide ion for the solvolysis of 9-acetoxy-9,10-dihydroan-
thracene in aqueous acetonitrile at 25 °C (0, 20% CH3CN; 50% CH3CN).
With this value of kaz, kH O for the benzofuranyl cation is
2
obtained as 1.1 × 107 s-1. The rate constant kp for deprotonation
of the carbocation to form benzofuran, based on kH O/kp ) 1.7,
2
sponds to a product composition of 64% anthracene and 36%
hydrate, which compares with 76% anthracene and 24% hydrate
previously determined by NMR measurements in 40:60 aqueous
acetone (v/v).18
is 6.5 × 106 s-1. Combining kH O with a best value of kH ) 4.1
2
× 10-3 M-1 s-1 for the acid-catalyzed formation of the
benzofuranyl cation from the hydrate3 yields KR ) kH O/kH
)
2
2.6 × 109 and pKR ) -9.4.
When the solvolysis was carried out in the presence of azide
ion, the absorbance from the anthracene was suppressed with a
linear dependence on the concentration of azide ion consistent
with trapping of an anthracenonium ion intermediate 11 to give
the azide (12, X ) N3). The azide adduct and anthracene were
too insoluble for a product analysis to be carried out by HPLC.
However, the analysis could be achieved spectrophotometrically
by making measurements in the acetonitrile-water mixtures.
Plots of the reciprocal of the absorbance against [N3-] gave
linear plots consistent with eq 2, on the basis of which values
Anthracene Hydrate. A rate constant for specific acid-
catalyzed dehydration of anthracene hydrate (9) in acetic acid
buffers at 0.5 M ionic strength was previously reported as 22.0
M-1 s-1 15
. In this work we found reaction in dilute HClO4 gave
a rate constant 9.5 M-1 s-1. Examination of the rate at different
ionic strengths of sodium chloride confirmed that the discrep-
ancy was substantially due to the difference in ionic strengths.
These results are reported in Table S1 of the Supporting
Information.
Anthracene hydrate 9 (Scheme 3) was acetylated18 to give
the acetate (12, X ) OAc), which was solvolyzed in the same
manner as the benzofuranyl chloroacetate to give a mixture of
hydrate and anthracene (11). The ratio of hydrate to anthracene
was determined from the fractional increase in absorbance of
the anthracene on addition of sufficiently strong acid to induce
acid-catalyzed dehydration of the hydrate. Because of the low
solubility of the acetate and anthracene in water, these measure-
ments were carried out in acetonitrile-water mixtures containing
of kaz/kH O were determined as 344, 306, 225, and 198 in aqueous
mixtures containing 33, 50, 67, and 80% acetonitrile, respec-
tively, making use of the previously established value of kp/
2
k
H O. Reciprocal plots of limiting absorbances against azide
2
concentration at two solvent compositions are shown in Figure
1 based on data recorded in Table S2 of the Supporting
Information. Similar measurements with sodium acetate, up to
0.64 M concentration, showed no appreciable effect on the
product absorbances, indicating that acetate ion does not catalyze
proton abstraction from the anthracenonium ion to form
anthracene. These measurements are in Table S3 of the
Supporting Information.
33, 50, 67, and 80% acetonitrile (v/v). A ratio of kH O/kp ) 0.57
2
( 0.03 with apparently no dependence on solvent composition
was derived from the absorbance measurements. This corre-
Scheme 3
9
J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002 8577

A R T I C L E S
McCormack et al.
The ratio kaz/kH O was extrapolated from acetonitrile-water
It should be noted that Toteva and Richard have argued that
where a rate constant for reaction with solvent is as large as
1.6 × 1010 s-1, it is difficult to exclude the possibility that a
fraction of the azide reacts by a preassociation mechanism.20,21
Ascribing all of the azide product formed to reaction of free
carbocations (kd ) 5 × 109 M-1 s-1) then underestimates the
rate constant for reaction with water and implies a more negative
value of pKR than is calculated (-6.7). However, as the limiting
rate constant for solvent relaxation is probably 1011 s-1, less
than an order of magnitude larger than the derived value, the
error introduced by the assumption of diffusion control is
small.21 Moreover, a correlation of log kp with pKa values for a
range of carbocations suggests qualitatively that kp might be
closer to its (apparent) measured value than its upper limit.22
2
mixtures to pure water by assuming that kH O and kaz had the
2
same dependence on solvent composition as that observed by
McClelland for reactions with benzhydryl and trityl cations.
Thus, kaz was found to be independent of solvent composition
between water and 50% acetonitrile but increased by a factor
of 1.25 in 67% and 1.5 in 80% acetonitrile.11 The value of log
k
H O was found to increase linearly with the volume fraction of
2
acetonitrile with slope 0.14 between 0 and nearly 80% aceto-
nitrile.17 Our own values of log kaz/kH O were corrected for these
2
variations in kaz and plotted against the fraction of acetonitrile.
A value of kaz/kH O ) 375 was extrapolated by drawing a best
2
straight line with slope 0.14 through the points.
To obtain the value of KR for the anthracenyl cation, kH O
2
must be combined with kH, the rate constant for carbocation
formation. The measured rate constant for acid-catalyzed
Aromatic Hydrogen Isotope Exchange. The carbocations
derived from aromatic hydrates are conjugate acids of their
parent aromatic molecules. This means that a rate constant for
deprotonation of the carbocation (kp) may be combined with a
rate constant for protonation of the aromatic molecule (kA) to
obtain an acid dissociation constant of the protonated species
as Ka ) kp/kA.
dehydration corresponds to kHkp/(kH O + kp) as may be seen
2
from inspection of Scheme 4
. From kH O/kp ) 0.57 we obtain kH ) 1.57 × 9.5 ) 14.9
2
M-1 s-1. Combining this with kH O )1.3 × 107 gives KR ) 8.9
2
× 105 and pKR ) -6.0 for the anthracenyl cation.
Naphthalene Hydrates. Measurements of rate constants for
dehydration of naphthalene hydrates from a previous study19
were combined with trapping experiments reported by Pirinc-
cioglu and Thibblin2 to obtain pKR values of the naphthale-
nonium ion protonated at the 1-position. Pirinccioglu and
Thibblin studied trapping of the 2-naphthalenonium ion 13,
generated from 1,2-dihydro-2-methoxynaphthalene (14, X )
OCH3), by azide ion to form 1,2-dihydro-2-azidonaphthalene
(14, X ) N3) in competition with deprotonation to naphthalene
and nucleophilic reaction with water to form the 1,2-naphthalene
hydrate, 1,2-dihydro-2-hydroxynaphthalene (14, X ) OH).
Measurements were reported for 75% aqueous acetonitrile in
the presence of 0.3 M sodium azide and gave a ratio of 2-azido
product to naphthalene of 1:5. In the absence of azide ion the
ratio of naphthalene to 1,2-naphthalene hydrate was 1600:1.
Rate constants for protonation of aromatic molecules are
available from measurements of hydrogen isotope exchange
corrected for isotope effects. Normally measurements have been
made in nonaqueous acidic media and must be extrapolated to
water.4-7 Thus, Cox5 has extrapolated reported rate constants
for deuterium or tritium exchange of a number of aromatic
molecules, including benzene and naphthalene, in concentrated
solutions of HClO4 or H2SO4, by using the medium acidity
function X23,24 to correlate the dependence of the measured rate
constants upon the acid concentration.5 Although tritium and
deuterium isotope effects have been fully evaluated in only a
few cases, those that have provide a basis for estimating
unmeasured isotope effects with a precision that is probably
within that of the extrapolation. Thus, most values of kH/kD fall
within a factor of 2 of each other and show a rather mild
dependence on reactivity.5
In this way a rate constant for protonation at the 1-position
of naphthalene in aqueous solution at 25 °C was obtained as
1.7 × 10-11 M-1 s-1. If this is multiplied by a statistical factor
of 4 to account for the number of equivalent positions, it may
be combined with kp from Pirinccioglu and Thibblin’s data2 to
obtain Ka ) 2.3 × 1020 (pKa ) -20.4).
By assuming that azide trapping was diffusion-controlled,
Pirinccioglu and Thibblin2 derived kp ) 1.6 × 1010 s-1 and
kH O ) 1.0 × 107 s-1. By combining kH O and kH for the acid-
The 2-naphthhalenonium ion is also the intermediate formed
in the dehydration of the 1,4-hydrate of naphthalene (15, X )
2
2
OH). Its pKa therefore may be combined with pKH O ) -16.9
2
catalyzed reaction of the 2-hydrate, pKR ) -7.8 was obtained
for the aqueous acetonitrile solvent.
estimated in the previous paper as the equilibrium constant for
conversion of naphthalene to its 1,4-hydrate, to obtain pKR )
We wished to obtain pKR in H2O and did so by using our
own measurement of kH ) 1.8 M-1 s-1 in water19 rather than
the value for aqueous acetonitrile. Since McClelland found little
-3.5 for this hydrate from the relationship pKR ) pK
pKa (cf. benzene hydrate in Scheme 1).
-
H2O
change in values of kH O between 75% aqueous acetonitrile and
2
(20) Toteva, M. M.; Richard, J. P. J. Am. Chem. Soc. 1996, 118, 11434.
water,11 this value of kH was combined with Pirinccioglu and
(21) Richard, J. P.: Amyes, T. L.; Lin, S.-S.; O’Donoghue, A. C.; Toteva, M.
M.; Tsuji, Y.; Williams, K. B. AdV. Phys. Org. Chem. 2000, 34, 67.
(22) MacCormack, A. C.; O’Donoghue, A. C.; More O’Ferrall, R. A. Modern
Problems of Organic Chemistry; St. Petersburg University Press: St.
Petersburg, Russia, 2002; Vol. 14, submitted for publication.
(23) Bagno, A.; Scorrano, G.; More O’Ferrall R. A. ReV. Chem. Intermed. 1987,
7, 313.
Thibblin’s value of kH O to give KR ) 5.5 × 106 and pKR )
2
-6.7.
(15) More O’Ferrall, R. A.; Rao, S. N. Croat. Chem. Acta 1992, 65, 593.
(16) McClelland, R. A. Tetrahedron 1996, 52, 6023.
(17) McClelland, R. A.; Kanagasabapathy, V. M.; Banait, N. S.; Steenken, S.
J. Am. Chem. Soc. 1991, 113, 1009.
(18) Leute, R.; Winstein, S. Tetrahedron Lett. 1967, 26, 2475.
(19) Boyd, D. R.; McMordie, R. A. S.; More O’Ferrall, R. A.; Sharma, N. D.;
Kelly, S. C. J. Am. Chem. Soc. 1990, 115, 5458
(24) Cox, R. A.; Yates, K. Can. J. Chem. 1981, 59, 2116. Cox, R. A. AdV.
Phys. Org. Chem. 2000, 34, 1.
(25) A rate constant of 640 M-1 s-1 for the acid-catalyzed dehydration of the
1,4-naphthalene hydrate is considered more reliable than that previously
reported (240 M-1 s-1).15
9
8578 J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002

Aromatic Hydrates Yielding pK and pKR
A R T I C L E S
a
Scheme 4
Scheme 5
For anthracene6 and benzofuran,4 measurements of tritium
exchange have been reported by Taylor for CF3COOH as
solvent. For anthracene, kA ) 3.27 × 10-3 s-1 at 25 °C. This
may be converted to a value in water by taking advantage of
the fact that a rate constant for exchange of 1-tritionaphthalene
has also been measured in CF3COOH7 and by assuming that
the ratio of the two rate constants (for anthracene and naph-
thalene) is unaffected by the change in solvent. A basis for this
approximation is the weak dependence of ratios of rate constants
for substrates of not too different reactivity upon the composition
of aqueous solutions of strong acids,5 and the general presump-
tion that partial rate factors are not sensitively dependent upon
solvent.26
constant pKH O for hydration of the aromatic molecule, by taking
advantage of the thermodynamic cycle illustrated for benzene
hydrate and the benzenonium cation in Scheme 1. Evaluation
2
of pKH O ) 7.4 for anthracene is described in the previous
2
paper.1 Combining this value with pKR ) -6.0 gives pKa )
-13.4. The cycle and relationship connecting these equilibrium
constants are shown in Scheme 5 and eq 3.
pKa ) pKR - pKH O
(3)
2
For benzofuran a value of pKH O is not available. However,
2
The rate constants for tritium exchange of naphthalene in CF3-
COOH were measured at 70, 100.7, and 125 °C.7 Extrapolation
of these to 25 °C using an Arrhenius plot gives a rate constant
1.4 × 10-7 s-1. From this value and the value in water we obtain
kA ) 3.27 × 10-3 × 1.7 × 10-11/(1.4 × 10-7) ) 4.0 × 10-7
M-1 s-1 for the protonation of anthracene by H3O+ in aqueous
solution at 25 °C, if we make no correction for a difference in
isotope effects for detritiation of anthracene and naphthalene
(vide supra). Multiplying this rate constant by a statistical factor
of 2 to allow for the two equivalent positions of protonation
and combining with kp ) 2.3 × 107 s-1 gives Ka ) 2.9 × 1013
(pKa ) -13.5) for anthracene.
Ka may be obtained directly by combining a rate constant kp )
1.9 × 107 s-1 for deprotonation of the 2-protonated benzofuran
4 with a rate constant kA for protonation of benzofuran at the
2-position, as shown in Scheme 6.
Scheme 6
For benzofuran Katritzky and Taylor4 have reported a partial
rate factor for the 2-tritiated compound in CF3COOH at 70 °C
as 1.7 × 104. Following Baker et al., this may be corrected to
2.0 × 104 to allow for the unfavorable effect of hydrogen
bonding by CF3COOH upon exchange of oxygen or sulfur
substituted substrates.26 This partial rate factor may be combined
with a value for 1-tritionaphthalene at 70 °C (1160) to give a
ratio of rate constants for the two substrates which may be
extrapolated to 25 °C by assuming that the difference in rates
stems from a difference in activation energies only. Finally, this
ratio may be combined with Cox’s rate constant for 1-proto-
nation of naphthalene in aqueous solution to give kA ) 3.0 ×
10-10 for protonation at the 2-position of benzofuran. With kp
) 6.5 × 106 M-1 s-1, this gives Ka ) 2.2 × 1016 (pKa ) -16.3).
In practice a rate constant for protonation of benzofuran in
aqueous solution at 25 °C must be extrapolated from the
measurement of a partial rate factor for isotope exchange of
2-tritiobenzofuran in trifluoroacetic acid solvent at 70 °C by
Taylor.4 The necessary corrections for an isotope effect and the
changes in solvent and temperature have been described above.
They yield a rate constant from which pKa ) -16.3 is obtained.
If this pKa is combined with pKR ) -9.4, we derive pKH O
)
2
6.9 by use of eq 3.
It is noteworthy that this value of pKH O is less than pKH O
2
2
) +7.4 for anthracene.1 This might suggest that the aromatic
stabilization of the heterocyclic ring of benzofuran is somewhat
smaller than that of the central ring of anthracene. However,
the relative heats of hydrogenation of benzofuran and anthracene
indicate that the benzofuran ring is the more stable by 2.6 kcal
Discussion
Equilibrium constants KR measuring stabilities of carbocations
derived from ionization of anthracene and benzofuran hydrates
in aqueous solution at 25 °C have been measured. Rate constants
kH for the acid-catalyzed conversion of the hydrate (ArHOH)
mol-1 27
. It is possible that this discrepancy reflects the relative
stabilities of the anthracene and benzofuran hydrates, and in
particular destabilization of the benzofuran hydrate by an
unfavorable repulsion between the endo- and exocyclic elec-
tronegative oxygen atoms.
For anthracene a pKa can be estimated in the same way as
for benzofuran based on measurement of a rate constant for
exchange of 9-tritioanthracene in CF3COOH at 25 °C.6 This
to the carbocation (ArH+) are combined with rate constants kH O
2
for the reverse reaction of the cation with water measured by
the azide clock method (Scheme 4). For the two carbocations,
values of pKR ) -9.4 and -6.0 are obtained. For anthracene a
pKa can also be found by combining pKR with an equilibrium
(27) Afeefy, H. Y.; Liebmann, J. F.; Stein, S. E. In Natural Thermochemical
Data in NIST Chemistry Webbook, July 2001; Linstrom, P. J., Mallard, W.
G., Eds.; NIST Standard Reference Base No. 69; National Institute of
Standards and Technology: Gaithersburg, MD (http://webbook.nist.gov).
(26) Baker, R.; Eaborn, C.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1972, 97.
Taylor, R. Electrophilic Aromatic Substitution; Wiley: Chichester, U.K.,
1990.
9
J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002 8579

A R T I C L E S
McCormack et al.
Table 1. Rate and Equilibrium Constants for Protonation and
yields pKa ) -13.5 which agrees well with the value of -13.4
Hydration of Aromatic Molecules
inferred from eq 3 and based on pKH O ) 7.4 from the previous
2
paper.1
The level of agreement between measured and calculated
values of pKH O for anthracene implied by this comparison must
2
be partly fortuitous. However, it suggests that the discrepancy
between the differences in heats of hydrogenation and equilib-
rium constants for hydration of anthrancene and benzofuran
probably does not arise from errors in the extrapolated rate
constants for hydrogen isotope exchange.
Naphthalene and Benzene. It is noteworthy that the conju-
gate acids of benzofuran and anthracene react in water at
comparable rates by deprotonation to form the aromatic
molecule and nucleophilic trapping to form the aromatic
hydrates. This allows both pKa and pKR to be evaluated directly.
For molecules with a greater aromatic stabilization, such as
naphthalene or benzene, the deprotonation pathway normally
dominates and the substitution reaction is not easily observed.15
Nevertheless, Pirinccioglu and Thibblin have recently reported
detectable nucleophilic trapping in competition with more rapid
deprotonation for the 2-naphthalenonium ion 13.2 From a study
of trapping by azide ion they obtained a value of pKR ) -7.8
in 75% aqueous acetonitrile. As described above, this value may
be extrapolated to pKR ) -6.7 for a fully aqueous medium.
a From proceeding paper1 unless indicated otherwise: the values of pKH O
2
for the 1,2- and 1,4-hydration of benzene are coincidentally equal.
b Measurements taken or interpolated from refs 4, 5, and 26. c Estimated
as described in text. d From difference in pKa and pKR, the calculated value
is -14.2.1 e From difference of pKa and pKR; the calculated value is 7.4.1
f From the difference of pKa and pKR.
We have chosen a rate constant of 5 × 1010 s-1 for
deprotonation of the benzenonium ion by recognizing that the
thermodynamic driving force for this reaction must be greater
than that for the 2-naphthalenonium ion but that the difference
will not be large. This means that the constant should be between
1.6 × 1010 and 1011 s-1 and that the uncertainty in the chosen
value should not be greater than a factor of 2 if that for the
naphthalenonium ion is correct. Combining 5 × 1010 s-1 with
the rate constant for protonation of benzene (2.6 × 10-14 M-1
s-1)5 gives Ka ) 1.9 × 1024 and pKa ) -24.3. This compares
with pKa ) -23 estimated by Kresge from extrapolation of a
correlation of pKas and rate constants for aromatic hydrogen
isotope exchange29 and pKa ) -25 extrapolated from a similar
correlation by Marziano et al.30 The latter value is corroborated
by further correlations between the aqueous pKas and pKas in
HF or proton affinities of aromatic molecules in the gas phase.30
Pirinccioglu and Thibblin also deduced a rate constant for
deprotonation of the 2-naphthalenonium ion as 1.6 × 1010 s-1
.
If this is combined with kA ) 6.9 × 10-11 M-1 s-1 for the
protonation of naphthalene,5 a value of pKa ) -20.4 is obtained.
From eq 3 this implies that pK
) -13.7, which is in
H2O
satisfactory agreement with -14.2 from the preceding paper.
The same pKa may also be combined with pK ) -16.9 from
H2O
the previous paper to obtain pKR ) -3.5 for formation of the
1,4-hydrate.
Combining the pKa ) -24.3 with values of pKH O ) 22.2 from
For benzene hydrate, derivations of pKa and pKR are more
speculative. The data available are rate constants for carbocation
formation from the 1,2-hydrate and for hydrogen isotope
exchange of benzene extrapolated from strongly acidic media
by Cox.5 In addition, the equilibrium constant for hydration has
been estimated in the preceding paper. To complete a thermo-
dynamic cycle, one more rate constant is needed: for reaction
of the benzenonium with water as either base or nucleophile.
2
the preceding paper allows pKR ) -2.1 to be inferred for
reaction of the benzenonium ion to form both the 1,2- and 1,4-
benzene hydrates.
Values of pKa, pKR, and pKH O for the molecules considered
2
so far are summarized in Table 1. Values of pKH O have been
2
discussed in the preceding paper and show the expected
dependence on the resonance stabilization of the parent aromatic
molecule. For benzene, pKR ) -2.1 may be compared with
the much more negative values of -16.4 for the tert-butyl,20
-16.2 for the R-methyl benzyl,13,31 and -9.6 for the benzo-
furanyl carbocations. From these comparisons it is clear that
the highly unfavorable pKas for protonation of benzene or
naphthalene reflect the thermodynamic stability of the aromatic
molecule rather than intrinsic instability of their conjugate acids
(see also below).
It seems certain that we can make a good estimate of the
rate constant for deprotonation of the benzenonium ion. The
value from Pirinccioglu and Thibblin’s study of the correspond-
ing naphthalenonium ion cation (1.6 × 1010 s-1) is close to that
expected for rate limiting relaxation of the aqueous solvent,
which provides a limiting value for deprotonation and which it
seems likely the benzenonium ion will approach. For this limit,
Richard et al. have suggested 1011 s-1 for nucleophilic attack
by water, which is close to the rate constant for proton transfer
between H3O+ and a water molecule.20,21 Warkentin has recently
measured a rate constant 2 × 1010 s-1 for reaction of an
isopropyl cation to form 2-propanol in aqueous acetonitrile
containing 10% (by volume) of water,28 which implies a rate
constant near this limit, especially if, as suggested, the value is
partly influenced by the rate of loss of nitrogen from the
diazonium ion precursor from which it is generated.
Vinyl Cation. Lucchini and Modena have measured rate
constants for acid-catalyzed hydrolysis of acetylene in aqueous
sulfuric acid using NMR to monitor the reaction.32 On the basis
of measurements of an isotope effect k
/k
) 2.2-2.5
H2SO4 D2SO4
and an analysis of solvent and substituent effects, they concluded
(29) Kresge, A. J.; Mylonakis, S. G.; Sato, Y.; Vitullo, V. P. J. Am. Chem. Soc.
1992, 114, 8032.
(30) Marziano, N. C.; Tortato, C.; Bertani, R. J. Chem. Soc., Perkin Trans. 2
1992, 955.
(31) A value of pKR in H2O is estimated as one log unit less than the
measurement in 50:50 (v/v) water-trifluorethanol in ref 13.
(32) Lucchini, V.; Modena, G. J. Am. Chem. Soc. 1990, 112, 6291.
(28) Pezacki, J. P.; Shukla, D.; Lusztyk, J.; Warkentin, J. J. Am. Chem. Soc.
1999, 121, 6589.
9
8580 J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002

Aromatic Hydrates Yielding pK and pKR
A R T I C L E S
a
Scheme 7
Table 2. Rate and Equilibrium Constants for Protonation and
Hydration of Carbon-Carbon Double and Triple Bondsa
that the reaction occurs with rate-determining protonation of
the acetylene to form a vinyl cation intermediate. Coincidentally
the rate constant extrapolated to water is similar to that for the
protonation of benzene. This suggests that a pKa for the vinyl
cation may be derived by assigning a rate constant near the
limit for solvent relaxation to the reverse reaction. If this rate
constant is taken to be the same as that for the benzenonium
ion (5 × 1010 s-1), then a similar value of pKa ) -24.8 is
obtained. Despite its large magnitude this rate constant is not
inconsistent with the measurement of a hydrogen isotope effect
and proton transfer being rate-determining. This is partly because
nucleophilic attack of water on the vinyl cation may occur at
the relaxation limit for the solvent, which has the larger value
of ∼1011 s-1, and partly because the effective basicity of the
water molecule will be reduced in the strong acid media of the
isotope measurement and the difference in rates between
deprotonation and nucleophilic attack by water may be en-
hanced.
a The benzenonium ion is included for comparison. b Measurements from
the literature; see text for references. c Measured experimentally only in
the case of the tert-butyl cation.20,41 d Includes a statistical factor of 9:
∼108 per H atom would be appropriate therefore.
For propene and methylacetylene rate constants for hydration
are considerably greater than for ethylene, acetylene, or benzene
and closer in value to that for 2-protonation of naphthalene,
i.e., ∼10-9 M-1 s-1 32
.
Approximate pKas were evaluated
therefore by combining these rate constants with reverse rate
constants based on short extrapolations from the value for
naphthalene using a correlation22 of log kp and pKa for other
carbocations21 and protonated aromatic molecules.35 The derived
pKa values, estimated values for pKH O and resultant values of
2
pKR are collected in Table 2 together with values for the tert-
butyl cation based on Richard’s assignment of pKR ) -16.4 20
and for the benzenonium, vinyl, and ethyl cations discussed
above. Although the correlation of log kp with pKA is subject to
significant deviations between subgroups of carbocation struc-
tures, it offers some support for the minor corrections of limiting
rate constants on which the pKas for the isopropyl and
methylvinyl cations are based.
The pKa for the vinyl cation may be converted into pKR by
using a thermodynamic cycle similar to that of Schemes 1 and
5 and combining pKa with pKH O, the equilibrium constant for
2
hydration of acetylene. As shown in Scheme 7, the hydrate of
acetylene is the enol of acetaldehyde. A value of pKH O ) -15.0
2
was estimated for this reaction in the previous paper. We thus
obtain pKR ) -39.8 for the vinyl cation.
Remarkably, although the vinyl cation and benzenonium ion
have practically the same acidities, their KR values differ by 37
orders of magnitude. Thus, the similar acidities do not imply
similar thermodynamic stabilities but that a large difference in
stabilities of the cations is offset by a corresponding difference
in the acetylenic and aromatic π-bonds subject to protonation.32
Recently Okuyama has argued that a vinyl cation is not
involved as a discrete intermediate in nucleophilic substitution
at a primary vinyl carbon atom, even when this bears a “super-
leaving group” such as the phenyliodonium ion.33 This is
consistent with the extreme energetic inaccessibility of the vinyl
cation from a functionalized vinyl alcohol, as implied by pKR
) -39.8. On the other hand, the instability of the double bond
of acetylene favors access to the carbocation from this precursor.
Alkyl Cations. In addition to acetylene Lucchini and Modena
studied the acid-catalyzed hydration of ethylene, butene, and
methylacetylene.32 For ethylene Tidwell et al. have summarized
arguments that this reaction proceeds with rate-determining
formation of an ethyl cation.34 The rate constant is again nearly
the same as for benzene and acetylene, and a similar value of
pKa ) -24.8 may be derived by the method already described.
Inspection of Table 2 shows a parallel and chemically
reasonable influence of methyl substituents upon the stability
of vinyl and ethyl cations. This offers further support for the
inference of Lucchini and Modena that like their homologues
acetylene and ethylene do indeed undergo hydration with the
formation of discrete carbocation intermediates. Lucchini and
Modena presented a similar argument based on a correlation of
rate constants for hydration of double and triple carbon-carbon
bonds. The results in Table 2 not only convert the correlation
of kinetic measurements to equilibria but allow comparison of
pKR as well as pKa values. The pKR values reflect explicitly
the difference in stability of alkyl and vinyl cations and support
the interpretation provided by Modena et al. of the surprising
similarity in reactivity of alkenes and alkynes,36 i.e., that this is
a consequence of compensating differences in stability between
the carbocations and carbon-carbon π-bonds in the reactants.
In conclusion, two further points deserve comment. The first
is the derivation of rate constants for carbocation formation in
water based on measurements in concentrated solutions of strong
acids. This does not imply that in practice the mechanism or
reaction in water is the same as in the more concentrated acid
solutions, but that the extrapolated rate constant refers to the
Combination with pKH O ) -4.8 for the hydration of ethylene
2
to ethyl alcohol yields pKR ) -29.6, a value intermediate
between that of the vinyl and benzenonium carbocations.
(35) Kresge, A. J.; Mylonakis, S. G.; Sato, Y.; Vitullo, V. P. J. Am. Chem. Soc.
1971, 93, 6181. Kresge, A. J.; Chen, H. J.; Hakka, L. E.; Kouba, J. E. J.
Am. Chem. Soc. 1971, 93, 6174.
(36) Modena, G.; Rivetti, F.; Scorrano, G.; Tonellato, U. J. Am. Chem. Soc.
1977, 99, 3392.
(33) Fujita, M.; Sakanishi, Y.; Okuyama, T. J. Am. Chem. Soc. 2000, 122, 8787.
(34) Nowlan, V. J.; Tidwell, T. T. Acc. Chem. Res. 1977, 10, 252. Chwang, W.
J.; Nowlan, V. J.; Tidwell, T. T. J. Am. Chem. Soc. 1977, 99, 7233.
9
J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002 8581

A R T I C L E S
McCormack et al.
a cuvette containing 2 mL of aqueous or mixed aqueous solvent with
or without added sodium azide. The change in absorbance accompany-
ing or at completion of the reaction was monitored.
process occurring in the acidic media. An alternative mechanism
would have a larger rate constant.
Also noteworthy is that the extrapolated rate constants yield
values of pKa and not pKR. In principle pKR for the ethyl cation,
for example, might have been derived by combining a rate
constant for acid-catalyzed oxygen exchange of 18O-labeled
ethanol with a rate constant for nucleophilic reaction of the
carbocation with water corresponding to the limit for solvent
relaxation (1011 s-1). In practice the exchange reaction would
be expected to proceed by a concerted SN2 mechanism.
By contrast reactions involving protonation of π-bonds
presumably proceed by a “stepwise” mechanism because,
despite the high energy of the reactive intermediate, the
alternative concerted reaction is less favorable. For electrophilic
aromatic substitution indeed it is doubtful if a concerted
mechanism exists at all. Even for addition reactions, such as
the hydration of ethylene, according to Dewar concerted
mechanisms are less favorable than for (SN2) nucleophilic
substitution because they require coupling of bond-making and
bond-breaking over four bonds rather than two.37
3-Chloroacetoxy-2,3-dihydrobenzofuran. For product analyses
from aqueous solvolysis of 3-dichloroacetoxy-2,3-dihydrobenzofuran,
6 µL of a 0.055 M stock solution in acetonitrile were injected into 1
mL of water or solution of sodium azide to give a final concentration
of 3.3 × 10-4 M. Aliquots of 25 µL were injected from the product
mixture to the HPLC column. Separation of components was achieved
by isocratic elution of aqueous solvent mixtures with 50-65% methanol
(v/v) and a flow rate of 1 mL/min. The analytical wavelength was
maintained at 600 nm for 4 min (to avoid overloading of the UV
detector during elution of azide ion) and at 220 nm for the remaining
elution time. The following ratios of peak areas for products 3-azido-
and 3-hydroxy-2,3-dihyrobenzofuran ([RN3]/[ROH]) were observed for
the indicated concentrations of sodium azide for reactions conducted
at 20 ( 0.2 °C and ionic strength µ ) 0.05 (NaClO4): 0.28, 1.0 mM;
0.55, 2.0 mM; 1.5, 5.0 mM; 2.7, 8.0 mM.
For the product analysis by spectrophotometry limiting absorbances
at 243 nm were measured for the reaction solution (a) before solvolysis
(or from complete conversion of the ester to alcohol by the addition of
sodium hydroxide), Ainit; (b) after solvolysis in water, Ao; and (c) after
solvolysis in the presence of sufficient acid to convert the benzofuran
hydrate product to benzofuran, Amax. The absorption at 243 nm arises
mainly from the benzofuran. The ratio of benzofuran to hydrate in the
solvolysis products is given by (Ao - Ainit)/(Amax - Ao). The ratio is
unaffected by partial conversion of the reactant solutions to benzofuran.
Addition of sodium azide to the reactant solution led to a reduction
in the limiting absorbance Ao to Alim. Reactions were carried out for
concentrations of sodium azide in the range 0-0.01 M at 0.1 M ionic
strength. At higher concentrations, the absorbance of the azide ion
swamped that of the benzofuran and the amount of benzofuran formed
in the product was small. Assuming that the drop in absorbance in the
presence of N3- is due to formation of azide trapped product, the ratio
of rate constants for trapping of the benzofuranyl carbocation intermedi-
ate by azide ion (kaz) to those for elimination to benzofuran (kp) and
Experimental Section
The preparation of the 2,3-hydrate of benzofuran and its chloroacetyl
derivative 3-chloroacetoxy-2,3-dihydrobenzofuran have been described.3
The 9,10-hydrate of anthracene38,393 and its acetoxy derivative18 have
also been reported. However, improved preparations are given below.
9-Hydroxy-9,10-dihydroanthracene. Anthraquinone (2 g, 9.6 mmol)
and zinc dust (4 g, activated with 5% hydrochloric acid) were dissolved
in aqueous ammonia (12 mL) and deionized water (8 mL) to give a
red solution which was heated at 60-70 °C for 3.5 h to give a yellow-
grey precipitate. This precipitate was filtered from the hot solution and
washed with two 10 mL portions of hot CH2C12. The filtrate was then
cooled and separated and the aqueous layer extracted with 2 × 10 mL
of CH2C12. The CH2C12 extracts were combined and dried with Na2-
SO4. The solvent was evaporated to give a yellowish solid which was
recrystallized from 40 to 60 petroleum spirits to give white needles
(1.7 g, 87%): 1H NMR (CDC13) δ 2.17 (d, 1H), 2.95 (d, 1H), 4.15 (d,
1H), 5.64 (d, 1H), 7.3-7.5 (m, 6H), 7.7 (br s, 2H).
9-Acetoxy-9,10-dihydroanthracene. 9-Hydroxy-9,10-dihydroan-
thracene (1.6 g, 8 mmol) was dissolved under nitrogen in a minimum
volume of dry pyridine. Acetic anhydride (0.98 g, 9.6 mmol) in dry
pyridine (1 mL) was added dropwise using a syringe. The solution
was left stirring under nitrogen overnight, and 20 mL of CC14 was
added. It was dried with Na2SO4 and evaporated to give a white solid
that was recrystallized from 40 to 60 petroleum spirits to give a white/
yellow solid (0.84 g, 53%) which was stored in a freezer: 1H NMR
(CDC13) δ 2.00 (s, 2H), 3.92 (d, 1H), 4.26 (d, 1H), 6.96 (s, 1H), 7.25-
7.37 (m,6H), 7.53 (m,2H).
nucleophilic trapping by water (kH O) is given by eq 2, in which ∆A )
2
Alim - Ainit and ∆Ao is the limiting absorbance in the absence of azide
ion (Ao) corrected for Ainit. For a substrate concentration 1.65 × 10-4
M and Ainit ) 0.264, the following limiting absorbances were obtained
at the indicated concentrations of NaN3: 0.0 M, 0.738; 0.001 M, 0.66;
0.003 M 0.565; 0.005 M, 0.503; 0.007 M, 0.445; 0.009 M, 0.360. The
value of Amax for complete conversion to benzofuran by H+ in the
absence of azide ion was 1.53. The ratio of slope to intercept of a plot
1/∆A (∆A ) Alim - Ainit) versus [N3-] (disregarding a 10% deviation
of Alim from the correlation at 0.009 M) gave kaz/(kH O + kp) ) 135.
2
The ionic strength was 0.01 M (NaCl).
9-Hydroxy-9,10-dihydroanthracene (kinetics). Kinetic measure-
ments of the acid-catalyzed dehydration of this substrate to anthracene
were based on measurements of the increase in absobance at 251 nm.
The low solubility of the product necessitated use of very dilute
solutions, <10-6 M, leading to absorbance changes of ∼0.2 ac-
companying reaction. The following first-order rate constants (102k,
s-1) were measured in aqueous solution at 25 °C at the indicated
concentrations of HClO4: 0.0008 M, 1.34; 0.00116 M, 2.24; 0.00252
M, 2.68; 0.00363 M, 3.56; 0.00503 M, 5.22; 0.00725 M, 7.43.
Measurements at ionic strengths 0.1 and 0.5 M (NaCl) showed a
2-fold increase in rate constant at the higher ionic strength, which was
confirmed by systematic measurements in this range summarized in
Table S1. The dependence of the rate constant for the acid-catalyzed
reaction upon sodium chloride concentration was given by 1.32(1 +
a[Cl-]) M-1 s-1, where a ) 1.25 M-1 and the concentration of HClO4
was 0.002 43 M. Suprisingly no increase in rate with buffer concentra-
tion was observed in acetic acid buffers if the buffer ratio was kept
constant despite the fact that in the absence of buffer deprotonation of
the protonated anthracene is partially rate determining. Apparently the
NMR spectra of the products were recorded on a JEOL JNMGX270
instrument operating at 270 MHz for proton NMR and 68 MHz for
13C NMR. HPLC analyses were carried out using a Waters 600E system
with a Waters 486 tunable UV detector and reverse-phase 10 µm
octadecylsilane column in a radial compression unit. Peak areas were
integrated using Millenium 2000 software and a 286 PC. Kinetic
measurements made use of a Phillips PU 8600 single beam UV-vis
spectrophotmeter or a Hitachi-124 double beam instrument. Water for
kinetic measurements was doubly distilled, and other reagents and
solvents were normally AR or HPLC grades.
For kinetic and product analyses based on UV-vis measurements
stock solutions of 10-2-10-4 M in (ester) substrate were freshly
prepared in methanol or acetonitrile solvents and 20 µL injected into
(37) Dewar, M. J. S. J. Am. Chem. Soc. 1984, 106, 209.
(38) Von Perger, H. R. J. Prakt. Chem. 1881, 23, 137
(39) Crump, S. L.; Netka, J.; Rickborn, B. J. J. Org. Chem. 1985, 50, 2746
9
8582 J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002

Aromatic Hydrates Yielding pK and pKR
A R T I C L E S
a
Bronsted exponent for the reaction is quite small or water is a more
efficient base than expected from its equilibrium basicity. These results
are shown in Table S4. Absence of catalysis of deprotonation by acetate
is confirmed below.
Equilibrium Constants for Dehydration. The equilibrium constant
for dehydration of the enol of acetone to form methylacetylene given
in Table 2 was based on free energies of formation in aqueous solution
taken from compilations by Guthrie: ∆Gï (aq) ) -27.02 kcal mol-1
f
9-Acetoxy-9,10-dihydroanthracene (kinetics). The solvolysis of this
substrate was studied in acetonitrile-water mixtures as its low solubility
precluded measurements in pure water. Rate constants for solvolysis
decrease with increasing fraction of CH3CN as indicated by the
following first-order rate constants (103k, s-1) measured at 25 °C by
monitoring the increase in absorbance of the anthracene product at 251
nm at the indicated volume percentages of acetonitrile: 142, 20%; 39.2,
30%; 10.7 40%; 3.37, 50%; 1.61, 60%; 0.60, 70%; 0.177, 80%.
Measurements of the ratio of hydrate to anthracene were based on
measurements of an initial absorbance, the absorbances on completion
of the solvolysis reaction, and the absorbance following addition of
acid to convert the hydrate formed to anthracene. The following
absorbances (in the order indicated above) were measured at 25 °C at
the volume percentages of CH3CN indicated: 0.269, 0.653, 0.87, 33%;
0.358, 0.89, 1.144, 50%; 0.269, 0.637, 0.861, 67%; 0.358, 0.767, 1.062,
(CH2dC(OH)CH3)40 and 47.93 kcal mol-1 (CH3CtCH).41 For liquid
water ∆Gof(aq) was taken as -56.69 kcal mol-1
.
In the hydration of isobutene a rate constant for deprotonation of
the tert-butyl cation (kp) was evaluated by combining the equilibrium
constant for hydration,1,42 KH O ) 9.9 × 103, with rate constants for
2
H+-catalyzed 18O exchange of tert-butyl alcohol, hydration (protonation)
of isobutene, and a value of kH O ) 1011 s-1 proposed by Toteva and
2
Richard20 for nucleophilic reaction of the carbocation with water. These
values are summarized in eqs 4 and 5 and yield kH O ) 8 × 108 s-1, a
2
value which is preferred to that derived by Richard et al. (3 × 109
21
s-1
)
based on different experimental data.42
80%. These yield values of kp/kH O ) 0.565, (0.48), 0.61, and 0.59,
2
respectively. The value of 0.48 for 50% acetonitrile was disregarded.
Six further measurements in dilute acetic acid buffers at 25 °C in 33%
acetonitrile gave kp/kH O ) 0.55.
2
Acknowledgment. We thank J. R. Keeffe and R. A. Mc-
Clelland for a number of helpful comments and criticisms.
Measurements of trapping by azide ion were also carried out in
acetonitrile-water mixtures in the same way as for the dihydroben-
zofuranyl acetate. The measured absorbances are reported in Table S2
of the Supporting Information for aqueous acetonitrile solutions with
Supporting Information Available: Tables S1-S4 listing
kinetic data and limiting absorbances for the dehydration of
9-hydroxy-9,10-dihydroanthracene and solvolysis of its acetoxy
derivative. This material is available free of charge via the
Internet at http://pubs.acs.org.
20, 33, 50, and 67% acetonitrile, and plots of 1/∆A (∆A ) Alim - Ainit
)
versus [N3-] for 20 and 50% acetonitrile are shown in Figure 1. Similar
measurements were carried out with sodium acetate and showed that
acetate ion did not appreciably increase the proportion of anthracene
in the product up to a concentration of 0.5 M. These results are reported
in Table S3 and are consistent with the kinetic measurements for
dehydration of the anthracene hydrate in acetic acid buffers mentioned
above. Both of these results indicate no contribution or a very minor
contribution from deprotonation of the anthracenonium ion by acetate
ion.
JA012613X
(40) Guthrie, J. P. In The Chemistry of Enols; Rappoport, Z., Ed.; Wiley: New
York, 1990.
(41) Guthrie, J. P. Can. J. Chem. 1992, 70, 1042.
(42) Gold, V.; Gruen, L. C. J. Chem. Soc. B 1966, 600.
(43) Eberz, W. F.; Lucas, H. J. J. Am. Chem. Soc. 1934, 56, 1230.
9
J. AM. CHEM. SOC. VOL. 124, NO. 29, 2002 8583